See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/348620545
An Econometric Study of Export or Import Led Growth hypothesis in India
Article in Delhi Business Review · January 2021
DOI: 10.51768/dbr.v22i1.221202106
CITATIONS READS
4 895
2 authors:
Anand Shankar Paswan Shankar Nath Jha
University of Allahabad Banaras Hindu University
12 PUBLICATIONS 20 CITATIONS 44 PUBLICATIONS 17 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Anand Shankar Paswan on 20 January 2021.
The user has requested enhancement of the downloaded file.
Delhi Business Review Vol. 22, No. 1 (January - June 2021)
AN ECONOMETRIC STUDY OF EXPORT OR IMPORT LED
GROWTH HYPOTHESIS IN INDIA
Anand Shankar Paswan*
S. N. Jha**
P
URPOSE
THIS paper aims at examining short run and long run relationship among GDP, export and
import in India for the study period 1991-2018 using annual data, various econometric tools are
employed to analyse whether there is association among the variables in the time series data or not.
Design/Methodology/Approach: Unit root test has been applied to test the stationary of the data in
the time series which find variables stationary at first level. Johansen’s Co-integration test has been
used to determine the long run common path among the variables which infers one Co-integration
equation. AIC and SC are used for the selection of lag length criteria, while VECM test infers long run
association among the variables but statistically insignificant.
Findings: It is suggested to the policy makers and government of India to promote economic activities
and pursue diversification in commodities and market along with trade integration for the expansion
of export and continue importing necessary raw material for value addition and needed technology to
expand the capacity to improve productivity. Granger causality exhibits short-run unidirectional
causality from GDP to export while bidirectional causality exists between import and GDP and there
exists no directional causality between export and import in India.
Research Limitations: The data have limitations as it is only for the period 1991-2018. Data is
restricted only to secondary sources.
Managerial Implications: This research would provide an impetus to the the policy makers and
government of India to promote economic activities in the country.
Originality/Value: There has not much studies in this area.
Key Words: Export Led Growth, Import Led Growth, Co-integration, VECM, Granger Causality,
Economic growth in India..
Introduction
The world has experienced an upward trend in economic performance over the past four decades.
International trade, whether bilateral trade or with trade blocs among developed, developing or under-
developed countries, plays prominent role in shaping any country economy especially in developing
nations. Different international bodies like IMF, WTO, World Bank and others, took numbers of
productive steps, to favour trade openness for boosting economic growth. As being one of the prominent
* Senior Research Fellow, Faculty of Commerce, Banaras Hindu University, Varanasi, Uttar Pradesh, India.
** Professor, Faculty of Commerce, Banaras Hindu University, Varanasi, Uttar Pradesh, India.
73
Anand Shankar Paswan and S. N. Jha
emerging nation in the world, India understood the importance of open international trade and adopted
the Liberalisation, Privatisation and Globalisation (LPG) in June, 1991 by executing trade Liberalisation
policy. To see the impact of trade openness in India from 1991 to 2018 in economic growth and to test
the trending hypothesis of export-led-growth (ELG) promoted by neo- classical economist (Bhagwati,
1978) works effectively in Indian scenario or not? There are ‘n’ numbers of sound researchers, institutes,
policy framer etc., which support that by free trade market, developing nations got assistant in overall
economic growth in their nations. Contrary to this others has the opinion that developing nations
rather concentrate on protecting home industries from outside industries and concentrate on promoting
their home industries that will provide as road map for economic upliftment (Mishra, 2011).While it is
also argued that increase in trade led to more income and more income facilities more trade(Bhagwati,
1978). Upliftment in export could also assist in having a comparative advantage over other nations in
which the country possesses upper hand in some specialised sector (Kunst & Marin, 1989). Diversification
in exports could increase the productivity of a country by offering larger scale of economies and the
merging the global market with the national market (Elhanan Helpman & Paul Krugman, 1985).
ELG leads to higher production of goods and services with quality control, assist in economies of scale,
encourages employment generation, technological upgradation, increase in economies efficiency, labour
training, well organised management system (Feder, 1983).Adding to this export also help in generating
foreign exchange to the exporting nation and assist in acquiring knowledge in respect to latest foreign
technology and production skills, prevailing market trend, culture (Grossman & Helpman, 1991).
While other side of the same coin, another peer of economist’s advocates that, India has been described
as an import substituting country in which import plays a decisive role in India’s growth (Xu, Guo,
Liang, & Yu, 1996). Import led growth (ILG), suggest that increase pattern of imports could be an
important engine in driving the economic growth of a country by providing home industries with the
latest and upgraded abroad technology packages which work as an intermates productions factors
(Coe, Helpman, & Hoffmaister, 2009). It also serves as a moderating factor by providing the upgraded
research and development (R&D) acquired competent knowledge and latest skill from the skilled
(developed)nations to the unskilled or semi-skilled nations (developing or under-developed) nations
(Mazumdar & Papatla, 2000). In present Indian economy scenario since the post liberalisation period
and import led growth (ILG) theory has been found significantly active in uplifting India’s growth.
While some other group of economists also support the existence of bi-directional causal relationship
between export and GDP; these economists has the opinion which favours trade openness and consider
it as one of the foremost options for boosting economic activities and increasing economic efficiency
across the world (Shan, Morris, & Sun, 2001). Further adding to this adoption of new policies assist
Indian government to bring down trade barriers, which helps in trade restriction and to facilities to
play a part at international magnitude (Manmohan Agarwal & John Whalley, 2013). India’s success
story has been based on the pillar of both export-led growth and import-led growth with application to
the latest high-tech techniques from globalisation.
This has been a topic of research for a long time whether export or import through trade openness
policy assists in rapid development of an economy or not? In context to Indian economy the turning
point towards globalisation took place in 1991 when Indian economy were facing critical problem of
balance of payment crisis and in the same year LGP policy were adopted which come up as a rescuer for
Indian economy. To check the robustness of different side of theory that favoured Export-Led Growth
(ELG) or Import-Led Growth (ILG), especially in one of fastest developing countries like India, this
paper has been framed which attempts to check the robust of both the theory in Indian economy
prospective after economic openness since 1991.
Review of Literature
Love and Chandra (2005) examined the export and GDP of Pakistan (1970-2000), Bhutan (1980-
1997) Maldives (1977-2000) and Bangladesh (1973-2000) using Co- integration and Error Correction
74
Delhi Business Review Vol. 22, No. 1 (January - June 2021)
Modelling. It was found that in India, Maldives and Nepal there exists a positive association between
export and GDP which indicates Export-Led Growth (ELG) found positive exhibits in these countries.
On the other hand, in Bangladesh and Bhutan shows inverse relationship.
Shirazi and Manap (2005) investigated the ELG in south Asian nations which includes India, Pakistan,
Bangladesh, Nepal and Sri Lanka by employing Co-integration, VAR model with multivariate granger
causality test by employing Toda and Yamamoto Model. Where output supports the long run association
among the GDP, Export and Import in every country except Sri lanka adding to this the result support
the ELG hypothesis for, Nepal Pakistan and Bangladesh but not support the same for Sri lanka and
India.
Sharma and Panagiotidi (2005) re-investigated the hypothesis of Export Led Growth (ELG) during
1971- 2001in Indian perspective, by employing Johansen cointegration test and granger causality. The
result doesn’t not favour the ELG hypothesis in the study period in-spite of economy reforms in 1991
India still retains some characteristic of an import substituting economy but study also reveals that
Indian economy does support ELG hypothesis in the 1980s period.
Awokuse (2006) analysed individually the ELG and ILG theory in Bulgaria, Czech Republic and
Poland by using multivariate cointegrated VAR method. The result support both Growth Led exports
(GLE)as well as Export Led Growth (ELG)theoryfor Bulgaria, while Czech Republic favours the existence
of ILG and ELG theory in difference to other result Poland support only the ILG hypothesis in the
study period.
Mishra (2011) reinvestigated the hypothesis of Export-Led Growth (ELG) in India during the period
from 1970-2009 using Co-integration, VECM from the defined time series econometric tools. The study
supports a long-term causal association between export and Gross Domestic Product (GDP) further
VECM found positive linkage with supported by Granger Causality and rejected the null hypothesis of
ELG.
Agarwal (2014) consider the role of export in India’s economic growth and categorized impact of export
in economic growth in two parts i.e., pre- liberalisation (1960-1991) and post- liberalisation period
(1994-2012). Result suggest that export and GDP were not co-integrated by themselves but they do co-
integrated when an additional variable Real Effective Exchange rate (REER) were taken into consideration.
In pre-liberalisation period export doses not led to growth rather output produced in India determine
export but post- liberalisation period shows bidirectional relationship between exports and non-exports
GDP and ELG found weaker even in post-liberalisation period.
Debnath et al., (2014) analysed does GDP is led by export by differencing them into exports and non-
exported GDP for the period 1981 to 2012 by applying the ARDL approach to check the potential long
run-equilibrium. The analysis reveals that at mass level export does not proof to be a determining
factor to affect the output of other sector or ELG hypothesis does not found significant at mass volume
in India, while at disaggregate exports in terms of merchandise and services export were found positive
impact but statistically insignificant.
Venkatraja (2015) tested the ELG hypothesis in India for the study period of (1970-2013) using
Econometric model and found prevalence in long-term equilibrium association between export and
economic enhancement. Whereas, VECM models estimates at lagged value (1) of GDP determine the
contemporary value of export and vice-versa. Granger causality test shows there exists unidirectional
relationship i.e., export cause GDP.
Gupta and Singh (2016) investigated the cause-and-effect relationship between FDI and GDP among
the BRICS countries by employing co-integration, VECM and Granger Causality. Result suggest there
75
Anand Shankar Paswan and S. N. Jha
prevails long term relationship from GDP to FDI only in India, Brazil and China and short run causality
between FDI and GDP in China whereas, South Africa and Russia don’t have any long run causality.
VECM model and Granger Causality test suggest in case of India, Brazil and China higher GDP rates
leads to higher inflow of FDI and it is contradicting in case of South Africa and Russia.
Rani and Kumar (2018) examined the performance of BRICs and linkage among growth, imports
and exports with reference to South Africa, India and Brazil countries employed by panel data for the
consideration period 1967-2014. The output supports the existence of long-term association with the
specified variables, while FMLOS and DOLS output supports gross capital formation and export are
positively related by economic development, whereas as import association with economic growth found
negative and significant. Granger causality suggests bidirectional causality between ELGand GLE
hypotheses.
Sehrawat and Giri (2019) investigated the role of globalisation, institutions on the economic
performance of Indian economy using unit root test (saikkonen and Lutkepohl), cointegration test
(Bayer-Hanck), ARDL model to check the robustness and granger causality. Major finding suggests
that there exists cointegration among the variables, whereas ARDL model suggest that globalisation
and institutions positively contributes towards Indian economy, granger causality does not reciprocate
the same it reveals that institutional ability does not alter economic growth in short-term.
Objective of the Study
1. To assess the long run and short run interdependence among the variables.
2. To analyse the causal association among the variables in short term.
Hypotheses
The null hypotheses of the study are mentioned below:
H1: There is no co integration among Import, Export and Gross Domestic Product (GDP).
H2: There is no short-run and long- run causality among the variables.
Research Methodology
The study includes the secondary data for the period starting 1991 to 2018 collected from the world
bank indicators. The focus of the study is to assess long term and short-term interdependence as well
as directional causality between the independent and dependent variables. The independent variables
for the study were export and import of India while the dependent variable was GDP of India. For
analysing the data, econometric tools like unit root test with support of ADF, Johansen test of Co-
integration, Vector error Correction Model (VECM) and Granger’s causality test were used. Augmented
Dickey Fuller (ADF) test of unit root was done both at intercept as well as trend and intercept at levels
and first difference. While Johansen co-integration test was run to check whether there are long run
variables follow the same path or not? Vector Error Correction Model (VECM) was used to analyse the
probability of long run equilibrium of the model and how much the variables affect each other in long
run. For assessing the direction of short run causality between GDP, export and import of India,
Granger’s causality model was applied. After fitting the model, robustness of the model was used
verifying serial correlation and normality of the model.
Unit Root Test
The empiric work is basically a times series working data; the first assumption of a time series
data works upon that the observed data should be static or stationary. Here, unit root tests are
used to establish stationary properties of the observed data or we can say to access if mean is equal
to one or a unit and that the variance is sustained. The data is said to be stationary when data
76
Delhi Business Review Vol. 22, No. 1 (January - June 2021)
mean and variance are constant over the study period and the value of covariance at different time
period were found dependent only on the distance or lagat different time period and not on the
actual time at which the covariance is compiled (Gujarati and Sangeetha, 2007). In order to check
whether the data is stationary or not? If the data is stationary at what levels? In order to get these
answers Augmented Dickey-Fuller (ADF) (1979) test is used. ADF test is the modified version of
Dickey-Fuller (DF) test, since the DF test assume that the error term is not correlated, in case the
error is correlated than the modified ADF test will work more effectively (Gujarati and Sangeetha,
2007). This test is more favourable since its facilities higher-level autoregressive stages for reverifying
stationarity of the observed data. The test uses the following equation.
yt = + t + yt-1 +………+ p-1yt-p+1 + t………………….(1)
where, is a sustained or constant, the coefficient time series data trend, p is the lag order
selected for autoregressive stages and is white Gaussian random error term. Imposing the
constraints that = 0 and = 0 corresponds to modelling a random walk. Whereas using the
constraint that = 0, corresponds to modelling a random walk with a drift and using the constraint
= 0, corresponds to modelling a random walk with time trend.
Johansen’s Co-integration Test
If the Unit root test (using ADF test in our case) exhibits that all the dependent and independent
variables in the time series data is found stationary, the next step is to check whether there exists
any long run relationship among the variables or not (GDP, Export and Import). Johansen
cointegration test is implied to check the possible existence of long -run relationship among the
economic variables. Johansen framework for co-integration is a multivariate unit root test which
valuate the co-integration (rank r) in the multivariate case and to estimate the parameters of co-
integration relationship (Nkuma-Udah et al., 2013) to examine the co-integration among GDP,
Export and Import, Johansen’s co-integration (1988) test is used. In the Johansen framework the
following system is estimate.
= + − + ………………….(2)
=
where, Xt is an n×1 vector of non-stationary I(1) variables, a is an n×1 vector of constants, p is the
maximum lag length, âj is an n×n matrix of coefficient and et is a n×1 vector of noise terms. The
coefficient value (â) indicates the degree of co-integration, while the sign preceding to the coefficient
indicates whether the long run relationship between the variables is positive or negative associated.
In order to decide the appropriate model, to study johansen co-integration test and to test the
hypotheses H0: r =0 versus H1: r + 1‘“0 trace and maximum eigen value statistics are used in
verifying theses hypotheses.
Vector Error Correction Model (VECM)
Johansen’s co-integration test reflects only one side of a coin among the variables i.e., whether
there is any relationship among the specified variables or not. One of the trending techniques to
ascertain the direction of causation among the variables is Vector Error Correction Model (VECM).
Adding to this VECM also assist in identifying and measuring the speed of adjustment among the
variables and helping to recover from the temporary shock, outlier or disequilibrium to maintain
the long-term association among the variables during the period data obtained or we can say, it
also assists in measuring speed or time taken in convergence to the long-term steady state of
equilibrium. In our case the dependent variables (GDP) and independent variables (Export, Import)
have co-integration, then Vector Error Correction Model (VECM) is used to check whether there is
77
Anand Shankar Paswan and S. N. Jha
long run association among the dependent (GDP) and independent variables (Export and Import).
Thus, the Johansen co-integration equation (2) has pivoted into a vector error correction equation
as follows.
=
∆ = ∆ − + − + ………………….(3)
=
Where Äis the first difference operator, is – and is equal to -1+
=1+1 =1+1
While, assumption of VECM model is that it automatically converts the data stationary at first
difference, so the ‘D’ in the VECM equation stand for the automatic first difference of the variables,
whereas the lag length has been determined by the Schwarz Information Criterion (SC)Akaike
Information Criterion (AIC) and HQ: Hannan-Quinn Information criterion (-1) and (-2) in the
variables are lag(1) and lag(2) respectively.
Granger Causality Test
After it is confirmed that the dependent and independent variables are co-integrated with each
other in long run/ short run. Next step is to check how much a dependent variable is helpful in
determining independent variables or vice versa, to check the causal relationship among the variables
Granger Causality Test (1969,1988) is widely used. Granger Causality Test assist in determining
whether previous/last value of a variables can assist in predicting changes in another variables or
adding to this we can say Granger Causality Test assist to measures the data provided by one
variable(dependent/independent) in illustrating the current value of another variable (dependent/
independent). Granger Causality Test also assist in determining the direction of causal relationship
among the variables i.e., unidirectional or bidirectional.
∆ = + − + ∆ − + ∆ − + ∆ − + …
………………….(4)
∆ = + − + ∆ − + ∆ − + ∆ − +
….
……………….(5)
∆ = + − + ∆ − + ∆ − + ∆ −
+
………………….(6)
Major Findings
Unit Root Test
The unit root test was done to examine the stationary of the data. In table-1, the series of the
variables were tested to check the stationary of data through both way intercept and trend and
intercept. At level all the three variables GDP, Export and import were found non-stationary both
at intercept as well as trend and intercept, further the data were tested at first difference and the
data were showing stationary behaviour at first difference with significant p-values. These p-values
78
Delhi Business Review Vol. 22, No. 1 (January - June 2021)
at intercept for GDP, export and import were 0.0063, 0.0016 and 0.0044 respectively, with significant
value at 5 per cent level of significance whereas trend and intercept also reflect similar result with
the p-value of 0.0016, 0.0074 and 0.0175 for GDP, export and import respectively at 5 per cent level
of significance.
Table No. 1: Augmented Dickey-Fuller test of unit root
At Level At first difference
Trend and
Series/ Intercept Intercept Decision Intercept Intercept
Variables (p-value) (p-value) (p-value) (p-value) Decision
GDP 1.0000 0.9191 Non- 0.0063** 0.0016** Stationary
Stationary
Export 0.9632 0.5963 Non- 0.0016** 0.0074** Stationary
Stationary
Import 0.9655 0.5998 Non- 0.0044** 0.0175** Stationary
Stationary
Source: Author’s calculation using E-views 10.
**rejection of null hypothesis at 5% level of significance.
Table No. 2: Johansen Cointegration test result
H0 H1 Trace 5% Critical p-value Maximum 5% Critical p-value
Statistics Value Eigen Value Value
r =0 r =1 55.03919 29.79707** 0.0000** 44.85713 21.13162** 0.0000**
rd”1 r=2 10.18206 15.49471 0.2670 9.848574 14.26460 0.2221
Source: Author’s calculation using E-views 10.
r denotes the number of co-integrating vector in the long run.
**rejection of null hypothesis of co-integration rank r at 5% level of significance.
Since the variables were non-stationary at level and become stationary at first difference so both
the condition to test Johansen Cointegration test were fulfilled. In table no. 2, Johansen cointegration
test unrestricted cointegration rank test (Trace Statistics and Maximum Eigen value) were examined
and it illustrate that there exists one co-integrating equation among the variables or we can say
that all three variables GDP, export and import have long run association and in long run they
move together or follow a common long-run path. As, r = 0 null hypothesis (H0), no cointegration
among variables has been rejected as the p-value is 0.0000 and the trace statistic> critical value
(55.03919>29.79707) alternative hypothesis (H1) there exists one co-integration equation among
variables is accepted. On the other hand, Johansen cointegration test unrestricted cointegration
rank test (maximum eigen value) also justifies the same as trace test that there exits one cointegration
equation among dependent and independent variables, p-value were 0.0000* as r = 0 null hypothesis
was rejected, whereas max eigen value > critical value (44.85713 > 21.13162) which also indicates
rejection of null hypothesis (Ho) and acceptance of alternative hypothesis (H1) there exists one co-
integration equation among variables is accepted. However, the null hypothesis of at most one
cointegrating equation cannot be rejected to favour the r = 2 Thus, Johansen Cointegration test
denotes that the null hypothesis H1: There is no Cointegration among variables i.e., GDP, Export
79
Anand Shankar Paswan and S. N. Jha
and Import in rejected at 5% level of significance and the empirical result support the one cointegrating
equation among all the variables.
Vector Error Correction Model (VECM)
Table No. 3: Lag Value Selection
Lag Value Selection
Lag AIC SC HQ
0 158.6035 158.7487 158.6453
1 153.7940 154.3746 153.9612
2 152.0549* 153.0710* 152.3475*
AIC: Akaike Information Criterion, SC: Schwarz Information Criterion, HQ: Hannan-Quinn Information
criterion.
Notes: * Selected lag order by the respective criterion.
Table No. 3 exhibits the process of selecting the lag value through the Vector Auto Regression
(VAR) lag order selection criteria using three popular different computing techniques all the respective
techniques favours the selection of lag value at 2 lag which further leads to the process employing
the VECM model in a more robust form.
Table No. 4: Vector Error Correction Model (VECM) Equation-
D(GDP) = C(1)*( GDP(-1) + 1092129.58994*EX(-1) – 5557993.56891*IM(-1) – 97212969626.3) +
C(2)*D(GDP(-1)) + C(3)*D(GDP(-2)) + C(4)*D(EX(-1)) + C(5)*D(EX(-2)) + C(6)*D(IM(-1)) + C(7)*D(IM(-
2)) + C(8)
Dependent Variable: D(GDP)
Sample (adjusted): 1994-2018
Vector Error Correction Model (VECM)
Coefficient Std. Error t-Statistic Prob.
C (1) -0.049099 0.108314 -0.453305 0.6561
C (2) 0.501047 0.209173 2.395369 0.0284
C (3) 0.515524 0.304643 1.692224 0.1088
C (4) 0.25167 2.506526 0.100406 0.9212
C (5) -5.330871 2.42957 -2.194163 0.0424
C (6) -1.833055 1.543791 -1.187372 0.2514
C (7) 2.880927 1.286398 2.239529 0.0388
C (8) 5.03E+10 3.06E+10 1.641269 0.1191
Source: Author’s calculation using E-views 10.
From the VECM result, it can be inferred that Coefficient of (1) [C(1)*( GDP(-1) + 1092129.58994*EX(-
1) - 5557993.56891*IM(-1) – 97212969626.3)] is the error correction term or in other words we can
say it is speed of adjustment towards equilibrium. As the assumption of the VECM model in the
long run causality is that the coefficient of the co integrating model i.e., C (1) must be negative and
80
Delhi Business Review Vol. 22, No. 1 (January - June 2021)
p-value must be less than 0.05 (<0.05) to established a statistically significant long run relationship
among the dependent and independent variables (in our case GDP as dependent variable and import
and export as independent variables). Even the C (1) is negative which reflect there is long run
association from export and import towards GDP but statistically p-value reflect it is insignificant
or in layman language we can say a combined increase in export and import by 1 percent there will
be an increase of 0.048 percent every year in GDP which is statistically insignificant. Whereas, the
coefficient of export C (5) towards GDP reveals that in long run export negatively impact over
India’s GDP with -5.330871 percent every year in a significant manner, on the other hand the
coefficient of import C (7) towards GDP reflect import assist in growth of Indian economy in the
long-run path and with positive impact over GDP with 2.880927 percent every year and statistically
significant.
Table No. 4: Granger Causality Test
Sample 1991-2018
Lags: 6
Null Hypothesis Obs. F-Statistic Prob. Decision
EX does not Granger Cause GDP 22 3.03791 0.0656 Accept
GDP does not Granger Cause EX 12.8885 0.0006* Reject
IM does not Granger Cause GDP 22 3.77398 0.0368* Reject
GDP does not Granger Cause IM 10.4876 0.0012* Reject
IM does not Granger Cause EX 22 2.38356 0.1165 Accept
EX does not Granger Cause IM 2.62412 0.0937 Accept
Source: Author’s calculation using E-views 10.
*denotes rejection of null hypothesis.
From the VECM we confined the existence of long run causality among the variables and to test the
short run direction we used the granger causality test. From the result it is clear that there exists
causality in short run between GDP and export, but does not exists causality between export and
GDP which support the hypothesis of Growth Led Export (GLE) but does not support the hypothesis
of Export Led Growth (ELG). In nutshell, the result indicates a unidirectional causality between
GDP and export in short run, or we can say GDP will assist in predicating the next year export but
export will not help in forecasting next year GDP. Contrary, there exists causality between import
and GDP and even exists causality between GDP and import and we can say both GDP and import
has bidirectional causal association in short run which favours the existence of both the hypothesis
Import Led Growth (ILG) and Growth Led Import (GLI) for the case of India. On the other hand,
there is no directional causality between import and export. It implies neither import nor export
effect each other in the short run. While, Schwarz Information Criterion (SC) suggest the selection
of 6 lag value for the Granger causality test to determine the short run direction.
Conclusion
This paper re-investigated the export/import led growth hypothesis in India after the post liberalisation
period from 1991 to 2018 through unit root test by checking the stationarity of the data by using
Augmented Dickey Fuller (ADF) test, Johansen Co-integration test were employed to examine the long
run Co-integration among the variables, and further VECM model were performed to check the speed
of adjustment towards equilibrium in long run path while multivariate Granger Causality tests were
performed to examine the short-run direction among the variables.
81
Anand Shankar Paswan and S. N. Jha
Result of unit root test, revealed the stationary of the variables i.e. GDP, export and import at first
difference [I (1)] using ADF test for both at intercept (without trend) and trend and intercept (with
trend) levels, Johansen Co-integration confirms the long-run equilibrium relationship among these
three variables. The empirical results of both the trace and max-eigen statistics suggest the existence
of one long-run equation among all three variables i.e., GDP, export and import. Similarly, Vector
Error Correction Model (VECM) finding suggests the existence of a long run relationship between GDP,
export and import but statistically insignificant manner while import plays more prominent role in
determining Indian economy growth as compare to export towards GDP of India or ILG hypothesis
works more effectively in Indian scenario. Granger Causality tests finding also provide strong evidence
against the ELG hypothesis in Indian growth scenario but empirical finding does favour the existence
of growth led export (GLE) hypothesis. In contrast, the study finds strong evidences favouring the
existence of ILG hypothesis and exhibits bidirectional causal relationship between import and GDP
which support the finding of Jung and Marshall (1985), Xu (1996), Anwar and Sampath (2000) and
Pradhan (2007) which oppose the ELG hypothesis in India. Further, adding to the finding the study
also reveals that there is no-directional relationship between export and import both in short and long
run path.
Some suggestion for the policy implications from the empirical results like export promotional activities
as a significant tool for expanding economic development and could be moderately effective if import
tariff is maintained and government should also promote economic and trade integration for the expansion
of export with accelerating rate. While, import openness is also worked as a significant tool in respect
to Indian economic upliftment as its accompaniment the role of exports by carry out as a supplier of
intermediate production process needed in export sector and government should also pursue diversification
in exports in terms of both commodities and market further continue importing required raw material
to work as a assets for the domestic firm and facilities with needed technology to expand the capacity to
improve productivity to ensure stabilise contribution towards growth.
References
Agrawal, P. (2014). Wp345. 345.
Agarwal, Manmohan, & Whalley, John (2013). The 1991 Reforms, Indian Economic Growth, and Social Progress. Cambridge.
Anwar and Sampath (2000), Exports and Economic Growth, Indian Economic Journal, 47(3), 79-88.
Awokuse, Titus O. 2008. ¯Trade Openness and Economic Growth: Is Growth Export-led or Import-led? Applied Economics,
40(2), 161-173.
Bhagwati, J. (1978). Foreign Trade Regimes and Economic Development: Anatomy and Consequences of Exchange Control
Regime, Working Paper Series, NBER, New York.
Coe, D., & Helpman, E. (1995). “International R&D Spillovers,” European Economic Review, 39, 859-887.
Coe, D. T., Helpman, E., & Hoffmaister, A. W. (2009). International R&D spillovers and institutions, European Economic
Review, 53(7), 723-741.
Debnath, A., Laskar, A. B., Bhattacharjee, N., & Mazmuder, N. (2014). Is India’s GDP Really Led by Export? A Further
Examination, Journal of Transnational Management, 19(4), 247-260.
Dickey, D. A., & Fuller, W. A. (1979). Distribution of the Estimators for Autoregressive Time Series With a Unit Root,
Journal of the American Statistical Association, 74(366), 427.
Elhanan, Helpman, & Paul, Krugman (1985). Market Structure and Foreign Trade: Increasing Returns, Imperfect Competition,
and the International Economy. United States of America: The MIT Press.
Feder, E. (1983). Plundering the poor: The role of the world bank in the third world, International Journal of Health
Services, 13(4), 649-660.
Granger, C., W., J. (1969). Investigating causal relations by econometric models and cross spectral methods, Econometrica,
37, 424-438.
Gujrati, N. Damodar & Sangeetha (2007). Basic Econometrics, Tata McGraw-Hill Publishing Company Limited, New
Delhi, 4th ed.
82
Delhi Business Review Vol. 22, No. 1 (January - June 2021)
Gupta, P., & Singh, A. (2016). Determinants of Foreign Direct Investment Inflows in BRICS Nations: A Panel Data
Analysis. Emerging Economy Studies, 2(2), 181-198.
Grossman, G. M., & Helpman, E. (1991). Trade, knowledge spillovers, and growth. European Economic Review, 35(2-3),
517-526.
Intra-BRICS Trade: An Indian Perspective Export-Import Bank of India Export-import Bank of India Intra-BrICS trade: An
IndIan PerSPeCtIve © Export-Import Bank of India, (2016).
Jabalameli, F., & Rasoulinezhad, E. (2018). BRICS-United Nations regional groups’ trade patterns: a panel-gravity approach,
Journal of Chinese Economic and Foreign Trade Studies, 11(2), 151-179.
Joshi, Amitabh (2013). Long-term Causality of GDP-led Export (GLE) Using VECM Model with Reference to India,
Prestige International Journal of Management & IT- Sanchayan, 2(1), 14-25.
Kumari, Deepika & Malhotra, Neena (2014). Export-Led Growth in India: Cointegration and Causality Analysis, Journal
of Economics and Development Studies, 2(2), 297-310.
Kunst, R. M., & Marin, D. (1989). On Exports and Productivity: A Causal Analysis Author (s): Robert M . Kunst and Dalia
Marin Source: The Review of Economics and Statistics, 71(4), (November, 1989), 699-703, Published by: The MIT Press
Stable URL/: http://www.jstor.org, The Review of Economics and Statistics, 71(4), 699-703.
Love, L., & Chandra, R. (2005): Testing Export – Led Growth in South Asia, Journal of Economic Studies.
Marshall, J. (1985). Exports, Growth and Causality in Developing Countries, Journal of Development Economics, 18,
1-12.
Mazumdar, T., & Papatla, P. (2000). An investigation of reference price segments, Journal of Marketing Research, 37(2),
246-258.
Mishra, A. K., Gadhia, J. N., Kubendran, N., & Sahoo, M. (2015). Trade Flows between India and Other BRICS Countries:
An Empirical Analysis Using Gravity Model, Global Business Review, 16(1), 107-122.
Mishra, P. K. (2011), The Dynamics of Relationship between exports and economic growth in India, International Journal
of Economic Sciences and Applied Research, 4(2), 53-70.
Nasim, Shah Shirazi, & Ali, T. U. A. (2005). Export-led growth hypothesis: Further econo- metric evidence from south asia,
4(December), 472-488.
Nkuma-Udah, K. I., Agoha, E. E. C., Ndubuka, G. I., Osuagwu, C. G., Iwuji, S. C., & Ejeta, K. (2013). Strengthening of
developing countries biomedical engineering by the developed countries: The engineering world health example, IFMBE
Proceedings, 39 IFMBE, 1704-1707.
Panda, Sethi, & Kumaran (2016). Working Paper No . 18 Potential for Enhancing India ’ S Trade With China: An update,
2015.
Panda, R., Sethi, M., & Kumaran, M. (2016). A Study of Bilateral Trade Flows of China and India (June).
Pradhan, E. K., Baumgarten, M., Langenberg, P., Handwerger, B., Gilpin, A. K., Magyari, T., Hochberg, M. C., & Berman,
B. M. (2007). Effect of mindfulness-based stress reduction in rheumatoid arthritis patients, Arthritis Care and Research,
57(7), 1134-1142.
Rani, R., & Kumar, N. (2018). Is There an Export- or Import-led Growth in BRICS Countries? An Empirical Investigation,
Jindal Journal of Business Research, 227868211876174.
Rasoulinezhad, E., & Jabalameli, F. (2018). Do BRICS countries have similar trade integration patterns?, Journal of
Economic Integration, 33(1), 1011-1045.
Sehrawat, M., & Giri, A. K. (2019). Globalization, role of institutions and economic performance in Indian economy:
Empirical evidence, Journal of Financial Economic Policy, 11(1), 82-100.
Shan, J. Z., Morris, A. G., & Sun, F. (2001). Financial development and economic growth: An egg-and-chicken problem?,
Review of International Economics, 9(3), 443-454.
Sharma & Panagiotidi (2005). An Analysis of Exports and Growth in India: Cointegration and Causality Evidence
(1971-2001), Review of Development (2005)Economics, 9(2), 232-248.
Takeshi, Inoue (2014). An Empirical Analysis of the Aggregate Export Demand Function in Post-Liberalization India,
Global Economy Journal, 14(1), 79-88.
Tahir, Khan, Israr, Qahar (2015). An Analysis of Export Led Growth Hypothesis: Cointegration and Causality Evidence
from Sri Lanka, Advances in Economics and Business, 3(2), 62-69.
83
Anand Shankar Paswan and S. N. Jha
Tayfur, B., Koçyiðit, A., Bayat, T., Kayhan, S., & Þentürk, M. (2015). Short and Long Term Validity of Export-Led
Growth Hypothesis in BRICS-T Countries: A Frequency Domain Causality Approach, Journal of Asian Development
Studies, 4.
Thakur, R. (2014). How representative are brics?, Third World Quarterly, 35(10), 1791-1808.
Venkatraja, B. (2015). Testing the Export-Led Growth Hypothesis for India: An Econometric Analysis, 37.
Wilson, D., & Purushothaman, R. (2003). Dreaming With BRICs: The Path to 2050.
Xu, J., Guo, Z., Liang, Y. Z., & Yu, R. (1996). Two new algorithms for resolution of two-way data, Journal of Chemometrics,
10(1), 63-76.
84
View publication stats
