Foreign Direct Investment, Export, External Debt and Economic Growth in India- A VECM Approach
Mr. Shivakumar
Research Scholar
Department of Management Studies
Centre for Post Graduation Studies,
Visvesvaraya Technological University,
Mysore
Dr. N Babitha Thimmaiah
Research Supervisor
Department of Management Studies
Centre for Post Graduation Studies,
Visvesvaraya Technological University,
Mysore
Abstract: The researcher attempted to analyse how FDI, export, external debt contribute to the economic growth of India. The author has developed various econometrics models to assess the relationship of the studied variables. The unit root tests, cointegration model and VECM revealed the short run and long run relationship of the studied variables. The study employed VECM to study the long run, short run and joint effect of the studied variables. The results showed significant effect from FDI, exports and external debt to economic growth. The Wald test result shows the short run relationship in export to economic growth and external debt to economic growth except FDI.
Keywords: Economic growth, export, external debt, VECM, FDI
FDI
Growth of the economy is the main thrust of the countries of developing tag, the study on FDI, export, external debt and economic growth has attracted the academicians and the researchers around the world. FDI helps the economy in reducing the imbalance of savings and investments (Vo et al. 2019a) and it also attract the revenue in the form of tax and capital in the form of skilled human resource(Buckley et al.2002).
In this globalization era, the countries are focusing on creating the attractiveness of FDI mainly due to positive effect and fewer commercial, technological and economical barriers. (Demirsel et al-2014).
The other determinant, exports also a significant contributor to the economic growth as studied by Jung and Mrshall-1995 and Kwan and Costomitis-1990. The external debt also play key role in the economic growth as indicated in the studies such as Levine and Renalt-1992, Remamurti (1992).
Export
Exports of tangibles and services are important aspects in economic growth of any country. A debatable issue is whether export drives growth or growth drives export. A number of empirical studies made an attempt to resolve it (Bhagwati 1978, Krueger 1978).
Export led growth provides benefits not only to exporters but also helps in various economic and government policies. Export helps the country to integrate with the world. The Asian country such as India is the best example of export led economic growth showing the export as an engine of economic growth. The export boosts economy with increase in total factor productivity from the economies of scale principle. It also improves skills of the workers, managers and overall productive capacity.
External Debt
The use of external debt in a productive manner enhances economic growth. It is argued (Todaro and Smith-2006) that the developing countries need to borrow the external debt in the initial stages of the economy as these countries generally faces the problem of shortage of domestic capital. The Harrod-Domar growth model highlighted the importance of external debt as it closes the gap between savings and investments. There is a general notation that the economy will be benefited reasonably from the external debt which leads to finance the productive investments. The studies in the literature have shown the inverse relation of external debt and economic growth.
Despite many studies conducted on assessing the FDI, export, and external debt on economic growth, still no common consensus has been established on the same. The main reason of this could be the country, time period, methods of study etc. FDI and economic growth an important factors have attracted the attention of the investors around the world in recent times (Basu et al. 2003; Vo et al. 2019a). The relationship between the above economic indicators have been assessed taking one country or multiple countries into consideration. The study (Edward E. Ghartey-2006) of export and economic growth using US, Japan and Taiwan shows the existence of causal effect in export and economic growth. The study (Shaista Alam -2000) shows the causal effect of external debt on economic growth and found the existence of unidirectional relationship. The positive effect of FDI on economic growth (Shahbaz and Rahman 2010) was found in the study. The FDI-growth study supports the negative effect of FDI on economic growth.
Pradhan (2002) made an analysis of FDI and economic growth in India. The result confirms no positive effect of FDI on economic growth. The results are as per the study of Agrawal (2005) that FDI had very little effect on growth.
Kapusuzoglu (2011) made an attempt to study the relation among the selected indexes of Istanbul Stock Exchange and price of Brent oil in international market and found the long run relationship in the studied variables.
Several studies in the literature through the light on individual and cross section export and economic growth. i.e. Rivera-Batiz and Romer (1991), Levin and Raut (1997), Crespo-Cuaresma and Wörz (2003).
Tang (2006) studied the relationship of the exports, real GDP and imports. The results of the study indicates no relationship between export and economic growth in China in both short run and long run. Srinivasan and Bhagwati (2001) made a research on the export and economic growth. The authors argument says the manufacturing products are less sensitive in exports in comparison with export of raw material and other goods in the international market.
Were (2001), and Hameed and Chaudhary (2008) studied debt and economic growth factors and found the negative relationship. Reinhart and Rogoff (2010) studied the relationship of external debt and economic growth with the sample of 20 developed countries. The result of the GDP to debt ratio indicates the negative relationship between debt and growth.
Musebu (2012) analysed external debt and economic growth. The result shows that the external debt hampers the economic growth. The results are consistent with the studies of Cohen (1993), Oteng-Abaye (2003).
The present study focuses on the FDI, export, external debt and economic growth of India.
The study made an attempt to examine the significance of FDI, export and external debt in contributing to the growth of Indian economy which is measured through GDP. The model of the research is as follows as per Shahbaz and Rahman (2010),
GDPt = β0 + β1FDIt + β2EXPt + β3EXDt + εt
Where, β0 is the intercept. β1, β2 and β3 are the coefficients. εt is the error term. Subscripts t denotes year. The variables are translated into log form in order to reduce the heteroskedasticity. The variables are defined as follows in the table 1.
Table-1: Variables Definitions
|
Factors |
Variables |
Symbol |
Description |
Sign |
|
|
|
|
|
|
|
Dependent Variable |
||||
|
Gross Domestic Product |
Real GDP |
GDP |
Log(GDP) |
+ |
|
|
|
|
|
|
|
Independent Variables |
||||
|
Investments |
FDI-Inflow |
FDI |
Log(FDI) |
+ |
|
Trade |
Total Export |
Export |
Log(Export) |
+ |
|
Financing |
External Debt |
EXDT |
Log(External debt) |
+ |
The researcher employed suitable methodologies to analyse the effect of study variables i.e. FDI, export, external debt and GDP. These variables passed through the stationary test, Vector Autoregressive model for lag selection, cointegration and VEC for relationship of the studied variables.
The above analyse is carried out for a period from 2000-2019(quarterly data) with totaling 80 observations. The data is gathered from the websites of RBI, World Bank, IMF, UNCTAD etc. The following table no 2 and 3 shows the descriptive statistics and multicollinearity of the selected variables.
Table no 2: Descriptive Statistics
|
Series: Residuals |
|
Sample 2000Q1 2019Q4 |
|
Observations 80 |
|
|
|
Mean 3.11e-15 |
|
Std. Dev 0.058799 |
|
Skewness 0.047743 |
|
Kurtosis 1.916022 |
|
|
|
Jarque-Bera 3.947089 |
|
Prob 0.138963 |
Table no 3: Multicollinearity Check
|
|
FDI |
Export |
External debt |
|
FDI |
1 |
||
|
Export |
-0.139776805 |
1 |
|
|
External debt |
0.056741155 |
0.183907019 |
1 |
The results from the table no 2 shows normal distribution of the residuals by accepting null hypothesis. i.e. normal distribution of residuals as it is acceptable at 5% level. Table no 3 shows the correlation between the FDI-export of -0.13, between FDI-external debt of 0.05 and between export-external debt of 0.18 indicating no multicollinearity problem.
4.1 Unit Root Test
The recent methods of stationary such as LLC test, IPS test, ADS test, and PP tests are performed to test the stationary of studied variables. The results are shown in the table no 4a and 4bas follows.
Table no 4a: Stationary Test at Level
|
Group Unit Root Test: |
||||
|
Series: GDP, FDI, Export, External Debt |
||||
|
Sample: 2000Q1 2019Q4 |
||||
|
|
||||
|
Methods |
Statistics |
Prob |
Cross sec |
Obs |
|
LLC |
-0.66954 |
0.2516 |
4 |
312 |
|
IPS |
1.68927 |
0.9544 |
4 |
312 |
|
ADF-Fisher |
2.67582 |
0.9530 |
4 |
312 |
|
PP-Fisher |
20.6294 |
0.0082 |
4 |
316 |
Table no 4b: Stationary Test at First Difference
|
Group Unit Root Test: |
||||
|
Series: GDP, FDI, Export, External Debt |
||||
|
Sample: 2000Q1 2019Q4 |
||||
|
|
||||
|
Methods |
Statistic |
Prob |
Cross sec |
Obs |
|
LLC |
-19.0567 |
0.0000 |
4 |
310 |
|
IPS |
-23.9279 |
0.0000 |
4 |
310 |
|
ADF-Fisher |
87.7047 |
0.0000 |
4 |
310 |
|
PP-Fisher |
87.8582 |
0.0000 |
4 |
312 |
The above result shows the non stationarity of the variables at level except for the test PP-Fisher Chi-Square shows the stationarity at level. The variables are stationary at first difference indicated by the significance at 5% level.
Figure 1: Stationary of Study Variables at First Difference
4.2 Cointegration Test
The cointegration test is used to examine cointegration of the variables. The said cointegatred test takes into account the statistics named λ trace and λ max statistics, where the model is:
Where,
r = number of individual series
T = number of usable observations
λ = Eigen values
H0: No cointegration equation
H1: H0 is not true
H0 shows the null hypothesis indicates the cointegrating vectors which is/are less than or equals r and H1 indicates greater than r. The results are computed by the comparison of t-statistics to critical values. H0can be rejected if the t statistics are significant at 5% and vice versa
Table no 5: Cointegration Test
|
Null Hypothesis |
Trace test |
Max eigen value test |
|
|
|
|
|
Null* |
51.95891 |
39.77123 |
|
Atmost 1 |
12.18768 |
7.489329 |
|
Atmost 2 |
4.698353 |
3.620729 |
|
Atmost 3 |
1.077624 |
1.077624 |
*denotes significant at 5% level
The next step after confirming the stationary of the variables at first order 1(I) perform the cointegration test to examine the cointegrating variables in the study. The cointegration test is performed on the level data as suggested by the literature.
The first hypothesis at (None *) is rejected since the value of trace test and max eigen value test are significant at 5% level. The cointegration test confirms that the study has cointegrating variables. The results are shown in the above table no 6 indicates at least 1 cointegration is found at 5% level. Sometime the shocks in the short run which may converge with long run. Hence, long and short run estimation is needed. The results of the cointegration test confirms the cointegration in the variables which are examined using the appropriate VECM model.
4.3 Vector Error Correction Model (VECM)
The next part of the analysis is to perform a restricted VAR .i.e. VECM. The cointegration term is called as error correction model as it gradually corrects the deviation from long run equilibrium. The cointegrated vectors show the long run effect and causality can be examined using error correction term. VECM is performed to analyse relationship in the variables.
The VECM is conducted on the level data using the software (Eviews-9) because the software automatically takes the first differences for estimating the variables.
VECM Model:
This model shows the error correction term where the first term is differenced log of GDP, followed by differenced FDI, differenced export and differenced external debt, ECT is error correction term signifies the long run effect and et is error term.
Model-I
VECM equation is given below:
|
Equation: D(GDP) = (1)*( GDP(-1) - 0.256521462296*FDI(-1)-0.485601555982 *EXPORT(-1) - 0.193488004837*EXDT(-1) -8.72629403816 ) +C(2) *D(GDP(-1)) + C(3) *D(GDP(-2)) + C(4) *D(FDI(-1)) + C(5) *D(FDI(-2)) +C(6) *D(EXPORT(-1)) + C(7) *D(EXPORT(-2)) +C(8) *D(EXDT(-1)) + C(9) *D(EXDT(-2)) + C(10) |
Table no 6: Results of VECM
|
Estimation Method: Least Squares Sample: 1 80
|
||||
|
|
||||
|
|
Coefficients |
Std. Error |
t-Statistic |
Prob. |
|
|
||||
|
Coefficient(1) |
-0.058364 |
0.017535 |
-3.328441 |
0.0019 |
|
Coefficient(2) |
-0.382215 |
0.264660 |
-1.444174 |
0.1565 |
|
Coefficient(3) |
0.394344 |
0.219472 |
1.796789 |
0.0799 |
|
Coefficient(4) |
0.006184 |
0.007164 |
0.863178 |
0.3932 |
|
Coefficient(5) |
0.006876 |
0.006784 |
1.013467 |
0.3169 |
|
Coefficient(6) |
0.138161 |
0.025328 |
5.454966 |
0.0000 |
|
Coefficient(7) |
0.074545 |
0.035819 |
2.081148 |
0.0439 |
|
Coefficient(8) |
-0.011308 |
0.007030 |
-1.608521 |
0.1156 |
|
Coefficient(9) |
4.22E-05 |
0.006093 |
0.006921 |
0.9945 |
|
Coefficient(10) |
0.011835 |
0.003752 |
3.154435 |
0.0030 |
|
Coefficient(11) |
-0.611049 |
0.488168 |
-1.251718 |
0.2179 |
|
Coefficient(12) |
-10.89131 |
7.368063 |
-1.478179 |
0.1472 |
|
Coefficient(13) |
-6.763966 |
6.110038 |
-1.107025 |
0.2749 |
|
Coefficient(14) |
-0.335421 |
0.199452 |
-1.681714 |
0.1004 |
|
Coefficient(15) |
-0.361679 |
0.188870 |
-1.914967 |
0.0627 |
|
Coefficient(16) |
1.605761 |
0.705113 |
2.277309 |
0.0282 |
|
Coefficient(17) |
0.019734 |
0.997197 |
0.019790 |
0.9843 |
|
Coefficient(18) |
-0.468685 |
0.195710 |
-2.394800 |
0.0214 |
|
Coefficient(19) |
0.173352 |
0.169617 |
1.022020 |
0.3129 |
|
Coefficient(20) |
0.330283 |
0.104453 |
3.162028 |
0.0030 |
|
Coefficient(21) |
0.165420 |
0.209277 |
0.790434 |
0.4339 |
|
Coefficient(22) |
-0.697762 |
3.158681 |
-0.220903 |
0.8263 |
|
Coefficient(23) |
2.988934 |
2.619367 |
1.141090 |
0.2606 |
|
Coefficient(24) |
0.065468 |
0.085505 |
0.765666 |
0.4484 |
|
Coefficient(25) |
0.058915 |
0.080968 |
0.727631 |
0.4711 |
|
Coefficient(26) |
0.351614 |
0.302281 |
1.163200 |
0.2516 |
|
Coefficient(27) |
0.190995 |
0.427498 |
0.446774 |
0.6574 |
|
Coefficient(28) |
0.041105 |
0.083901 |
0.489924 |
0.6269 |
|
Coefficient(29) |
0.020225 |
0.072715 |
0.278137 |
0.7823 |
|
Coefficient(30) |
-0.005831 |
0.044779 |
-0.130218 |
0.8970 |
|
Coefficient(31) |
1.593986 |
0.424734 |
3.752904 |
0.0006 |
|
Coefficient(32) |
-5.432794 |
6.410633 |
-0.847466 |
0.4018 |
|
Coefficient(33) |
4.540444 |
5.316080 |
0.854096 |
0.3981 |
|
Coefficient(34) |
0.418314 |
0.173535 |
2.410552 |
0.0206 |
|
Coefficient(35) |
0.347790 |
0.164327 |
2.116447 |
0.0406 |
|
Coefficient(36) |
-1.065744 |
0.613489 |
-1.737187 |
0.0900 |
|
Coefficient(37) |
1.417691 |
0.867618 |
1.634003 |
0.1101 |
|
Coefficient(38) |
-0.838529 |
0.170279 |
-4.924455 |
0.0000 |
|
Coefficient(39) |
-0.141127 |
0.147577 |
-0.956297 |
0.3447 |
|
Coefficient(40) |
0.191670 |
0.090880 |
2.109045 |
0.0412 |
|
|
|
|
||
|
R2 |
0.875501 |
|||
|
Adj R2 |
0.763451 |
|||
|
S.E. of reg |
0.002882 |
|||
Table no 7: Wald Test
|
Variables |
Chi-squ |
Prob. |
|
D(FDI) |
0.605314 |
0.7389 |
|
D(EXPORT) |
11.13907 |
0.0038 |
|
D(EXDT) |
5.565677 |
0.0619 |
The results of the above table no 6 show the estimates of VECM and effect on GDP. The table no 6 shows coefficients from 1 to 40. The Coefficient(1) shows the long run relationship from FDI, export and external debt to GDP as the Coefficient(1) value is negative (-0.1044) and acceptable at 5% level. The negative value of Coefficient (1) and significant result shows the long run relationship. The findings are inconsistent with (Aitken and Harrison 1994; Alfaro et al. 2004; Konings 2001; Lyroudi et al. 2004;)
The table no 7 indicates the results of short run relationship of the selected variables. For the short run relation, Wald test results show between GDP-FDI, the chi square value is 0.605314 (0.7389) not significance at 5% level. It indicates no short run relationship between GDP-FDI. These results are consistent with the previous studies (Chandana Chakraborty & Parantap Basu -2010, Tang, T.C- 2006). The results of the wald test shows significant short run relationship between GDP-Export, where the chi square value is 11.13907 ( 0.0038) significance at 5% level. Lastly, there is a significant short run relationship between GDP-external debt, where the chi-square value is 5.565677 (0.0619) significant at 10% level.
4.4 Diagnostic Check for the Model
The model that is used in the research is employed for the diagnostic checks to find out the correctness of the model specification. Firstly the normality of the residuals has been checked and found the residuals of the model show the data is normally distributed. The H0 of ‘residuals are normally distributed’ is accepted as the results are not significant at 5% level. (0.250873>0.05) as per the table no 2. It is confirmed the normal distribution of residuals.
Table no 8: Autocorrelation Results
|
Obs*R-squared |
5.699461 |
|
Prob |
0.0579 |
The table no 8 result shows the observed R-squared value is 5.699 and significant at 5%. i.e 0.0579. The H0 of ‘no autocorrelation’ is not rejected. It shows the data is free from the serial autocorrelation.
The study attempted to analyse how FDI, export and external debt add value to the economic growth of India in both the short and long run. The econometric models such as VAR, Johansen Cointegration, VECM and Wald test are employed for the analysis. The findings of this study significantly contribute to the literature. The study employed the quarterly data during the period 2000-2019 which is collected from the RBI database.
The analysis is concluded as follows. Firstly, FDI inflows have shown positive effect on the GDP in the long run, but no significant effect in the short run. Secondly, exports also have shown significant effect on the GDP in the long run and also contributes significantly in the short run. Finally, external debt has also shown significant effect on economic growth in long run and short run.
From the above results, it can be inferred that the FDI, exports and external debt are key factors in determining the economic growth of India. It is suggested that the government should design the optimal policies on FDI, export and external debt to create healthy economic environment in the country. This will boosts the confidence of the investors and attract the new investors to India. The liberalised BOT policies, flexible human capital will tends to have positive effect of FDI on economic growth (Zhang-2001). There are many factors determines the economic growth of India apart from FDI, export and external debt. Further researcher can look into other factors to understand the effect on the economic growth.
References
Alfaro, Laura, Areendam Chanda, Sebnem Kalemli-Ozcan, and Selin Sayek. 2004. FDI and economic growth:The role of local financial markets. J. Int. Econ. 64: 89–112.
Bhagwati, J. (1978). Foreign Trade Regimes and Economic Development: Anatomy and Consequences of Exchange Contrast Regimes. MA, Ballinger Publishing Company,
Chandana Chakraborty &Parantap Basu -2010, Applied Economics 34(9):1061-73,
Crespo-Cuaresma, J. and J. Wörz (2003) "On Export Composition and Growth", University of Vienna, Department of Economics, Vienna, Austria; and Vienna Institute for International Economic Studies (WIIW), Vienna, Austria,
Demirsel, Mustafa Tahir, Adem Ögüt, and Mehmet Mucuk- 2014. The Effect of Foreign Direct Investment on Economic Growth: The Case of Turkey. International Academic Conferences, Prague, Czech Republic, September 1-4,
Hameed, H. A. and Chaudhary M. A. (2008). External Debt and its Impact on Economic and Business Growth in Pakistan. International Research Journal of Finance and Economics, 20: 132-140,
Konings, Jozef. 2001. The effects of foreign direct investment on domestic firms. Econ Transit. 9: 619–33.
Koojaroenprasit, Sauwaluck. 2012. The impact of foreign direct investment on economic growth: A case study ofSouth Korea. Int. J. Bus. Soc. Sci. 3: 8–19.
Levin, A. and L.K. Raut (1997) "Complementarities between Exports and Human Capital: Evidence from the Semi-Industrialized Countries" Economic Development and Cultural Change, Vol. 46: 155-174,
Musebu S. (2012). Debt Overhang and Economic Growth in HIPC Countries: The Case of Southern African Development Community (SADC), International Journal of Economics and Finance, 4(10): 82-92, ISSN 1916-971X,
Nguyen, Ha Minh, Ngoc Hoang Bui, and Duc Hong Vo. 2019. The Nexus between Economic Integration andGrowth: Application to Vietnam. Annals of Financial Economics 14: 1950041.
Phillips, Peter C. B., and Pierre Perron. 1988. Testing for a unit root in time series regression. Biometrika 75: 335–46.
Reinhart, C.M. and Rogoff, K.S. (2010). "Growth in a Time of Debt". American Economic Review, 100 (2): 573-578,
Rivera-Batiz, L.A. and P.M. Romer (1991) "Economic Integration and Endogenous Growth" Quarterly Journal of Economics, Vol. 206:531-555,
Stanisic, Nedad. 2008. Do foreign direct investments increase the economic growth of Southeastern Europeantransition economies? South.-East. Eur. J. Econ. 6: 29–38.
Tang, T. C., 2006, 'New Evidence on Export Expansion, Economic Growth and Causality in China', Applied Economics Letters, 13, pp. 801-803,
Todaro, P. M. and Smith, S. C. (2006). Economic Development, (9th edition). Washington D.C.: Pearson Education, Harlow,
Were, M. (2001). The Impact of External Debt on Economic Growth in Kenya: An Empirical Assessment, UNU-WIDER Research Paper.
