Description
Financial Reports on Direction of Causation between FDI and Economic Development in India during Post-reforms Era (1991-2010): A VAR Modeling Approach, Economic development generally refers to the sustained, concerted actions of policymakers and communities that promote the standard of living and economic health of a specific area. Economic development can also be referred to as the quantitative and qualitative changes in the economy.
FINANCIAL REPORTS ON DIRECTION OF CAUSATION BETWEEN FDI AND ECONOMIC DEVELOPMENT IN INDIA DURING POST-REFORMS ERA (1991-2010): A VAR MODELING APPROACH
Abstract: This paper empirically studies whether simultaneous equation modeling is possible between Foreign Direct Investment (FDI) and economic development (GDP) in India, that is, whether or not, both the variables had an endogenous relationship between them during the period (19912010), on the backdrop of a common knowledge that FDI causes economic development (FDIled growth) through development of the real sector and FDI inflow also comes in at a huge volume at the time of economic prosperity (Growth-led FDI). From the analysis of data during this period, we saw that each variable had an endogenous relationship between them, as confirmed by 'Hausman Test of Endogeneity'. Moreover, change in FDI Granger-caused change in economic development as well as change in economic development Granger-caused FDI inflow to have a cascading effect on each other, which is evident from 'VAR Granger Causality/Block Exogeneity Wald Tests', explicitly establishing bi-directional causality between FDI and economic development during the study period'.
Key Words: FDI, Economic Development, VAR Granger Causality, Endogeneity
1. Introduction: The role of Foreign Direct Investment (FDI) in the growth process of an economy has been a burning topic of debate in several countries, including India. FDI is a vital ingredient of the globalization efforts of the world economy. The growth of international production is driven by economic and technological forces. It is also driven by the ongoing liberalization of FDI and trade policies. One outstanding feature of the present-day world has been the circulation of private capital flow in the form of FDI in developing countries, especially since 1990s. Governments around the world, in both advanced and developing countries, have been attracting MNCs to come to respective countries with FDI. In this context, globalization offers an unparalleled opportunity for developing countries like India to attain quicker economic growth through trade and investment. In the 1970s, the world FDI has increased its importance by transferring technologies and establishing marketing and procuring networks for efficient production and sales internationally. FDI) has appeared as the most significant source of external resource flows to developing countries over the years and has become a significant part of capital formation in these countries, though their share in global distribution of FDI continuing to remain small or even declining. The effects of FDI in the host economy are usually believed to be through increase in the employment, augmenting productivity, boost in exports and amplification of pace of technology transfer. The effect of FDI on growth rate of output is generally believed to be constrained by the existence of diminishing returns of physical capital. Consequently, FDI could only put forth an effect on the level of output per capita, but not on the growth rate. In other words, it was unable to modify the growth of output in the long run. On the contrary, when there is economic development or prosperity, the rate of interest increases, which in turn increases the return on capital employed. This higher return on capital employed makes foreign direct investment profitable and attract more and more FDI inflow into the economy. So, it is also true that economic development causes FDI to flow into an economy.
2
2. Literature Review: Solow (1956) proposed that long-run growth can be enhanced through technological and population growth and moreover, if FDI positively influences technology, then it will be growth advancing. Somwaru and Makki (2004), indicated that FDI can be growth advancing if it results in increasing returns in production through spillover and technological transfers via diffusion processes. Yangruni Wu (1999,) emphasized that learning process through FDI has a role in the growth of a country. Charkovic and Levine (2002,) claimed that effect of FDI on growth is either insignificant or negative because FDI creates the crowding out effect on domestic capital. Petri and Plummer (1998), expressed confusion over whether FDI causes exports or exports cause FDI. Gray (1998), expressed confusion over the role of FDI i.e. whether FDI is market-seeking (substitute) FDI or efficiency-seeking (complement). Kjima (1973), analyzed whether FDI is trade-oriented or anti trade-oriented. Alguacil et al. (2002), Baharumshan and Thanoon (2006), Balasubramanyam et al. (1996 & 1999), Bende-Nabende and Ford (1998), Borensztein et al.(1998), Chakraborty and Basu (2002), De Mello (1997 & 1999), Liu et al. (2002) and Wang (2005), indicated that through such spillover effect as new technologies, capital formulation, the expansion of international trade and the development of human capital (labor skills and employment), FDI can stimulate the economic growth. However, Bende-Nabende et al. (2003), Carkovic and Levine (2005) and Bende Nabendem et al. (2003), pointed out that FDI can offset then economic growth and found that FDI in some countries had a negative relation with economic growth also. Balasubramanyam et. al. (1996), tested the hypothesis that exports promoting FDI in countries like India confer greater benefit than FDI in other sectors. Borensztein et al. (1998),examining the absorptive capacity of FDI recipient country, measured by stock of human capital required for technological progress, through 'capital deepening' associated with new capital goods brought into an economy by FDI, found out that adequate infrastructure is a pre requisite if the fructification of growth effect of FDI is required to be there in place. Bosworth and Collins (1999), taking a sample of fifty-eight developing countries during 1978-95, found out that an increase of a dollar in capital inflows is associated with an increase in domestic investment of about fifty cents. Pradeep Agrawal (2000), on economic impact of foreign direct investment in south Asia (India, Pakistan, Bangladesh, Sri Lanka and Nepal), found that that there existed complementarily and linkage effects between foreign and national investment and the impact of FDI inflows on GDP growth rate is negative prior to 1980, mildly positive for early eighties and strongly positive over the late eighties and early nineties. Brecher and Diaz-Alejandro (1977), gave us evidence that foreign capital can lower the economic growth by earning excessive profits in a country with severe trade distortions such as high tariffs. Maria Blomström teals (1994), found that FDI inflows had a significant positive effect on the average growth rate of per capita income for a sample of 78 developing and 23 developed countries, but
3
when the sample of developing countries was split between two groups based on level of per capita income, the effect of FDI on growth of lower income developing countries was not statistically significant although still with a positive sign. Borensztein, et al.( 1998), found that stock of human capital is pivotal for the effect of FDI on host country growth. Rajit Kumar Sahoo (2005), has pointed out that FDI has a direct and indirect impact on a certain particular sectors of the economy. Feenstra and Markusen (1994), asserted that FDI can lead to a higher growth by incorporating new inputs and techniques. Chakraborty and Basu (2002), examined the direction of causality between FDI and GDP for India between 1974 and 1996 and showed that FDI stimulates GDP and not the other way round. Liu, Burridge and Sinclair (2002), taking data for the period 1981-1997, predicted a bi-directional causal long term relationship between FDI, trade and economic growth in China. Chowdhury and Mavrotas (2005), examined the causality between growth and investment for Chile, Malaysia and Thailand for the period, 1969-2000 and pointed out that there is a uni-directional causality between GDP and FDI from GDP to FDI in case of Chile, while for Malaysia and Thailand, bi-directional causality exists. In this context, Athreye and Kapoor (2001), supported that growth-led FDI is more likely than FDI-led growth.
3. Motivation: Empirical researches suggest that FDI leads to economic development through capital formation, technology transfer, employment generation, labour productivity and infrastructure development. But, a plethora of empirical studies, which are not mutually exclusive, have been advanced to suggest that FDI flows into an economy when there is development or prosperity, establishing reverse causality between FDI and economic development. This empirical bi-directional causality motivated us to evince special to interest to look into the fact that which caused what between FDI and economic development during our study period 1991—2010.
4. Objective: To see, whether or not, FDI and Economic Development (GDP) in India had an endogenous relationship between them and which caused what (direction of causation) between FDI and economic development during the 20-year-period (1991-2010).
4
5. Methodology: FDI and GDP data, both of which have been collected from RBI Bulletin for 20 years (1991 to 2010), are as below. Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 FDI (Rs Crores) 375 965 1838 4126 7172 10015 13220 10358 9338 18406 29235 24367 19860 27188 39674 103367 140180 173741 179059 138462 GDP at market price (Rs Crores) 1503337 1,585,755 1,661,091 1,771,702 1,905,899 2,049,786 2,132,798 2,264,699 2,456,363 2,554,004 2,680,280 2,785,013 3,006,254 3,242,209 3,544,348 3,812,974 4,253,184 4,462,967 4,780,179 5,236,823
5
GDP and FDI, both being exponential functions, have been transformed into their logarithmic forms. But, since GDP and FDI are both time series data, in order to see the relationship between them first we have to check whether both the series are stationary or not. This can be checked with the help of simple graphs as well as Correlogram, as shown below;
GDP at market price (Rs Crores)
5,500,000 5,000,000 4,500,000 4,000,000 3,500,000 3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 2 4 6 8 10 12 14 16 18 20
6
FDI(Rs Crores)
200,000
160,000
120,000
80,000
40,000
0 2 4 6 8 10 12 14 16 18 20
Correlogram of GDP Included observations: 20 Autocorrelation . |******| . |***** | . |**** | . |*** | . |**. | . |* . | .|. | . *| . | . *| . | .**| . | .**| . | .**| . | Partial Correlation . |******| .|. | . *| . | . *| . | .|. | .|. | .|. | .|. | .|. | . *| . | . *| . | . *| . | 1 2 3 4 5 6 7 8 9 10 11 12 AC 0.826 0.669 0.516 0.360 0.231 0.114 0.012 -0.076 -0.145 -0.213 -0.273 -0.331 PAC 0.826 -0.044 -0.078 -0.109 -0.032 -0.061 -0.055 -0.061 -0.042 -0.088 -0.081 -0.100 Q-Stat 15.806 26.738 33.629 37.200 38.760 39.165 39.170 39.381 40.228 42.221 45.868 51.894 Prob 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
7
Correlogram of FDI Included observations: 20 Autocorrelation . |******| . |**** | . |*** | . |* . | .|. | .|. | .|. | . *| . | . *| . | . *| . | .**| . | .**| . | Partial Correlation . |******| ****| . | .|. | .|. | .|. | .|. | .**| . | .|. | .|. | .|. | .|. | .**| . | 1 2 3 4 5 6 7 8 9 10 11 12 AC 0.866 0.618 0.354 0.145 0.019 -0.025 -0.057 -0.093 -0.141 -0.185 -0.206 -0.226 PAC 0.866 -0.526 -0.039 0.064 0.043 0.048 -0.236 0.003 -0.054 -0.006 0.011 -0.216 Q-Stat 17.362 26.699 29.942 30.519 30.530 30.549 30.659 30.976 31.768 33.272 35.354 38.165 Prob 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
From the graphs as well as Correlograms (where we have tested, with the help of Q-statistic, the joint significance of autocorrelation up to 12 lag order, since data is annual), we can see that both the series are non-stationary. So, both of them are to be made stationary first so that they don't form a spurious relationship between them. For checking stationarity statistically, we go in for Unit Root Test and with the help of 'Augmented Dickey Fuller Test', we check stationarity in the level first including an intercept in the equation, then including trend for the purpose of de- trending and at last taking 1st as well as 2nd differencing. Here, both the series, through 'Augmented Dickey Fuller Test', become stationary after 2nd differencing, as shown below;
Null Hypothesis: D(FDI,2) has a unit root Exogenous: Constant Lag Length: 3 (Automatic based on SIC, MAXLAG=3) t-Statistic Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level -3.699263 -4.121990 -3.144920 -2.713751 Prob.* 0.0201
Null Hypothesis: D(GDP,2) has a unit root Exogenous: Constant Lag Length: 0 (Automatic based on SIC, MAXLAG=3) t-Statistic Prob.*
8
Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level
-6.288777 -3.959148 -3.081002 -2.681330
0.0002
Moreover, since FDI causes economic development through development of the real sector and FDI inflow also comes in at a huge volume at the time of economic prosperity, so here, the model may be a simultaneous-equation regression model of the form; lnGDP2 =? lnFDI 2=?
1
+?
1*lnFDI2
+ ut1 + ut2
2
+?
2*lnGDP2
Simultaneous equation relationship between these two variables, if not considered, may lead to simultaneity bias. Whether or not, these two variables can be treated as endogenous, that can be tested through 'Hausman test of Endogeneity'. Since, it's a two-variable simultaneous equation system, so even if the system doesn't suffer from identification problem on account of omitted variables as instruments, we should use 2SLS (Two-Stage Least Squares) method for finding the fitted values of GDP as well as FDI from the reduced form equations, for the Hausman test to be applied. Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 GDP2_fit 15323.2 12916.2 14313.2 16356.6 15155.2 28825.4 12008.3 -5525.41 12180.5 FDI2_fit 428.579 -3700 -2560.66 -1206.21 5671.84 -5026.97 -6086.87 8902.81 -3052.79
9
2002 2003 2004 2005 2006 2007 2008 2009 2010
49301.9 15157.3 -9240.1 4957.37 -92957.7 73080.5 22839.8 75978.7 113555
1838.11 -11617.9 -1695.9 -6712.74 3004.84 -16986.2 22198.3 -10732.9 -13852.3
Dependent Variable: GDP2 Method: Two-Stage Least Squares Instrumental Variables: C, FDI2, FDI2_fit Variable C FDI2 FDI2_FIT Coefficient -2684.964 -3.19E-06 -10.25944 Std. Error 0.030621 1.72E-06 3.78E-06 t-Statistic -87684.74 -1.850498 -2712917 Prob. 0.0000 0.0840 0.0000
Dependent Variable: FDI2 Method: Two-Stage Least Squares Instrumental Variables: C, GDP2, GDP2_fit Variable Coefficient Std. Error C FDI2 FDI2_FIT 7489.431 -8.32E-08 -0.470295 0.003924 4.36E-08 9.57E-08
t-Statistic 1908598 -1.911350 -4915794
Prob. 0.0000 0.0753 0.0000
10
Here, since the instruments are variables in the reduced form equation, so the method of instrumental variables (IV) is equivalent to 2SLS. From the output, we see that p-values of both GDP2_fit as well as FDI2_fit are significant, which implies that both of them can be considered as endogenous. This clearly shows that both FDI as well as GDP had an impact on the each other. Now, to check which variable had a chronological precedence, a VAR Model (Vector Autoregressive Model) consisting of the variables and their lags, is considered for estimation of coefficients of the variables with lags. In this case, VAR model is of the form; y1t =? y2t =?
10
+? +?
11
y1t-1 + ??.+? y2t-1 + ??.+?
1k y1t-k +? 2k y2t-k +?
y2t-1+ ?..+? y1t-1+ ?..+?
1k y2t-k + u1t 2k y1t-k + u2t
20
21
It is a case of simple bi-variate VAR model, which is estimated with VAR specification and optimum lag-length is decided with the help of optimum information criteria. But before that we, estimate the model, taking lag-length of order (2). Vector Auto regression Estimates: GDP2 GDP2(-1) -0.556763 (0.29930) [ -1.86020] -0.645514 (0.28888) [-2.23452] 2.077743 (1.18107) [ 1.75921] FDI2 0.180439 (0.08207) [ 2.19859] 0.183923 (0.07921) [2.32186] 0.060226 (0.32386) [-0.18597]
GDP2(-2)
FDI2(-1)
FDI2(-2)
-2.577883 -0.089916 (1.38290) (0.37920) [ -1.86411] [ -0.23712] 39917.27 (18278.7) [ 2.18381] -6670.120 (5012.13) [-1.33080]
C
11
Now, we go in for VAR lag order selection, whereby we can cross-check that whether lag-order selection is optimum or not through such information criteria as Akaike information criterion (AIC), Schwarz Bayesian information criterion (SIC), Hannan-Quinn Information Criteria (HQ) etc, amongst which AIC is least punitive. VAR Lag Order Selection Criteria: Lag 0 1 2 3 LogL -362.6576 -359.3196 -352.6006 -351.4040 LR NA* 5.340815 8.958751 1.276388 FPE 4.48e+18 4.94e+18 3.59e+18* 5.78e+18 AIC 48.62102 48.70928 48.34674* 48.72053 SC 48.71543* 48.99250 48.81878 49.38137 HQ 48.62001 48.70627 48.34171* 48.71349
Though, we went in for 2SLS estimation with lag order (2), without knowing the actual lag order for efficiency of coefficients, but from the lag-order selection criteria, we come to know that AIC is optimum at lag-order (2). So, the coefficients of the variables of VAR estimation with lagorder (2) are efficient. Next, Granger Causality Test, which doesn't necessarily signify that one variable causes the other but only establishes chronological precedence, is applied to check for direction of causation between GDP and FDI. From the output, given below, we see that FDI is Granger-causing GDP as well as GDP is Granger-causing FDI, establishing a bi-directional causality between them. This finding can be strengthened by the plots of 'Impulse Responses' and 'Variance Decomposition', as shown below. Though ordering of the variables can have impact on the impulse response as well as variance decomposition (Ordering for Cholesky in Cholesky Decomposition), but Granger Causality remains the same.
VAR Granger Causality/Block Exogeneity Wald Tests: Dependent variable: GDP2 Excluded FDI2 All Chi-sq 7.089566 7.089566 df 2 2 Prob. 0.0289 0.0289
Dependent variable: FDI2 Excluded GDP2 All Chi-sq 6.491436 6.491436 df 2 2 Prob. 0.0389 0.0389
12
Variance Decomposition of GDP2
100
80
60
40
20
0 1 2 3 4 5 GDP2 6 7 FDI2 8 9 10
Variance Decomposition of FDI2
100
80
60
40
20
0 1 2 3 4 5 GDP2 6 7 FDI2 8 9 10
13
6. Conclusion: From the analysis of data during the study period, we come to the conclusion that simultaneous equation modeling is possible between FDI and economic development, that is both the variables had an endogenous relationship between them during this period (1991-2010), implying that each had impact on the other, as confirmed by 'Hausman Test of Endogeneity'. Moreover, change in FDI Granger-caused change in economic development as well as change in economic development Granger-caused FDI inflow to have a cascading effect on each other, which is evident from 'VAR Granger Causality/Block Exogeneity Wald Tests', explicitly establishing a bidirectional causality between FDI and economic development during the study period'. This interdependence between FDI and economic development is possible only when an economy is growing or developing. Therefore, there is no gainsaying the fact that as evident from statistical significance of our study output, development is seen to have taken place in India during the period 1991-2010. References: 1. Aitken, B., and A. Harrison (1999). 'Do Domestic Firms Benefit from Direct Foreign Investment? Evidence from Venezuela'. American Economic Review, 89 (3): 605-18. 2. Anitha & Dr. K. Maran(2011), Recent trends in foreign direct investment, International Journal of Research in Commerce & Management, vol. no. 2 (2011), issue no. 8 (August), pp 28- 34. 3. Alguacil, M. T. and V. Orts (2002), "A Multivariate Co-integrated Model Testing for Temporal Causality between Exports and Outward Foreign Investment: the Spanish case" Applied Economics, 34, 119-32. 4. Alguacil, M. T., A. Cuadros and V. Orts (2002), "Foreign Direct Investment, Exports and Domestic Performance in Mexico:A Causality Analysis," Economic Letters, 77, 371-76. 5. Bende-Nabendem, A., J. L. Ford, B. Santoso and S. Sen (2003), "The Interaction between FDI, Output and the Spillover Variables: Co-integration and VAR Analysis for APEC, 1965- 1999," Applied Economics Letters, 10, 165-72. 6. Balasubramanyan, V., N. M.A. Salisu and D. Sapsford. (1996), "Foreign Direct Investment and Growth in EP and IS Countries", Economic Journal, 106: 92-105. 7. Balasubramanyam, V. N., M. Salisu and D. Sapsford (1999). "Foreign Direct Investment as an Engine of Growth," The Journal of International Trade & Economic Development, 8,1, 27-40. 8. Blomstrom, M., and A. Kokko (1998). 'Multinational Corporations and Spillovers'. Journal of Economic Surveys, 12: 247-77. 9. Brecher, R.A. and Diaz-Alejandro, C.F. (1977) "Tariffs, Foreign Capital and Immiserizing Growth", Journal of International Economics, Vol.7, pp.317-22.
14
10. Bende-Nabendem, A. and J. L. Ford (1998), "FDI, Policy Adjustment and Endogenous Growth: Multiplier Effects from a Small Dynamic Model for Taiwan, 1959-1995," World Development, Vol. 26, pp. 1315-1330. 11. Borensztein, E., J. De Gregorio, J-W. Lee (1998), "How does Foreign Direct Investment Affect Economic Growth," Journal of International Economics, 45, 115-35. 12. Christopoulos, D. K. and Tsionas, E. G. (2004), Financial development and economic growth: evidence from panel unit root and co-integration tests. Journal of Development Economics, 73(4), 55- 74. 13. Chakraborty, C. and P. Basu (2002), "Foreign Direct Investment and Growth in India: a Cointegrating Approach," Applied Economics, 34, 1061-73. 14. Duasa, J.(2007): Malaysian Foreign Direct Investment and Growth: Does Stability Matter? , The Journal of Economic Cooperation, Volume 28,No 2,pp.83-98. 15. De Mello, Jr. (1997), "Foreign Direct Investment in Developing Countries and Growth: A Selective Survey," Journal of Development Studies, 34, 1, 1-34. 16. De Mello, Jr. (1999), "Foreign Direct Investment-led Growth: Evidence from Time Series and Panel Data," Oxford Economic Papers, 51, 133-51. 17. Dickey, D.A and W.A.Fuller (1979), Distribution of estimators of Autoregressive Time series with a Unit Root, Journal of the American Statistical Association,74,427-31. 18. Findlay, R. (1978), "Relative Backwardness, Direct Foreign Investment and the Transfer of Technology: A Simple Dynamic Model." Quarterly Journal of Economics, 92: 1-16. 19. Gray, H. P. (1998): International trade and foreign direct investment: the interface. In: J. H. Dunning (Ed.), Globalization, trade and foreign direct investment, Oxford: Elsevier, pp. 19-27. 20. Granger, C.W.J (1986), Developments in the Study of Co-integrated Economic Variables. Oxford Bulletin of Economics and Statistics, nr. 48. 21. Hermes, N., R. Lensink (2003), Foreign Direct Investment, Financial Development and Economic Growth, The Journal of Development Studies, 142-163. 22. Hsiao, F.S.T and Hsiao, M. C.W.(2006), 'FDI, exports, and GDP in east and southeast Asia Panel data versus time series causality analyses', Journal of Asian Economics,17(2006):10821106. 23. Jayashree Bose(2007), "FDI Inflows in India and China - A Sectoral Experiences", ICFAI University Press, Hyderabad. 24. Johansen, S., Juselius, K.(1992), Structural hypotheses in a multivariate co-integration analysis of the PPP and UIP for UK. J. Economics. 53,pp 211-44. 25. Khawar, M. (2005). 'Foreign Direct Investment and Economic Growth: A Cross Country Analysis'. Global Economy Journal, 5 (1): 1-11. 26. Liu, Xiaohui, P. Burridge and P. J. Sinclair (2002), "Relationships between Economic Growth, Foreign Direct Investment and Trade: Evidence from China," Applied Economics, 34, 1433-40.
15
27. Makki,S.S and Somwaru,A(2004),Impact of Foreign Direct Investment and trade on Economic Growth :Evidence from Developing Countries, American Journal of Agricultural Economics,vol.86(3),pp795-801. 28. Sim, C. (1992), Money, Income and Causality. American Economic Review, 62: 251-266.
16
doc_274844808.docx
Financial Reports on Direction of Causation between FDI and Economic Development in India during Post-reforms Era (1991-2010): A VAR Modeling Approach, Economic development generally refers to the sustained, concerted actions of policymakers and communities that promote the standard of living and economic health of a specific area. Economic development can also be referred to as the quantitative and qualitative changes in the economy.
FINANCIAL REPORTS ON DIRECTION OF CAUSATION BETWEEN FDI AND ECONOMIC DEVELOPMENT IN INDIA DURING POST-REFORMS ERA (1991-2010): A VAR MODELING APPROACH
Abstract: This paper empirically studies whether simultaneous equation modeling is possible between Foreign Direct Investment (FDI) and economic development (GDP) in India, that is, whether or not, both the variables had an endogenous relationship between them during the period (19912010), on the backdrop of a common knowledge that FDI causes economic development (FDIled growth) through development of the real sector and FDI inflow also comes in at a huge volume at the time of economic prosperity (Growth-led FDI). From the analysis of data during this period, we saw that each variable had an endogenous relationship between them, as confirmed by 'Hausman Test of Endogeneity'. Moreover, change in FDI Granger-caused change in economic development as well as change in economic development Granger-caused FDI inflow to have a cascading effect on each other, which is evident from 'VAR Granger Causality/Block Exogeneity Wald Tests', explicitly establishing bi-directional causality between FDI and economic development during the study period'.
Key Words: FDI, Economic Development, VAR Granger Causality, Endogeneity
1. Introduction: The role of Foreign Direct Investment (FDI) in the growth process of an economy has been a burning topic of debate in several countries, including India. FDI is a vital ingredient of the globalization efforts of the world economy. The growth of international production is driven by economic and technological forces. It is also driven by the ongoing liberalization of FDI and trade policies. One outstanding feature of the present-day world has been the circulation of private capital flow in the form of FDI in developing countries, especially since 1990s. Governments around the world, in both advanced and developing countries, have been attracting MNCs to come to respective countries with FDI. In this context, globalization offers an unparalleled opportunity for developing countries like India to attain quicker economic growth through trade and investment. In the 1970s, the world FDI has increased its importance by transferring technologies and establishing marketing and procuring networks for efficient production and sales internationally. FDI) has appeared as the most significant source of external resource flows to developing countries over the years and has become a significant part of capital formation in these countries, though their share in global distribution of FDI continuing to remain small or even declining. The effects of FDI in the host economy are usually believed to be through increase in the employment, augmenting productivity, boost in exports and amplification of pace of technology transfer. The effect of FDI on growth rate of output is generally believed to be constrained by the existence of diminishing returns of physical capital. Consequently, FDI could only put forth an effect on the level of output per capita, but not on the growth rate. In other words, it was unable to modify the growth of output in the long run. On the contrary, when there is economic development or prosperity, the rate of interest increases, which in turn increases the return on capital employed. This higher return on capital employed makes foreign direct investment profitable and attract more and more FDI inflow into the economy. So, it is also true that economic development causes FDI to flow into an economy.
2
2. Literature Review: Solow (1956) proposed that long-run growth can be enhanced through technological and population growth and moreover, if FDI positively influences technology, then it will be growth advancing. Somwaru and Makki (2004), indicated that FDI can be growth advancing if it results in increasing returns in production through spillover and technological transfers via diffusion processes. Yangruni Wu (1999,) emphasized that learning process through FDI has a role in the growth of a country. Charkovic and Levine (2002,) claimed that effect of FDI on growth is either insignificant or negative because FDI creates the crowding out effect on domestic capital. Petri and Plummer (1998), expressed confusion over whether FDI causes exports or exports cause FDI. Gray (1998), expressed confusion over the role of FDI i.e. whether FDI is market-seeking (substitute) FDI or efficiency-seeking (complement). Kjima (1973), analyzed whether FDI is trade-oriented or anti trade-oriented. Alguacil et al. (2002), Baharumshan and Thanoon (2006), Balasubramanyam et al. (1996 & 1999), Bende-Nabende and Ford (1998), Borensztein et al.(1998), Chakraborty and Basu (2002), De Mello (1997 & 1999), Liu et al. (2002) and Wang (2005), indicated that through such spillover effect as new technologies, capital formulation, the expansion of international trade and the development of human capital (labor skills and employment), FDI can stimulate the economic growth. However, Bende-Nabende et al. (2003), Carkovic and Levine (2005) and Bende Nabendem et al. (2003), pointed out that FDI can offset then economic growth and found that FDI in some countries had a negative relation with economic growth also. Balasubramanyam et. al. (1996), tested the hypothesis that exports promoting FDI in countries like India confer greater benefit than FDI in other sectors. Borensztein et al. (1998),examining the absorptive capacity of FDI recipient country, measured by stock of human capital required for technological progress, through 'capital deepening' associated with new capital goods brought into an economy by FDI, found out that adequate infrastructure is a pre requisite if the fructification of growth effect of FDI is required to be there in place. Bosworth and Collins (1999), taking a sample of fifty-eight developing countries during 1978-95, found out that an increase of a dollar in capital inflows is associated with an increase in domestic investment of about fifty cents. Pradeep Agrawal (2000), on economic impact of foreign direct investment in south Asia (India, Pakistan, Bangladesh, Sri Lanka and Nepal), found that that there existed complementarily and linkage effects between foreign and national investment and the impact of FDI inflows on GDP growth rate is negative prior to 1980, mildly positive for early eighties and strongly positive over the late eighties and early nineties. Brecher and Diaz-Alejandro (1977), gave us evidence that foreign capital can lower the economic growth by earning excessive profits in a country with severe trade distortions such as high tariffs. Maria Blomström teals (1994), found that FDI inflows had a significant positive effect on the average growth rate of per capita income for a sample of 78 developing and 23 developed countries, but
3
when the sample of developing countries was split between two groups based on level of per capita income, the effect of FDI on growth of lower income developing countries was not statistically significant although still with a positive sign. Borensztein, et al.( 1998), found that stock of human capital is pivotal for the effect of FDI on host country growth. Rajit Kumar Sahoo (2005), has pointed out that FDI has a direct and indirect impact on a certain particular sectors of the economy. Feenstra and Markusen (1994), asserted that FDI can lead to a higher growth by incorporating new inputs and techniques. Chakraborty and Basu (2002), examined the direction of causality between FDI and GDP for India between 1974 and 1996 and showed that FDI stimulates GDP and not the other way round. Liu, Burridge and Sinclair (2002), taking data for the period 1981-1997, predicted a bi-directional causal long term relationship between FDI, trade and economic growth in China. Chowdhury and Mavrotas (2005), examined the causality between growth and investment for Chile, Malaysia and Thailand for the period, 1969-2000 and pointed out that there is a uni-directional causality between GDP and FDI from GDP to FDI in case of Chile, while for Malaysia and Thailand, bi-directional causality exists. In this context, Athreye and Kapoor (2001), supported that growth-led FDI is more likely than FDI-led growth.
3. Motivation: Empirical researches suggest that FDI leads to economic development through capital formation, technology transfer, employment generation, labour productivity and infrastructure development. But, a plethora of empirical studies, which are not mutually exclusive, have been advanced to suggest that FDI flows into an economy when there is development or prosperity, establishing reverse causality between FDI and economic development. This empirical bi-directional causality motivated us to evince special to interest to look into the fact that which caused what between FDI and economic development during our study period 1991—2010.
4. Objective: To see, whether or not, FDI and Economic Development (GDP) in India had an endogenous relationship between them and which caused what (direction of causation) between FDI and economic development during the 20-year-period (1991-2010).
4
5. Methodology: FDI and GDP data, both of which have been collected from RBI Bulletin for 20 years (1991 to 2010), are as below. Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 FDI (Rs Crores) 375 965 1838 4126 7172 10015 13220 10358 9338 18406 29235 24367 19860 27188 39674 103367 140180 173741 179059 138462 GDP at market price (Rs Crores) 1503337 1,585,755 1,661,091 1,771,702 1,905,899 2,049,786 2,132,798 2,264,699 2,456,363 2,554,004 2,680,280 2,785,013 3,006,254 3,242,209 3,544,348 3,812,974 4,253,184 4,462,967 4,780,179 5,236,823
5
GDP and FDI, both being exponential functions, have been transformed into their logarithmic forms. But, since GDP and FDI are both time series data, in order to see the relationship between them first we have to check whether both the series are stationary or not. This can be checked with the help of simple graphs as well as Correlogram, as shown below;
GDP at market price (Rs Crores)
5,500,000 5,000,000 4,500,000 4,000,000 3,500,000 3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 2 4 6 8 10 12 14 16 18 20
6
FDI(Rs Crores)
200,000
160,000
120,000
80,000
40,000
0 2 4 6 8 10 12 14 16 18 20
Correlogram of GDP Included observations: 20 Autocorrelation . |******| . |***** | . |**** | . |*** | . |**. | . |* . | .|. | . *| . | . *| . | .**| . | .**| . | .**| . | Partial Correlation . |******| .|. | . *| . | . *| . | .|. | .|. | .|. | .|. | .|. | . *| . | . *| . | . *| . | 1 2 3 4 5 6 7 8 9 10 11 12 AC 0.826 0.669 0.516 0.360 0.231 0.114 0.012 -0.076 -0.145 -0.213 -0.273 -0.331 PAC 0.826 -0.044 -0.078 -0.109 -0.032 -0.061 -0.055 -0.061 -0.042 -0.088 -0.081 -0.100 Q-Stat 15.806 26.738 33.629 37.200 38.760 39.165 39.170 39.381 40.228 42.221 45.868 51.894 Prob 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
7
Correlogram of FDI Included observations: 20 Autocorrelation . |******| . |**** | . |*** | . |* . | .|. | .|. | .|. | . *| . | . *| . | . *| . | .**| . | .**| . | Partial Correlation . |******| ****| . | .|. | .|. | .|. | .|. | .**| . | .|. | .|. | .|. | .|. | .**| . | 1 2 3 4 5 6 7 8 9 10 11 12 AC 0.866 0.618 0.354 0.145 0.019 -0.025 -0.057 -0.093 -0.141 -0.185 -0.206 -0.226 PAC 0.866 -0.526 -0.039 0.064 0.043 0.048 -0.236 0.003 -0.054 -0.006 0.011 -0.216 Q-Stat 17.362 26.699 29.942 30.519 30.530 30.549 30.659 30.976 31.768 33.272 35.354 38.165 Prob 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
From the graphs as well as Correlograms (where we have tested, with the help of Q-statistic, the joint significance of autocorrelation up to 12 lag order, since data is annual), we can see that both the series are non-stationary. So, both of them are to be made stationary first so that they don't form a spurious relationship between them. For checking stationarity statistically, we go in for Unit Root Test and with the help of 'Augmented Dickey Fuller Test', we check stationarity in the level first including an intercept in the equation, then including trend for the purpose of de- trending and at last taking 1st as well as 2nd differencing. Here, both the series, through 'Augmented Dickey Fuller Test', become stationary after 2nd differencing, as shown below;
Null Hypothesis: D(FDI,2) has a unit root Exogenous: Constant Lag Length: 3 (Automatic based on SIC, MAXLAG=3) t-Statistic Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level -3.699263 -4.121990 -3.144920 -2.713751 Prob.* 0.0201
Null Hypothesis: D(GDP,2) has a unit root Exogenous: Constant Lag Length: 0 (Automatic based on SIC, MAXLAG=3) t-Statistic Prob.*
8
Augmented Dickey-Fuller test statistic Test critical values: 1% level 5% level 10% level
-6.288777 -3.959148 -3.081002 -2.681330
0.0002
Moreover, since FDI causes economic development through development of the real sector and FDI inflow also comes in at a huge volume at the time of economic prosperity, so here, the model may be a simultaneous-equation regression model of the form; lnGDP2 =? lnFDI 2=?
1
+?
1*lnFDI2
+ ut1 + ut2
2
+?
2*lnGDP2
Simultaneous equation relationship between these two variables, if not considered, may lead to simultaneity bias. Whether or not, these two variables can be treated as endogenous, that can be tested through 'Hausman test of Endogeneity'. Since, it's a two-variable simultaneous equation system, so even if the system doesn't suffer from identification problem on account of omitted variables as instruments, we should use 2SLS (Two-Stage Least Squares) method for finding the fitted values of GDP as well as FDI from the reduced form equations, for the Hausman test to be applied. Year 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 GDP2_fit 15323.2 12916.2 14313.2 16356.6 15155.2 28825.4 12008.3 -5525.41 12180.5 FDI2_fit 428.579 -3700 -2560.66 -1206.21 5671.84 -5026.97 -6086.87 8902.81 -3052.79
9
2002 2003 2004 2005 2006 2007 2008 2009 2010
49301.9 15157.3 -9240.1 4957.37 -92957.7 73080.5 22839.8 75978.7 113555
1838.11 -11617.9 -1695.9 -6712.74 3004.84 -16986.2 22198.3 -10732.9 -13852.3
Dependent Variable: GDP2 Method: Two-Stage Least Squares Instrumental Variables: C, FDI2, FDI2_fit Variable C FDI2 FDI2_FIT Coefficient -2684.964 -3.19E-06 -10.25944 Std. Error 0.030621 1.72E-06 3.78E-06 t-Statistic -87684.74 -1.850498 -2712917 Prob. 0.0000 0.0840 0.0000
Dependent Variable: FDI2 Method: Two-Stage Least Squares Instrumental Variables: C, GDP2, GDP2_fit Variable Coefficient Std. Error C FDI2 FDI2_FIT 7489.431 -8.32E-08 -0.470295 0.003924 4.36E-08 9.57E-08
t-Statistic 1908598 -1.911350 -4915794
Prob. 0.0000 0.0753 0.0000
10
Here, since the instruments are variables in the reduced form equation, so the method of instrumental variables (IV) is equivalent to 2SLS. From the output, we see that p-values of both GDP2_fit as well as FDI2_fit are significant, which implies that both of them can be considered as endogenous. This clearly shows that both FDI as well as GDP had an impact on the each other. Now, to check which variable had a chronological precedence, a VAR Model (Vector Autoregressive Model) consisting of the variables and their lags, is considered for estimation of coefficients of the variables with lags. In this case, VAR model is of the form; y1t =? y2t =?
10
+? +?
11
y1t-1 + ??.+? y2t-1 + ??.+?
1k y1t-k +? 2k y2t-k +?
y2t-1+ ?..+? y1t-1+ ?..+?
1k y2t-k + u1t 2k y1t-k + u2t
20
21
It is a case of simple bi-variate VAR model, which is estimated with VAR specification and optimum lag-length is decided with the help of optimum information criteria. But before that we, estimate the model, taking lag-length of order (2). Vector Auto regression Estimates: GDP2 GDP2(-1) -0.556763 (0.29930) [ -1.86020] -0.645514 (0.28888) [-2.23452] 2.077743 (1.18107) [ 1.75921] FDI2 0.180439 (0.08207) [ 2.19859] 0.183923 (0.07921) [2.32186] 0.060226 (0.32386) [-0.18597]
GDP2(-2)
FDI2(-1)
FDI2(-2)
-2.577883 -0.089916 (1.38290) (0.37920) [ -1.86411] [ -0.23712] 39917.27 (18278.7) [ 2.18381] -6670.120 (5012.13) [-1.33080]
C
11
Now, we go in for VAR lag order selection, whereby we can cross-check that whether lag-order selection is optimum or not through such information criteria as Akaike information criterion (AIC), Schwarz Bayesian information criterion (SIC), Hannan-Quinn Information Criteria (HQ) etc, amongst which AIC is least punitive. VAR Lag Order Selection Criteria: Lag 0 1 2 3 LogL -362.6576 -359.3196 -352.6006 -351.4040 LR NA* 5.340815 8.958751 1.276388 FPE 4.48e+18 4.94e+18 3.59e+18* 5.78e+18 AIC 48.62102 48.70928 48.34674* 48.72053 SC 48.71543* 48.99250 48.81878 49.38137 HQ 48.62001 48.70627 48.34171* 48.71349
Though, we went in for 2SLS estimation with lag order (2), without knowing the actual lag order for efficiency of coefficients, but from the lag-order selection criteria, we come to know that AIC is optimum at lag-order (2). So, the coefficients of the variables of VAR estimation with lagorder (2) are efficient. Next, Granger Causality Test, which doesn't necessarily signify that one variable causes the other but only establishes chronological precedence, is applied to check for direction of causation between GDP and FDI. From the output, given below, we see that FDI is Granger-causing GDP as well as GDP is Granger-causing FDI, establishing a bi-directional causality between them. This finding can be strengthened by the plots of 'Impulse Responses' and 'Variance Decomposition', as shown below. Though ordering of the variables can have impact on the impulse response as well as variance decomposition (Ordering for Cholesky in Cholesky Decomposition), but Granger Causality remains the same.
VAR Granger Causality/Block Exogeneity Wald Tests: Dependent variable: GDP2 Excluded FDI2 All Chi-sq 7.089566 7.089566 df 2 2 Prob. 0.0289 0.0289
Dependent variable: FDI2 Excluded GDP2 All Chi-sq 6.491436 6.491436 df 2 2 Prob. 0.0389 0.0389
12
Variance Decomposition of GDP2
100
80
60
40
20
0 1 2 3 4 5 GDP2 6 7 FDI2 8 9 10
Variance Decomposition of FDI2
100
80
60
40
20
0 1 2 3 4 5 GDP2 6 7 FDI2 8 9 10
13
6. Conclusion: From the analysis of data during the study period, we come to the conclusion that simultaneous equation modeling is possible between FDI and economic development, that is both the variables had an endogenous relationship between them during this period (1991-2010), implying that each had impact on the other, as confirmed by 'Hausman Test of Endogeneity'. Moreover, change in FDI Granger-caused change in economic development as well as change in economic development Granger-caused FDI inflow to have a cascading effect on each other, which is evident from 'VAR Granger Causality/Block Exogeneity Wald Tests', explicitly establishing a bidirectional causality between FDI and economic development during the study period'. This interdependence between FDI and economic development is possible only when an economy is growing or developing. Therefore, there is no gainsaying the fact that as evident from statistical significance of our study output, development is seen to have taken place in India during the period 1991-2010. References: 1. Aitken, B., and A. Harrison (1999). 'Do Domestic Firms Benefit from Direct Foreign Investment? Evidence from Venezuela'. American Economic Review, 89 (3): 605-18. 2. Anitha & Dr. K. Maran(2011), Recent trends in foreign direct investment, International Journal of Research in Commerce & Management, vol. no. 2 (2011), issue no. 8 (August), pp 28- 34. 3. Alguacil, M. T. and V. Orts (2002), "A Multivariate Co-integrated Model Testing for Temporal Causality between Exports and Outward Foreign Investment: the Spanish case" Applied Economics, 34, 119-32. 4. Alguacil, M. T., A. Cuadros and V. Orts (2002), "Foreign Direct Investment, Exports and Domestic Performance in Mexico:A Causality Analysis," Economic Letters, 77, 371-76. 5. Bende-Nabendem, A., J. L. Ford, B. Santoso and S. Sen (2003), "The Interaction between FDI, Output and the Spillover Variables: Co-integration and VAR Analysis for APEC, 1965- 1999," Applied Economics Letters, 10, 165-72. 6. Balasubramanyan, V., N. M.A. Salisu and D. Sapsford. (1996), "Foreign Direct Investment and Growth in EP and IS Countries", Economic Journal, 106: 92-105. 7. Balasubramanyam, V. N., M. Salisu and D. Sapsford (1999). "Foreign Direct Investment as an Engine of Growth," The Journal of International Trade & Economic Development, 8,1, 27-40. 8. Blomstrom, M., and A. Kokko (1998). 'Multinational Corporations and Spillovers'. Journal of Economic Surveys, 12: 247-77. 9. Brecher, R.A. and Diaz-Alejandro, C.F. (1977) "Tariffs, Foreign Capital and Immiserizing Growth", Journal of International Economics, Vol.7, pp.317-22.
14
10. Bende-Nabendem, A. and J. L. Ford (1998), "FDI, Policy Adjustment and Endogenous Growth: Multiplier Effects from a Small Dynamic Model for Taiwan, 1959-1995," World Development, Vol. 26, pp. 1315-1330. 11. Borensztein, E., J. De Gregorio, J-W. Lee (1998), "How does Foreign Direct Investment Affect Economic Growth," Journal of International Economics, 45, 115-35. 12. Christopoulos, D. K. and Tsionas, E. G. (2004), Financial development and economic growth: evidence from panel unit root and co-integration tests. Journal of Development Economics, 73(4), 55- 74. 13. Chakraborty, C. and P. Basu (2002), "Foreign Direct Investment and Growth in India: a Cointegrating Approach," Applied Economics, 34, 1061-73. 14. Duasa, J.(2007): Malaysian Foreign Direct Investment and Growth: Does Stability Matter? , The Journal of Economic Cooperation, Volume 28,No 2,pp.83-98. 15. De Mello, Jr. (1997), "Foreign Direct Investment in Developing Countries and Growth: A Selective Survey," Journal of Development Studies, 34, 1, 1-34. 16. De Mello, Jr. (1999), "Foreign Direct Investment-led Growth: Evidence from Time Series and Panel Data," Oxford Economic Papers, 51, 133-51. 17. Dickey, D.A and W.A.Fuller (1979), Distribution of estimators of Autoregressive Time series with a Unit Root, Journal of the American Statistical Association,74,427-31. 18. Findlay, R. (1978), "Relative Backwardness, Direct Foreign Investment and the Transfer of Technology: A Simple Dynamic Model." Quarterly Journal of Economics, 92: 1-16. 19. Gray, H. P. (1998): International trade and foreign direct investment: the interface. In: J. H. Dunning (Ed.), Globalization, trade and foreign direct investment, Oxford: Elsevier, pp. 19-27. 20. Granger, C.W.J (1986), Developments in the Study of Co-integrated Economic Variables. Oxford Bulletin of Economics and Statistics, nr. 48. 21. Hermes, N., R. Lensink (2003), Foreign Direct Investment, Financial Development and Economic Growth, The Journal of Development Studies, 142-163. 22. Hsiao, F.S.T and Hsiao, M. C.W.(2006), 'FDI, exports, and GDP in east and southeast Asia Panel data versus time series causality analyses', Journal of Asian Economics,17(2006):10821106. 23. Jayashree Bose(2007), "FDI Inflows in India and China - A Sectoral Experiences", ICFAI University Press, Hyderabad. 24. Johansen, S., Juselius, K.(1992), Structural hypotheses in a multivariate co-integration analysis of the PPP and UIP for UK. J. Economics. 53,pp 211-44. 25. Khawar, M. (2005). 'Foreign Direct Investment and Economic Growth: A Cross Country Analysis'. Global Economy Journal, 5 (1): 1-11. 26. Liu, Xiaohui, P. Burridge and P. J. Sinclair (2002), "Relationships between Economic Growth, Foreign Direct Investment and Trade: Evidence from China," Applied Economics, 34, 1433-40.
15
27. Makki,S.S and Somwaru,A(2004),Impact of Foreign Direct Investment and trade on Economic Growth :Evidence from Developing Countries, American Journal of Agricultural Economics,vol.86(3),pp795-801. 28. Sim, C. (1992), Money, Income and Causality. American Economic Review, 62: 251-266.
16
doc_274844808.docx