Globalization , Foreign Direct Investment and Economic Growth in Sub Saharan Africa

The paper examined contributions of foreign direct investment, globalization to real economic growth fluctuation in selected sub-Saharan Africa countries. Adopting the conventional vector autoregressive mechanism the time series data from the selected countries, the result showed that out of the eleven countries studied, foreign direct investment explained the highest proportion in just three countries, Morocco, Ethiopia, and Zimbabwe. Except in Tunisia, Tanzania and Kenya, where the degree of economic openness explained substantial proportion of the output fluctuations, the variations in most of the countries were explained by factors beyond foreign direct investment and economic openness. The result supports the existing finding on African economies that trade liberalization had not substantially impaired economic growth process of the sub African economies as alluded to by previous studies. The upsurge in the capital flows to African economies was also insufficient insulate the economic from the global meltdown and furthermore kick start post crisis economy recovery in Southern African countries. Therefore, the paper concludes that fluctuations in real economic growth in these countries might be beyond the external shocks from the capital inflows and trade flows.


INTRODUCTION
Until the very early of 1990s cross border trade as a percentage of global GDP struggled to get to the level of 1913 when over one third of what was produced in the world flew across border (Kutznets 1967).While it is true that the growth of world trade during the 1990s has surged ahead of world output much faster than in the 1970s and 1980s.Data on trade participation by various groups of countries show that intensity of trade is not the same as extensity of trade (Hoogvelt 2001).There was a turnaround of fortunes in the developing world as a whole, receiving not less than 38 percent of the total of world foreign direct investment by 1997 (UNCTAD, 1990).The direction of this flow however was extremely selective, with 60 percent going to 6 major recipients in the developing world, and 94 percent going to 20 countries including for so called transition economies in Central and Eastern Europe (Human Development Report, 1999).The UNDP Human Development Report noted that only 25 developing countries have access to private markets for bonds, commercial banks loan and portfolio equity, other sub Saharan countries included are shut out for lack of credit rating.In other words, what prospects do exist for sub Saharan African nations?
The concept of globalization which refers to the increasing integration of national economies significantly impacted positively on the world economy than its envisaged negative aspects.Among these positive aspects of globalization are most frequently mentioned factors of globalization as internationalization of production and services, international division of labour, global world trade, transfer of knowhow but also intercontinental transport and especially communications represented mainly by the latest information and communication technologies.In this latest case is the most visible aspect-Internet World Wide Web, e-mail but also still more and more widespread e-commerce, e-trade, e-banking, efinance, e-education, etc.In this sense, the contemporary world is really becoming more interrelated, integrated into one entity often called "a global village" all these most positively perceived aspects of globalization are in general directly contributing to the acceleration of the overall socio-economic development on the global scale.
In view of the above synopsis, inflow of foreign direct Investment and globalization significantly impacted positively on the real gross domestic product in developing countries, the basic issue remains however, how much of the inward FDI to developing countries comes to sub Saharan Africa and secondly, to what extent has openness in trade in the region contributed to economic growth of the countries?Though, some studies have been conducted into the subject matter in the region, this paper re-examines the relationship between growth and globalization with a view to determine the relative contribution of capital inflow and trade on the growth process of countries within the sub Saharan Africa.
The results of the paper supports the existing finding on African economies that trade liberalization though might not be the main driver of growth but it has not impaired the growth process either.Similar, the upsurge in the capital flows to African economies in recent times before the 2008 financial crisis was not sufficient enough to stimulate sustainable economic growth that could withstand the global economic shocks and kick start post crisis recovery in Southern African countries.The post crisis fluctuations in real economic growth in these countries might be beyond the external shocks from the capital inflows and trade flows.Therefore there is no sufficient evidence to claim that either foreign direct investment or trade policy orientation adopted by the African countries contributed negatively to real economic Saibu and Akinbobola 63 growth in African countries.The rest of the paper is divided into five sections.

Survey of Literature
An extensive literature has developed on the influence of openness on foreign across countries.A number of researches, using different approaches, have found growth to be enhanced by foreign trade, or opens, or trade liberalization (Dollar 1992;Sachs andWarner 1995, Ben-David et al., 2000;Edwards 1998, based on earlier work; among others).
A general methodology problem arises in determining the impact of trade on growth because trade and output are determined simultaneously.Each researcher has developed surrogates for measuring the degree and character of openness, and each surrogate is open to disputation.Indeed, Rodriguez and Rodrick (2001) provided a withering critique of the studies mentioned in the preceding paragraph, raising serious doubts about whether the authors have demonstrated their claim that pursuits of liberal trade policies have enhanced growth.Rodriguezair Rodrick persuasively find fault with the surrogates, with choice of data, or with specifications of the model to be fitted.
Frankel and Romer (1999) also found a significant impact of openness on levels of per capita income.To avoid the problem of simultaneity, they constructed an index of trade possibility ban geographic factors and found that it is strongly correlated with per capita income.They also found that actual trade is positively correlated (r=0.62) with trade possibilities enhance income through three diverse channels, greater stock of capital, greater stock of education, and higher total factor productivity.But they explicitly cautioned against using their results to draw inferences for trade policy, which brings different issues into play.Ades and Edward (1999) conjectured that greater openness, by relaxing constraints imposed by the extent of the domestic market, should be associated with higher growth.More particularly, they hypothesized that initial levels of per capita income should have greater (positive) impact on growth for more closed economies than for more open ones, since more open economies are less bound by domestic market size.Their hypothesis was broadly confirmed with the use of data for 66 countries and within 1960-85.They confirmed that the relationship of growth to initial per capita income is statistically significant for closed economies and insignificant for open ones.
Far less empirical work has been done on foreign investment than foreign trade, in part because data were neither so copious nor so detailed.Dobson and Hufbauer (2001) estimated conservatively that cumulative foreign investment (Mainly FDI) contributed over 60 percent to the GDP of emerging market countries by 2000.This significantly outweighs the damage that foreign investment (mainly bank loans) may have contributed through banking or foreign exchange crises, which amounted to about 3 percent of the GDP of emerging markets (if half the estimated loss in GDP were attributed to foreign investment, an estimate the authors consider high).
Borensztein, De Gregorio and Lee (1998) examined the influence of FDI on economic growth in 69 developing countries for 1970 89, they found, after controlling for other variables, that FDI makes a positive effect on domestic investment.Kant (1996) found FDI to be associated with a significant reduction in capital flight as well.A World Bank (1998) found that economic aid alone did not foster economic growth-an appalling result, even allowing for the fact that much aid was given for political support to particular countries or governments, not necessarily to increase growth or reduce poverty.From the perspective of economic development, much aid seems to have simply been wasted.However, aid given to countries that pursue effective economic significant policies can boost economic growth significantly.Aid can contribute to economic growth in a policy environment that encompasses good management of economic policy and the setting of suitable development objectives.Aid alone cannot ensure the right policy environment, the government must desire economic development or improvements in health or education, and act accordingly.Vigorous economic growth in turn, always reduces poverty, even when it enriches some people more than the others.
The phenomenal difference between the growth rates of the East Asian economies and the Latin American economies over the last several decades had stimulated renewed interest in the determinants of economic growth.A prominent and important hypothesis is that these differences in growth rates can be explained by differences in the degree of openness to international commerce.Many suppose that the successful East Asian economies are open, and the unsuccessful Latin American economies are closed.But clear empirical support for this proposition is not easy to come by.Studies by Tyler (1981), Feder (1983), Karoussi (1984), Balassa (1985) and Ram (1985) have examined the relationship between trade and growth in a cross-section of countries by regressing the rate of growth of GNP on the rate of growth of trade and the rate of growth of certain measurable inputs.Generally, the coefficient on the growth rate of trade is positive and statistically significant for all these studies.These studies are however dated; it reflects the base in which this study is anchored.Asiedu (2002), Globerman andShapiro (2002) Fernandez-Arias (2000) found a positive relationship between the degree of openness and foreign direct investment (FDI).thoughRoot and Ahmed (1979), Tumman and Emmert (1979) finds insignificant relationship, this could be adduced to paucity or the unreliability of data because these studies are quite dated.The key policy issue however, is whether for each country, starting where it is, some liberalizations of trade (or foreign investment) would improve its economic performance.In recent times attempt has also being made to examine the causal nexus between trade and capital inflow.Liargovas and Skandalis (2012) examines the importance of trade openness for attracting Foreign Direct Investment (FDI) inflows, using a sample of 36 developing economies (Latin America, Asia, Africa, CIS (Commonwealth of Independent States) and Eastern Europe and found that though trade might not contribute significantly to economic growth in this region there wa sufficient evident to support in the long run, trade openness contributes positively to the inflow of FDI in developing economies.
In view of the discussions above on the influence of trade on growth and openness on the inflow of foreign direct investment, this study intends to fill the gap, not only in the area of scanty literature that exists for sub-Saharan Africa, but also in the area of methodology.This study uses the vector autoregressive modeling approach to estimate the relationship between these variables for some selected countries within the region.

Definition and Measurement of variables
The study examines the impact of globalization and foreign direct investment on economic growth in 12 African countries in the period 1986-2004.The twelve African countries included in the study are Nigeria, Cote d'Ivoire, Ghana, Morocco, Tunisia, Egypt, Zimbabwe, Botswana, South Africa, Kenya, Tanzania, and Ethiopia.The choice of countries was determined primarily by the availability of data and also by categorization, in the sense that , three each were chosen from the four sub region within Africa ,that is west Africa, north Africa, south Africa, and east Africa respectively.The choice of period is premised on the fact that virtually all these countries started economic reform programme during this period.Detailed information on the definitions of variables used in the analysis is presented below: (a) Globalizations (GLO): In the literature, the popular measure of globalization is the degree of openness.The use of this measure is premised on the argument that the more a country opens, the higher the level of its integration with global economy and consequently, a resultant increase in economic growth.This actually explains the various liberalization policies in sub Saharan Africa (SSA) since mid 1980s.There are several measures of openness in literature.These include ratio of trade (exports + imports) to GDP, increase in export, Sach Warner index, and export-import ratio.Following Cigno et al (2002), Ramirez (2001) among others, we used the ratio of trade to GDP as our measure of degree of openness.Since globalization involves dispersion of production activities and location of different segment of the same process in different countries, It is assumed that globalization boosts foreign direct investment (FDI) and equally, positively impact on growth of these economies.(b) Gross Domestic Product (GDP): this is defined as the rate of growth of GDP and is used as a measure of the attractiveness of the host country's market.Theoretically, investment will go primarily to markets that are large enough to support the scale economies needed for production.This simply means the higher the rate of growth of the GDP, the greater the possibility of increased inwards FDI, however considering the lackluster performance of African economies in the last three decades, GDP growth might not have a significant effect on inward FDI (c) Foreign Direct Investment (FDI): we used the stock FDI based on the fact that the positive contribution of the surge in FDI in period under consideration is better measured by the stock variables.The stock FDI values used in this study were generated using a standard perpetual inventory model of the form: Where K t-1 is the flow of gross investment during period t, and s is the rate at which capital stock depreciates in period t-1.In this study, the initial stock of foreign capital was estimated by aggregating over 5years of gross investment (inward inflows)-1980-86 assuming 5percent depreciation rate.We however, anticipate a positive impact on growth of GDP.

Analytical techniques
The causal nexus between globalization (openness), foreign direct investment (FDI) and economic growth is examined within the context of a three-variable vector autoregressive (VAR) system.The model is specified and estimated using quarterly data for 1986(1)-2004(4).Virtually all sub Saharan countries were either implementing economic reforms or about to start economic reforms.Quarterly data were used for two reasons.First, the size of our system requires quarterly data in order to have enough degree of freedom for estimation.The second is based on a desire to minimize any problem with temporal aggregation (see Christiano and Eichenbaum, 1987) that might arise with the use of annual data.A vector autoregressive of order β, VAR (p), for a system of k variables can be written as: Where Xt is a (k × 1) vector of system variables, A is a (k × 1) vector of constants, B(L) is a ( k × k) matrix of polynomials in the lag operator L, and Ut is a ( k × 1) Saibu and Akinbobola 65 vector of serially uncorrelated white noise residuals.The standard Sims (1980) VAR is an unrestricted reduced form approach and uses a common lag length for each variable in each equation.Likewise here, no restrictions are imposed on coefficient matrices to be null, and the same lag length is used for all system variables.Three variables are included in the model: degree of openness (DO), foreign direct investment (FDI) and growth rate of GDP (∆GDP).The data for all the variables are obtained from the International Financial Statistics (IFS).Prior to estimation of the VAR, augmented Dickey-Fuller test were employed to check for the first-order unit roots.
These tests suggested that the first differences of the logs of DO, FDI and GDP should be used in specifying and estimating the model.Based upon the arguments of Engle and Granger (1987), co integration tests were also performed for the three variables that required differencing to achieve their stationarity.Since no evidence of co integration was found, the system was estimated with differences of all system variables.

Empirical Model
The model represented by a three-component vector is thus defined as: Where V is the vector containing the three variables, DOdegree of openness, FDI-foreign direct investment, and GDP-growth rate of GDP.Equation ( 2) is an identity that would be estimated using the VAR technique.The impulse response functions (IRFs) and the variance decompositions (VDCs) are based on the moving-average representations of the VAR model and they reflect short-run dynamic relationships between variables.The VDCs show the percentage of the forecast error variance for each variable that may be attributed to its own innovations and to fluctuations in other variables in the system.The IRFs indicate the direction and size of the effect of a one standard deviation shock to one variable on other system variables over time.Since model variables are converted to first differences prior to estimation of the model, the VDCs and IRFs reported here indicate the effects of a shock to the changes in the growth rates on the changes in foreign direct investment and the degree of openness.The equations of the VAR contains only lagged values of the system variables, it is assumed that the residuals of the VAR model are purged of the effects of the past economic activity.Any contemporaneous relations among the variables are reflected in the correlations of residuals across equations.The Choleski decomposition is used to orthogonalize the variance-covariance matrix.The variables are ordered in a particular fashion, and as such, some structure is imposed in computation of the VDCs and IRFs.The economic intuition of such ordering is that when a variable higher in the order changes, variables lower in the order are assumed to change.The extent of the change depends upon the covariance of the variables higher in the order with that lower in the order.Therefore, the order in which the variables enter the VAR system affects the outcome of the analysis.The preferred ordering in this paper is LFDI, LDO and LGDP.Accordingly, an increase in the foreign direct investment is assumed to stimulate investment within the economy, apparently improving exports and opening up the economy (LDO), and consequently, leads to an increase in economic growth (LGDP).

RESULTS
The causal nexus and sources of variation in globalization (openness), foreign direct investment and economic growth are examined through the computation of impulse response functions (IRFs), and the variance decomposetions (VDCs), which in turn, are based on the movingaverage representations of the VAR model and they reflect short-run dynamic relationships between variables.
The VDCs shows the percentage of the forecast error variance for each variable in the system.The IRFs indicates the direction and size of the effect of a one standard deviation shock to one variable on other system variables over time.Since model variables are converted to first difference prior to estimation of the model, the VDC and IRFs reported here indicate the effect of a shock to the changes in globalization and foreign direct investment on the changes in economic growth rates.More importantly, the equations of the VAR contain only lagged values of the system variables; it is assumed that the residuals of the VAR model are purged of the effect of past economic activity.Any contemporaneous relations among the variables are reflected in the correlation of residuals across equations.The Choleski decomposition is used to orthogonalize the variancecovariance matrix.The variables are ordered in a particular fashion, and, in this way, some structure is imposed in computation of the VDCs and IRFs.The extent of the change depends upon the covariance of the variables higher in the order with that lower in the order.Therefore, the orders in which the variables enter the VAR model affect the outcome of the analysis.The preferred ordering in this paper is LFDI, LOPEN and, LGDP.As established by previous studies (for example Akinlo 2005), an increase in net foreign direct investment (LFDI) inflow is assumed to lead to increase in external trade that boost the ratio of trade to GDP, hence enhancing degree of economic openness.The degree of economic openness has also been found to have a positive relationship with economic growth, so increase in capital inflow and greater access to international goods and factor markets not only lead to increase productivity but also promote transfer of technology and knowledge spillover that bring about higher economic growth.
The determinant of economic growth varied across the eleven African countries considered in the paper.Own shock explained highest proportion of the variation in economic growth in eleven African countries (Morocco, Ethiopia, Zimbabwe, Kenya, Nigeria, Lesotho, South Africa, Egypt, Tunisia, Senegal and Tanzania).Out of these eleven countries studied, foreign direct investment explained the highest proportion in just three countries, Morocco, Ethiopia, and Zimbabwe.Except in Tunisia Tanzania and Kenya, where the degree of economic openness explained the substantial proportion of the output fluctuations, the variations in most of the countries were explained by factors beyond foreign direct investment and economic openness (Table 1).This result supports the existing finding on African economies (Rodrinez andRodrik 1998, Saibu 2004), that trade liberalization had not substantially impacted on the growth rate of the African economy.Though there was an upsurge in the capital flows in to the African economies but the inflow is not sufficient to kick start the economy to recovery in the Southern African countries, neither openness nor foreign direct investment explained any appreciable proportion in the variation in growth fluctuation thus, this implies that fluctuations in real economic growth in these countries should be seen beyond the external shock from the capital inflows or trade flows.

CONCLUSION
From the above result, some deductions can be made as regards economic growth dynamics in African countries.Policies that will improve the foreign direct investment and economic openness will not necessarily improve economic growth in the SADC area, while such policy might be effective in stimulating economic growth in the central and northern African countries.Therefore, trade policies, which encourage capital inflow and increase in volume of external trade might not necessary implied economic growth in developing countries especially African countries, rather the growth enhancing policy must be the one that promote domestic economy, strengthened capacity building and shock absorbing capacity that will allow them withstand the externally induced shocks from trade and capital flows.However, this result does not necessarily imply that trade and foreign direct investment are irrelevant in growth dynamics of African economy but they may be complementary to other factors beyond trade and FDI.

Table 1 .
Variance Decomposition of Economic Growth in Selected African Countries