1 Student number: S2377721
Name: Hermen Dam
Study Programme: MSc IFM
Field Key Words: Foreign Direct Investment; Developing Countries
Do FDIs create Economic Growth in
Developing Countries?
Abstract
2 1. Introduction
In a time where globalisation is undeniable, Foreign Direct Investment (FDI) is an important factor FDI is an integral part of an open and effective international economic system and a major catalyst to development (OECD; 2002). The role of developing countries in globalisation is significant. They accounted for more than half (52%) of the US$1.35 trillion global FDI inflows for the first time in 2012 (Yeung; 2017). FDI means a lot for the investor, but also for the host economy. For example, Santangelo (2018) has found for agriculture in developing countries, that FDIs from developed countries investors increases the food security in the host country.
For countries it can be very useful to know what foreign investments create for the host country. Their activities cause many different effects on the host country in areas such as economic growth, technology and innovatory capacity, employment and market structure, among others (Forte; 2016).
The impact of FDI has been researched a lot and many different contrary conclusions were drawn. for example the effect of FDI on entrepreneurship. In the hosting countries it is found that foreign ownership has helped restructure and enhance the productivity of domestic firms, FDI has positive influence in reinforcing the creation of new firms, and in line with the established literature, a foreign investment is likely to influence the job seeker to get employed rather to start their own business (Apostolov; 2017). In contrast, another research by Franco & Weche Gelübcke (2015) concluded that industrial FDIs (FDIs where the manufacturing company is owned by another industrial company, instead of a foreign bank with financial purposes) always contribute to crowd out domestic firms, but not when high-tech sectors are examined. But also many other determinants of FDI have been studied. Dlamini, Masuku and Raufu (2015) found that openness to trade was positively correlated with FDI inflows in Swaziland’s agricultural sector. Braga Nonnemberg and Cardoso de Mendonca (2004) came up with a relationship between FDI and marco-economic performance, for example inflation and Aziz and Mishra (2016) stated that market size has an positive impact on the FDI inflows in Arab Economies.
3 East and North Africa) but found that for 12 of the 17 countries there was no significant relationship. But they also found that FDI does create Economic growth for three countries (Algeria, Jordan and Libya).
Questions that arise are why would multinationals want to invest in foreign countries? And how can developing countries attract foreign multinationals? Multinationals want to invest in other countries, because for instance FDI has offered multinationals a wider range of choices on how to serve international markets, gain access to immobile resources and improve the efficiency of production systems (Nunnenkamp; 2002).
It looks likes developing countries know the benefits of FDI because they are competing with each other to attract FDI by liberalizing their policy regimes and offering various incentive packages, such as tax rebate, trade liberalization, establishment of special economic zones and incentive packages to the foreign investors. This is important because the conclusion is that small developing countries across the globe can attract substantial amount of FDI just by adopting more outward oriented trade policy and by providing more business friendly environment to the foreign investors (Mottaleb & Kalirajan; 2010).
This research wants to answer the question if FDIs create economic growth in developing countries. The main research question is: Does Foreign Direct Investments (FDI) affect the economic growth in developing countries?
This research adds to the existing literature, by expanding the research on sectoral FDI flows in developing countries. This paper is also a contribution, because it uses an additional measure of economic growth. Where most papers focused on GDP per capita, real GDP etc. this research will not only focus on GDP per capita but also on the Real Welfare Total Productivity Factor, which is introduced by Feenstra, Inklaar & Timmer (2015). This way of calculating growth is different because it looks at living standards within an economy.
4 2. Literature review and hypotheses
Economic growth is determined by a lot of different factors. Petrakos and Arvanitidis (2008) concluded that political and institutional aspects of an economy, as well as development of the country affects the growth in a country. Barro (1996) found that growth rate is enhanced by multiple variables like life expectancy, rule of law and higher initial schooling, while real per capita GDP was negatively correlated to growth.
FDI is also being mentioned as a determinant of economic growth, for example by Ayanwale (2007) who found that FDI impacts the economic growth in Nigeria. But on the other hand Nonnemberg and Cardoso de Mendonca (2008) found that there is no evidence that FDI leads to GDP, which is often seen as a proxy for growth.
In previous literature there has been a debate on the impact of FDI in the host country. Two main theoretical perspectives are widely discussed: the modernization theory and dependency theory. The modernization perspective is based on a fundamental principle in economics that economic growth requires capital investment (Alemu; 2017). The perspective of the new growth theories, the transfer of technology through FDI in developing countries, is especially important because most developing countries have lack of the necessary infrastructure. The lack is there in terms of adequate human capital, economic stability and liberalized markets. This is needed to benefit from long-term capital flows (Bengoa & Sanchez-Robles; 2003).
On the opposite, dependency theorists argue that dependence on foreign investment is expected to have a negative effect on growth and the distribution of income. According to the dependency theorists, FDI could be a threat to young growing companies/firms with limited capital outlays as compared to the multi-national corporations (MNCs); since the young domestic firms will be unable to compete with the MNCs with huge capital outlays (Alemu; 2017).
So on one hand there is expected that FDI generates transfer of technology and human capital, which supposed to generate value and growth in a country, while on the other hand FDI could be a threat to the host country, because domestic companies will not be able to grow if too much multinationals will be established. In the existing literature there is found that FDI contributes positively to economic growth in Nigeria (Ayanwale; 2007). Also for MENA countries the FDI inflows create economic growth (Abdouli and Hammani; 2017). Because the focus of this paper is on the developing countries, we conclude and expect that FDI in developing countries can be positively affected by FDI inflows. Thus the hypothesis, is as follows:
5 Both the dependent and independent variable are influenced by other variables. Some factors have a direct impact on FDI, like trade openness and investment protection which are positive related to FDI inflows in agriculture in Swaziland (Masuku & Raufu; 2015) while some other factors have a direct impact on the economic growth, like population growth and export performance of a country. (Otani & Villanueva; 1990)
Another interesting option is factors influencing the relationship of the independent on the dependent variable, FDI and economic growth respectively. So these factors can strengthen or weaken the effect that FDI has on economic growth, thus after the FDI has arrived in the country. For instance Luiz & De Mello (1999) found that the impact of FDI on growth depends on the technological gap between leaders (capital exporters) and followers (capital importers). The aim of this part of the research is to find if a determinant of infrastructure impacts the relationship of FDI and economic growth.
To introduce this factor we show that the effect of infrastructure has been researched on both FDI inflows and economic growth. For example Weseka (2015) found that quality of infrastructure lowers the cost of doing business and improves the investment climate, thus attracting FDI. While Czernich, Falck, Kretschmer & Woessmann (2009) found that broadband infrastructure (enabler of high-speed internet) creates economic growth in OECD countries. Empirical evidence has also shown that educational infrastructure in India shows an impact on FDI and that would help to sustain the economic growth (Kaur, Khatua and Yadav; 2016).
In this research we want to test whether infrastructure influences the relationship between FDI and economic growth. We want to find if the effect of the FDI inflows on economic growth is better or worse if infrastructure impacts the relationship. Kamara (2013) found for instance that infrastructure has a negative effect on the FDI-growth relationship. This contrasts the findings of Borensztein, Gregorio & Lee (1997) found that it contributes the effect of FDI on economic growth. Measuring infrastructure can be done in multiple ways, because the literature provides soft and hard infrastructure variables. In this research a hard infrastructure is used: the use of electric power in the host country.
6 developed the country is. Electric power as a proxy for infrastructure has been researched before in combination with economic growth, for example by Javid, Javid & Awan (2013) who examined the relationship between electricity and economic growth, which was also done by Lee (2005), Shiu & Lam (2004) and many more. Therefore the hypothesis is as follows: H2: The use of electric power strengthens the relationship between FDI inflows and economic growth in the host country.
For now, the conceptual model will be as follow:
+
+/-
A factor which may influence the impact of the relation of FDI on growth, is the different sectors in which companies and countries operate. Alfaro (2003) shows that the benefits of FDI vary greatly across sectors by examining the effect of foreign direct investment on growth in the primary, manufacturing, and services sectors. She showed that FDI has a negative relationship on growth in the primary sector, but a positive relationship in the manufacturing sector. So, if a country is more oriented on manufacturing, is it possible, that it has a more positive impact on growth? This question is answered by Wang (2009) who concluded that FDI in manufacturing sector has a significant and positive effect on economic growth in the host economies, using data from 12 Asian countries (time period: 1987 – 1997).
Knowing that different sectors have contradicting effects on the economic growth in the host country, it is interesting to test these sectors in developing countries. We make a distinction between the agriculture, manufacturing and service sector, which can be seen as the primary, secondary and tertiary sector. Evaluating the results of prior research, we think that FDI flows in the manufacturing sector (secondary) can positively impact FDI and economic growth more than FDI flows in primary and tertiary sector. This leads to the following hypothesis:
FDIs
Economic growth
in developing
countries
Infrastructure in
7 H3: The FDIs in the secondary sector have an positive impact on the relationship between FDIs and the economic growth in the host country. This impact is larger than FDI flows in the two other sectors.
For the part in which the sectors are measured the model looks like:
+/-
3. Methodology
This research will perform four regressions. The first and second regression will include all data, except the data that contain the information for the different sectors. We will test the impact of FDI on economic development and the moderating effect of infrastructure on this relationship. The third and fourth regression involves the moderating effect of the sectors on the relationship between FDI and economic development. Regression one and three will be measured using RWTFP as a proxy for economic growth, while regression two and four measure economic growth using GDP per capita as a parameter.
For the first two regressions data is used from 48 countries in Africa, Asia and Latin-America, excluding all high-income countries. Data about FDI per sector is not easy to find, but UNCTAD provided two documents, for Africa and Latin-America, about sectoral FDI data per country, so 16 countries of these two regions are combined. All regressions have a time period of 1996 to 2002. The countries which are included in the third and fourth regression are: Argentina, Bolivia, Brazil, Colombia, Costa Rica, Dominican Republic, Ecuador, Honduras, Mauritius, Mexico, Morocco, Nicaragua, Nigeria, Paraguay, Peru and Venezuela.
FDIs
Economic growth
in developing
8 3.1 Dependent variable
Our dependent variable economic growth is measured as the growth of welfare-relevant productivity over time in each country.
To measure economic growth a proxy has to be taken. Many researchers choose GDP or Real GDP as their variable, but GDP is a flawed measure of economic welfare, according to many economists (Jones & Klenow; 2016). The total factor productivity (TFP) is taken.
“The new version of the Penn World Tables represents an important step forward by including measures of physical and human capital and estimates of productivity based on the translog production function which allows for substitution elasticities to differ across countries and over time. The first novelty is to estimate physical capital stocks for all countries in PWT based on data of investment by asset. The second novelty is to estimate the share of labor income in GDP for a large majority of PWT countries. These are combined with (more standard) measures of human capital to arrive at measures of total factor productivity (TFP).” (Feenstra et al.; 2015)
So the Total Factor Productivity thus includes the share of labour income in GDP and measures of human capital. Kato (2016) stated that TFP is an indispensable procedure for the analysis of economic growth. But they also stated that both labour and capital are also essential in establishing growth factors, and that a relationship exists between this labour input and technological progress (measured as TFP). These different factors are combined in the Real Welfare Total Factor Productivity, as is explained in part 3.1 (dependent variable).
Thereby TFP is able to help explaining both the growth of output as well as the growth of welfare. All these parts together is the reason why in this research RWTFP is picked over GDP. For the robustness the impact of the variables on GDP growth is also measured.
9 Both variables measure growth of the economic situation. Growth is measured by comparing current year with the year before year t, so growth, G, it’s calculated as:
G = (Value at t – Value at t-1) / Value at t-1 (1)
3.2 Independent variable
The independent variable Foreign Direct Investment (FDI) is measured by using the FDI inflows as a percentage of the GDP of a country. The data is reported by the World Bank.
3.3 Moderating variables
Many determinants of FDI have been researched before and one of them is infrastructure. Even for infrastructure a lot of variables are available. Rehman, Ilyas, Mobeen Alam & Akram (2010) used telephone mainlines as a proxy for infrastructure in their research about the impact of Infrastructure on FDI. In a study of African countries, it was found that transportation has a positive relationship on FDI, meaning that an improvement of the railroads etc. makes a host-country more attractive for FDI inflows (Khadaroo & Seetanah; 2009). Kaur et al. (2016) argue that a lack of proper infrastructure in terms of inadequate transport facilities, telecommunication services and electricity transmission and distribution can act as a fetter to productivity and increase the cost of doing business. There is also found for developing countries that electric power consumption has a positive effect on FDI (Kachoo and Khan; 2012). That’s why in this paper infrastructure is measured as the electric power consumption (kwh per capita) using the World Bank data. Because the dataset contained lots of big differences, the Log of electric power is taken.
For the regression of the sectors, the percentage FDI flows of the total FDI of the three sectors is computed per sector for each year and country in this regression. The sectors (primary, secondary and tertiary) had available data in the UNCTAD World Investment Directory.
3.4 Control variables
Two control variables are used for all regressions. The first one is the population growth of the countries. The mutually effect of population and economic growth has been researched before, for example by Herzer, Strulik & Vollmer (2012) who found that an increase in income per capita was in line with a decline in population growth. Li & Zhang (2007) found also a negative relationship for birth rate and economic growth.
10 measures inflation as the growth in the GDP deflator. This is the ratio of the GDP in current local currency to GDP in constant local currency. Using inflation as a proxy for macroeconomic stability is also done by Alfaro (2003). For these two control variables the growth method, explained in formula 1 which compares year t with year t-1, is applied.
To sum up all the variables we now present Table one, which includes all variables. Table 1
Variable definitions and data sources
Variables Definition Data source
Dependent variable Economic
Growth (1)
growth of Real Welfare Total Factor Productivity
Penn World Table version 9.0 Economic
Growth (2) Growth of the GDP per capita growth World Bank Independent variable
FDI FDI inflows in percentage of GDP World Bank
Moderator
Electric power
Log of the electric power consumption
(kwh per capita) World Bank
FDI primary
FDI flows in the primary sector as a percentage of the total FDI flows
UNCTAD Investment Directory
FDI secondary
FDI flows in the secondary sector as a percentage of the total FDI flows
UNCTAD Investment Directory
FDI tertiary
FDI flows in the tertiary sector as a percentage of the total FDI flows
UNCTAD Investment Directory
Control variables
Population Growth in total population World Bank
11 4. Analysis
To analyse the relation between FDI and economic growth and to see if the moderating effect of infrastructure influences this relationship, the model for our first and second regression is built up as follows:
Econ. Growthit = β0 + β1*FDIit + β2*Infrastructureit + β3*Populationit
+ β4*Inflationit + β5*(FDI*Infrastructure) + ε_(i,t) + ai (2) With i indicating the countries and using t to index time.
The model which is used for the third and fourth regression, including the sectors, is constructed as follows:
Econ. Growthit = β0 + β1*FDIit + β2*sectoritl + β3*Populationit
+ β4*Inflationit + β5*(FDI*Sectorl) + ε_(i,t) + ai (3)
Whereas i is used to index countries, t indicates time and l is linked to the sectors primary, secondary and tertiary respectively.
β0 is the constant and ε is the error term. In both cases the β5 is the interacting factor of infrastructure and the sectors respectively. The data consists of cross-sectional panel data. The Hausman test will determine if we take a fixed or random effects model (ai).
5. Results
12 5.1 Regression 1
The descriptive statistics show an average growth of less than 1% per year for economic growth and the maximum economic growth (1) is 0.298, which shows us that the growth of Total Factor Productivity from a country in this research grew by almost 30% in one year. The maximum FDI is 12.715, which indicates that the FDI inflows of a country in one year was 12.715 times the GDP. A reason can be that a country becomes more attractive for FDI. Especially developing countries, because like noted before, these countries have become more and more interesting and valuable for multinationals to invest in. Negative minimum values of economic growth (1), population and inflation implies a fall of the values, which mean they did not grow, but decreased.
For Infrastructure the values are the result of a log-function. This is also done by Canning and Bennathan (2007).For example he mean value of 1.426 is a log return of electric power. The values of the variable electric power are being logged with the following formula:
Electric power = 100 * log(electric power / electric power-1) (4)
This is done because the differences in the values of electric power were big, and to scale it, the values are being logged, so the differences in the values are smaller and better to test. This is also beneficial if more proxies of infrastructure would be introduced, because then the log return would be able to compare these variables and their results better.
Population is measured the same way economic growth is measured. Therefore the outcome of the maximum of 0,258 indicates a growth of 25,8% of the population on a 1 year basis. This is very large, because this would implicate that the population increased with more than a quarter of the whole population of the country.
Inflation is measured as the growth of the GDP deflator. This is done before, as is mentioned in the data part. Again the growth formula is applied and we see that the mean is 0.019 which
Table 2
Descriptive statistics
Variables Mean Median Maximum Minimum Stand. Dev.
Economic growth (1) 0.002 0.004 0.298 -0.275 0.064
FDI 2.713 2.487 12.715 -8.589 2.657
Electric power 1.426 1.289 34.396 -7.740 3.882
Population 0.019 0.018 0.258 -0.017 0.015
13 indicates that the mean inflation is 1,9%. The median inflation from one year to the next seems to be 7%. This looks a lot but, this is due rising demand of goods and services in a developing market, which leads to higher prices and that results in higher inflation.
Table 3
Correlations RWTFP
Variables Eco. Grow. FDI Infrastr. Population Inflation
Economic growth (1) 1
FDI 0.080 1
Electric power 0.083 0.022 1
Population -0.022 -0.090 0.035 1
Inflation -0.193 -0.038 -0.042 -0.010 1
Table 3 shows the correlations between all the variables. The highest correlation is -0.193, which is not high, indicating that there are no correlations that could bias the research. There are no problems expected by using these variables, looking at their correlation. The results of the correlation tables Table 3, Table 6, Table 9 and Table 12 show if the variables are correlated. A quick look is taking at the strength and direction of the linear relationship between to variables. A value of 0.080 of economic growth and FDI shows that they have a small positive relationship. If a correlation is small or zero, this does not mean that there is no relationship, because this relationship could be nonlinear. For this reason the regressions are performed.
14 Table 4
Panel regression results of a pooled OLS: Economic growth (1), FDI inflows and Infrastructure Variables (1) (2) (3) FDI 0.003 0.003 0.004** (0.184) (0.190) (0.034) Population -0.504* -0.507* -0.492* (0.077) (0.076) (0.082) Inflation -0.084*** -0.085*** -0.083*** (0.000) (0.000) (0.000) Electric power 0.000 0.004** (0.869) (0.032)
FDI Inflows*Electric power -0.001**
(0.015) Constant 0.013 0.013 0.007 (0.113) (0.122) (0.441) Countries 48 48 48 Periods 7 7 7 Observations 336 336 336 R-squared 0.226 0.227 0.243 Adjusted R-squared 0.091 0.088 0.104
Note: In all models the dependent variable is Economic Growth (1) measured as Real
Welfare Total Factor Productivity growth. All models are estimated using panel least squares regressions, with fixed effects. The statistical significant is presented as: ***, **,* for 1%, 5% and 10% levels respectively.
Table 4 provides the outcomes of the first regression. We computed three models. The first model is done with only the dependent variable, the independent variable and the control variables. Model two extended model one with our moderating variable Infrastructure. Model three is an extension of model two, because it includes the interaction between the moderator and the relationship of the dependent en independent variable. The models are all balanced panel datasets, with 336 observations of 48 countries in 7 years (1996 – 2002).
15 economic growth in a negative way too. The p-value of -0.084, which is significant on a 1% level, indicates that when inflation rises, the economic growth will drop.
In the second model (2) the moderating variable infrastructure is added. Electric power shows no relationship with economic growth and it is also insignificant. We see that all other variables stay the same, in comparison with the first model, so no significant relationship between FDI and economic growth, but population and inflation are still significant and negative.
The third model (3) includes all variables and also includes the interaction variable, where infrastructure is linked to FDI and economic growth (1). This moderating effect of electric power has a value of -0.001 and is also significant on a 5% significance level, so infrastructure has a small negative affect on the relationship between FDI inflows and economic growth, which illustrates that the amount of electric power used per capita, lowers the effect of FDI on economic growth (1). Including all the variables in the model shows us that the direct relationship of FDI on economic growth (0.004) and infrastructure on economic growth (0.004) both became significant on a 5% significance level. In other words, FDI inflows have a small positive impact on the economic growth (1) of developing countries, and also the use of Electric power is positively correlated with economic growth (1), thus an increase in FDI inflows and/or Electric power increases the economic growth (1) measured in Real Welfare Total Factor Productivity. The control variables stay negative and significant, so population (-0.508*) and inflation (-0.089***) both affect the economic growth of the countries in this research.
So what does these outcomes practically mean? Looking at the first model, where the initial relationship of FDI on economic growth (1) is measured, we see no significant relationship. Therefore there can be concluded that FDI inflows in a country does not significantly impact growth (1). This indicates that if a government looks for economic growth, FDI does not creates wealth. This does not mean, that FDI inflows are harmful for the country, it’s neutral. A reason why it’s not harmful, is that there is also no negative relationship found between FDI inflows and economic growth (1). FDI can influence other things, for example transfer of knowledge which may not immediately leads to economic growth. When other variables are included the relationship of FDI inflows and economic growth (1) becomes significant and is still positive (model 3). This indicates that, in the presence of these other variables FDI inflows create a little growth in the economy of the host country.
16 approximately 0.500 that is significant on a 10% level in all models. This shows that when the population growth drops, the economic growth increases. This sounds like a logical relationship, because when the growth of population is smaller, there are less people to share the economic wealth with, while an increased population growth indicates that the country has more inhabitants, who have to share the same wealth. Brander and Dowrick (1994) indeed found that high birth rates appear to reduce economic growth through investment effects. A similar argumentation is provided for the control variable inflation. The negative relationship of inflation on economic growth (1) indicates that less inflation leads to more growth. And more inflation, so the currency is worth less, leads to less economic wealth. The inhabitants of the country have less to spend, because of higher inflation.
17 5.2 Regression 2
Now the results of the descriptive statistics, correlations and regression analysis of the second regression will be presented. The second regression is similar to the first model, the only difference is that we used GDP per capita growth in the second regression, as a substitute of Real Welfare Total Factor Productivity growth.
As the descriptive statistics of Table 5 are compared with the descriptive statistics of Table 2, the only differences occur at the variable economic growth, because economic growth (2) is measured as GDP per capita growth. The growth of GDP has a median of 1.756 indicating that the growth of the GDP was more than 100%. We see a large maximum and minimum growth rate for GDP per capita. A reason can be the up effect of developing countries. The catch-up effect is that due to diminishing returns on capital, with all other things being equal, it is easier to achieve higher rates of economic growth in countries with relatively low levels of economic development than in those with a more advanced economy (Papava; 2016).
The correlation matrix shows the correlation of one variable with the other variables, and we see that the highest correlation exists between Economic growth (2) and Inflation: -0.253. This is relatively low and therefore no problems are expected in computing the regression model.
Table 5
Descriptive statistics GDP GROWTH
Variables Mean Median Maximum Minimum Stand. Dev.
Economic growth (2) 1.706 1.756 23.027 -17.908 3.949 FDI 2.713 2.487 12.715 -8.589 2.657 Electric power 1.426 1.289 34.396 -7.740 3.882 Population 0.019 0.018 0.253 -0.017 0.015 Inflation 0.107 0.070 4.306 -0.403 0.265 Table 6 Correlations GDP
Variables Eco. Grow. FDI Infrastr. Population Inflation
Economic growth (2) 1
FDI 0.136 1
Electric power 0.205 0.022 1
Population -0.094 -0.090 0.035 1
18 The test of the first model (4) is done by using the fixed effects model. All other models are done with random effects. The measurement of economic growth (2) using GDP per capita growth shows really different outcomes of the model. Again three tests are performed. The first test (4) shows only the relationship between FDI inflows and economic growth (2). This relationship is positive (0.155) and significant on a 10% level, showing that if FDI inflows increases with 1, the economic growth (2) increases with 0.155. Also the control variables Population and Inflation are significant, on a 10% and 1% significance level, respectively. Population shows a high negative relationship (-12.711) with economic growth (2), indicating that a drop of the population will increase the GDP per capita, which seems completely logic. Also inflation shows a negative relationship (-4.490), which means that a deflation, increases the GDP per capita. During the time of inflation the value of money drops, therefore the welfare of people in a country decreases, because they have less purchasing power.
Table 7
Panel regression results of a pooled OLS: Economic growth (2), FDI inflows and Infrastructure Variables (4) (5) (6) FDI 0.155* 0.118 0.117 (0.075) (0.262) (0.230) Population -12.711* -3.907 -3.911* (0.371) (0.802) (0.802) Inflation -4.490*** -5.204*** -5.204*** (0.000) (0.000) (0.000) Electric power 0.092* 0.091 (0.092) (0.369)
FDI Inflows*Electric power 0.000
(0.992) Constant 2.007 1.884 1.886 (0.000) (0.000) (0.000) Countries 48 48 48 Periods 7 7 7 R-squared 0.104 0.393 0.393 Adjusted R-squared 0.096 0.283 0.280
Note: In all models the dependent variable is Economic Growth (2) measured as GDP per
19 Looking at the second test (5) the variable electric power use per capita is introduced and some changes occur. FDI inflows no longer shows a significant relationship with economic growth (2), so the effect of FDI inflows is gone. Electric power shows a small positive and significant sign (0.092), concluding that when the use of electric power increases the economy is growing. The control variable population became insignificant and plays no role in the determining of the economic growth (2). The other control variable, inflation, stayed significant and became more negative (-5.204).
The third test (6) includes all variables. The interacting effect of infrastructure measured in electric power use is examined in this test. Looking at the results, while using all variables, the effect of FDI inflows on economic growth (2) is insignificant and therefore irrelevant, because FDI inflows do not effect the GDP per capita growth of the developing countries. The variable electric power is not significant, thus also plays no role in the economic growth (2). The interacting effect that the use of electric power affects the relationship between FDI inflows and economic growth (2) is 0.000 as well being insignificant, meaning that it does not impact the relationship. The control variables are both negative, -3.911 and -5.204 respectively, and insignificant in the last model. The value of the population is less negative in model 2 and 3 than in model 1, indicating that the effect of the growth of population is less harmful for economic growth (2), but still quite influencing.
A recap of the outcomes and the interpretation of the second regression shows the positive relationship between FDI and economic growth (2). This indicates that FDI inflows have a positive effect on the economic growth measured as GDP. This effect means that FDI inflows create wealth for the host country. Thus, when more foreign multinationals invest in the countries in Africa and Latin-America, the economy of the host country will benefit. This outcome needs to be an indicator for parties who decide on the policies about attractiveness of foreign investors. Because FDI inflows create economic growth, they should try to get more FDI in to the country, to create more wealth, so it’s important to be attractive to multinationals. When other variables are introduced to the model, FDI is no longer significant, but the initial test of FDI inflows on economic growth (2) is significant and therefore relevant.
20 relationships are very negative and have thus an big impact on the economic growth. So inflation makes that economic growth (2) drops dramatically, while the growth of population also show a big reduction of economic growth (2).
The variable electric power shows a positive relationship with economic growth (2) This is the same result as in model 3 of regression 1. The impact of electric power gives a small growth of the economy of the country. Other researchers found that there is a unidirectional causality running from electricity consumption to economic growth (Shengfeng, Sheng, Tianxing & Xuelli; 2012). Looking at the moderating effect of electricity on the FDI-economic growth relationship, a coefficient of 0..000 was found which was also insignificant, indicating that the use of electric power does not influence the relationship between FDI inflows and economic growth (2). Thus the amount of electricity used by the population of the host country does not strengthen or weaken the effect of FDI on the economic growth (2) in that country.
Now the third and fourth regression will be examined. In these models the dependent and independent variable are still economic growth(1 & 2) and FDI respectively, but the moderating effect is changed. Instead of looking at the infrastructure, this regression looks at the impact of FDI in three different sectors within a country: primary, secondary and tertiary sector.
Just like the first two regressions, these models are also tested with the Hausman test, to examine if random or fixed effects are preferred. The results of this Hausman test show no rejection of the null hypothesis (p-value 0.287), and so the random effects model is used.
5.3 Regression 3
Table 8
Descriptive statistics
Variables Mean Median Maximum Minimum Stand. Dev.
21 Table 8 shows the descriptive statistics of the third model, where a smaller dataset is used. Using 16 countries from Africa and Latin America, this unbalanced dataset consists of 109 country – year combinations. The outcomes of the descriptive statistics show the change in economic growth from one year to another , whereas the maximum is still 29,8%. Looking at the variable FDI we see other values as in the first two regressions. This is because for the regressions which includes the sectoral variables, not the FDI inflows is taken, but the total FDI flows, because FDI inflows in combination with sectoral FDIs was not found in the World Investment Directory or other databases. The median FDI is 2,4% in a percentage of GDP per capita. The values of the primary, secondary and tertiary sector are percentages of the total FDI flows. Therefore the median of the FDI in the primary sector of 0.086 indicates that the value in the middle of the dataset is 8,6% of the total FDI flows in a country. Negative values and values more than one are possible because the total flows are tested so a value of – 1036 lowers the total value to 785 (Argentina, 2002). The values of the control variables can be a different because of a smaller dataset, but the calculation is still the same as in the first regressions.
The correlations between all the variables are still low for the most part. The correlations of the FDI in the different sectors show higher values. For example the FDI flows in the primary sector correlates negatively with the FDI flows in the secondary sector (-0.362) and also in the tertiary sector (-0.831). These values aren’t tested in relationship to each other and therefore should not give any problems. Most of the variables are less than 0.200 correlated, and the variables above are still stay below the 0.400 (positive or negative), meaning that they should not give any problems in the variables that are used.
Next up is the third regression. This model contains 5 models. The first model (7) shows the outcomes of the relation between the independent and dependent variable, the control variables Table 9
Correlations
Variables Eco. Grow. FDI
FDI primary
FDI
secondary FDI tertiary Population Inflation
22 and the FDI flows in the primary and secondary sector. The FDI flows in the tertiary sector is not included, because than the total of the sectoral FDI flows would be 1. Models (8), (9) and (10) examines the moderating effect of the FDI flows in each sector. The last model (11) is an additional test where the moderating effect of combining the FDI flows in the primary and secondary sector is tested.
Table 10
Panel regression results of a pooled OLS: Economic growth, FDI inflows and Infrastructure Variables (7) (8) (9) (10) (11) FDI 0.026 -0.365 0.057 0.355 -0.053 (0.916) (0.409) (0.859) (0.475) (0.894) Population 1.768 1.847 1.283 1.755 1.242 (0.250) (0.215) (0.390) (0.241) (0.382) Inflation 0.004 0.005 -0.008 0.007 -0.007 (0.933) (0.912) (0.857) (0.881) (0.867)
FDI primary sector -0.023 -0.039
(0.266) (0.149)
FDI secondary sector -0.017 0.007
(0.595) (0.903)
FDI tertiary sector 0.036
(0.178) FDI*primary 0.939 (0.294) FDI*secondary -0.415 (0.839) FDI*tertiary -0.713 (0.418) FDI*primary + FDI*secondary -0.128 (0.838) Constant -0.025 -0.021 -0.025 -0.052 0.104 (0.362) (0.417) (0.378) (0.119) (0.867) Countries 16 16 16 16 16 Periods 7 7 7 7 7 Observations 109 109 109 109 109 R-squared 0.019 0.027 0.021 0.025 0.008 Adjusted R-squared -0.028 -0.021 -0.027 -0.022 -0.031
Note: In all models the dependent variable is Economic Growth measured as RWTFP
23 The table above, Table 10, provides the outcomes of the third regression. This regression includes 5 models. The first model (7) includes the dependent variable, independent variable, control variables and the both the sector variables of the primary and secondary sector. If the tertiary sector would also be included, an near singular matrix would occur, because summing up the three sectors would be a total of 1. Model two (8), three (9) and four (10) are the models in which each sector is being tested for an interaction in the relationship between FDI and economic growth. This is done separately, per sector. The last model (11) introduces the combined effect of the primary and secondary sector on the relationship between FDI and economic Growth (1).
The first model (7) provides a small positive relationship between FDI and economic growth (1) of 0.011, but it’s not significant. The primary (-0.023) and secondary (-0.017) sector show a negative and also insignificant relationship with economic growth (1). Neither of the variables show a significant relationship towards the dependent variable. What is worth noticing, is that the control variables changed, when comparing it to the first regression. Both variables changed from a negative to a positive correlation (Population: 1.768 and Inflation: 0.004) , which would imply that an increase population and inflation would also increase the economic growth (1). The second model (8) includes the relationship between FDI and economic growth, the variable FDI in the primary sector, as well the moderating effect of FDI flows in the primary sector on that relation. When the FDI flows from the primary sector interact with FDI flows, the relationship between FDI and economic growth(1) becomes a negative one (-0.365) but it is still insignificant on all levels. The interaction of FDI in the primary sector on the FDI relationship shows a positive sign of 0.939, which also is insignificant. The control variables have the same direction as before, but are nevertheless insignificant.
24 The fourth test (10) examines the effect of FDI flows in the tertiary sector and contains again the moderating effect of this sector on the relationship between FDI and economic growth. The outcomes of the direction of the same as model (7) and (9), showing a positive relationship of 0.355 between FDI and economic growth, but it stayed insignificant. The control variables also stayed insignificant. The moderating effect of FDI flows in the tertiary sector on the relationship above is negative (-0.713) and not significant, and therefore can be concluded that the FDI flows of that sector do not effect the relationship of FDI on the economic growth of a country. Looking at model (8), (9) and (10) the direct relationship of the primary, secondary and tertiary sector shows small relationships, which are insignificant. This indicates that none of the flows of FDI in the specific sectors impact Economic Growth (1).
The last model (11) shows the effect of combining the moderating impact of the FDI flows in the primary and secondary sector on economic growth (1). All other variables in this test show a insignificant coefficient. The effect of the primary and secondary sector together have a negative relationship (-0.128) towards the relationship of FDI and economic growth (1). The coefficient is insignificant, so there is no impact of the two sectors on this relationship.
Summarizing the outcomes of the third regression all relations are insignificant, the conclusion can be drawn that FDI does not create wealth in the host country. Also the FDI flows in the different sectors have no impact on the economic growth (1) of the host country. While the first two regressions showed significant relationships for the control variables, in the third there is no relationship found. Indicating that the growth of the population and inflation does not affect the economic growth (1) of the host country.
25 5.4 Regression 4
The fourth and last regression is executed by using GDP per capita growth as the value driver of economic growth (2). All other variables and tests stayed the same. First the descriptive statistics are shown, afterwards the correlations and finally the regression analysis.
All the values of the descriptive statistics remained the same as in Table 8, except economic growth. Economic growth (2) in this table shows the growth in GDP per capita and, just like comparing Table 2 and 5, Table 11 shows higher and lower numbers for economic growth. The median of economic growth (2), indicating that the value in the middle of the dataset, is now 1.437. Consequently with the higher volatility in values, the standard deviation of economic growth increased.
Again the values of the sectoral FDI flows show high correlations, but aren’t tested on each other. The other values show a stronger correlation, but there are still no excessive red flags.
Table 11
Descriptive statistics RWTFP SECTOR
Variables Mean Median Maximum Minimum Stand. Dev.
Economic growth (2) 1.040 1.437 10.777 -11.877 3.445 FDI 0.030 0.024 0.132 0.000 0.026 FDI primary 0.292 0.086 1.561 -0.045 0.383 FDI secondary 0.201 0.120 0.759 0.000 0.218 FDI tertiary 0.507 0.564 1.000 -1.320 0.366 Population 0.017 0.017 0.028 0.007 0.005 Inflation 0.105 0.075 1.155 -0.263 0.143 Table 12 Correlations
Variables Eco. Grow. FDI
FDI primary
FDI
secondary FDI tertiary Population Inflation
26 The last regression models all the test as done in regression model three, but in this fourth regression model the GDP per capita is used in measuring economic growth. The same 5 models are provided as in regression 3.
Table 13
Panel regression results of a pooled OLS: Economic growth, FDI inflows and Infrastructure
Variables GDP (12) (13) (14) (15) (16) FDI 43.668** 21.136 -21.068 22.422 58.216** (0.024) (0.393) (0.525) (0.516) (0.045) Population 494.489** 486.382** 483.120** 470.115** 475.381* (0.019) (0.023) (0.047) (0.045) (0.052) Inflation 1.031 2.140 0.054 2.545 0.940 (0.662) (0.360) (0.984) (0.330) (0.725)
FDI primary sector -10.365*** -10.810***
(0.000) (0.000)
FDI secondary sector 4.802* 4.293
(0.061) (0.160)
FDI tertiary sector 5.274***
(0.003) FDI*primary 40.064 (0.410) FDI*secondary 95.259** (0.042) FDI*tertiary -3.371 (0.946) FDI*primary + FDI*secondary -76.448* (0.088) Constant -6.928 -5.472 -9.114 -8.796 -7.993 0.070 (0.154) (0.0401) (0.860) (0.068) Countries 16 16 16 16 16 Periods 7 7 7 7 7 Observations 109 109 109 109 109 R-squared 0.524 0.508 0.360 0.408 0.246 Adjusted R-squared 0.416 0.400 0.215 0.274 0.206
Note: In all models the dependent variable is Economic Growth measured as GDP per
27 The first model (12) shows that FDI flows in developing countries impacts the economic growth (2) positively in these regions, with a value of 43.668, which is significant on a 1% significance level. Also the Population is highly positive (494.489) and significant, indicating that an increase of the population would increase the GDP per capita. These high outcomes are a result of the small dataset in combination with the GDP per capita. But showing these outcomes does give relevant information about the growth of an economy and which factors influence this growth. The control variable Inflation is insignificant, and therefore plays no role. Both the FDI flows in the primary sector and the FDI flows in the secondary sector have an significant impact on the economic growth (2). For FDI flows from the primary sector a negative coefficient of 10.365*** is estimated, showing that if FDI flows are from the primary sector it has a negative influence on the economic growth (2). A possible reason can be that employees in the agricultural sector of developing countries are underpaid. The secondary sector shows a positive value of 4.802 indicating that the FDI flows from this sectoral are creating growth for the economy of the host developing countries.
The second model (13) gives the outcomes of the moderating effect of FDI flows in the primary sector on the relationship between FDI flows and GDP per capita growth, as a proxy of Economic growth. The direct relationship between FDI and economic growth (2) becomes insignificant. The other variables, population, inflation and FDI inflows in the primary sector, stayed the same, meaning that FDI inflows in the primary sector still affect the economic growth (2) negatively. The moderating effect of the FDI flows in the primary sector is positive (40.064) but insignificant, so does not impacts the relationship.
Up next is the third model (14) in which FDI inflows in the secondary sector replaces the moderating effect of FDI flows in the primary sector, as is done in the previous model. The outcomes show an insignificant and negative FDI flows value (-21.068). Population is still highly significant and positive: 483.120***. The direct effect of FDI flows in the secondary sector is insignificant, but the moderating effect on the relationship of FDI and the growth of the economy is significant and positive (95.259). Therefore can be concluded that if FDI flows are in the secondary sector, the relationship between FDI flows and economic growth (2) is strengthened.
28 economic growth (2) is positive (5.274) and significant. There can be concluded that FDI flows in this sector influences the economic growth (2) positively.
The last model (16) provides the results of the interaction of the FDI flows in the primary and secondary sector combined. In this last model the FDI flows show a positive correlation of 58.216 towards economic growth, which is significant on a 5% level. All other variables stayed the same. The interacting effect of combining the primary and secondary sector shows that when FDI flows come in the primary and secondary it has a negative (-76.448) impact on the relationship between FDI and economic growth (2), so the relationship is weakened. This is possible, because of the fact that the FDI flows in the tertiary sector becomes increasingly important nowadays, leading to the fact the primary and secondary sector create less or no wealth for the developing country.
The practical outcomes of the last regression are now summarized. This regression shows a positive relationship between FDI flows and the GDP growth, indicating that FDI flows in the host country creates economic growth (2). This relationship is significant in model (12) where no other relationships are tested. When the sectoral effects are tested, this relationship is no longer significant, only in the last model (16) where the impact of the primary and secondary sector is tested. Thus looking at the initial relationship, we see that FDI creates economic growth, if it is measured using GDP.
29 sector strengthens the relationship of FDI on economic growth (2). In model 14 the FDI-economic growth(2) relationship is unfortunately an insignificant relationship. Combining the primary and secondary sector weakens the relationship of FDI on economic growth (2). This is caused by the primary sector, because the secondary sector itself impacts that relation positively. We can therefore say that the primary sector FDIs are harmful for the economic growth of a developing country.
Looking at the control variables there is found that inflation has no relationship with economic growth (2), which is in line with the findings of regression three, but the growth of population has now become positive related to economic growth (2). So a growth of the population will also result in a growth of the economy.
6. Conclusion 6.1 Outcomes
The findings belonging to the first regression show that FDI inflows have a positive impact on the economic growth, measured as the Real Welfare Total Factor Productivity. Therefore hypothesis 1 can be confirmed. When looking at economic growth measured as GDP per capita growth, there is no evidence that FDI inflows impact the economy positively and for this part hypothesis 1 would be rejected.
The interaction of electric power has a very small effect on the relationship between FDI and economic growth looking at the first regression, where RWTFP is used, while the second regressions shows no impact. This outcome is similar to the outcomes of Blonigen and Piger (2011) who found no support in their analysis, that infrastructure affects FDI. That is why hypothesis 2, that predicted a stronger relationship between FDI and economic growth if electric power use increases, can be rejected when GDP per capita growth is used.
30 secondary sector impacts the relationship between FDI flows and economic growth positively. Therefore hypothesis 3 is confirmed. We also see that FDI in the last model (16) has a positive impact on economic growth. This is in a situation where the FDI flows of both the primary sector and the secondary sector are combined, to find the effect on the relationship of FDI on economic growth. It can be concluded that this two sectors are weakening the positive relationship between FDI and economic growth.
6.2 Practical implications
The results of the calculations with economic growth measured as RWTFP, provides us with the facts that increasing FDI inflows means that the economic growth will also gets higher and the host country will benefit. Therefore developing countries should try to attract FDIs from multinationals. For example Borensztein, Gregorio and Lee (1997) found that FDI contributes to economic growth only when a sufficient absorptive capability of the advanced technologies is available in the host economy. This indicates that for example electric power is needed to profit from FDIs, this can unfortunately not be confirmed by the calculations of RWTFP and GDP per capita growth, because for RWTFP a negative relationship was found, indicating that electric power use impacts the effect of FDI inflows on economic growth negatively, while no relationship was found for economic growth measured as GDP. Using the GDP per capita growth only indicates that population and inflation both impact the economic growth negatively, which is logical, because when there are less people in a country and all other factors stayed the same, these people have more to spend. And the same relation applies for Inflation. The results of the FDI flows per sector show that developing countries should focus on getting tertiary FDI flows into the country. These create value for the economy. FDIs from the secondary sector would impact the effect of FDI on economic growth positively. The combination of primary and secondary sector FDI flows will lower the effect of FDI on economic wealth.
6.3 Limitations
31 done. Because of less datapoints, the standard errors are higher and the power of the outcomes is smaller. A bigger dataset would give more reliable outcomes and can strengthen the power of the research.
The data is from 1996 to 2002, which is not recent because the World Investment Directory only provides data up to and including 2006 for Africa and 2002 for Latin America. This therefore gives an explanation of the situation in that period, but would maybe not be relevant anymore because of changed circumstances. For example the crisis of 2008 can have impacted the relationship between FDI and economic growth. So more recent data would give a better display of the current situation.
The amount of energy used may not be the best proxy for measuring infrastructure. The amount of energy used can be biased by the quality of the electricity. If the quality of the electric power is not as good as thought the effect of electric power on the relationship between FDI inflows and economic growth could be weakened. Therefore a possible solution for further research could be to measure the quality of electric power in developing countries, to get a more clear vision in which way electric power impacts the effect of FDI on economic growth. Other proxies of infrastructure could also be tested, like soft infrastructure proxies, which is stated in the part 6.4 Further research.
These limitations impact the results, but nevertheless the outcomes show interesting results about the sectoral impact of FDI flows on the economic growth in developing countries.
6.4 Further research
Concluding the previous paragraph a new research with data that consists of post-crisis FDI data per sector, would be very useful for the developing host countries to respond to behaviour of multinationals that are interested in investing overseas, especially to know which sector is profitable for the host economy.
32 Whereas this research focused on electric power while looking at infrastructure affecting FDI, future research on FDI in developing countries could focus on soft infrastructure variables, because Fung, Garcia-Herrero, Iizaka and Siu (2005) found that soft infrastructure, as more transparent institutions and deeper reloads, have outperformed the hard infrastructure variables. Finally an interesting question is, whether different variables have effects on the different sectors that countries are working in. like IT infrastructure in the tertiary sector, or openness to trade in the primary sector.
6.5 Summary
33 7. References
Abdouli, M., Hammami, S., 2017. The Impact of FDI Inflows and Environmental Quality on Economic Growth: an Empirical Study for the MENA Countries. J Knowl Econ. 8: 254 – 278. Alemu, A.M., 2017. Analyzing the Effects of Foreign Aid vs FDI on Economic Growth n Africa. Journal of Economic Development, Management, IT, Finance and Marketing. 9 (2): 1 – 13.
Alfaro, L., 2003. Foreign Direct Investment and Growth: Does the sector Matter? Harvard Business School. 1 – 31.
Apostolov, M., 2017. The impact of FDI on the performance and entrepreneurship of domestic firms. Journal of International Entrepreneurship. 15: 390 – 415.
Ayanwale, A.B., 2007. FDI and Economic Growth: Evidence from Nigeria. African Economic Research Consortium Research Paper NO. 165. 1 – 42.
Aziz, O.G., Mishra, A.V., 2016. Determinants of FDI inflows to Arab economies. The Journal of International Trade & Economic Development. 25 (3): 325 – 356.
Barro, R.J., Determinants of Economic Growth: a Cross-Country Empirical Study. NBER Working Paper Series NO. 5698. 1 – 118.
Basu, S., Ghosh, S. 2017. FDI in India: Understanding the Implications for Growth and Job Creation. Journal of Management & Public Policy. 9 (1): 5 – 16.
Bengoa, M., Sanchez-Robles, B., 2003. Foreign direct investment, economic freedom and growth: new evidence from Latin America. European Journal of Political Economy. 19: 529 – 545.
Blonigen, B.A., Piger, J., 2011. Determinants of Foreign Direct Investment. NBER Working Paper Series NO. 16704. 1 – 47.
Borensztein, E., De Gregorio, J., Lee, J-W., 1997. How does foreign direct investment affect economic growth? Journal of International Economics. 45: 115 – 135.
34 Brander, J.A., Dowrick, S., 1994. The role of fertility and population in economic growth. Empirical results from aggregate cross-national data. Population Economics. 7. 1 – 25.
Canning, D., Bennathan, E., 2007. The Rate of Return to Transportation Infrastructure. 29 – 68. CDCGroup. Development Impact Evaluation. What are the links between power, economic growth and job creation? 2016. 1 – 30.
Dlamini, M.S., Masuku, M.B., Raufu, M.O., 2015. Determinants of Foreign Direct Investment Inflows in Swaziland’s Agricultural Sector. Journal of Economics and Sustainable Development. 6 (15): 150 – 158.
Dritsaki, C., Stiakakis, E., 2014. Foreign Direct Investments, Exports, and Economic Growth in Croatia: A Time Series Analysis. Procedia Economics and Finance. 14: 181 – 190.
Feenstra, R.C., Inklaar, R., Timmer, M.P., 2015. The Next Generation of the Penn World Table. American Economic Review. 105 (10): 3150 – 3182.
Forte, R.P., 2016. Multinational firms and host country market structure: a review of empirical literature. The Journal of International Trade & Economic Development. 25(2): 240 – 265. Franco, C., Weche Gelübcke, J.P., 2015. The Death of German Firms: What role for Foreign Direct Investment? The World Economy. 677 – 703.
Fung, K.C., Garcia-Herrero, A., Iizaka, H., Siu, A., 2005. Hard or soft? Institutional Reforms and Infrastructure Spending as Determinants of Foreign Direct Investment in China. The Japanese Economic Review. 56 (4): 408 – 416.
Herzer, D., Strulik, H., Vollmer, S. 2012. The long-run determinants of fertility: one century of demographic change 1900-1999. Journal of Economic Growth. 17: 357 – 385.
Javid, A.Y., Javid, M., Awan, Z.A., 2013. Electricity consumption and economic growth: evidence from Pakistan. Economics and Business Letters. 2 (1). 31 – 32.
Jones, C.I., Klenow, P.J., 2016. Beyond GDP? Welfare across Countries and Time. American Economic Review. 106 (9). 2426 – 2457.
Kamara, Y.U., 2013. Foreign Direct Investment and Growth in Sub-Saharan Africa What are the Channels? University of Kansas. 1 – 29.
35 Kaur, M., Khatua, A., Yadav, S.S., 2016. Infrastructure Development and FDI Inflow to Developing Economies: Evidence from India. Thunderbird International Business Review. 58 (6): 555 – 563.
Khachoo, A.Q., Khan, M.I., 2012. Determinants of FDI inflows to developing countries: a panel data analysis. Munich Personal RePEc Archive Paper no. 37278. 1 – 19.
Lee, C.C., 2005. Energy consumption and GDP in developing countries: A cointegrated panel analysis. Energy Economics. 27. 415 – 427.
Li, H., Zhang, J., 2007. Do High Birth Rates Hamper Economic Growth? Review of Economics and Statistics. 89 (1): 110 – 117.
Luiz, R., de Mello, Jr., 1999. Foreign direct investment-led growth: evidence from time series and panel data. Oxford University Papers 51. 133 – 151.
Mottaleb, K., Kalirajan, K., 2010. Determinants of Foreign Direct Investment in Developing Countries: A Comparative Analysis. ASARC Working Paper. 13: 1 – 27.
Nunnenkamp, P., 2002. Determinants of FDI in Developing Countries: Has Globalization Changed the Rules of the Game? Kiel Working Paper. 1122: 1 – 47.
OECD. 2002. Foreign Direct Investment for Development. Maximising Benefits, Minimising Costs. 1 – 32.
Otani, I., Villanueva, D., 1990. Long-Term Growth in Developing Countries and Its Determinants: An Empirical Analysis. World Development. 18 (6). 769 – 783.
Petrakos, G., Arvanitidis, P., 2008. Determinants of Economic Growth. Economic Alternatives. 11 – 30.
Papava, V., 2016. The Problem of the Catch-Up Effect and Post-Crises Economic Growth in the World Leading Countries. Bulletin of the Georgian National Academy of Sciences. 10 (4). 97 – 104.
Rehman, A., Ilyas, M., Mobeen Alam, H., Akram, M., 2011. The Impact of Infrastructure on Foreign Direct Investment: The Case of Pakistan. International Journal of Business and Management. 6 (5): 268 – 276.
36 Sehleanu, M., 2016. Empirical Study on the Determinants of FDI Inflows in Romania. The Annals of the University of Oradea, Economic Sciences. 2: 253 – 261.
Shengfeng, X., Sheng, X.M., Tianxing, Z., Xuelli, Z., 2012. The Relationship between Electricity Consumption and Economic Growth in China. Physics Procedia. 24. 56 – 62. Shiu, A., Lam, P.L., 2004. Electricity consumption and economic growth in China. Energy Policy. 32. 47 – 54.
Walsh, J.P., Yu, J., 2010. Determinants of Foreign Direct Investment: A Sectoral and Institutional Approach. IMF Working Paper. 1 – 27.
Wang, M., 2009. Manufacturing FDI and Economic Growth: Evidence from Asian Economies. Economics Faculty Research and Publications. 1 – 23.
