Foreign direct investment and its drivers: an empirical comparative
study on developed and developing countries
NEVILA BACI, GJERGJI MULLA, DENISA MILLO, KLEI XHYHERI
Department of Statistics and Applied Informatics
University of Tirana
Mother Teresa Square, 4
ALBANIA
Abstract: Foreign Direct Investment plays an important role in a country’s economic activity because of its posi-
tive effect on economic growth. In developing or transition countries foreign capital is considered one of the most
important sources of economic growth, helping in creating new jobs and influencing technological innovation.
Many governments design and implement strategies to attract Foreign Direct Investment over time. Therefore,
determining the role of Foreign Direct Investment in different economies has become an important issue for de-
signing policy responses. This paper aims to determine through empirical analysis, the determinants of Foreign
Direct Investment flows in developing and developed countries and make policy recommendations for the pro-
motion of Foreign Direct Investment in these countries. Based on the selected data period collected by the World
Bank repository, we applied pooled regression models and panel data analyses to determine the factors influencing
FDIs. Applying the fixed effect model and the random one, we identified the important factors impacting the FDI
for developing and developed countries. Based on the results obtained by applying the random-effects model,
among effective factors on Foreign Direct Investment inflows, we could mention Gross Domestic Product (GDP)
growth, Official Development Assistance, Trade, inflation, regulatory quality, government effectiveness, political
stability, and population. From all these factors, only inflation tends to decrease the Foreign Direct Investment
inflows in a hosting country, and hence, governments in developing countries must give more attention to these
factors.
Key-Words: Foreign Direct Investment, economic, growth, development, panel data, fixed and random models.
Received: September 11, 2021. Revised: June 17, 2022. Accepted: June 26, 2022. Published: July 20, 2022.
1 Introduction
Foreign Direct Investments (FDI) have been starting
earlier at the beginning of the 19-th century, but after
World War II-nd, they exploded. There are different
reasons behind this increase such as changes in the
market toward international trade. Later, the devel-
opment in technology contributed more to spreading
FDI rapidly around the world, as globalization has
changed the rule of the game. With their potential,
today FDI is considered one of the key factors in eco-
nomic development for developing countries.
Authors in [1] concluded that companies have two
ways to enter a foreign market either by exporting or
by FDI. The first way, exporting is the simplest, it is
risky on the other side since companies have no op-
portunity to control their production or take advantage
of opportunities that can only be created by a concrete
presence in a foreign market. As the process of ex-
porting deals with the production of goods at home
first and second shipping them to the receiving coun-
try for sale, it is associated with constraints of trans-
portation costs and trade barriers.
In comparison to exporting, FDI has an obvious ad-
vantage as it provides greater opportunities for the
companies to gather information about local demand
and costs. FDI brings valuable opportunities for a
country driving technological enhancement, increas-
ing competition in the global market, and diversifi-
cation of the economy. Governments of developing
countries give priority to new investment strategies
to attract new investment and improve the investment
climate. To absorb and attract new foreign invest-
ments, governments have to promote and pursue poli-
cies and investment strategies, take appropriate ac-
tion toward liberal regulatory frameworks, improve-
ment of market penetration, and institutional capac-
ity buildings, improvements of the business environ-
ment, and improvements in ”ease of doing business”.
Foreign investments play a crucial role in the over-
all economic development of the hosting country, be-
cause of their positive effect on accelerating economic
growth [2]. Foreign capital is considered the most im-
portant source of economic growth in many develop-
ing or transition countries. This source helps govern-
ments continually lead the creation of new jobs and
influence new waves of technological innovation [3].
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
992
Volume 18, 2022
Table 1: List of countries.
Developing Countries Developed Countries
Albania Indonesia Canada United States
Argentina North Macedonia Chile United Kingdom
Bosnia and Herzegoniva Malaysia France
Brasil Mesico Germany
Bulgaria Serbia Italy
Columbia Thailand Spain
Croatia China South Corea
Therefore, determining the role of FDI for different
economies is not an easy task for many governments
becoming an important challenge to design appropri-
ate policy responses.
This paper aims to explore through empirical evi-
dence, the determinants of FDI flows in hosting coun-
tries (developed and developing) and make policy
recommendations for the promotion of FDI. The se-
lected countries in our research are listed in Table 1.
The countries’ classification in developing and devel-
oped countries is based on the World Bank classifica-
tion.
We will focus our study on identifying the main fac-
tors that attract FDI in two main groups of countries,
developed countries and developing ones.
Most of the research conducted in this area is fo-
cused on economic factors such as macroeconomic
indicators, inflation indexes, economy size, and trade
[1, 4]. Based on the research found in this field, there
is yet no consensus on the relationship between FDI
and its determinants as they are depending on country
specifics. There is not a consensus on the list of fac-
tors (political, economic, and institutional) that can be
considered significant determinants of FDI [5]. De-
spite many studies in this field, only a few studies can
be found that are focused on additional factors other
than economic ones.
The main contribution of this study will be the con-
tribution of analyzing not only the economic factors
affecting FDI inflows in developing countries but also
institutional and political factors.
The remaining part of the paper is organized as fol-
lows. Section two presents the literature review and
the background to our study regarding the FDI in-
flows and their determinants. The research methods,
dataset, and variables as well as the empirical frame-
work are presented in section three. In section four
authors present the empirical. Finally, the conclusions
and future work are given in section five.
2 Background
In this section of the paper, we provide an overview
of FDI, including different definitions of what FDI is
based on different sources. We will also provide its
classification into several types and classifications ac-
cording to the purpose of investing in a foreign coun-
try.
2.1 International capital flow
Protsenko [6] classified the cross-border capital flow
into three main categories: portfolio, FDI, and others
such as bank loans. Portfolio investment is a collec-
tion of groups of assets such as stocks, bonds, and eq-
uities becoming a source of income for investors. In
this type of investment as shown in Figure 1, investors
should mitigate the interest-rate risk. While FDI and
portfolio investment are both ways of international
capital flow, and both involve investing in a foreign
country, the two approaches differ significantly. FDI
should not be seen as an initiative to add to the stock
of capital but as an international mechanism for mo-
bilizing assets and expanding to the global market.
Figure 1: Capital flow Source [16].
2.2 Definitions of FDI
While the definition of FDI has shifted over time, it
essentially holds the same concept. According to the
classic definition, FDI will be considered the invest-
ments done by a company of a given country in estab-
lishing an enterprise of the same type but in another
country (host country) controlling ownership in it [7].
According to the definition provided by OECD [8],
FDI scope is to ensure a stable interest of an enterprise
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
993
Volume 18, 2022
(direct investor), in a resident enterprise. The agree-
ment must be focused on the long-run sustainable in-
terest of the country. Both the investor and the direct
investment enterprise should be in closer collabora-
tion, to provide a positive effect in decision-making.
A definition of FDI is provided in the statistical man-
ual of the Balance of Payments [9], where the FDI is
defined as an investment in another economy where
the investors held 10 percent or more of the shares.
There are three main categories of FDI as they are de-
fined by Dunning [10] as follows:
(a) market seeking- This type of investment tends to
exploit new markets or new opportunities to the
host countries with the main focus on reducing
the export costs.
(b) resource seeking- The investor gain privilege ac-
cess to resources at a lower price such as cheap
labour, natural resources, physical infrastructure,
cheap raw materials etc.
(c) efficiency seeking- This is the case when in-
vestors tend to gain benefits by increasing their
efficiency while they reduce the production cost,
avoid trade barriers etc.
The five major categories of FDI defined by authors
[11] are as follows:
Investment done to benefit from cheaper fac-
tors of production This kind of investments are
more prone to the host countries where govern-
ments promote financial assistance for securing
foreign markets. In cases where the governments
discourage imports, then investors may penetrate
foreign markets with relevant assets.
Investments done to benefit from specific prod-
uct factors These factors might be scarce re-
sources, technological skills, licenses, or patents
owned by the host company. Usually, companies
invest in this type of investment, where the re-
quired assets cannot be found or are not available.
Investments done to access consumers in the
country of origin- Investors, when decided to go
on for this type of investment, they intend to pro-
duce the same goods and to offer the same ser-
vices to the clients following standard specifica-
tion requirements.
Joint-venture FDI - This type of investment is re-
lated to those firms which compete internation-
ally and is considered as the most common type
of FDI since they intensify the global competition
between products, markets or infrastructures.
FDI aimed at regional diversification and integra-
tion - Diversification and integration are found to
be beneficial to the countries of origin by ensur-
ing expansion of the capacity of the existing mar-
kets, and growth of the business values as well.
Thus, FDI is a movement of capital that allows either
the creation of an overseas production structure, or
the acquisition of what already exists abroad, or the
merger of so-called joint ventures. Foreign investors
can choose to invest in different ways, depending on
the host country and the goal they want to achieve.
2.3 Factors impacting FDI
Research has been working on identifying the deter-
minants and their impacts on FDI attracting great at-
tention, but still, there is not a consensus about them
among them [12]. They all conclude that FDI over-
all affects the economic growth of hosting countries
positively but most of these studies are based on the
investigation of data coming from a specific country.
The author in [13] was focused on investigating the
patterns of FDIs flow and economic growth in the
country of Korea. Their main findings revealed that
different important economic factors such as eco-
nomic growth, human capital, and export generate e
strong positive impact on FDI. Also, in similar stud-
ies, the same conclusions were drawn [14].
Furthermore, evidence can be found that FDI does not
impact positively the economic growth. Such stud-
ies are based on countries like Poland, Romania, and
Bulgaria [15]. Also, some authors argue that there is
a strong correlation between foreign investments and
economic growth rates. Not only does FDI promote
economic growth but also vice versa is true [16]. Less
research is done on the role of the labor cost in deter-
mining FDI. The main reasons behind this, it is the
business environment and social-political stability of
these countries bringing barriers to the investors due
to trade imbalances, monopolies, or legal barriers.
Many researchers tried to investigate the motivation
behind the differences in the quantity of FDI in dif-
ferent countries. Some of the identified determinants
are political stability, country infrastructure, market
size, investment climate, etc. [17].
We will focus our analyses on the three main cate-
gories of determinants impacting FDI that we will use
in our analyses, economic factors, political factors,
and institutional factors.
2.3.1 Economic factors
The economic conditions of a given country play a
crucial role in shaping the trends of FDI. The eco-
nomic factors that may affect the FDI inflow are
macroeconomic factors such as market size, labor
cost, GDP, etc. Among all these factors, market
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
994
Volume 18, 2022
volume or market value known as market size are
widely used factor in many empirical studies [18, 19].
A larger market will allow better utilization of re-
sources, especially for the market seeking the type
of FDI. Even though different researchers use a dif-
ferent notation for market size, in general, it refers
to the Gross Domestic Product (GDP). Some authors
refer to it as the size of the economy [20], others as
economic size, and some others as GDP directly [21].
Almost all the studies revealed a positive correlation
and casual relationship as well, between FDI and eco-
nomic size are evident, concluding that this factor is
one of the most important factors affecting the behav-
ior and decisions taken by investors. Another eco-
nomic factor that impacts FDI is the fluctuations in
the inflation rate, measuring economic instability. in
the literature, there are numerous cases where infla-
tion negatively impacts the FDI inflows but only in a
few cases, this correlation is statistically significant.
While in [22] and [23] the FDI concept for devel-
oping and emerging countries is seen about debt and
life insurance expenditure. In these studies, the em-
pirical evidence demonstrated that, mixed financing
(debt and FDI) remains more profitable for devel-
oping countries because of the inverted U-shaped
growth effect of the FDI-to-debt ratio and that coun-
tries with higher FDI tend to have higher life insur-
ance penetration. The latter concepts will not be the
scope of our work but might be an incentive for future
works in the field.
2.3.2 Political factors
The impact of government regulations is defined as
effective rules and is expressed in several ways such
as policy bounds and legal behavior are also studied
in different empirical studies. In case the regulatory
institutions are increased, the FDI flows are decreased
as governments are putting more pressure and control
on investors [24]. Authors in [25] tried to measure the
relationship between FDI and institutions’ determi-
nants in 83 developing countries. This empirical re-
sult covers the period from 1984 up to 2003 and found
that factors such as political stability and government
effectiveness are key factors of FDI inflows. Ahlquist
[26] found evidence that countries with democratic
institutions attract more FDI. The analyses of a re-
cent study, based on the collected data of 110 coun-
tries during the period from 2002 to 2012, showed
that developed countries still attract a higher share of
worldwide FDI inflows than developing countries.
2.3.3 Institutional factors
The most important factor in this group is the cor-
ruption index. Many studies show a negative im-
pact of corruption on FDI. Different indexes for mea-
surement of corruption exists based on countries’
specifics. One of the main indexes used from dif-
ferent countries, also for comparison reasons among
them is the Corruption Perception Index, released an-
nually by the global organization Transparency Inter-
national. Based on the authors’findings in [?] discov-
ered a positive relationship between FDI and Corrup-
tion Perception Index. This relationship is positive
because this index uses a scale of 0 (zero) to 100 (one
hundred) when 0 (zero) means a high corruption level
and 100 (one hundred) no corruption at all. Also, ear-
lier in this study, authors in [?] provided their consent
that the corruption index negatively impacts FDI in-
flows in a country, such as if the corruption index is
increased by one standard deviation, this will be as-
sociated with a decrease in the investment flows by
2.9% of GDP. Mostly in developing countries com-
panies complain about corruption as an obstacle af-
fecting the length of licensing procedures and mak-
ing them quite costly. Also, most of these companies
must pay a bribe to win contracts in the hosting coun-
try, as bribery is still a common practice [?]. So, the
presence of corruption in society harms all aspects in-
cluding economic development and consequently for-
eign investments.
In conclusion, analyzing the literature, authors argue
on different factors such as economic, institution, and
political ones, factors that are likely to affect FDI in-
flows, and factors that investors should consider at
the time of making the investment decisions. Some
of these factors we will consider in our empirical re-
search.
2.4 Conditions for attracting FDI
As we mentioned, some important economic factors
are market size, labour costs, employees’ skills, and
qualifications, etc. But there are also other important
factors such as: infrastructure, political stability in the
hosting country, corruption, and democratic situation
etc. However, the right conditions to attract FDI de-
pend on the types of FDI themselves. According to
a study conducted by the Global Economic Forum in
1997, the market size of the FDI in the country of des-
tination in CAP reports and its growth rate, appear as a
determining factor in deciding on the location of FDI.
In general, the larger the market size, higher inflow of
FDI are going to be attracted by the hosting country.
This factor is decisive when a foreign investor decides
to change the location of his production. But this fac-
tor is less important for investors who have decided
to produce in the host country and then export their
production.
Political stability and economic environments of a
country directly affects the decisions of investors in
investing while making their decisions. Any miscon-
ception in this area could jeopardize the implementa-
tion of the investment plan and negatively affect the
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
995
Volume 18, 2022
projected benefits.
Labour force and labour cost are two other factors di-
rectly affecting the attraction of FDI. An investor can
be attracted to invest by the low labour cost or high
skills. Often, investors prefer to invest in countries
characterized by skilled and cheap workforce ate the
same time. Other factors that dominate FDI are insti-
tutional and legal environment, infrastructure, techni-
cal progress, political stability, etc.
Recent economic improvements make developing
countries attractive for foreign investors offering a
wide range of opportunities from new manufacturing
initiatives through joint ventures and direct invest-
ments in the establishment of trade structures or the
signing of licensing agreements to traditional exports.
Some of the main factors affecting the decisions to po-
tentially invest in these countries are:
Low labour cost and flexible labour market -
The low cost does not match the quality and tech-
nical capacity of the labour force, which on the
contrary is often better than in other countries
analogous labour cost levels such as Moldova
and Serbia.
Availability of natural resources - Different de-
veloping countries, such as Albania or Kosovo
have significant mineral recourses, including
chromium, copper, nickel, oil and coal. Also,
thanks to the favourable climate conditions in
most of these developing countries, agriculture
offers interesting opportunities.
Entering in interesting local market - Some of
the developing countries are small countries in
comparison to the developed ones, classified as
continues with middle-low income. However, all
those activities that do not require a large produc-
tion scale can find an interesting local market.
Still limited level of domestic production open
opportunities for foreign investors interested in
setting up new production units or distribution
centres.
Adequate regulatory frameworks In general,
the developing countries allows foreign investor
to build different types of legal entities. Usually,
they chose to invest in individual companies, or
other form of partnerships. These forms might
be limited partnerships where one of the partners
is considered as limited partner, limited liability
companies as an hybrid form of partnership, as
well as joint stock companies owned collectively
by its shareholders. Governments should put ef-
forts to improve the regulatory frameworks in the
short term, especially in developing countries.
Identifying the main factors impacting FDI in devel-
oping countries will help governments to attract new
foreign investment as an important action taken in de-
veloping countries that results in consequently reduc-
ing poverty.
Due to the nature of the data (Time-series cross-
sectional), we will compare the linear regression mod-
els and the panel data analyses. Even though the
dataset selected for this study is longitudinal data,
for comparison reasons, the Ordinary Least Square
method will be considered.
The study addresses the research question: “What are
the factors affecting FDI in the developing and devel-
oped countries based on the selected cross-sectional
times series data?”
3 Methodology
The methodology of this research started with liter-
ature review where our focus has been in identify-
ing the main factors affecting FDI flows. The liter-
ature review process is based on exploring scientific
research papers, articles and publications by different
authors working in the same field. In conclusion to
the literature review, less focus has been given to de-
termination of the factors impacting FDI in develop-
ing countries, leaving a gap regarding the importance
of FDI in these countries. These data have two di-
mensions; the cross-sectional information which re-
flects in the differences between countries selected in
this study and time series that reflects in the change
within countries over time.
3.1 Methods
First, we applied multiple linear regression models
to model the relationship of FDI and its determinants
for all the countries involved in this study. In this
study, we also apply panel data methodology that
helps to identify some determinants that are supposed
to influence the structure of FDI.
Based on the main finding of literature review,
multiple linear regression models most of the time
are not suggested to be used for dataset characterised
by time series and cross-sectional data. In literature,
authors suggest to us panel data analyses as the
estimators obtained by these analyses are more
accurate [30]. Panel data analyses provide efficiency
in the estimation of parameters and allows dynamic
modelling of economic behaviours by capturing
intertemporal information. By applying panel data
analyses to our study, we can better discover causal
relationship. Panel data analysis is a combination of
cross-sectional data with time series data used for
dynamic analyses for the treatment of variables [31].
The data used in this research are panel data. The
dataset contains information of the selected countries
as listed in Table 1 over time. These data have two di-
mensions; the cross-sectional information which re-
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
996
Volume 18, 2022
Table 2: Variables affecting FDI inflows including in analysis.
Explanatory Variables Indicators Data sources
Economic variables
GGP per capita
World development indicators, World Bank
GDP growth
Inflation
Unemployment
Net ODA received per capita
Trade(% of GDP)
Population
Institutional variables Corruption World development indicators, World Bank
Political risk variables
Regulatory quality
World development indicators, World BankPolitical stability
Goverment effectiviness
flects in the differences between countries selected in
this study and time series that reflects in the change
within countries over time. There exit two types of
effects models used in panel data: fixed effects model
and random effects models. The fixed model used
Least Squares Dummy Variables (LSDV). This model
considers constant values for each group instead of a
constant for the whole model, by including dummy
variables in the model for each of the groups. The
Random effects model is a mixed approach of estima-
tion in which the constant for each country is regarded
as random, meaning that systematic effects are con-
sidered random instead of a fixed value. Hausman test
is a test used as an instrument allowing researchers to
differentiate the two panel data models, fixed effects
model and random effects model. It is used to test the
significance of the difference between the coefficients
estimated obtained by the two models.
In our study, the countries selected do not have very
similarity between them, implying a random effects
model as more appropriate to be used. However, we
applied the non-parametrical test of Hausman to de-
cide between fixed or random effects.
3.2 Variables and Data Collection
The variables in this research are selected based on
the previous theoretical and empirical research on
this field. This study examined any potential corre-
lational relationships between FDI inflows and GDP
Per Capita, GDP growth, Inflation, Unemployment,
Net ODA received per capita, Trade (% of GDP),
Population, Corruption, Regulatory Quality, Politi-
cal Stability No Violence and Government Effective-
ness. Variables selected as explanatory variables in
the models we run are presented in Table 2. To re-
spond to the second research question, the correla-
tion coefficients of Pearson were used to test the rela-
tionship between the FDI inflows and the explanatory
variables.
The data are downloaded from the World Bank and
are processed using SPSS. As we have data for dif-
ferent countries as listed in the Table 1, and for
each country we have time series date from 2000-
2019. The number of observations collected from
data source available belongs to 25 countries, includ-
ing Albania. The data used in the empirical analy-
ses are data during the last 19 years between 2001 to
2019. So, in total the dataset has 475 records. Even
if they are not normally distributed, the sample size
which is more than 30 observations helps us to use
data on the variables.
3.3 Empirical framework
Multivariate regression technique is used to analyse
our panel data with the scope of identifying the fac-
tors affecting FDI inflows in the selected countries
over the time period 1996-2019. In Figure 2 it is illus-
trated a factor relationship conceptual model of differ-
ent factors that impact FDI. These factors are grouped
into three main categories. Based on this conceptual
Figure 2: Factors affecting FDI inflows.
model we will state different hypothesis used to test
the significancy of the relationships of the presented
factors with FDI inflows for the developing countries
and developing countries too. All the hypotheses will
be tested for two different groups, developed coun-
tries, and developing ones to test their significance in
these groups of countries. As the scope of this work is
to investigate the relationship of FDI and other poten-
tial determinant variables that influences it, we con-
ducted a linear regression analysis. In this study, the
statistical analyses are carried out using SPSS soft-
ware and R. Using OLS (Ordinary Least Square), we
discover which of the variables are statistically sig-
nificant in explain the variation in the FDI inflows by
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
997
Volume 18, 2022
economic, institutions and political risk variables.
We state the first model as follows:
F DI =F(GDPgrowth, U nemployment, T rade,
Inf lation, Regulatory, Corruption, Effecti
veness, Stability, lnP op, lnGDP pc)
(1)
where lnPop and lnGDPpc are logarithmic transfor-
mation of the variable Population and GDP per capita
respectively. This can be econometrically modelled
thus:
Model 1- developing countries.
F DI =α0+α1GDPgrowth +α1U nemployment+
α3T rade +α4Inf lation +α5Regulatory+
α6Corruption +α7Ef f ectiveness +α8Stability+
α9lnP op +α10 LnGDP pc
(2)
Model 2- developed countries.
F DI =α0+α1GDPgrowth +α1U nemployment+
α3T rade +α4Inf lation +α5Regulatory+
α6Corruption +α7Ef f ectiveness +α8Stability+
α9lnP op +α10 LnGDP pc +α11 ODA
(3)
In the model 2 we introduce the Official Develop-
ment Assistance (ODA) variable. This is an impor-
tant variable mostly for developing countries where
the investments come in forms of foreign aids or in-
ternational assistance.
4 Empirical Results and Discussion
In this section the focus will be given to the presen-
tation of the main results analysed through statistical
techniques. First, we present graphically the FDI in-
flows of the selected countries grouped by developing
countries and not developing ones. The second ses-
sion provides an overview of all the statistical analy-
ses used to test the significance of the variables into
the FDI flows in hosting countries.
4.1 FDI inflow
Figure 3 shows the latest situation of FDI for all
the countries in the world classified into one of the
three categories: developed economies, developing
economies, and economies in transition. Analysing
the trends in the data, it is evident that FDI has de-
clined from 2019 to 2019, respectively from 1.41 tril-
lion US$ to $1.39 trillion US$. Also, from the se-
ries of data it is clear that FDI inflows to developed
countries shrank by 6%, while the inflows in the de-
veloping countries demonstrated to be more resistant.
Their flows were unchanged (Figure 3). This find-
ing provides evidences that the FDI determinants im-
pact differently the FDI inflows based on categories
in which the country belongs: developed, developing
or transition. The data source of Figure 3 is UNCTAD
and the data are preliminaries estimates for the period
2008–2019 and the FDI inflows are reported in Bil-
lions of US dollars. In Figure 4 and 5 it is shown the
Figure 3: FDI inflows over years.
temporal evolution of FDI inflow by countries. As
we can see, among the developing countries only a
couple of countries stable FDI inflow growth provid-
ing evidences that Indonesia and Argentina registered
the lowest FDI inflows, whereas Bulgaria and Mon-
tenegro the highest values. Also, the FDI inflows is
not stable in the developed countries as it is shows in
Figure 5. Chile and United Kingdom registered the
highest value.
Figure 4: FDI inflow by developing countries.
Figure 5: FDI inflow by developed countries.
4.2 Results of descriptive statistics
Table 3 reveals the summary statistics for all the coun-
tries selected in this study.
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
998
Volume 18, 2022
Table 3: Descriptive statistics result.
Minimum Maximum Mean Std. Deviation
GDP growth (annual %) -10,89 14,233 3,70 3,35
Unemployment, total (% of total labor force) 2,60 37,25 12,27 8,25
Net ODA received per capita (current US$) -3,85 278,08 53,10 65,57
Trade (% of GDP) 21,85 210,37 77,09 38,96
Inflation, GDP deflator (annual %) -5,99 86,83 6,20 8,52
Foreign direct investment -1,86 37,27 4,62 4,56
Regulatory Quality: Estimate -1,07 0,84 0,06 0,41
Control of Corruption: Estimate -1,18 0,41 -0,29 0,29
Government Effectiveness: Estimate -0,96 1,27 0,03 0,42
Political Stability and Absence of Violence/Terrorism -2,37 0,82 -0,27 0,62
logPop 13,32 21,06 16,79 2,10
InGDPpc 6,62 9,70 8,59 0,63
Table 4: OLS Results of the Regression Model (1).
Correlations
FDI, net inflows (%
of GDP)
GDP growth (annual
%)
Unemployment, total
(% of total labor
force) (national
estimate)
Trade (% of GDP)
Inflation, GDP
deflator (annual %)
Regulatory Quality:
Estimate
Government
Effectiveness:
Estimate
Control of
Corruption: Estimate
Political Stability and
Absence of
Violence/Terrorism:
Estimate
lnPop
lnGDPpc
FDI, net inflows (%
of GDP)
Pearson Correlation
1 ,143** ,204** ,242** -,042 ,096*,191** -,117*,065 ,449** ,238**
Sig. (2-tailed)
,002 ,000 ,000 ,364 ,039 ,000 ,012 ,160 ,000 ,000
N 463 463 453 463 463 463 463 463 463 463 463
** Correlation is significant at the 0.01 level (2-tailed).
*Correlation is significant at the 0.05 level (2-tailed).
Table 5: OLS Results of the Regression Model (1).
Coefficientsa
Unstandardized Coefficients Standardized Coefficients
Model variable B Std. Error Beta t Sig.
(Constant) 32,251 4,684 6,885 ,000
GDP growth ,180 ,058 ,145 3,092 ,002
Unemployment -,092 ,035 -,174 -2,622 ,009
Trade 9,136E-5 ,007 ,001 ,013 ,989
Inflation -,051 ,038 -,070 -1,345 ,179
Regulatory Quality: Estimate 1,075 ,765 ,191 1,405 ,161
Control of Corruption 1,634 ,703 ,365 2,323 ,021
Government Effectiveness 2,684 ,825 ,506 3,253 ,001
Political Stability and Absence of Violence/Terrorism ,842 ,364 ,145 2,314 ,021
lnPop 1,125 ,177 ,486 6,359 ,000
lnGDPpc ,916 ,323 ,238 2,841 ,005
F= 187,213613 R-Squared=0.809
aDependent Variable: Foreign direct investment
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
999
Volume 18, 2022
Based on the descriptive statistics is observed that
FDI inflows values to the developed and develop-
ing countries selected in this study between 2001 and
2019 range from the maximum and minimum of 37.27
and -1.86 respectively with an average of 4.62 and
standard deviation of 4,56. Further, GDP growth has
a range from -10.89 to 14.23, and a mean and standard
deviation 3.7 and 3.35 respectively.
As indicated in Table 4, some of the variables re-
ported positive relation and some others negative re-
lationship. Only 7 out of 10 report significant cor-
relations (GDP growth, Unemployment, Regulatory
Quality, Government Effectiveness, Control of Cor-
ruption, lnPop, lnGDPpc).
By applying the Model 1 we obtain the following es-
timated coefficients presented in Table 5. The fol-
lowing variables are statistically significant at the
0.005 level: GDP growth, control of corruption, un-
employment, government effectiveness, political sta-
bility and absence of violence/terrorism, total country
population and GDP per capita.
In the multiple regression model, R-squared is 0.809,
which shows that around 80% of the data fits the
model. 80% variation in FDIs inflow might be ex-
plained by the by the variables included in the regres-
sion. Based on the F-statistics value (187.2,) and p-
value (<0.05) we conclude that model 1 is overall sta-
tistically significant.
Unemployment and inflation have a negative influ-
ence on FDI inflows. GDP growth, control of corrup-
tion, unemployment, government effectiveness, polit-
ical stability and absence of violence/terrorism, total
population of the country, and GDP per capita are sta-
tistically significant in the model. So, these findings
are in line with general conclusions drawn from other
similar studies found in literature. The GDP per capita
of the country of destination is statistically significant
and in a positive relationship with bilateral FDI flows.
This positive impact is with elasticity 0.916 which
means that 10% increase in the level of GDP is ac-
companied by an average increase of 9.16% of HDI
flows.
4.2.1 FDI inflows means comparison between
developed and developing countries
The first model applied is not considering the dif-
ference between developed and developing countries.
Therefore, different countries belonging to two cate-
gories (10 developed economies and 16 developing
economies) are part of the dataset used to derive in-
sights of potential factors that attract FDIs in these
countries.
The null and alternative hypotheses can be stated as
below:
H0: There is no significant difference in FDI be-
tween developed and developing economies.
H1: There is significant difference in FDI between
developed and developing economies.
Levene’s test is used to check the homogeneity of
variances. By applying this test, equality of means
in two or more groups is tested. The results are given
in table 6 and 7.
Since p-value of Levene’s test 0.002 (<0.005), al-
low the rejection of the null hypothesis. So, the al-
ternate hypotheses is accepted concluding that there
is significant difference in FDI between developing
and developed countries. Based on this, we conclude
that ”Equal variances not assumed” option should be
selected.
4.2.2 Multiple Regression for the developing
countries
As previously stated, after finding empirical evi-
dences that the means inflows of FDI in the devel-
oping economies and developing ones differs signifi-
cantly, we will apply the other model (model 2). In
this model we will introduce the Official Develop-
ment Assistance (ODA) variable, an instrument used
by developing countries to promote welfare and eco-
nomic development. This variable is very important
for the developing economies as it is considered as
important instrument for promoting economic devel-
opment in these economies and contributing directly
to the economic growth. In developing countries,
many important projects are implemented as ODA
and positive impact of ODA in FDI inflows is ex-
pected.
In the model two (2) we have excluded all the data of
the developed countries. We run the regression only
on the records belonging to developing countries. The
results of are given in table 8.
The determinants that are statistically significant
for developing countries are unemployment, ODA,
population and political stability and absence of Vi-
olence/Terrorism. Based on the obtained result, it is
evident that GDP growth and GDP per capita are not
statistically significant in these countries. This might
be explained by the fact that investors in general do
not have high expectations related to economic situa-
tion in these countries.
The regression results reveal a significant negative
relationship at 3,4% between the dependent variable
(FDI) and unemployment. This suggests that a de-
crease in the unemployment rate in one of these coun-
tries will be associated with an increase in the FDI in-
flows.
Also, the empirical results show a negative correla-
tion between trade and FDI inflows, but not signifi-
cant. The same conclusion is drawn for inflation and
regulatory quality variables. Furthermore, the results
indicate that variable political stability and absence of
violence/terrorism has significant and positive impact
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
1000
Volume 18, 2022
Table 6: Results of Levene Test.
Group Statistics
D1 N Mean Std.
Deviation
Std. Error
Mean
Foreign Direct Investment, net inflows (% of GDP) 0 292 4,2609 4,3829 ,2564
1 171 2,7420 2,3844 ,1823
Table 7: Levene’s Test result for equality of means.
t-test for Equality of Means
95% Confidence Interval of the Difference
Sig. t df Sig. (2-
tailed)
Mean
Differ-
ence
Std.
Error
Differ-
ence
Lower Upper
Equal variances
assumed
,002 4,1 461 ,000 1,51888 ,363159 ,805228 2,2325
82
Equal variances
not assumed
4,8 458,8 ,000 1,51888 ,314703 ,9004 2,1373
26
Table 8: OLS Results of the Regression Model (2).
Coefficientsa
Unstandardized Coefficients Standardized Coefficients
Model variable B Std. Error Beta t Sig.
(Constant) 23,397 7,298 3,206 ,002
GDP growth ,059 ,075 ,049 ,783 ,435
Unemployment -,158 ,052 -,329 -3,040 ,003
Net ODA received per capita ,034 ,008 ,501 4,346 ,000
Trade -,019 ,011 -,185 -1,789 ,075
Inflation -,008 ,047 -,013 -,175 ,861
Regulatory Quality -,066 1,100 -,006 -,060 ,952
Control of Corruption ,166 1,273 ,012 ,131 ,896
Government Effectiveness ,576 1,420 ,059 ,405 ,686
Political Stability and Absence of Violence/Terrorism 1,234 ,510 ,184 2,421 ,016
logPop ,790 ,257 ,409 3,078 ,002
lnGDPpc -,404 ,515 -,060 -,785 ,433
F=75,965 R-Squared=0.803
aDependent Variable: Foreign direct investment
in FDI inflows.
In this model, R-squared is 0.803, which shows that
around 80% of the data fits the model. 80% variation
in FDIs inflow might be explained by the by the vari-
ables included in the regression. The F statistics is
75.964 (p<0.005) which shows that the overall Mod-
ell 2 is significant.
Regarding the population, in both models it is stated
that it is a significant determinant in FDI. This con-
clusion agrees with the findings in previous studies
conducted by different authors [32].
If we compare both models, the determinates of FDI
that are statistically significant differs. More deter-
minants are significant in the model one in compari-
son to the model two. This might be explained by the
fact that in the model one the FDI inflows in devel-
oped countries and its determinants values are more
stable in time in comparison with developing coun-
tries where these indicators vary in time. Also, as the
results obtained by the two models differ quite sig-
nificantly, we applied panel data analyses in order to
benefit from their capability to provide more leverage
for causal inference.
4.2.3 The Hausman Specification Test: Fixed
Effects or Random Effects
During the panel analyses, the Hausman test is used
in order to test for difference whether random effects
or fix effects model should be used.
In order to decide which model to use, we applied the
Hausman test. We run this test in R . The obtained re-
sults are given in table 9. The p-value of greater than
0.005 was observed (p=0.8749) implying acceptance
of the null hypothesis (H0). Thus, the Hausman test
shows that the random effects model is more reliable
than the fixed one. Based on this evidence, we select
the random effects model that takes into considera-
tion the random deviation for each of the countries.
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
1001
Volume 18, 2022
Table 9: Hausman specification test.
Test summary Chi-square p-value
Cross-section random 5.9766 0.8749
As we have previously explained, the data consists of
26 countries during the period 1996-2019.
The following Table 10 shows the results of random
effects model. Evidence from the results shows that
the third model differ form he previous ones. The
random effect model estimators and their significance
differ quite significantly from that of the pooled re-
gression models (Model 1 and Model 2). The signif-
icant determinants are GDP growth, ODA, Trade, in-
flation, regulatory quality, government effectiveness,
political stability, and population. Among all these
variables, only inflation shows a negative relation-
ship with FDI inflows. Also, if you compare the co-
efficients estimated by the models, their magnitude is
different in each of the model. The differences of the
three models are shown in Table 11.
5 Conclusions and Further Work
Due to the increasing role of FDI inflows, in job
creating, in increasing the host country welfare and
technical enhancements, changing the rule of the
game in the global market, we aimed in this paper
to determine the factors affecting FDI inflows in de-
veloping and developed countries based on empirical
analyses. First, based on similar studies in literature,
we defined some important determinants classified
into three main categories: economic, institutional,
and financial. Identifying main factors impacting FDI
in developing countries, will help governments to
attract new foreign investment as an important action
taken in country level that results in consequently
reducing poverty. In this research, based on the
recent data collected by the World Bank repository,
we applied pooled regression models and panel data
analyses to determine the factors influencing FDIs.
Multivariate regression technique is used to analyse
the panel data with the scope of identifying the
factors affecting FDI inflows in the selected countries
over the time period 1996-2019. We applied three
types of models: (1) pooled regression OLS for all
countries, (2) pooled regression OLS for developing
countries, and (3) panel data.
Using the Levene’s test, we conclude that the mean of
FDI inflows for developed countries and developing
ones is significantly different (model 1 vs model2).
When researchers study the determinates of FDI
inflows, they should separate during the analyses the
developing countries vs developed ones as these two
categories do not have very similarity behaviour in
FDI trends and factors affecting it.
As the data used in this study are panel data, com-
prising time series observations across a collection of
countries, panel analyses are more efficient models
to be used. These data have two dimensions; the
cross-sectional information which reflects in the
differences between countries selected in this study
and time series that reflects in the change within
countries over time. Random effects models are
capable to control the unobserved heterogeneity bye
estimating the mean of distribution effects. Also, as
the countries selected do not have very similarity
between them, implying the random effects model is
more appropriate to be selected. For the panel data
analyses, to decide which model to use (random or
fixed effects model), Hausman test shows that the
random effects model is appropriate.
Based on the results obtained by applying random
effects model, among effective factors on FDI in-
flows, we could mention GDP growth, ODA, Trade,
inflation, regulatory quality, government effective-
ness, political stability, and population. From all
these factors, only inflation tends to decrease the
FDI inflows in a hosting country. Hence, we suggest
governments in developing countries to give more
attention to these factors.
In the future work, we would like to add more vari-
ables that impact FDI inflows such social exclusion
(percentage), Rule of Law competitiveness, tax re-
forms, financial debts, life insurances, etc. Also, we
will enhance our analyses towards identifying which
countries are the most sensitive and not sensitive to
changes in FDI from the factors that influence them.
The main limitations of this work are two : 1) some
of the factors mentioned in literature were dropped
from our analyses due to their unavailability of data,
multi-collinearity concerns or time limitations to
conduct this research, 2) other countries data might
be taken into consideration for further analyses in
order to generalise the conclusions.
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
1002
Volume 18, 2022
Table 10: Results of Random Effects Model 3.
t Sig.
(Constant) 1.01263386 0.628139
GDP growth 0.13240424 1.657e-05
Unemployment 0.05507111 0.497312
Net ODA 0.00095856 0.048258
Trade 0.00132203 0.032319
Inflation -0.02610114 0.002195
Regulatory Quality 0.00132203 0.027333
Control of Corruption 0.02131442 0.938948
Government Effectiveness 0.68787923 0.024885
Political Stability and Absence of
Violence/Terrorism
0.32095936 0.005938
logPop 0.08266574 0.035310
lnGDPpc 0.13240424 0.183104
Table 11: Summary of models’output.
Determinants Model 1 Model 2 Model3
sign significant sign significant sign significant
GDP growth + yes + no + yes
Unemployment - yes - yes + no
Net ODA NA NA + yes + yes
Trade + no - no + yes
Inflation - no - no - yes
Regulatory
Quality
+ no - no + yes
Control of
Corruption
+ yes + no + no
Government
Effectiveness
+ yes + no + yes
Political
Stability and
Absence of
Violence/Ter-
rorism
+ yes + yes + yes
logPop + yes + yes + yes
lnGDPpc + no - no + no
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
E-ISSN: 2224-3496
1003
Volume 18, 2022
References:
[1] Antràs, P. and Yeaple, S.R., “Multinational
Firms and the Structure of International Trade”,
Handbook of International Economics, Vol. 4,
2014, pp. 55‑130.
[2] Mühlen, H. , & Escobar, O. (2020). The role of
Foreign Direct Investment in structural change:
Evidence from Mexico. The World Economy ,
43(3), 557585.
https://doi.org/10.1111/twec.12879.
[3] Żukowska-Gagelmann K., (2000), Productivity
spillovers from Foreign Direct Investment in
Poland, Economic Systems, Vol. 24, No. 3,
September 2000, pp. 223-256.
[4] Karbalaei, A. & Ho, C.S.F. 2013.
“Macroeconomic Factors and Country Specific
Indicators on FDI Developed and Developing
Countries”, 15th Malaysian Finance Association
Conference 2013.
[5] Pham, L.H. and Wongsurawat, W. (2020),
”Foreign Direct Investment determinants: dynamic
extreme bounds analysis”, International Journal of
Emerging Markets, Vol. ahead-of-print No. ahead-
of-print.
[6] Protsenko, A. (2004). Vertical and horizontal
Foreign Direct Investment s in transition countries,
Ludwig-Maximilians-Universität München.
[7] Wang, M. (2009), “Manufacturing Foreign
Direct Investment and economic growth: evidence
from Asian economies”, Applied Economics, 41
(8), 991-100
[8] OECD. (2008). Benchmark Definition of
Foreign Direct Investment Fourth Edition 2008.
Retrieved from http://www.oecd.org/ pp. 17.
[9] IMF, (1993), Balance of Payments Manual, 3rd
edition, Washington, D.C.: IMF Publications.
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2022.18.95
Nevila Baci, Gjergji Mulla, Denisa Millo, Klei Xhyheri
1004
Volume 18, 2022
[10] Dunning, J.H. (1993) Multinational
Enterprises and the Global Economy. Addison
Wesley, New York.
[11] Sabir, S., Rafique, A., & Abbas, K. (2019).
Institutions and Foreign Direct Investment:
evidence from developed and developing countries.
Financial Innovation, 5(1).
https://doi.org/10.1186/s40854-019-0123-7.
[12] Moosa, I. A.; Cardak, B. A. The determinants
of Foreign Direct Investment: an extreme bound
analysis. Journal of Multinational Financial
Management, v. 16, p. 199-211, 2005.
[13] McMillan, Susan M. 1999. FDI in Three
Regions of the South at the End of the 20th Century.
Basingstoke: McMillan Ltd.
[14] Chakrabarti, Avik. 2001. “The Determinants of
Foreign Direct Investment: Sensitivity Analyses of
Cross-Country Regressions.Kyklos, 54:1:89- 113.
[15] Konings, Jozef. 2001. The effects of FDI on
domestic firms. Econ. Transit. 9: 61933.
[16] Moran, T. (2002), Beyond sweatshops: Foreign
Direct Investment and globalization in developing
nations, Washington, D.C.: the Brookings
Institution.
[17] Pantelidis, P. and Nikolopoulos, E. (2008).
Foreign Direct Investment Attractiveness in
Greece. International Advances in Economic
Research, 14(1), 90-100.
[18] Chakrabarti, A. (2001). The Determinant of
Foreign Direct Investment: Sensivity Analysses of
Cross-Country Regression. Kyklos, 54 (1), 89- 114.
[19] Haaland, J. I., and Wooton, I. (2007). Domestic
labor markets and Foreign Direct Investment.
Review of International Economics, 15, 462480.
[20] Jurcau, A.-S., Andreicovici, I., & Matis, D.
(2011), The macroeconomic determinants of
Romanian cross-border mergers and acquisitions.
International Journal of Business Research.
[21] Aitken, B. J., Hanson, G. H., & Harrison, A. E.
(1997). Spillovers, foreign investment, and export
behavior. Journal of International Economics, 43,
103132.
[22] Brahim Gaies and Mahmoud-Sami Nabi
(2019), “Financial Openness and Growth in
Developing Countries”, Journal of Economic
Integration Vol. 34, No. 3 (September 2019), pp.
426-464.
[23] James M. Carson, Pei-Han Chen and J.
François Outreville (2021), “Foreign Direct
Investment and the Supply of Life Insurance in
Emerging Countries”, Journal of Insurance Issues.
Vol. 44, No. 1 (Spring 2021), pp. 38-64.
[24] Contractor, Farok J. & Dangol, Ramesh &
Nuruzzaman, N. & Raghunath, S., (2020). ”How do
country regulations and business environment
impact Foreign Direct Investment (Foreign Direct
Investment) inflows?” International Business
Review, Elsevier, vol. 29(2).
[25] Busse, M., and Hefeker, C. (2007). ”Political
Risk, Institutions and Foreign Direct Investment ”,
European Journal of Political Economy, Vol. 23,
pp. 397-415.
[26] Ahlquist, J.S. (2006). Economic policy,
institutions, and capital flow: portfolio and direct
investment flows in developing countries.
International Studies Quarterly, 50, 681-704.
[27] Mateev, M. (2008), Determinants of FDI in
Central and Southeastern Europe: New Empirical
Tests. 8th Global Conference on Business &
Economics. Florence, Italy.
[28] Mauro, P. 1995. Corruption and Growth. The
Journal of Quarterly Economics, August: 686- 706.
[29] Bray, J. (2006). Agents, consultants, and
jointventure partners in international business
transactions, in B. Errath (ed.), business against
corruption: case stories and examples. United
Nations Global Compact Office, New York), 108-
118.
[30] Baltagi, B. H. (2008). Econometric analysis of
panel data. New York: John Wiley and Sons.
[31] Okoroafor, I. B., Nwabueze, J. C. and
Okoroafor-Iyke, F. E., (2012), Panel Data Analysis
of National Accounts of Nigeria: Random Effect
Model Approach, Indian Journal of Education,
Information Management Vol. 1(4) 170 173.
[32] Abdul Aziz and Bilal Makkawi (2012),
“Relationship between Foreign Direct Investment
and Country Population”, International Journal of
Business and Management, Vol. 7, issue 8, pp: 63-
70.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the Creative
Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en_US