Dynamic Relationship between Foreign Direct Investment & Economic Growth in Iraq: Evidence from ARDL Bound Testing Approach

The connection between economic development and FDI has been very popular among economists and academia. The importance of this relationship is more significant in developing countries particularly with high unemployment rate and low technological advancement. This investigation is an effort to highlight the issue in Iraqi context by applying ARDL bound test estimation approach. For this purpose, yearly time series data over the period of 1971 to 2016 has been utilized to apply ARDL long run estimation. The results provide robust evidence supporting the view that Iraqi GDP is affected by foreign direct investment in long run. Further, bound test affirm the non presence of short-run causality relationship from foreign direct investment to GDP.


Introduction
Economic development has been a prime attention in the previous couple of decades particularly in underdeveloped nations. Investment is a way to lubricate the economic system especially foreign investment which is known as FDI. Countries are always finding new ways to attract FDI to increase national productivity. The productivity increase always stimulates the economic growth hence overall efficiency of country increases (Findlay, 1978). Furthermore, FDI builds the rate of specialized advancement in the host nation through a learning dispersion and administration rehearses.
Conclusively, inward FDI is a gigantic source to elevate the domestic funds for investments. Hence, new investment directly affects the economic growth in many ways like Research and Development investments to enhance the technical innovation. Human capital also strengthens with the increase in productive efficacy due to foreign firms which increase exports level. Therefore, holistically FDI effect the domestic growth by uprising the domestic investment level by technological advancement in human capital. The technological advancement stimulates the export level due to competitive advantage over technologically deprived nations. The increase in export level introduces new employment opportunities (Ewe-Ghee, 2001).
It is worth mentioning that the domestic players also improve their quality and quantity using the human capital which is developed by foreign firms. Furthermore, the spillover effect created by foreign firms creates the environment of competition for local firms (Glass and Saggi, 2002). The diverse effect of foreign direct investment on economic growth has been witnessed around the globe. Previous studies have also reported negative effect of FDI on economic growth. The focus of past researches was on the co-integration of FDI-growth using different statistical tools. In addition to this, previous researches focused on cross country data where country specific characteristics were ignored. Furthermore, there is a dearth of literature on the causal relationship among FDI and economic growth.
Therefore, this study is an effort to bridge the gap in literature to investigate the FDI and growth relationship using ARDL-bound testing approach. There is a little attention drawn on the subject in Iraqi context. Although, many of the studies have focused on developing countries, there is negligible empirical evidence in Iraqi context. By investigating the role of FDI for economic growth this study can draw important guidelines for policy makers to cope the economic challenges faced by Iraq. Iraq is facing a challenging time since after a long period of war. A high unemployment ratio is also the key challenge for Iraq.
The rest of this study structures as section 2 provides conceptual insights. Section 3 focused on the method used to investigate the research phenomenon. Section 4 presents and discusses the results and section 5 concludes the research.

Literature Review
The impact of FDI on the host country's economic growth has been debated because of inconclusive evidence. The definition of FDI in previous literature is merely different. FDI reflects the "fairness of foreign investors (residents) in establishing and developing companies in various fields of activity in other countries" . The capitals emphasize that entities wishing to gain long-term benefits reside in an economic business resident in another economy and investors are engaged in the business of influencing and controlling it. It is evident that the developing and transition countries benefiting from FDI penetration have a strong investment leap: technology transfer, Industrial restructuring, labour force development skills, impact on production, incomes, prices and the economy as a whole (Ranjan and Agrawal, 2011). It is conducive to improve the quality of production factors, enhance the "based on professional production factors of some competitive advantage" (Anghel, 2002). The active role of the host country is determined by long-term implementation to ensure high levels of stability. A significant advantage to host countries by preferring FDI over other sources of funding is, it is helpful reducing deficit by increasing by increasing balance of payment. Hence, public debt is negatively affected by balance of payment.
According to the basic determinants and types of investment, the impact of FDI on the welfare of host countries is significant through the creation of new productive capacity, new working capacity. Although most of the time foreign investors take over the business means a significant reduction in employment (Bonciu, 2003). Studies at the macroeconomic level emphasize that FDI externalities are affected by the absorptive capacity of the host country, constrained by local conditions, financial market development or educational levels (De Mello, 1999;Johnson, 2006). The existence of direct correlation between the absorbing forces of the foreign capital by the authority and the higher economic expansion level is evident. The quadratic relationship of FDI-economic growth allows researchers to approach the issue (Chong and Baharumshah, 2010;Moran et al., 2005). The development of beneficiary countries highlights the role of foreign direct investment in the development of human capital skills, technology and knowledge transfer, expansion of financial markets. Furthermore, the enhancement of local enterprise productivity increases due to foreign competition to ensure equitable distribution of income (Abzari et al., 2011). Similarly, the relationship between growth rate and rate of national income positively impact the balance of payments (Ericson and Irandoust, 2001).
Conversely, a Negative connections between FDI and growth has also reported (Carkovic and Levine, 2005;Lyroudi et al., 2004). Similarly, an insignificant relationship between the foreign investment inflows to GDP and the rate of national growth has been reported by (Lipsey and Sjoholm, 2005). However, A large number of studies have shown that FDI and economic growth are positively correlated between developed and developing countries. Blomstrom asserted a positive correlation between FDI and growth requires the lowest level of human capital, economic and financial stability and market liberalization in the host country. The foreign capital contributions stimulate the GDP rates by improving competitiveness, human resource development and influencing capital formation (Blomstrom et al., 1994).
There is a general recognition that FDI has a positive and negative impact, involving costs and benefits, but the latter is higher than the cost of investment. Foreign investment has a strong impact on national welfare by promoting competition in the domestic market by increasing innovation (Herman et al., 2004). The view of the relationship between FDI and economic growth reveals the important position of foreign capital in the host economy, and different methods vary depending on the choice of variables. The goal of developing countries is to create an environment that attracts FDI and may lead to technology transfer, increased production, creating new employment opportunities and exports. This objective is the main motive for our investigation of the relationship between FDI and Iraq's economic growth. In this study, we tried to find out whether changes in FDI had a favorable impact on economic growth of Iraq.

Data and Model Estimation
This study aimed to investigate the relationship between foreign direct investment and economic growth variables. Following equation is used to measure the model of this study GDP = f (FDI,H,K) (1) Where FDI stands for foreign investment and measured as Gross FDI inflows to GDP ratio, H is understood as human capital and K is Gross Fixed Capital Formation. GDP is considered as GDP per Capita. This equation explains that economic growth is the function of FDI. The annual data on FDI, GDP and Gross Capital Formation and Human Capital was utilized to estimate the underlying relationship aimed at time period of 1971 to 2016. The data has been extracted from World Bank open access database.

Econometric Model
The study uses the ARDL technique to measure the relationship between the underlying variables i.e. GDP and FDI. For different integration orders and small samples, this is an important method. Even though some explanatory variables are endogenous (Odhiambo, 2009), the method provides long-term unbiased estimations. Before ARDL assessment, the no co-integration need to be assess at second order difference I(2). To obtain vigorous results ADF and PP unit root tests has been utilized. To avoid ambiguous results, Zivot-Andrews test (ZA) is useful to consider break point. The traditional unit root is unable to detect the breaks in unit root test whereas; ZA detects the breaks of the unit root structure. We utilize Baum's (2015) modification method to detect a unit root in an alternate trend for a time series with a steady change in average, changing the time trend and shifting in the slope and interception. The unit root and stationary test results are illustrated in Table 2. The ADF and PP tests show that the correlation variable is the integral of GDP in first difference I(1), and mixed integration has been recorded by all independent variables, i.e. I(0) or I(1). Therefore, the prerequisite that applies the ARDL co-integration, i.e. the variable of interest must be at I(1), and the explanatory variable I(0) or I(1). The null hypothesis in the unit root is also denied at the existence of all variables at level 1%, 5%, or 10% as resulted by ZA test. In addition, the series has not been found significant at second order I(2) integration, It also satisfies an important condition to execute the ARDL program (Kouakou, 2011). In the first stage of ARDL long run estimation is to bound the test process. The statistical significance of variables at lagged level has been examined to estimate long run relationships by unrestricted ECM. Following equations is used to explain ECM to explore the relationship between GDP and FDI: In Equation (2), no co-integration is defined as H 0: ϑ 2 =……… = ϑ 5 = 0 whereas, H1 states: {ϑ2≠0} ….. ... U (ϑ 5 ≠0). Likewise, for the equation (3), no co-integration has defined as H 0 : 2 = ….…= 5 = 0 whereas, H 1 : { 2 ≠0} ……U { 5 ≠ 0}. In the constraint process, the joint meaning of the lag layer is tested by using the F-test method with non-standard asymptotic distribution. When an independent variable is integrated into a combination of order I (0) or I(1) or both, the two asymptotic thresholds provide a test for co-integration. When all series are assumed to be I(0), the lower limit is appropriate while the upper limit is I(1). The null hypothesis is therefore rejected if the calculated F statistic value is higher than the upper bound, hence, indicating co-integration. The null hypothesis of no cointegration is accepted if the calculated F statistic value is lower than the lower limit. In contrast, if the calculated F statistic is in the boundary range, the inference will be Indeterminate. Further, Breusch-Godfrey LM test to establish length of lag.  The t-statistics are presented in parenthesis. AIC is utilized for selection of lag. Table 3 shows the results of bound test on the basis of variables selected for this model. Model 1 Controls H and K, and we fail to reject the hypothesis of no co integration. Hence, Table 3 show that proposed model presented in equation (2) has co integration among variables of the study. Therefore, long run relationship among FDI and GDP has been confirmed when GDP is considered a dependent variable. Consequently, equation (3) consider the FDI as a dependent variable, we find that there is absence of co integration in the long run. In next stage Granger causality test is executed based on the following specifications; The equation (4) consists to diagnose long run causality on the basis of Granger Causality test on the basis of FDI as dependent variable. The equation (5) argues that FDI may affect GDP in the short term without a long-term co-integration. Table 4 shows the results of the long-term causal relationship. This discovery provides strong evidence of the long-term causal relationship between independent variables and foreign direct investment. These results confirm the results of bound tests that establish a long-term relationship between GDP and foreign direct investment only when GDP is considered as dependent variable. The short run causality results are presented in table 6. F-statistics is used to determined causality by testing the significance of all lagged values at I(1). Furthermore, by adding up the coefficients of all independent variables lag, direction of relationship is determined (Lee, 2010;Zhang, 2001). As can be learned from the significance of Fstatistics, short run Granger causality between GDP and FDI is not observed. In general, we have found that in the long run, Iraq's GDP to FDI positive and unidirectional causal relationship. Conversely, no causal relationship has noticed in short-term. The induction of control variables i.e. H and K provides a powerful result and also improves overall model fitness. Conclusively, we can say that in the long run, gross domestic product is positively correlated with foreign direct investment. The result of this investigation is consistent with previous literature on FDI anf Economic Growth (Hisarciklilar et al., 2006;Narayan, 2005). Hence GDP and FDI has a positive and unidirectional casual relationship in the long run for Iraq. At the same time we did not notice any short run causality among the variables under study. Inclusion of Hand K provides robust results and also improves the goodness of fit of the models.

Conclusion
This paper analyzes the dynamic causal connection between foreign direct investment and economic growth since 1970 to 2016 in Iraq. It actualizes an ARDL co-integration model to look at the presence of long run connections between FDI and GDP and the Granger causality is utilized to inspect the direction of causal relationship. This subject merits extraordinary consideration due to the conceivable interrelationship between this arrangement and the effect on economic progress. The outcomes demonstrate that when FDI is a related variable, there is a co-integration connection between the GDP and FDI. The outcomes demonstrate that there is Granger causality among FDI and GDP in the presence of underlying control variables.
The fact that there is an across the board conviction that FDI can create positive efficiency externalities for the host nation, our experimental outcomes affirm to this. Foreign capital flow is the principle driver of economic development in Iraq. These outcomes can produce critical ramifications and proposals for policymakers in Iraq. The study recommends that for FDI to anticipate positive effect on economic development, Iraq should embrace genuine changes with clear goals. It needs to enhance its fascination of FDI through a progression of more eager auxiliary arrangements. The findings of this investigation could be extremely fascinating for developing nations to take in the exercise that the fascination of FDI is essential to advance economic development.