Institutions, internationalization and FDI:

the case of economies in transition

Mike Pournarakis and Nikos C. Varsakelis *

The attempts in the literature to explain the uneven allocation

of foreign direct investment in the economies in transition for
the most part stress the role of the market as the most significant
factor in the attraction of such investment. This article attempts
to verify empirically the argument that institutional factors such
as civil rights and internationalization of the national economy
are critical in explaining the behaviour of foreign direct
investment inflows in the economies in transition. It uses a
panel data set for the economies in transition, which are to
become member states of the European Union. The findings
show that market size and degree of internationalization of the
host economy explain a significant part of the cross-country
variation of foreign direct investment. However, institutional
factors related to investment decisions strengthen these location
advantages and help a country become an attractive location
for such investment.

Key words: foreign direct investment, transition economies,

institutions, economic integration.

Introduction

The interlinkages of trade and foreign direct investment

(FDI) influence the economic growth and welfare of countries
in a global environment, which undergoes continuous change.

* The authors are, respectively, Professor of Economics, Department

of Economics, Economic University of Athens, Athens, Greece and Associate

Professor of Industrial Policy, Department of Economics, Aristotle University
of Thessaloniki, Thessaloniki, Greece. The authors would like to express
their gratitude to the participants of the EIBA Annual Conference, December
2002, Athens, Greece, Marina Papanastasiou and two anonymous referees
of this journal for their helpful comments. The usual disclaimer applies.
Correspondence: [email protected].
In this sense, FDI inflows are viewed as a measure of the extent
to which a country or a region is integrating into the world
economy. Therefore, policies to attract FDI are included in the
governmental agenda of many countries. Despite these policies,
FDI growth is unevenly distributed among the economic regions
of the world. Recent statistics (UNCTAD, 2001) show that 80%
of total world FDI inflows are accounted for by the Triad
(European Union-Japan-United States) which also hosts 90%
of the worlds largest (in terms of foreign assets) transnational
corporations (TNCs). In fact, the top 30 host countries account
for 95% of the total world FDI inflows and 90% of the total
stock of FDI.

Economies in transition in Central and Eastern Europe

(CEE) attract a small share of the worlds FDI, a share that is
moreover unevenly distributed in the region. Central Europe
and the Baltic States have received more FDI per capita than
South-Eastern Europe and the Commonwealth Independent
States (Sengenberger, 2002).

The recent literature in the field has tried to explain this

uneven allocation of FDI in the economies in transition. Most
of the studies, whether descriptive (for example, Glaiser and
Atanasova, 1998; Tuselmann, 1999; Pournarakis, 2001;
Sengenberger, 2002; Barry, 2002), or empirical (Tondel, 2001;
te Velde, 2001), stress the market as being the most significant
factor for the attraction of FDI in the economies in transition,
while institutions are hardly included in their analysis.

Economies in transition have only recently received the

attention of researchers regarding the impact of institutional
change on their economic performance. A. Brunetti, G. Kisunko
and B. Weder (1997), using the findings of a survey, have
verified that institutions are correlated with economic growth
and foreign direct investment. To explain the cross-country
variation in economic performance, C. Zinnes, Y. Eilat and J.
Sachs (2001) have emphasized the deep privatization, while L.
Grogan and L. Moers (2001) have emphasized the role of quality
institutions.

78 Transnational Corporations, Vol. 13, No. 2 (August 2004)

 This article aims to verify empirically whether the deep
determinants (geography, integration and institutions) proposed
by D. Rodrik and A. Subramanian (2003) explain the behaviour
of FDI inflows in a sample of economies in transition, using a
set of institutional variables that differ from the ones found in
other studies. The article uses a panel data set for eleven
countries of the region for the period 1997-2001. The first group
of countries includes the Czech Republic, Estonia, Hungary,
Latvia, Lithuania, Poland, Slovakia and Slovenia, which have
already been accepted to join the European Union (EU) by the
year 2004. The second group of countries includes Bulgaria,
Croatia and Romania, which may join the EU in the second
enlargement phase of the Union. Finally, Albania is also included
in the sample. The period 1997-2001 has been chosen as the
reference period because, during the first half of the 1990s, most
of the countries in our sample were subject to the turbulence of
transition from the old planning regime to a market economy
(war in Croatia, the political split of Czechoslovakia and political
instability in Albania, Bulgaria and Romania). All the countries
of our sample now enjoy vigorous democratic institutions.
However, there are still substantial differences in their relative
success in building quality institutions, especially those related
to law enforcement.

This article is organized as follows: in the second section,

the theoretical framework is presented; the third section presents
the data used in our empirical analysis; the fourth section
presents the empirical results and related discussion; and, finally,
the fifth section consists of concluding observations.

The theoretical framework

The flow of FDI is viewed as an integrating factor in the

world economy. The economies in transition, having lagged
behind in the race for FDI attraction, are called upon to agree
on a more efficient investment regime. Among other things, this
means the recognition and exploitation of their location
advantages, while the progressive removal of their disadvantages
should be incorporated more fully into their development

Transnational Corporations, Vol. 13, No. 2 (August 2004) 79

strategies. According to John H. Dunning (1993), the location
advantages derive from the supply side (labour skills and costs,
corporate taxation), the demand side (market size and growth)
and the political and social infrastructure. To date the literature
on FDI in economies in transition has focused on both the supply
and demand side location advantages, but market size stands
out as being the most significant factor in the attraction of FDI.
There is a widespread argument that most CEE bound FDI has
been market seeking (Sengenberger, 2002; Tuselmann, 1999;
Tondel, 2001; Te Velde, 2001). In this respect the Balkan
countries are in a disadvantageous position for various reasons.
The region suffers from the syndrome of fragmentation rather
than unification. Long-standing rivalry and instability have
created distances among the countries of the area and thus make
crossborder trade and FDI activity more difficult.

Of course, good market performance does not exclusively

depend on market liberalization and privatization. The enabling
market setting requires, among others, an appropriate kind and
degree of regulation, effective law enforcement and qualitative
public services (Sengenberger, 2002). Thus, excessive
bureaucracies, delays in privatization, unclear and arbitrarily
enforced rules, monopoly control of the real sector (Glaister
and Atanasova, 1998), the lack of tripartite social dialogue, and
unsatisfactory industrial relations (Sengenberger, 2002) could
constitute strong investment barriers even for market-seeking
TNCs (Ekholm and Markusen, 2002). Brunetti, Kisunko and
Weder (1997) found that differences in the degree of
predictability of the institutional framework might explain to a
significant degree the differences in FDI across economies in
transition. Zinnes, Eilat and Sachs (2001) present empirical
evidence that the change of ownership resulting from
privatization programmes was not an adequate explanation of
the variation of economic performance. The firms objective
function (profit maximization), the severity of the budget
constraints and the legal and institutional framework that enables
a firms agents to monitor and control enterprise managers, are
also critical in explaining the variation in economic performance,
as measured by economic growth and FDI inflows. Finally,

80 Transnational Corporations, Vol. 13, No. 2 (August 2004)

Grogan and Moers (2001), using different institutional measures,
verify the previous findings that quality institutions are important
for economic growth and FDI in economies in transition, for
the period 1990-1998.

This article seeks to explain the cross-country variation

of FDI inflows for the period 1997-2001 in those countries that
will become full members of the EU in 2004. Rodrik and
Subramanian (2002) proposed three fundamental factors
influencing the economic performance of a country: geography,
economic integration in the global economy and institutions.
This article will test for the last two, integration and institutions,
controlling for geography.

Integration in the global economy is proxied by two

alternative variables, exports and international trade as a
percentage of GDP. Countries that have become more integrated
in the global economy the level of their annual exports is higher
are expected to attract more FDI than countries with a lower
degree of integration. Thus, the following hypothesis is posited:

H1: The more integrated a country is in the global

economy the higher are FDI inflows.

Quality institutions, and the rules of the game in a country

are defined in terms of the degree of property rights protection,
the degree to which laws and regulations are fairly applied and
the extent of corruption (IMF, 2003). Civil liberties, the freedom
of expression, freedom of association and organization rights,
and the rule of law and human rights, are all important influences
on business decision-making. For example, the lack of free trade
unions and collective bargaining is an important factor since
feelings of job insecurity inevitably run high and FDI
performance worsens in the host country. Empirical research,
as noted by Sengenberger (2002), has shown that this feeling is
present in the CEE countries. The independence of the judiciary
system, the prevalence of the rule of law in civil and criminal
matters, the treatment of population under law with equality,
and other related issues such as corruption and the mafia, may

Transnational Corporations, Vol. 13, No. 2 (August 2004) 81

also significantly influence the decision of a TNC to enter a
new market. As K.W. Glaister and M. Atanasova (1998) point
out, even though Bulgaria has adopted a very liberal legal
framework for FDI, the endemic nature of organized crime in
business and the strength of the official bureaucracy have
resulted in a lagging behind of Bulgaria when compared to
other CEE countries in terms of FDI. Thus, the following
hypothesis is posited:

H2: The higher the institutional quality that a

country exhibits the higher FDI inflows.

The data

In the following section, the two hypotheses stated in the

previous section are tested, using data from the aforementioned
sample of economies in transition for the period 1997-2001.
Recent studies (Mauro, 1995; Brunetti, Kisunko and Weder,
1997; Zinnes, Eilat and Sachs, 2001; Grogan and Moers, 2001)
used information about the perceived quality of institutions
trying to explain the cross-country variation of economic
performance. Subjective institutional measures, as Grogan and
Moers (2001) point out, could lead to more interesting
conclusions about the mechanism at work and the policies
needed. These subjective measures are constructed by
commercial, international country-risk agencies using surveys
of the opinions of economic agents who make investment
relevant decisions, and for that reason are more relevant to FDI.

The first source of our data for institutional measures is

Freedom House. Three indices are used to measure the quality
of political institutions (Freedom House, 2002b). These indices
are constructed using survey methodology. The first is an index
of political rights. The index ranges from one to seven. In
countries that receive a rating of one, elections are free and fair;
those who are elected rule the country, the opposition plays a
significant role in the political system and citizens enjoy self
determination. The countries that receive a rating of two are

82 Transnational Corporations, Vol. 13, No. 2 (August 2004)

less free, and factors such as political corruption, political
discrimination against minorities, or foreign or military influence
may be present. In countries that receive a rating of three, four
or five, the presence of military involvement, unfair elections,
one party dominance and civil war are considered harmful to
civilians freedom.

The second index involves civil liberties. The index ranges

from one to seven. Countries receiving a rating of one are
distinguished by an equitable system of rule of law, are free of
corruption and enjoy free economic activity. Countries that
receive a rating of two exhibit some deficiencies in civil liberties
but could still be characterized as free. Finally, countries
receiving ratings of three, four or five present significant
deficiencies in terms of free association and limitations in
business activity imposed either by governmental institutions
or non-governmental agents (that is, terrorists, mafia).

The third index is related to freedom of the press. The

data reported in Freedom House (2000a) are used. They measure
the degree to which each country of our sample permits the free
flow of information. The free press is the sum of ratings for the
news delivery system as functioning under countrys laws and
administrative decisions; the degree of political influence over
the content of news media; and the economic influence on media
content (that is, government funding, corruption). The free press
index ranges from zero to one hundred, zero indicating a
completely free press and one hundred a completely non-free
press.

The second source of data for the institutional measure is

derived from Transparency International. The Transparency
Internationals Corruption Perception Index is used as a measure
of corruption. This is a composite index based on international
surveys of the perception that business people and country
experts have regarding corruption in over fifty countries. The
results of individual surveys are standardized, that is, they are
expressed in standard deviations from the mean. The index is

Transnational Corporations, Vol. 13, No. 2 (August 2004) 83

the simple average of these standardized values and it is a
continuous scale from 0 representing an absolutely corrupted
state to 10 representing a completely clean one. The sample
consists of data for the period 1998-2001, but the data are not
available for the full range in all countries of the sample.

The degree of internationalization of an economy is

captured by two measures, the share of exports in gross domestic
product and the ratio of international trade to gross domestic
product. The data are drawn from the World Bank Economic
Indicators for the years 1997-2001.

The endogenous variable is FDI inflows per capita. The

data for FDI inflows are published by UNCTAD for the period
1997-2001. The control variables that a priori may be expected
to matter for the behaviour of FDI in economies in transition
are drawn from the literature gathered in the field. A proxy for
market size is per capita gross national income. Data are used
(Atlas method in current dollars) as reported in World Bank
Economic Indicators for the years 1997-2001. The annual
inflation rate for a country as reported in World Bank Economic
Indicators for the period 1997-2001 is used as a proxy for
macroeconomic stability.

The cluster typology proposed and developed by Zinnes,

Eilat and Sachs (2001) who assign countries based on
similarities in variables at the start of transition is used to
control for the initial conditions. This clustering exercise resulted
in seven clusters of economies in transition, and four of them
are used: Cluster 1: Albania; Cluster 2 (Baltic States): Estonia,
Latvia, Lithuania; Cluster 3 (the Balkans): Bulgaria and
Romania; and Cluster 4 (EU Border States): Croatia, Czech
Republic, Hungary, Poland, Slovakia, Slovenia.

Empirical results and discussion

Based on the discussion of the previous section, equation

(1) is set, which is the model to be estimated:

84 Transnational Corporations, Vol. 13, No. 2 (August 2004)

LFDICij = a +b1INSTij + b2 INTij + b3LGNI1ij + b4 INFLij + Cluster Dummies (1)

where LFDICij is the logarithm of per capita FDI inflows for

country i in year j, INTij the internationalization degree for
country i and the year j, and INSTij stands for the institutional
variables used in our paper to capture the impact on FDI. The
CPI ij , the corruption perception index, the LPRESS ij , the
logarithm of the press freedom index, the POLIT ij, the index for
the political rights, and the CIVILij, the index for the civil rights,
are used alternatively. In order to deal with macroeconomic
stability, the annual inflation rate for the country i and the year
j, INFLij is used. Two alternative indices to capture the degree
of internationalization, INTij, are used: first, the percentage of
exports to gross domestic product in logarithms, denoted as
LEXPij in the data set; second, the percentage of international
trade to gross domestic product in logarithms, denoted as LINTRij
in our data set.

In order to face the endogeneity problem between FDI and

per capita gross national income (GNI) the log of the per capita
GNI with one lag, LGNI1ij in equation (1) is used. Considering
that FDI takes one to two years to affect the national income of
a host country, the use of per capita GNI with one lag solves the
endogeneity problem satisfactorily (Griffiths, Hill and Judge,
1993).

The expected signs, according to the hypotheses set in the

previous section and the literature, are: b2, b3>0, b4<0. The signs
of the institutional variable depend on the measurement scale.
Thus, we expect positive sign for CPI and negative for LPRESS,
POLIT and CIVIL.

Table 1 presents the descriptive statistics and table 2 the

correlation coefficients. Table 3 evaluates a series of pair-wise
relationships between LFDIC and the individual determinants.
Individually, only LGNI1 explains almost 45% of the overall
variation of the LFDIC, while the individual contribution of the
other determinants ranges between 11% and 27%. The individual
estimated parameters of the institutional variables and

Transnational Corporations, Vol. 13, No. 2 (August 2004) 85

integration are statistically significant and have the expected
signs. The LFDIC is correlated with lower corruption, more press
freedom and better political and civil rights. Finally, the LFDIC
is correlated with the degree of internationalization. However,
as these are only univariate regressions, the conclusions must
be viewed with caution.

Table 1. Descriptive statistics

Variable Mean Max Min s.d.

COR 3.95 6 2.30 1.035

PRESS 31.76 75 17 14.34
FDIC 143.29 614.22 12.12 128.17
CIVIL 2.45 5 2 .87
POLIT 1.62 4 1 1.01
GNI 3787.13 10070 750 2277.47
INFL 27.83 949 -1 122.75
EXP 48.79 95 9 19.20
INTR 104.73 192 41 35.41

Source: Authors own calculation.

Table 2. Correlation matrix

Variable COR PRESS FDIC CIVIL POLIT GNI INFL EXP

COR
PRESS -.583**
FDIC .309 -.353**
CIVIL -.558** .768** -.282*
POLIT -.673** .915** -.238 .880**
GNI .706** -.302* .379** -.553** -.366**
INFL -.190 .140 -.136 .064 .074 -.193
EXP .541** -.504** .463** -.493** -.542** .354** .037
INTR .507** -.479** .443** -.399** -.492** .323* -.031 .983**

Source: Authors own calculation.

* correlation is significant at the 0.05 level.
** correlation is significant at the 0.01 level.

86 Transnational Corporations, Vol. 13, No. 2 (August 2004)

 Table 3. Single OLS estimations: dependent variable LFDIC

Variable (1) (2) (3) (4) (5) (6) (7) (8)

Constant 2.404 7.340 0.761 -1.204 6.106 4.959 -2.864 4.736

(0.001) (0.000) (0.326) (0.364) (0.000) (0.000) (0.011) (0.000)
COR 1.707
(0.001)
LPRESS -0.772
(0.006)
LEXP 1.039
(0.000)
LINTR 1.286
(0.000)
CIVIL -1.534
(0.001)
POLIT -0.739
(0.001)
LGNI1 0.944
(0.000)
INFL -0.0009
(0.332)
R2-adj 0.25 0.110 0.307 0.249 0.27 0.155 0.447 -0.001
F-stat 12.664** 2.256** 26.74** 20.27** 13.7** 11.78** 48.78** 0.959
Observations 35 59 58 58 59 59 59 59

Source: Authors own calculation.

Note: p-values in parenthesis.
** statistically significant at the 0.01 level

Table 4 presents the estimates of the LFDIC equation.

Eight different modes of equation (1) were estimated.1 As a

1 Heterogeneity between countries has been tested, that is whether

equation (1) is the adequate one by assuming that all parameters are equal
for the ten-cross-country units. If the assumption is correct, there are no
behavioural differences across countries and the data can be treated as one
sample of 58 observations. In order to test for common or different intercepts
in individual countries the least squares dummy variable model (Griffiths,
Hill and Judge, 1993) was applied. The estimated F-statistic suggests that
the null hypothesis, the constant terms of the individual countries are equal,
could not be rejected at p=0.01. Thus the constant term is the same across
countries and it was possible to proceed by considering the data set as one
sample.

Transnational Corporations, Vol. 13, No. 2 (August 2004) 87

control variable, the LGNI1, INFL and the cluster dummies were
used, and tested for the overall significance of the institutional
variables and internationalization. The conclusions that can be
inferred from these regressions are two. First, the LGNI1, LEXP,
and LINTR are statistically significant in all regressions (with
the exception only of LINTR with COR and CIVIL as regressors).
Second, for all specifications institutional variables have a
statistically insignificant effect on FDI inflows, even though
the estimated coefficients have the expected signs (with the
exception of LPRESS with LEXP as regressor). Thus, the results
verify the previous findings in the literature that market size, as
captured by LGDI1, and internationalization, as captured either
by LEXP or LINTR, are very important factors influencing the
decision of TNCs to enter the host countries. The results also
suggest that institutional variables do not contribute substantially

Table 4. OLS estimations: dependent variable LFDIC

Variable (1) (2) (3) (4) (5) (6) (7) (8)

Constant -2.730 -3.408 -3.248 -4.224 -4.677 -3.612 -0.866 -1.713

(0.186) (0.165) (0.071) (0.045) (0.008) (0.014) (0.623) (0.391)
COR 0.067 0.151
(0.927) (0.837)
LPRESS 0.0115 -0.031
(0.963) (0.899)
CIVIL -0.422 -0.624
(0.270) (0.147)
POLIT 0.051 0.112
(0.816) (0.617)
LGNI1 0.673 0.711 0.744 0.804 0.822 0.768 0.476 0.550
(0.041) (0.033) (0.000) (0.000) (0.000) (0.000) (0.046) (0.027)
LEXP 0.522 0.517 0.562 0.593
(0.138) (0.032) (0.021) (0.065)
LINTR 0.489 0.565 0.605 0.573
(0.244) (0.068) (0.05) (0.196)
INFL 0.0006 0.0009 -0.0005 0.0001 0.0001 -0.0001 -0.0004 -0.0001
(0.886) (0.849) (0.914) (0.866) (0.886) (0.883) (0.596) (0.822)
Cluster Dummy Yes Yes Yes Yes Yes Yes Yes Yes
R2-adj 0.36 0.34 0.47 0.46 0.46 0.475 0.513 0.485
F-stat 5.901** 5.53** 13.97** 13.326** 13.35** 14.1** 10.21** 9.224**
Observations 35 35 58 58 58 58 58 58

Source: Authors own calculation.

Note: p-values in parentheses.
** statistically significant at the 0.01 level.

88 Transnational Corporations, Vol. 13, No. 2 (August 2004)

to the explanation of the cross-country variation of FDI inflows,
beyond the control variables. The results, however, do not imply
that institutional variables have no impact on FDI and that only
market or internationalization matters. It would be more
appropriate to test whether FDI decisions require simultaneous
improvements in markets, internationalization and institutions.
These simultaneous improvements were tested by adding
interaction terms to the model. The interaction terms are the
products of the LGNI1, LEXP and LINT with the variable CIVIL.
The corresponding results for LPRESS, COR and POLIT are not
statistically significant and therefore they are not reported here.

Table 5 presents the estimation results of an alternative

specification of the model. The synergetic effects are given by
the LGNI1*CIVIL, LEXP*CIVIL and LINTR*CIVIL interactive
terms. The strong conclusion of these regressions is the role of
CIVIL in support of LGNI1, LEXP and LINTR, on FDI inflows.2
The interpretation of these results is that, the better the civil rights
level of a country, the more positive is the impact of an increase in
per capita income on FDI. Thus, countries in our sample that have
promoted economic growth would have attracted more FDI if this
had been followed by reforms to improve civil rights.

These results, using different measures for institutional

quality, verify previous findings in the literature. Quality
institutions render a country attractive for TNCs beyond its
market size, its productive endowments and internationalization.
The countries in the sample have made significant progress, in
terms of political stability, privatization, macroeconomic
stabilization and the adoption of laws protecting property rights.
However, this progress should be accompanied by improvements
in civil rights. Countries that are distinguished by a more
equitable system of rule of law, lower corruption and more
freedom in economic activity achieved much better performance
than countries that are characterized by significant deficiencies.
Countries that suffer from limitations in economic activity either
2 To check the robustness of the result we repeated the regressions

for various specifications and methods (random effects, inclusion of

quadratic terms, i.e. CIV squared, cluster dummies and country dummies).

Transnational Corporations, Vol. 13, No. 2 (August 2004) 89

by governmental institutions or non-governmental agencies (that
is, the mafia, armed groups) exhibit the worst performance in
attracting FDI.

Table 5. Synergetic effects of the interaction between LGNI1,

LEXP, LINTR and CIVIL: dependent variable LFDIC

Variable (1) (2) (3) (4)

Constant -13.701 -17.824 -9.448 -13.424

(0.003) (0.003) (0.064) (0.002)
CIVIL 0.500 0.431 0.0511 -0.818
(0.038) (0.07) (0.887) (0.145)
LGNI1 2.088 2.074 2.127 2.521
(0.000) (0.000) (0.000) (0.000)
LEXP 0.468 -0.812
(0.037) (0.322)
LINTR 0.461 -0.480
(0.101) (0.305)
INFL 0.0003 0.0004 -0.00003 0.00001
(0.722) (0.563) (0.965) (0.983)
LGNI1*CIVIL -0.0001 -0.0001 -0.0001 -0.0205
(0.01) (0.015) (0.008) (0.002)
LEXP*CIVIL 0.0127
(0.108)
LINTR*CIVIL 1.005
(0.017)
Cluster Dummy Yes Yes Yes Yes
R2-adj. 52.7 51.2 54.2 55.5
F-Statistic 13.942** 13.170** 12.429 ** 13.054**
Nr. Obs 58 58 58 58

Source: Authors own calculation.

Note: p- values in parentheses.
** statistically significant at 1%.

Even though the data span is rather narrow (five years for
each country), however, they are the most stable for our sample
countries. Furthermore, taking into consideration previous
findings in the literature, the findings of this article are an initial
hint for policy orientation in these countries in order to catch
up with their Western neighbours. Since good institutions
guarantee property rights and minimize transaction costs they
create an environment conducive to investment. Thus, the

90 Transnational Corporations, Vol. 13, No. 2 (August 2004)

application of policies that aim at the stabilization of the political
and social environment and the implementation of an efficient
judiciary and bureaucratic system will help these countries to
increase FDI inflows. As FDI is a major integrative factor, the
new members of the EU will integrate faster into the European
economy. As far as the countries of the next wave of enlargement
are concerned, the EU should focus not only on policies that
will help these countries to improve their physical infrastructure,
internationalization, privatization and macroeconomic stability,
but also on policies to help them substantially to improve the
quality of their institutions, and on creating an environment in
which property rights and entrepreneurship are well protected.
In this way these countries could be rendered more attractive to
foreign firms and they would be enabled to improve their
economic performance and to converge faster towards meeting
the European standards.

Conclusions

This article, using a data set for a sample of economies in

transition for the period 1997-2001, has verified empirically that
market size and the internationalization of the host economy
explain a significant part of the cross-country variation of FDI
inflows. This is the strong point of the region, and in this respect
it can be taken as advantageous. However, civil rights, which
are related to investment decisions, strengthen these location
advantages and also help a country to become a more attractive
location for FDI.

The above discussion suggests that a great deal of

groundwork, in terms of policy changes, is needed for the new
and future members of the EU to increase substantially their
share of FDI activity. If there is a role for the host country
government, it surely lies in the creation of the necessary
preconditions for FDI inflows. A significant component of this
changed economic policy must be concerned with the
development of political and civil institutions together with an
efficient bureaucratic system. Emphasis should be given not only
to the attraction of TNCs, e.g. by providing tax incentives, but

Transnational Corporations, Vol. 13, No. 2 (August 2004) 91

also after entrance-care programmes for foreign affiliates and
this certainly would include the adoption of efficient institutions
(Williams, 1997). In view of their expected EU membership,
some of these countries could aim at creating a transparent legal
environment (Barry, 2002) by adopting the EU institutional
framework. The EU could also help its new members in this
respect, especially those with lower quality institutions. On the
other hand, the countries of the next wave should now start the
process of development of quality institutions and the EU should
support this process.

References

Barry, F. (2002). EU accession and prospective FDI flows to CEE countries:

A view from Ireland, (Dublin: University College Dublin), working
paper, (August), mimeo.

Brunetti, A., G. Kisunko, and B. Weder (1997). Institutions in transition:

Reliability of rules and economic performance in former socialist
countries, (Washington, D.C.: World Bank), Working Paper, No. 1809,
mimeo.

Dunning, J. (1993). Multinational Enterprises and the Global Economy

(Berkshire: Addison Wesley).

Ekholm, K. and J.R. Markusen (2002). Foreign direct investment and EU-
CEE integration, (Copenhagen: Centre for Economic and Business
Research), CEBR, Working Paper, (May), mimeo.

Freedom House (2002a). Press Freedom Worldwide, www.freedomhouse.org/.

Freedom House (2002b). Annual Survey of Freedom Country Scores,

www.freedomhouse.org/.

Glaister, K.W. and H. Atanasova (1998). Foreign direct investment in

Bulgaria: patterns and prospects, European Business Review, 98 (2),
pp. 122-134.

Griffiths, W.E., R.C. Hill, and G.G. Judge (1993). Learning and Practicing
Econometrics (New York: John Wiley and Sons).

92 Transnational Corporations, Vol. 13, No. 2 (August 2004)

Grogan, L. and L. Moers (2001). Growth empirics with institutional
measures for transition economies, Economic Systems, 25, pp. 323-
344.

International Monetary Fund (2003). World Economic Outlook (Washington

D.C.: IMF).

Mauro, P. (1995). Corruption and growth, Quarterly Journal of Economics,

110 (3), pp. 681-712.

Pournarakis, M. (2001). Development integration and FDIs in the Balkans,

Paper presented at the International Conference on Current Issues for
Transition Economies in the Broad Balkan Region, Xanthi (December),
mimeo.

Rodrik, D. and A. Subramanian (2003). The primacy of institutions,.

Finance and Development, 40 (2), pp. 31-35.

Sengenberger, W. (2002). Foreign direct investment, industrial restructuring

and local labour markets in the transition economies of Central and
Eastern Europe: issues for debate, (Geneva: United Nations Economic
Commission for Europe), (May), mimeo.

Tondel, L. (2001). Foreign direct investment during transition: determinants

and patterns in Central and Eastern Europe and the former Soviet Union,
(City: Chr. Michelsen Institute), working paper No. 2001:9, mimeo.

Transparency International (2002). Transparency International Corruption

Perception Index, www.transparency.org/.

Tuselmann, H.J. (1999). German direct foreign investment in Central and

Eastern Europe: relocation of German industry? European Business
Review, 99 (6), pp. 359-367.

United Nations Conference on Trade and Development (UNCTAD) (1999)

World Investment Report 1999, Transnational Corporations, 8 (3),
pp. 55-100.

________ (UNCTAD) (2001) World Investment Report 2001,

Transnational Corporations, 11 (1), pp. 77-111.

Zinnes, C., Y. Eilat and J. Sachs (2001). The gains from privatization in
transition economies: is change of ownership enough? International
Monetary Fund Staff Papers, 48 (Special Issue), pp. 146-170.

Transnational Corporations, Vol. 13, No. 2 (August 2004) 93

Te Velde, Willem D. (2001). Policies towards foreign direct investment in
developing countries: emerging best-practices and outstanding issues,
(London: ODI), working paper (March), mimeo.

Williams, D. (1997). Strategies of multinational enterprises and the

development of the Central and Eastern European economies, European
Business Review, 97 (3), pp. 134-138.

World Bank (2001). World Development Report 2000/2001: Attacking

Poverty (New York: Oxford University Press).

94 Transnational Corporations, Vol. 13, No. 2 (August 2004)