Increasing economic growth constitutes one of priorities for all countries. Stimulating foreign investors, and increasing trade may contribute to this purpose.
Purpose: This paper aims to investigate the relationship between foreign direct investment (FDI), export (EXP) and gross domestic product (GDP). The impact of interaction between EXP and FDI on GDP was also examined.
Design/methodology/approach: For this purpose, quantitative approach was adopted. Secondary data for 49 countries whose gross national income per capita was less than 6 000 $ were collected. Spearman correlation, robust regression and causal mediation analysis were performed.
Findings: Spearman correlation showed very strong correlation among GDP-FDI-EXP. Robust regression indicated that all regression coefficients are statistically significant indicating a positive moderation effect of the interaction between EXP and FDI on GDP. Causal mediation effect indicated that average causal mediated effect is statistically significant while average direct effect is not statistically significant, indicating full mediation. The effect of FDI on GDP is transmitted to GDP through increasing EXP. The effect of the interaction of FDI and GDI was found statistically significant. The results are consistent with empirical studies and existing theories.
Originality/value: This study provided a better understanding of the impact of FDI on GDP by combining various statistical analyses methods. While most studies focused on analyzing direct relationship between FDI and GDP, they failed to explain how such relationship occurred.
Key words: Gross Domestic product, Export, Foreign direct investment.
Increasing economic growth is one of the priorities for all countries. Nevertheless, the determinants of economic growth are still unknown as indicated by Christian (2010). The relation between trade, FDI and GDP has received a great deal of attention both in the theoretical and empirical literature. FDI and EXP were found to be among factors that increase GDP as predicted by Bhagwati hypothesis (Sakyi & Egyir, 2017). However, there is on consensus on the impact of FDI on GDP as findings are still contrasting. Some researchers found that FDI had positive impact on GDP, (Nadeem, Naveed, Zeeshan & Sonia, 2013; Qaiser, Salman, Ali, Hafiz & Muhammad, 2011; Sauwaluck,2012; Jugurnath, Chuckun and Fauzel, 2016). Whereas others found negative impact (Dierk;2010; Abdelbagi,2015). Using correlation analysis and regression model, they identified a positive relationship between FDI and GDP. Similarly, findings are inconsistent regarding the impact of trade on economic growth. Some researchers indicated that trade had positive impact on economic growth while other find negative impact. For instance, Keho (2017) identified that trade openness had positive effects on economic growth both in the short and long-run in Côte d’Ivoire. Kore and Pierre (2019) indicated that trade openness had a negative impact on GDP per capita in the long-run in economic community of Western African States. The conflicting findings on the effect of trade on economic growth makes impossible prediction of the impact of trade on economic growth. There is still a gap knowledge on the impact of trade, FDI on economic growth.
Consequently, the enhancing economic growth effect of FDI and trade predicted by Bhagwati hypothesis (Sakyi & Egyir, 2017) becomes doubtable as the direction of the impact of trade, and FDI on economic growth is still inconsistent. Moderation and mediation effects of FDI and trade on GDP was not estimated in previous researchers in developing countries.
Thus, towards these limitations, it is important to investigate the impact of FDI and EXP on GDP in developing countries. The research question to be answered in this study becomes “to what extent FDI and EXP were related to GDP in developing countries?”
The general objective of this study is investigating to what degree FDI, EXP and GDP were related. Specifically, the study will answer the following question: what was the relationship between FDI, and EXP and GDP. It also analyses the mediation and moderation role of EXP in the relationship between FDI and GDP. The study will address the following questions:
- To what degree FDI-EXP-GDP were related in developing countries?
- What was the impact of EXP, FDI, the interaction terms of FDI-EXP on GDP in developing countries?
- What was the indirect relationship between FDI and GDP in developing countries?
The paper is organized as follows. Section two provides a review of the theoretical and empirical literature on the relationship between FDI, EXP and GDP. Section Three presents the methodology and data used. Section four discusses the empirical results, and provides some policy recommendations.
2. LITERATURE REVIEW
There is number of literatures analyzing the impact of FDI and EXP on GDP. However, researchers did not answer the question related to how FDI increases economic growth/GDP. However, Mamingi and Martin (2018) indicated that the direct effect of FDI on GDP is small when FDI is investigated in isolation, while indirect effect becomes more significant than direct effect. This idea is also supported by Casi and Resmini (2012), based on endogenous growth models, these authors indicated that the impact of FDI may be more significant because of the presence of indirect effects which are capable to potentially affect all variables included in the production function. Muthusamy and Raghuveer (2019) examined the relation between FDI and economic growth including Bangladesh, China, India, the Lao PDR, Mongolia, the Republic of Korea Sri Lanka as well. Least Ordinary Squares, Dicky-Fuller and Granger Augmented were performed to estimate the effect of FDI on economic growth. In spite of the stable trend in FDI, the outcome showed that not all countries analyzed have seen significant effect of FDI on economic growth.The relationship between FDI and GDP in Pakistan was examined by Nadeem, Naveed, Zeeshan and Sonia (2013), and a positive relationship between FDI and GDP was found. Qaiser, Salman, Ali, Hafiz and Muhammad (2011) examined the impact of FDI on GDP of South Asian Association for Regional Co-Operation (SAARC), and found out a positive and significant relationship between GDP and FDI. Sauwaluck (2012) analyzed the impact of FDI on economic growth in South Korea. The result of multiple regression showed a strong and positive impact of FDI on South Korean economic growth. George, James and Peter (2013) showed the mechanism through which FDI contributed to economic growth They indicated FDI increases resources available for investment, and capital formation; transfers technology; increases skills; creates innovative capacity; enhances organizational and managerial practices. In the same line, the Organization for Economic Co-operation and Development (“OECD”, 2002) indicated that FDI comes with technology spillovers, contributes in assisting human capital formation, increases international trade integration, contributes improvement in environmental and social conditions, aids creating a more competitive business environment, leading to more socially responsible corporate policies and reinforces enterprise development. The study of the key positions of proponents of neoclassical theory (Solow), endogenous approach (Romer), and evolutionary growth theory (Freeman) has led to the conclusion that these approaches accept that the category of technological changes is a key driver of economic growth (Dragoslava, Slobodan & Gorica, 2016). Their supported their argumentation based on Solow’s model of production function, written as follows Y = TF (K, L), where Y refers to production (gross domestic product), T refers to technology, K represents physical capital, and L is the amount of work. Dierk (2010) found a negative impact of FDI in 44 developing countries over the period 1970 to 2005. Abdelbagi (2015) found negative and statistically significant impact of FDI on economic growth in low income and middle-income countries. The inconsistency in research findings limits the generalization of the impact of FDI on GDP.
The relation between FDI and EXP attracted researchers. Mukhtarov, Alalawneh, Ibadov and Huseynli (2019) found a positive and statistically significant impact of FDI on export in the long-run in Jordan. They findings showed that increasing 1% in FDI increased exports by .13%. Nasir, Kushtrim and Luljeta (2016) used panel regression techniques and least square dummy variable (LSDV) regression method, and found that FDI have positive effect on export performance in the western Balkan Countries in various model specifications.The Bhagwati hypothesis estimated a growth enhancing effects resulting from the combination of EXP and FDI, (Daniel & John, 2017).
Relationship between EXP and economic growth is supported by previous findings. Sayef and Mohamed (2017) used Johansen co-integration analysis of Vector Auto Regression Model, and the Granger-Causality tests, and found that EXP constituted the source of economic growth in Panama. Nevertheless, Pam (2017) studied 42 SSA countries using data from 1980 to 2012. He concluded there could be a trade threshold below which greater trade openness has beneficial effects on economic growth, and above which the impact on growth declines. Moreover, for SSA, he found no evidence supporting the linear association between openness to trade and economic development.
After evaluating 169 countries between 1988 and 2014, Marilyne, Chantal and Mariana (2018) concluded that countries exporting higher quality goods and new varieties have developed more rapidly. However, when countries specialize in low-quality goods, openness to trade can have a negative effect on growth. The report of the United Nations (2001) enumerated supporting reasons that exports will stimulate economic growth by generating greater use of capacity; taking advantage of economies of scale; bringing technological progress; creating employment and growing labor productivity; enhancing the distribution of limited resources in the economy; reducing the current account pressures on foreign capital goods by gradually reducing the current account pressures on foreign capital goods.
3. METHODOLOGY and DATA
This study used secondary data collected on World Bank web site https://data.worldbank.org/indicator ). Annual data 2018 related to GDP, FDI and EXP was collected for countries whose gross national income is below 6000 $. After checking for missing data, 49 countries were included in the study. To understand the nature of the data, descriptive statistics was calculated. Shapiro test was used to test data normality. Collected data were checked for outliers by using univariate method that examine each variable individually and multivariate method that looks for unusual combinations on all the variables (Irad, 2005). An outlying observation refers to one that appears to deviate markedly from other members of the sample in which it occurs (Irad, 2005). It is important to identify outliers/extreme variables before modeling because they can lead to adversely to model misspecification, biased parameter estimation and incorrect results, (Irad, 2005). Spearman correlation analysis was also conducted to find the nature, the strength and the direction among variables. Robust regression was used to investigated the moderation effects of EXP and FDI on GDP as it gives better coefficients and estimators protect against bias under contamination and breakdown point (Eva, 2017). These robust-regression methods were developed between the mid-1960s and the mid-1980s and are insensitive to outliers and possibly high-leverage points (John & Sanford, 2018).
Data were analyzed using R programming environment. Package robusbase was used to perform moderation analysis. Moderation analysis consists of regression FDI, EXP and FDI*EXP to test whether the interaction term is significant. The model can be written as follows GDP ~ FDI + EXP + FDI:EXP. Whereas mediation package was used to depict direct and indirect relationship between GDP and FDI. The analysis used bootstrapping procedures in simple mediation analysis which does need data to be normally distributed. Data were standardized with scale function. According to Carsten, Steffen and Yasemin (2017) bootstrapping refers to a non-parametric approach that avoids the issue of conventional techniques' doubtful distributional assumptions and makes an effective test of the indirect effect. According to Koopman, Howe, Hollenbeck, and Sin (2015) bootstrapping can be useful when cases are between 20 to 80.
4. RESULT DISCUSION
Before performing correlation, mediation and moderation analysis descriptive statistics are presented. Mean, median, standard deviation, skewness, range, minimum, and maxim are presented. Information provided by this analysis allows to determine statistical methods to be used in data analysis.
Abbildung in dieser Leseprobe nicht enthalten
Source: Author’s calculation
The Table one presents the summary of central and dispersion characteristics of the data. The table shows big difference between mean, and median. This implies that data are not normally distributed. The skewness coefficient which should be near zero for normally distributed data, is greater than two. This confirms the lack of normality. The violation of the assumption of data normal distribution implies that the analysis must be performed using non-parametric test. Todd and Karen (2007) indicated that parametric tests are more robust, and need fewer data to make a stronger conclusion than nonparametric tests. Nevertheless, to use a parametric test, data must be normally distributed, have equal variance and have the same standard deviation, and must be continuous.
Correlation analysis was performed to investigates the direction and the strength of the relationship between GDP-FDI-EXP. The results of the pairwise correlation analysis allow to answer the question related to the relationship between all variables. The question to be answered is the following: to what degree FDI-EXP-GDP were related in developing countries?
Tableau 2: Spearman correlation result
Abbildung in dieser Leseprobe nicht enthalten
Source: Author’s computation
The correlation analysis indicates a very strong correlation among all the four variables. A high correlation means that there is a close association between two or more variables, while a weak correlation means the variables are barely related (Monica & Antonella, 2019). The correlation between GDP and FDI is 0.847. The correlation between GDP and EXP is 0.922. The correlation between FDI and EXP is 0.897. The correlations are very strong. The result of correlation is helpful in forecasting dependent variable with lowest possible errors. Additionally, regression, mediation, and moderation can be performed because there is a relationship between variable.