Foreign Direct Investment and Economic Growth in Tanzania: The Granger-Causality Analysis

Transcription

1 Foreign Direct Investment and Economic Growth in Tanzania: The Granger-Causality Analysis Lema, Nicodemas Christopher Department of Economics and Statistics, The University of Dodoma, P.O. Box 395, Dodoma, Tanzania & Romanus Dimoso Faculty of Social Sciences, Mzumbe University, P.O. Box 5, Morogoro, Tanzania Abstract The main objective for this paper was to study the causal link between Foreign Direct Investment inflows and economic growth (GDP growth) for Tanzania. We also studied the direction of causality between the two variable based on the granger causality test. Annual time-series data covering the period between 1970 and 2007 was used. The study proves that there is a two-way causality (bilateral) between FDI inflows and Economic Growth. Key Words: Economic Growth (GDP growth), Foreign Direct Investment Inflows, Granger Causality and Tanzania Introduction In the face of inadequate resources to finance long-term development in Africa and with poverty reduction and other Millennium Development Goals (MDGs) looking increasingly difficult to achieve by 2015, the issue of attracting foreign direct investment (FDI) has assumed a prominent place in the strategies of economic renewal being advocated by policy makers at the national, regional and international levels (UNCTAD, 2005). 22

2 FDI according to Dunning (1970) involves equity ownership that is extended across national borders and that comes accompanied by a significant degree of managerial control. OECD (1999) benchmarks it and says that; FDI is established when the direct investor has acquired 10 percent or more of the ordinary shares or voting power of an enterprise abroad. It is the net inflows of investment to acquire a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of the investor. It is the sum of equity capital, reinvestment of earnings, other long-term capital and short-term capital as shown in the balance of payments. However, DeMello (1997) argues that the definition of a FDI should be broader and also include non-equity cooperation in which a foreign firm supplies a domestic firm with tangible as well as intangible assets. These assets come in the shape of collaboration concerning for example licensing, leasing, franchising, start-up and international production-sharing agreements. Joint ventures with a limited foreign equity participation, as well as R&D cooperation are also included in DeMello s broader definition of FDI. On the other hand Economic growth is a steady process by which the productive capacity of the economy is increased over time to bring about rising levels of national output and income. (Todaro M.P. and Smith S.C., 2008) According to Todaro and Smith in their book Economic Development, Economic growth consists of three components that are of prime importance in any society which are Capital accumulation which includes all new investments in land, physical equipment and human resources through improvements in health, education and job skills, Growth in population and hence eventual 23

3 growth in the labour force and Technological progress which broadly means new ways of accomplishing tasks. Economic growth is a term used to indicate an increase in total GDP, often measured as the rate of change of gross domestic product (GDP). It refers only to the quantity of goods and services produced but say nothing about the way in which they are produced. (Encyclopedia Britannica, 2008). This paper is organized as follows: Section 3.0 offers a Literature Review. Section 4.0 presents Data and Methodology to be applied in the analysis. Section 5.0 gives Estimation Results and Discussion arising from the results. Section 6.0 is the Conclusion and Policy Implication Literature Review 2.1. FDI and Economic Growth in Tanzania Tanzania is one of Africa s best performing countries in terms of GDP growth and attracting foreign direct investment. With its remarkable set of natural assets, long-term stable democracy (for the past 45 years) and strong macroeconomic performance, Tanzania has the potential to attract much greater FDI inflows. (MIGA, 2007) Tanzania remains one of the top destinations for FDI in Africa, and inflows have increased dramatically in the past 15 years. From a low level of USD 12 million in 1992, FDI inflows reached USD 517 million in 1999, due to mass privatization, and adjusted to USD 249 million in Based on 2003 figures, FDI is relatively diversified in mining (39 percent), followed by manufacturing (22 percent), tourism (13 percent) and agriculture (7 percent). Investment inflows are primarily from South Africa, the UK, Ghana, Australia, Canada, Kenya and the US. Most of this investment (36 percent) is located in or around Dar es Salaam. One hundred percent foreign ownership is allowed in almost every sector and for every type of business and land laws have 24

4 been amended to permit 99-year leases for foreign companies. The National Investment Act of 1997 and its amendments provide favorable conditions for investors, including incentives for tax relief and concessional tax rates. (MIGA, 2007) Foreign Direct Investment (FDI) has emerged as the most important source of external resource flows to developing countries over the 1990s and has become a significant part of capital formation in the developing countries despite their share in global distribution of FDI continuing to remain small or even declining. (Pradhan J.P. 2001). Despite many studies done on the direction of the causal link between Foreign Direct Investment and economic growth, the empirical evidence is not clear for country groups. Following the criticisms in recent studies such as the one done by Kholdy S. (1995), Ngowi H.P. (2001) and the like, of the traditional assumption of a one-way causal link from FDI to growth, new studies have also considered the possibility of a two-way (bidirectional) or non-existent causality among variables of interest. In other words, not only FDI can lead to economic growth (with either positive or negative impacts), but also economic (GDP) growth can affect the inflow of FDI or there could be no causal link at all. For example Frimpong J. M. and Oteng-Abayie E. F. (2008), conducted a study on bivariate causality between Foreign Direct Investment and Economic Growth in Ghana and found out that there is FDI-led growth in the post SAP period i.e. years between 1984 and However, from numerous existing studies, the causal link between FDI and economic growth as an empirical question seems to be dependent upon the set of conditions in the specific host country economy. 25

5 Chowdhury A. and Mavrotas G. (2005) suggested in their study that, individual country studies must be done to examine the causal links between FDI and economic growth since the relationship depends on the characteristics of a particular country. Thus this study is developed to find out the causal relationship between Foreign Direct Investment and Economic Growth, as to whether FDI is Economic Growth driven or Economic Growth is Foreign Direct Investment driven in Tanzanian context Conceptual Framework Figure 1.0. Direction of Causality Foreign Direct Investment Capital Formation Economic Growth Conducive Investment Climate Employment Generation Skills and management Technology Transfer Fiscal Revenue/Tax Revenue Good Infrastructure Law and Order Good governance Low inflation Low unemployment rates etc. Source: Own s Construction The conceptual framework on figure 1.0 shows the causal relationship that exists between Foreign Direct Investment and Economic Growth. From literatures and previous studies as postulated in the introductory part, Foreign Direct Investment may lead to economic growth and vice versa. 26

6 The Foreign Direct Investment comes with capital formation which brings in invisible financial resources to capital scarce countries of developing world; employment generation which is a result of Greenfield investment whereby foreign companies employ people thus generating new jobs in their new projects; Skills and management that comes with Foreign Direct Investment addresses the problem of idea gap that most of developing countries including Tanzania face today by bringing experts and setting up training facilities, technology transfer resulting from FDI raise efficiency; and fiscal revenue are the main factors that contribute to economic growth as far as FDI is concerned. On the other hand, economic growth creates conducive investment climate for Foreign Direct Investment to take place. This includes good infrastructure which make production easy; maintenance of law and order which provides for business practices to take place smoothly; good governance; Peace; low inflation which ensures stable prices; and low levels of unemployment all of which ma y contribute in attracting FDI to a country Theoretical Framework In order to test for causality between Foreign Direct Investment and Economic Growth the study employed the Granger causality test Granger Causality Test The Granger causality test used in time series analysis to examine the direction of causality between two economic series has been one of the main subjects of many econometrics studies for the past three decades. Recent studies have shown that the conventional F-test for determining joint significance of regression-derived parameters, used as a test of causality, is not valid if the 27

7 variables are non-stationary and the test statistic does not have a standard distribution (Gujarati, 1995). The causality between two economic variables will be tested using Granger causality test. Within a bivariate context, the Granger-type test states that; if a variable x Granger causes variable y, the mean square error (MSE) of a forecast of y based on the past values of both variables is lower than that of a forecast that uses only past values of y. (Granger 1969). Following Frimpong J. M. and Oteng-Abayie E. F. (2008), the Granger test will be implemented by running the following regression: Δyt = α + β iδyt-1 + iδxt-1 + µt. (1) And testing the joint hypothesis Ho: γ1=γ2=.γp=0; against H1: γ1 γ2.γp 0 Granger causality from the y variable to the coincident variable x is established if the null hypothesis of the asymptotic chi-square (X ²) test is rejected. A significant test statistic indicates that the x variable has predictive value for forecasting movements in y over and above the information contained in the latter s past Data and Methodology 3.1. Data The study used secondary data for annual series of real GDP growth which is a proxy for economic growth and Foreign Direct Investment net inflows as a % of GDP (FDI ratio) for Tanzania for the period 1970 to Data for GDP growth rate were obtained in the IMF website at from the World Economic Outlook database where as data for Foreign Direct Investment net Inflows (in US $ at current prices) and the exchange rate (local currency per US $) where obtained in the 28

8 world bank database at from the African Development Indicators (ADI) and thus FDI ratio was computed using the formula (FDIt/GDPt) multiplying by 100% The Model In order to test for FDI influence on Economic Growth, Economic Growth was regressed as a function of FDI and in order to test Economic Growth influence on FDI, FDI was regressed as a function of Economic Growth. The model is as follows; Where; lngdpt = α0 + 1ilnGDPt-1 + 1ilnFDIt-1 + µ1t..(2a) lnfdit = α0 + 2ilnFDIt-1 + 2ilnGDPt-1 + µ2t (2b) 1. lngdpt is the natural logarithm of real GDP growth rate (proxy for economic growth), 2. lnfdit is the natural logarithm of foreign direct investment net inflows as a % of GDP (FDI ratio), and 3. k is the optimal lag order, d is the maximal order of integration of the variables in the system and µ1t and µ2t are error terms that are assumed to be white noise. Equation (2a) postulates that current GDP is related to past values of itself as well as that of FDI and equation (2b) postulates a similar behavior for FDI. 29

9 The following four cases are distinguished from the two equations: 1. One direction causality from FDI to GDP exists if the estimated coefficients on FDI in equation (1) are statistically different from zero as a group (i.e. γ1i 0) and the set of estimated coefficients on GDP in equation (2) is not statistically different from zero (i.e. γ2i = 0). 2. Conversely, one direction causality from GDP to FDI exists if the estimated coefficients on FDI in equation (1) are not statistically different from zero as a group (i.e. γ1i = 0) and the set of estimated coefficients on GDP in equation (2) is statistically different from zero (i.e. γ2i 0). 3. Two way, or bilateral causality, is suggested when the sets of FDI and GDP coefficients are statistically significantly different from zero in both regressions. 4. Finally, independence i.e. neither FDI nor Economic growth causes one another, is suggested when the sets of FDI and GDP coefficients are not statistically significant in both the regressions. More generally, since the future cannot predict the past, if variable FDI (Granger) causes variable GDP, then changes in FDI should precede changes in GDP. Therefore, in a regression of GDP on other variables (including its own past values) if we include past or lagged values of FDI and it significantly improves the prediction of GDP, then we can say that FDI (Granger) causes GDP. A similar definition applies if GDP (Granger) causes FDI. 30

10 4.0. Estimation Results and Discussions The Estimation results for this study are presented in the following four stages. The first stage was to establish the maximal order of integration (d) for both real GDP growth and Foreign Direct Investment as a % of GDP in the model. This was done by the use of Augmented Dickey Fuller (ADF) unit root test. The second stage was to find out the optimal lag order (k) using Akaike information criteria (AIC), Schwartz Bayesian Information Criteria (SBIC), Likelihood Ratio (LR), Hannan Quinn Information Criteria (HQIC) and Forecast Prediction Error (FPE) Criteria. Thirdly we used the established maximal order of integration and the selected VAR length to estimate the augmented VAR model using SURE technique and finally we conducted the Granger Causality test in order to test for the pre-stated hypotheses. Augmented Dickey Fuller Unit Root Test Before applying the granger causality test in the augmented VAR(k+dmax), we established the maximal integration order (dmax) of the variables by carrying out Augmented Dickey-Fuller (ADF) unit root tests on the GDP growth and FDI as a % of GDP series in their natural loglevels. The results, reported in Table 1, indicate that real GDP growth is stationary at 5% degree of significance at lag 0 i.e. I (0). Similarly, results in Table 2 indicate that FDI as a % of GDP is stationary even at 1% of significance at lag 0 i.e. I (0). Therefore; both real GDP growth and FDI ratio are integrated of order 0. 31

11 Table 1 dfuller lngdp, trend lags (0) Interpolated Dickey-Fuller Number of observation = 37 Test Statistics 1% Critical Value 5% Critical Value 10% Critical Value Z(t) Mackinnon approximate p-value for Z(t) = Source: Stata Output Table 2 dfuller lnfdi, trend lags (0) Interpolated Dickey-Fuller Number of observation = 37 Test Statistics 1% Critical 5% Critical 10% Critical Value Value Value Z(t) Mackinnon approximate p-value for Z(t) = Source: Stata Output Optimal Lag Order (k) Next we employed Akaike information criteria (AIC), Schwartz Bayesian Information Criteria (SBIC), Likelihood Ratio (LR), Hannan Quinn Information Criteria (HQIC) and Forecast Prediction Error (FPE) Criteria to establish and select the optimal lag length of the VAR(k) Table 3 presents the output of the choice criteria for selecting the order of VAR model. 32

12 Table 3 Varsoc lngdp lnfdi Selection order criteria Sample: Number of Observation = 34 Lag LL LR df P FPE AIC HQIC SBIC * * * * * Endogenous: lngdpt lnfdit Exogenous: _ cons. Source: Stata Output On the basis of the results, the LR information criteria, the FPE information criteria and the AIC selects 2 lags while HQIC and SBIC selects 1 lag. The SBIC one (1) lag order of the VAR model is selected due to small sample of series in order to preserve some degree of freedom for the estimation. Model Estimation (Seemingly Unrelated Regression) Using the established maximal order of integration (dmax = 0) and the selected VAR length (k = 1), the following augmented VAR (1) model was estimated using the SUR technique: lngdpt = α0 + 1ilnGDPt-1 + 1ilnFDIt-1 + µ1t..(3a) lnfdit = α0 + 2ilnFDIt-1 + 2ilnGDPt-1 + µ2t (3b) 33

13 The following Table 4 and Table 5 give a summary of the output from the Seemingly Unrelated Regression: Table 4 SURE Equation Obs Parms RMSE R-sq Chi2 P lngdpt lnfdit Source: Stata Output Table 5 lngdp Coef. Std. Err z P> z [95% Conf. Interval] lnfdit lngdpt _cons lnfdi lngdpt lnfdit _cons Source: Stata Output Granger Causality-Test and Hypothesis Testing Finally we conducted the Granger Causality test using F-test and chi2 to verify if the coefficients γ1i and γ2i of the lagged variables are significantly different from zero in the respective equations 3(a) and 3(b). The results of the granger causality test are as follows: 1) lnfdit-1 lngdpt, lags (1) H0: lnfdit-1 does not Granger-cause lngdp, F(1, 35) = , Prob > F =.488, and; Chi2 (1) = , Prob > chi2 =

14 2) lngdpt-1 lnfdit, lags (1) H0: lngdpt-1 does not Granger-cause lnfdit, F( 1, 35) = , Prob > F =.276, and ; Chi2 (1) = , Prob > chi2 = From the results shown above it can be clearly seen that the computed F and chi2-values exceed the critical values of F and chi2 values in both cases therefore null hypothesis is rejected; i.e. overall, we find clear evidence of two way causality from FDI to GDP and from GDP to FDI Conclusion and Policy Implication The main objective of this paper was to assess the Foreign Direct Investment inflows as a % of GDP (FDI ratio) and economic growth (GDP growth rate) in Tanzania and test for the causal relationship between the two focusing mainly on the period between 1970 and The study employed granger causality test procedure, an innovative and more efficient econometric methodology to test the direction of causality between the FDI ratio and GDP growth proxy for economic growth. This study simulated discussion on FDI ratio and Economic growth by envisaging four possible types of relationship between the variables; (1) FDI-led growth i.e. to see if Foreign Direct Investment inflows as a % of GDP result into Economic Growth, (2) Growth-driven FDI i.e. finding out if a certain level of economic growth is essential for Foreign Direct Investment inflow, (3) See if there is a two way or bilateral causality between Foreign Direct Investment inflows and Economic Growth, and (4) Finding out if there is independence between Foreign Direct Investment inflows and Economic Growth i.e. neither FDI inflows nor Economic Growth cause one another. 35

15 Our empirical findings based on Granger causality test clearly suggest that there is a two way or bilateral causality between FDI inflows and Economic Growth. The fact that there is both growthdriven FDI and FDI-driven growth clearly shows that in Tanzania Foreign Direct Investment inflows have led to economic growth and similarly GDP growth in Tanzania has been attracting FDI inflows. Finally from our findings the traditional belief that the direction of causality runs from FDI to economic growth is confirmed in the case of Tanzania and furthermore the vice versa also. These findings lend support to the validity of policy guidelines which emphasize the importance of FDI for economic growth and stability in developing countries under the assumption of FDIled growth. On the other hand the findings emphasize on the importance of economic growth to attract FDI inflows. Understanding the bi-directional of causality between the two variables it is crucial to formulate policies that would encourage more private investors in Tanzania on one hand, and providing enabling economic environment by providing the required economic growth since they depend on each other. Future research in this area may analyze the causal link in a multivariate VAR system by taking into account other vital determinants of FDI inflows and GDP growth of which may lead to improvements of the results and even provide more sturdy conclusions. 36

16 References Chowdhury A. and Mavrotas G., (2005). FDI and Growth: A Causal Relationship. UNUWIDER Research Paper No. 2005/25, UNU-WIDER. DeMello L.R., Jr., (1997). Foreign direct investment in developing countries: A selective Survey. Journal of Development Studies, 34(1), Dunning J. H., (1970). Studies in International Investment. London: Allen & Unwin. Economics, 27, Encyclopedia Britannica Economic Growth. Encyclopedia Britannica 2006 Ultimate Reference suite DVD. Frimpong J.M. and Oteng Abiye E.F., (2008). Bivariate causality analysis between FDI inflows and Economic Growth in Ghana. International Research Journal of Finance and Economics. EuroJournals Publishing, Inc Granger C. W. J. (1969). Investigating Casual Relationship by Econometric Models and Cross Spectral Methods. Econometrica, 37, Gujarati D., (1995). Basic Econometrics 3rd Edition. McGraw-Hill, New York. Kholdy S., (1995). Causality between foreign investment and spillover efficiency. Applied Multilateral Investment Guarantee Agency (MIGA) (2007). Benchmarking FDI competitiveness: Snapshot Africa-Tanzania. World Bank Group. Ngowi H.P., (2001). Attracting New Foreign Direct Investment to Tanzania. Tanzanet Journal, Volume 1. OECD (Organisation for Economic Co-operation and Development), (1999). OECD benchmark definition on Foreign Direct Investment. Third Edition. Paris: OECD Publications. Pradhan, J. P., (2001). Foreign Direct Investment and Economic Growth: The Case of Developing Countries. Jawarhlal Nehru University, New Delhi. Todaro M.P. and Smith S.C., (2008). Economic Development 10 th Ed. Addison-Wesley Publishers, Edinburgh Gate, England. UNCTAD, (2005). Economic Development in Africa: Rethinking the Role of Foreign Direct Investment. UNCTAD/GDS/AFRICA/2005/1, United Nations, Geneva. 37