Description
WEATHER MAKERS PRESENTS RAINWATER MINERAL WATER WANTED FRENDS TO JOIN THIS VENTURE
Export and Economic Growth of India -
A Cointegration and Causality Evidence
Prepared by:
! C! Pradhan
"esearch #fficer
$inancial %ar&ets 'epartment
"E(E")E *A+ #$ I'IA
Central #ffice, $ort
%umbai - -..../
for presentation at the -0rd Annual Conference of 1he Indian Econometric
(ociety 21IE(3 to be held during 4anuary 5-6, 7..6 at the 'epartment of
8umanities and (ocial (ciences, II1 *ombay, Powai, %umbai - -....69 !
Abstract
The export led growth (ELG) hypothesis for India is tested with the data at
aggregate level covering post-liberalisation period. Stdy presents a test of the
ELG hypothesis by adopting co-integration hypothesis (!ohansen "ethod) for the
case of India. #ollowing hypotheses are tested$ (i) whether exports% i"ports and
G&' are cointegrated% (ii) whether export growth Granger cases G&' growth%
and (iii) and whether export growth Granger cases invest"ent. The finding
strengthens the arg"ent against the ELG hypothesis for the case of India and
strengthens the arg"ent that inspite of refor"s% it still retains so"e
characteristics of an i"port sbstitting econo"y. Since aggregate export data
for India incldes sectors% sch as% software exports% the ephoria abot
perceived sccesses in the I(T (infor"ation and co""nication technology)
sector for India see" so"ewhat pre"atre% given that at an aggregated level
there is little evidence to spport the export led growth hypothesis% which brings
into dobt the i"plicitly ass"ed prodctivity differentials and reslting positive
spillovers into the rest of the econo"y.
2
Export and Economic Growth of India -
A Cointegration and Causality Evidence
?
! C! Pradhan
:
The independent India had faced foreign exchange problem that leads to
export stagnation, as was happened in other underdeveloped countries. In this
context, Ragnar Nurkse
1
observed that, stagnation in underdeveloped countries
export was mainl! attributed to slowl! expanding world markets for traditional
products. The same argument was also corroborated b! Indian "overnment
#ommittee
2
and $. %. &atel
'
. The "eneral (greement on Tariff and Trade )"(TT*
report
+
illustrated the phenomenon of stagnant world export during 1,-.s mainl!
due to relativel! inelastic suppl! of exports as a result of slowl! expanding
domestic output and rapidl! expanding domestic demand in underdeveloped
countries, including India. In 1,/1, India was third onl! to 0ene1uela and 2ra1il
as an exporter among the underdeveloped countries )#ohen, 1,/+*. 2etween 1,+3
and 1,/1, while annual value of Indian exports had fluctuated around 4$ 5 1.'
billion, the world export had more than doubled from 4$ 5 -, billion to 4$ 5 12,
billion. The share of exports declined continuousl! from 2.2 per cent to 1.1 per
cent during the period )(ppendix Table 1*. 2en6amin #ohen )1,/+* had examined
the situation to conclude that declining market share was caused b! an increase in
the price of Indian exports relative to her competitors prices )deterioration of
terms of trade*.
7
8arlier version of the article was presented as a progress seminar at IIT 2omba! in $eptember 2../.
9
Research :fficer, ;inancial <arkets =epartment, R2I, <umbai and Research $cholar, =epartment of
>umanities and $ocial $ciences, IIT 2omba!, &owai, <umbai.
1
Ragnar Nurkse, ?&atterns of Trade and =evelopment@, E)ilibri" and Growth in the *orld Econo"y,
eds. "ottfried >aberler and Robert <. $tern )#ambridgeA >arvard 4niversit! &ress, 1,/1*.
2
+eport of the I"port and Export 'olicy (o""ittee )New =elhiA <inistr! of #ommerce and Industr!,
1,/2*.
'
$. %. &atel, ?8xport &rospects and 8conomic "rowthA India@, The Econo"ic !ornal % BCIC )$eptember
1,-,*.
+
Trends in International Trade% ( Report b! a &anel of 8xperts )"enevaA "(TT, 1,-3*.
'
The "overnment planners might have felt that a dramatic decline in Indias
foreign exchange reserves was a preDreEuisite for obtaining foreign assistance
called for the $econd ;ive Fear &lan. 8ven if a polic! of export promotion had
been considered generall! desirable, its implementation for traditional Indian
exports conflicted with other economic goals of the Indian "overnment G
increasing the flow of foreign aid, controlling imports, increasing government
revenues, maintaining shortDrun labour emplo!ment, stabilising the domestic price
of important consumer goods, encouraging smallDscale producers who have high
production costs and conserving natural resources. The policies of the Indian
government freEuentl! affected production costs and domestic demand so as to
raise the suppl! price for Indian exports. <oreover, Indias bitter experience of
colonial rule and s!mpath! for socialist beliefs resulted in a cautious polic!
environment where selfDreliance and indigenous efforts were vigorousl!
encouraged b! the government.
In addition, the grand economic theories )H2ig &ush theories and
unbalancedIstrategic growth models* attributed variousl! to RosensteinDRodan
)1,+'*, >arrod )1,',* and =omar )1,+/* and >irschmann )1,-3* led to a
dominant role for state in most areas of industrial activit!. The need to correct the
HantiDexport bias was documented in the Third &lan and export subsidisation was
introduced in 1,/2. <an! economic and political happenings in the late 1,/.s,
however, subseEuentl! stalled the process of opening up of the econom!. 8xport
subsidies were reinstated and even augmented and industrial licensing reverted to
more restrictive mode. (s a result of tightness in the export and import policies of
the government, Indias export share slips down to a level of ..J per cent in 1,J..
=uring the 1,J.s and later, influenced b! the ideas forwarded b! &rebisch )1,J.*
and ;rank )1,/,*, the Indian state eventuall! developed an intricate bod! of rules
and regulations which led to a highl! protected econom! where government
departments displa!ed increasing levels of interventionism in the basic
functioning of the econom!. The state sector grew at the cost of a large private
sector. The ke! outcome was that private industr! lobbied for and received
protection behind tariffs and Euota walls, which ultimatel! undermined the
+
competitiveness of Indian industr! in general and led to highDcost inefficient
production. This was accompanied b! rent seeking behaviour b! agents of state
K2hagwati )1,32*, Lrueger )1,J-* and $rinivasan )1,3-*M. Inspite of this, India
has managed to create a highl! diversified industrial base and it has managed to
develop competencies in a wide range of industrial activities KBall )2..1*M.
India has been described as an Himport substituting countr! par excellence
KRodrik )1,,/A1-*M. ( balance of pa!ments crisis in 1,,1 led to the initiation of
an ongoing process of trade liberalisation. These events corrected the inDbuilt
s!stemic bias against exports and the! have led to a degree of correction of the
price distortions in the Indian econom! through the creation of a more open
econom!. <ore importantl!, increased competition and the presence of firms
from foreign markets has in6ected a greater degree of Eualit! consciousness and
customer orientation, which had hitherto been largel! absent due to the lack of
competitive pressures. In the past there were few foreign firms present in the
protected domestic Indian market. These changes have reduced the tendenc! of
Indian firms to seek and obtain protection from foreign imports. &olic! reform has
also reduced the effectiveness of attempts b! Indian firms to hide behind high
tariff barriers and it has challenged interests that have attempted to perpetuate
inefficient production.
In recent !ears, Indias percentage share in world exports has been
increasing steadil!, though at a slow pace. ;urther, India is building up new areas
of strength in export markets b! moving to computer software exports, exports of
pharmaceuticals and engineering manufactures in addition to traditional export
strengths in gems N 6eweller!, textiles and primar! products KN($$#:<
-
and
="#I$
/
)various issues*M. These events have succeeded in reducing the
ideological opposition to trade which derived in part from Indias colonial
experience combined with an obsession with selfDsufficienc! at an! cost. The
purpose of this paper is to investigate the following h!pothesesA )1* whether
exports, imports and "=& are cointegrated using the %ohansens approachO )2*
-
N($$#:<, $tatistics of the Indian IT Industr! )online*, )downloaded from www.nasscom.org*,
New =elhiA National (ssociation of $oftware and $ervice #ompanies.
/
=irectorate "eneral of #ommercial Intelligence and $tatisticsO :fficial Trade statistics.
-
whether export growth "ranger causes "=& growthO and )'* whether export
growth "ranger causes investment. This paper is structured as follows. $ection 2
provides a review of previous studies as well as a surve! of the work done for the
case of India. $ection ' outlines the data sources and provides a description of the
specific time series investigated in this stud!. It also presents in detail the
methodolog! and formal techniEues emplo!ed in the empirical anal!sis, as well as
the results obtained. $ection + summarises the main findings.
7! ;iterature "eview
7!/ 1he Export-;ed Growth 2E;G3 8ypothesis
There are fair amount of literature on the empirical investigation of the
export led growth )8B"* h!pothesis as well as investigations using "ranger
)1,/,* causalit!. There is the well known argument about the greater
effectiveness of export oriented industrialisation )8:I* K2hagwati )1,32*, Lrueger
)1,J-*, and $rinivasan )1,3-*M as compared to import substituting
industrialisation )I$I* K&rebisch )1,J.*O ;rank )1,/,*O <!rdal
J
)1,-J*M. The
opposing views on trade as an Hengine of growth KBewis )1,3.*M or a
Hhandmaiden of growth KLravis )1,J.*O Riedel )1,3+*M are also well known.
There have been several studies that have found some association between
exports )or export growth* and output )"=&* levels )or output growth*. ;or the
case of developing countries anal!tical work originall! focused on correlations
between exports and income K8mer! )1,/J*, <ai1els )1,/3*, Lravis )1,J.*M,
moving on to studies with limited samples K2alassa )1,J3*M, followed b! studies
focusing on aggregate production functions that included exports as an
explanator! variable K;eder )1,32*M. There have been studies on the existence of a
threshold effect as well KLavoussi )1,3+*, <oschos )1,3,*, Lohli and $ingh
)1,3,*M. These have been supplemented b! causalit! tests K%ung and <arshall
)1,3-*O #how )1,3J*M. The econometric methods emplo!ed in this anal!sis have
been significantl! influenced b! the work of "ranger )1,/,, 1,33*, 8ngle and
J
Econo"ic Theory and ,nderdeveloped +egions, 1,-J.
/
"ranger )1,3J*, %ohansen )1,33, 1,,-*, and %ohansen and %uselius )1,,.*, among
others.
The idea that export growth is one of the ma6or determinants of output
growth Kvi1. the export led growth )8B"* h!pothesisM is a recurrent one. 8xport
growth ma! effect output growth through positive externalities on nonDexports,
through the creation of more efficient management st!les, improved production
techniEues, increased scale economies, improved allocative efficienc!, and better
abilit! to generate d!namic comparative advantage. If there are incentives to
increase investment and improve technolog! this would impl! a productivit!
differential in favour of the export sector )in other words, marginal factor
productivities are expected to be higher in the export sector than in the other
sectors of the econom!*. It is thus argued that an expansion of exports, even at the
cost of other sectors, will have a net positive effect on the rest of the econom!. It
ma! also ease the foreign exchange constraint. There could also be positive
spillover effects on the rest of the econom!. These factors notwithstanding, the
empirical evidence for the 8B" h!pothesis is mixed. Time series evidence fails to
provide uniform support to the 8B" h!pothesis whereas a wide bod! of literature
appl!ing a range of cross section t!pe methodologies strongl! supports an
association between exports and growth. In other words, cross section results
appear to find a close and robust relationship, while time series results are less
conclusive.
$tudies such as %ung and <arshall )1,3-*, #how )1,3J*, >siao )1,3J*,
=arrat )1,3J*, (fxentiou and $erletis )1,,1*, 2ahmaniD:skooee et al )1,,1*,
=odaro )1,,1*, "reenawa! and $apsford )1,,+* and Bove )1,,2* have cast some
doubt on the validit! of the 8B" h!pothesis. :thers such as $erletis )1,,2*,
>enriEue and $adorsk! )1,,/*, 2ahmaniD:skooee and (lse )1,,'*, "hatak et al
)1,,-* and Nidugala )2..1* provide fairl! robust evidence in favour of the 8B"
h!pothesis. <ost of the time series studies emplo! the "ranger method, while
onl! a few studies combine "rangers test with the (kaikes Information
#riterion )(I#* to determine the optimal lag length in the "ranger causalit! test.
J
The latter approach removes the ambiguit! involved in the arbitrar! choice of lag
lengths. ;urther, most studies Kwith exceptions like (fxentiou and $erletis )1,,1*
and 2ahmaniD:skooee and (lse )1,,'*M do not consider whether exports and
income are themselves cointegrated. Thus there ma! not exist a genuine long term
relationship between exports and outputA the results ma! indicate a pure short run
relationship.
7!7 India<s Case
There are a few studies on this sub6ect for the case of India as well.
=hawan and 2iswal )1,,,* investigate the 8B" h!pothesis using a vector
autoregressive )0(R* model b! considering the relationship between real "=&,
real export and terms of trade for India for the period 1,/1 to 1,,'. The! emplo!
a multivariate framework using %ohansens cointegration procedure and find a
longDrun eEuilibrium relationship between the three variables and the causal
relationship flows from the growth in "=& and terms of trade to the growth in
exports. >owever, the! conclude that the causalit! from exports to "=& appears
to be a short run phenomenon. In a similar framework, (safuD(d6a!e et al )1,,,*
consider three variablesA exports, real output and imports )for the period 1,/.D
1,,+*. The! do not find an! evidence of the existence of a causal relationship
between these variables for the case of India and no support for the 8B"
h!pothesis, which is not too surprising given Indias economic histor! and trade
policies. (nwer and $ampath )2..1*, also find evidence against the 8B"
h!pothesis for India. In contrast, Nidugala )2..1* builds on 8sfahanis )1,,1*
model and uses an augmented production function with exports as a regressor.
Nidugala finds evidence in support of the 8B" h!pothesis for the case of India,
particularl! in the 1,3.s. >e finds that export growth had a significant impact on
"=& growth. ;urther, his stud! reveals that growth of manufactured exports had a
significant positive relationship with "=& growth, while the growth of primar!
exports had no such influence. "hatak and &rice )1,,J* test the 8B" h!pothesis
for India for the period 1,/.D1,,2, using exports as regressors and measure of
"=& that nets out exports, along with exports and imports as additional variables.
Their results indicate that real )aggregate* export growth is "rangerDcaused b!
3
nonDexport real "=& growth in India over 1,/.D,2. Their cointegration tests
confirm the long run nature of this relationship. >owever, imports do not appear
to be important for the case of India. (s corroborated subseEuentl! b! Nidugala
)2..1*, their disaggregated anal!sis shows that nonDtraditional manufactured
exports )such as machiner! and transport eEuipment* are found to "ranger cause
output growth, while traditional manufactures )such as textiles, wood, paper* have
little effect.
7!0 Excluding growth accounting effects
In empirical anal!sis of trade data a ma6or problem arises from the fact that
exports are themselves a component of output, via the national income accounting
identit! Ksee <ichael! )1,JJ, 1,J,*, >eller and &orter )1,J3*, ;eder )1,32*,
(fxentiou and $erletis )1,,1*, Bove )1,,2*, 8sfahani )1,,1*, "reenawa! )1,,+*,
"hatak and &rice )1,,J* and $heehe! )1,,.*M. The results of such a model are
likel! to suffer from a simultaneit! bias since export growth ma! itself be a
function of the increase in output. ;ollowing ;eder )1,32*, the framework of the
econom! can be divided into two sectors D export and nonDexport. In this anal!sis,
there has been a separation of economic influence of exports on output from that
incorporated in the HNational Income (ccounting relationship b! using a measure
of "=& )F* that netted out exports )C*.
)1* ..... u C a a F 1 . + + =
Phere Fstands for rates of change of "=& while C stands for export growth.
0 Empirical Analysis
0!/ 'ata (ource
There are two basic sources for data on Indias exports. :ne set is
compiled b! the =irectorate "eneral of #ommercial Intelligence and $tatistics
)="#IN$*, <inistr! of #ommerce, "overnment of India and the other is
compiled b! the Reserve 2ank of India )R2I*. The ="#I$ compiles information
on real transactions, reporting EuantitiesIvolumes of exports as well as export
earnings in Indian rupees. 8xports are decomposed into headings matching with
,
the IT# )>$*
3
$tandard Industrial #lassification )$I#* codes. Thus exports are
broken down b! $I# categories and b! destination. R2I compiles export data on
its balance of pa!ment statistics component b! aggregating the econom! wide
financial transactions related to exports. 8xporters and financial intermediaries
have to provide this information to the R2I b! statute. ="#I$ data has been used
much more freEuentl! in the literature and the R2Is data has been relativel! less
freEuentl! referred to. In this stud!, R2Is data sets used for anal!sis, in part to
correct the above mentioned lacuna and to include invisible receipts and pa!ment
along with current account to capture recent upsurge in services export.
(ccordingl!, the data used in this exercise has been compiled from the Reserve
2ank of Indias H>andbook of $tatistics on the Indian 8conom!, 2..+D.- starting
from 1,J.DJ1 )referred to as 1,J1* to 2..+D.- )referred to as 2..-*. The national
income and trade data at aggregate level have been used in order to facilitate
further disaggregation subseEuentl!. The following time series data at aggregate
level are adapted for the period 1,J1D2..-A
1. "=& at constant market prices )gross domestic product*
2. "=& net of exports
'. Real exports )exports deflated b! the time series of unit price index of exports*
+. Real imports
-. "ross domestic capital ;ormation as prox! for Investment
/. 8mplo!ment in the formal sector
The constant "=& estimates are used in this stud! to permit intertemporal
comparisons. The time series of unit price index of exports is used to deflate the
export series. The prefix HB stands for the natural logarithm of the concerned time
series. (ll econometric estimations in this paper have been carried out using
Eviews -.
The data emplo!ed in this stud! are graphicall! displa!ed in (ppendixA
logarithmic transformations of time series data. In all the cases except "=& and
3
International Trade #lassification )>armonised $!stem*.
1.
"=& without exports, the probabilit! of the %arEueD2erra test statistic provides
evidence in favour of the null h!pothesis of a normal distribution. (dditionall!,
simple correlations are estimated. It is pertinent to note the negative correlations
between emplo!ment and all economic variables )income, income without
exports, real export and investment*.
0!7 =nit "oots and Cointegration
In investigating the export led growth )8B"* h!pothesis, the traditional
approach of first differencing disregards potentiall! important eEuilibrium
relationships among the levels of the series to which the h!potheses of economic
theor! usuall! appl! K8ngle and "ranger )1,3J*M. The first test for an ordinar!
unit root D single series unit root tests include (ugmented =icke!D;uller )(=;*.
Table 1 summarises the results for unit root tests on levels and in first differences
of the data. $trong evidence emerges that all the time series are I )1*. In Table 1,
for the (ugmented =ick! =icke!D;uller test, the lag length is based on the
$chwar1 Information #riterion, while for the &hillipsD&erron 4nit Root Test
bandwidth selection is based on Newe!DPest.
1able /: =nit "oot 1ests
Bevel (=;
test
$tatistics
Bevel &&
Test
$tatistics
;irst =ifference
(=; Test
$tatistics
;irst
=ifference &&
Test $tatistics
"=& without
8xports
D..J.+ D....1 D2.-., D/.''J
8xports J.,+/ 2,.'/+ 1.//+ D1.'31
Imports 1..21/ +..'3 1.+22 D..'-3
Investment +.11+ 1-.'3/ ..J1, D+.+2-
8mplo!ment D'.+.J D2.J1J D'..2- D2.,+,
1Q critical value D'./', D+.2-2 D+.2/2 D2.1'.
1Q critical value
)Trend and
Intercept*
D2..+, D2..+, D+.2J' D+.2/'
;ollowing a multivariate approach the estimation proceed with considering
the cointegration h!pothesis between output )"=&*, real exports and real imports.
11
These variables have been chosen for anal!sis for three reasons. ;irst, Rie1mann
et al )1,,/* have suggested that imports are an important variable while
considering causalit! between exports and growth, and omission of imports could
lead to biased results
,
. $econdl!, testing the 8B" h!pothesis is an explicit
ob6ective and the chosen variables seem to be appropriate for such an exercise.
;inall!, given the set of variables for which time series data is available for India,
both investments and emplo!ment seem less appropriate. In the case of
investments, foreign direct investment is excluded and the series are thus
underestimates for total investment in India, especiall! during the 1,,.s.
8mplo!ment data is ambiguous because of definitional and reporting issues. <ore
seriousl!, with a large informal, unorganised sector emplo!ment data also suffers
from underestimation problems. In this anal!sis, two cases are considered. ;irst,
using the %ohansen method, the test has been done whether there is a cointegrating
relationship between exports, imports and "=&. $econdl!, estimation used b!
considering the case of exports, imports and "=& net of exports in order to avoid
the Haccounting effect.
1able 7: 4ohansen Cointegration 1est >lnG'P, lnExport, lnImport?
r 8igen
0alue
Trace
$tatistics
-Q #0
&rob
?
<axD
8igen
$tatistic
-Q #0 &rob
None ..',+''J 2/.3323. 2,.J,J.J 1/.-+J2+ 21.1'1/2 ..1,+-
(t most 1 ..2.2-32 1..''--- 1-.+,+J1 J.+J.+2J 1+.2/+/. ..+'-1
(t most 2 ...3'1/. 2.3/-12+ '.3+1+// 2.3/-12+ '.3+1+// ...,.-
The results for the first case are presented in Table 2. Then repeat the test
replacing "=& with "=& less exports )Table '*. (s Tables 2 and ' shows, the null
h!pothesis can not be re6ected at the -Q significance level. The %ohansen procedure, like
man! others, reEuires estimation of various structural and nuisance parameters. ;or
,
Inclusion of imports )along with exports and output* in this anal!sis allows to examine the notion that
imports relieve the foreign exchange constraint that developing countries often face. This referred to as
import compression K)8sfahani)1,,1*M.
7
<ackinnonD>augD<ichelis )1,,,* pDvalues.
12
example, a vector autoregressive )0(R* log order must be specified and then the lag
parameters are estimated.
1able 0: 4ohansen Cointegration 1est >lnG'Pless export, lnExport, lnImport?
r 8igen
0alue
Trace
$tatistics
-Q #0
&rob
<axD
8igen
$tatistic
-Q #0 &rob
None ..'//,/- 1,.++J,J 2,.J,J.J 1-..33-, 21.1'1/2 ..2323
(t most 1 ...JJ,,, +.'-,'J- 1-.+,+J1 2./J,,1+ 1+.2/+/. ..,/-3
(t most 2 ...+,/1, 1./J,+/1 '.3+1+// 1./J,+/1 '.3+1+// ..1,-.
0!0 Granger causality
#orrelation does not necessaril! impl! causation in an! meaningful sense of the
word. The econometric anal!sis has full of magnificent correlations, which are simpl!
spurious or meaningless. To investigate the causalit! between "=& )and "=& less
exports* on the one hand and exports on the other, simple "ranger causalit! test has been
performed b! estimating the bivariate autoregressive processes for "=& )and "=& less
exports* and exports. The ob6ective of this exercise is to test the export led growth )8B"*
h!pothesis for India empiricall!. ;urthermore, building on preceding anal!sis, it can be
argued that export growth can stimulate investments )gross domestic capital formation*,
especiall! if there exists a productivit! differential between the export sector and the
nonDexport sector. In such cases, investment would be expected to increase in those
sectors of the econom! where productivit! and returns are higher )the export sector*.
8Euall! well, it is theoreticall! plausible to expect the reverseA the case where increased
investment would also stimulate export growth. Phether investments are in social
overhead capital )infrastructure* or in specific industries, there could be an overall
beneficial effect of investments on exports. 8mpirical anal!sis of the h!pothesis has been
presented in Table 3.
Thus the estimated eEuationA
Ryt S a. T a1Ryt.1 T ... T a1Ryt.1 T b1Rxt.1 T ... T blxt.l ///..)-*
Rxt S a. T a1Rxt.1 T ... T a1Rxt.1 T b1Ryt.1 T ... T blyt.l ///.. )/*
1'
The reported ;Dstatistics are the Pald statistics for the 6oint h!pothesisA
b1 S ... S bl S . UUU )J*
The null h!pothesis is therefore that C does not "rangerDcause F in the first regression
and that F does not "rangerDcause C in the second regression.
1able -: Granger Causality 2"eal Export: G'P net Export3
Null >!pothesisA :bs ;D$tatistic &robabilit!
R8(B8C& does not "ranger #ause "=&N8T8C& '+ -.J'+// .!.77@9
"=&N8T8C& does not "ranger #ause R8(B8C& '+ ....1,' ..,/-2.
1able 5: Granger Causality 2"eal Export: G'P3
Null >!pothesisA :bs ;D$tatistic &robabilit!
R8(B8C& does not "ranger #ause "=& '+ ../+/'3 ..+2J-'
"=& does not "ranger #ause R8(B8C& '+ ..12+22 ..J2/3,
1able 9: Granger Causality 2"eal Export: "eal Investment3
Null >!pothesisA :bs ;D$tatistic &robabilit!
IN08$T does not "ranger #ause R8(B8C& '+ ..-+21. ..+/J1.
R8(B8C& does not "ranger #ause IN08$T '+ +.32-3/ .!.059-
In all the cases of Tables +, - and /, the reported probabilities are greater than
...- except for real export and "=& net of export and Investment. $ince annual
observations have been used, onl! one lag is emplo!ed. The h!pothesis that exports
"ranger cause investment )or vice versa* can not be full! re6ected. (t this significance
level, we could marginall! accept the h!pothesis that growth in investment and income
would "ranger causes growth in real exports. The evidence in this section does not
provide an! support for the causalit! relationship between exports and output )"=&*.
There is weak evidence suggesting that the direction of causalit! runs from "=& to
1+
exports, which further strengthens the case against the 8B" h!pothesis for the case of
India.
-! Conclusion
This stud! has examined export led growth )8B"* h!pothesis for India
using different approaches b! emplo!ing data at the aggregate level covering the
postDliberalisation period. It presents a more upDtoDdate test of the 8B" h!pothesis
for India and able to capture the effects of liberalisation on exports and output
growth. 2! emplo!ing cointegration h!pothesis )following the %ohansen method*
for the case of India, the stud! investigate the following h!pothesesA )i* whether
exports, imports and "=& are cointegrated using the %ohansen approach, )ii*
whether export growth "ranger causes "=& growth, and )iii* and whether export
growth "ranger causes investment. ;or the first two cases, strong evidence is
found against the cointregration h!pothesis. In this anal!sis, it also fail to find
support for the h!pothesis that exports "ranger cause "=&, using two measures
for "=& )"=& with exports and "=& without exports*. The same holds for the
relationship between exports and investment. This strengthens the argument
against the 8B" h!pothesis for the case of India and strengthens the argument that
inspite of reforms, it still retains some characteristics of an import substituting
econom!. $ince aggregate export data for India includes sectors such as software
exports, the euphoria about perceived successes in the I#T )information and
communication technolog!* sector for India seem somewhat premature, given that
at an aggregated level there is little evidence to support the export led growth
h!pothesis, which brings into doubt the implicitl! assumed productivit!
differentials and resulting positive spillovers into the rest of the econom!.
VVVVVVVVVVV
*ibliography and "eferences
1-
(fxentiou, & # and ( $erletis. )1,,1*, H8xports and "N& #ausalit! in the
Industrial #ountriesA 1,-.D1,3-, L!klos, ++, 2, pp. 1/JDJ,.
(hluwalia, <ontek $. )2..2*A ?8conomic Reforms in India since 1,,1A >as
"radualism PorkedW@, !ornal of Econo"ic 'erspectives, 0ol. 1/, No J, pp. /JD
33.
(lston, Richard <.O %. R. Learl, and <ichael 2. 0aughan. )1,,2*A XIs There a
#onsensus (mong 8conomists in the 1,,.YsWX 0"er. Econ. +ev. )&apers and
&roceedings*, 32A2, pp. 1-3DJJ.
(nwer, < $ and R L $ampath. )2..1*, H8xports and 8conomic "rowth, Indian
8conomic %ournal, +J, ', pp. J,D33.
(safuD(d6a!e, % and = #hakrabort!. )1,,,*, H8xportDled "rowth and Import
#ompressionA ;urther Time $eries 8vidence ;rom B=#s, (ustralian 8conomic
&apers, '3, pp. 1/+DJ-.
2ahmaniD:skoee, < and % (lse. )1,,'*, H8xport "rowth and 8conomic "rowthA
(n (pplication of #ointegration and 8rrorD#orrection <odelling, %ournal of
=evelopment (reas, 2J, %ul, pp. -'-D+2.
2ahmaniD:skoee, <, > <ohtadi and " $habsigh. )1,,1*, H8xports, "rowth and
#ausalit! in B=#sA ( ReDexamination, %ournal of =evelopment 8conomics, '/,
pp. +.-D1-.
2alassa, 2. )1,J3*, H8xports and 8conomic "rowthA ;urther 8vidence, !ornal
of &evelop"ent Econo"ics, -, pp. 131D3,.
2haduri, (mit and =eepak Na!!ar )1,,/*A ?The Intelligent 'erson1s Gide to
Liberalisation@, &enguin 2ooks, New =elhi.
2hagwati, %. )1,32*, H=irectl! unproductive profit seeking )=4&* activities,
%ournal of &olitical 8conom!, ,., -, pp. ,33D1..2.
2reitung, %. )2..2*A HNonparametric Tests for 4nit Roots and #ointegration,
!ornal of Econo"etrics, 1.3, '+'D'/+.
#how, & # F. )1,3J*, H#ausalit! 2etween 8xport "rowth and Industrial
=evelopmentA 8mpirical 8vidence from the NI#s, !ornal of &evelop"ent
Econo"ics, 2/, pp. --D/'.
#ohen, 2en6amin I. )1,/+*A ?The $tagnation of Indian 8xports, 1,-1D/1@, The
2arterly !ornal of Econo"ics, 0ol. J3, No. +, &&. /.+D/2..
=arrat, (li ;. )1,3J*, H(re exports engine of growthW (nother look at the
evidence, (pplied 8conomics, 1,, pp. 2JJD3'.
=hawan, 4 and 2 2iswal. )1,,,*, HReDexamining export led growth h!pothesisA a
multivariate cointegration anal!sis for India, (pplied 8conomics, '1, pp. -2-D'..
=odaro, $anto. )1,,1*, H#omparative (dvantage, Trade and "rowthA 8xportDBed
"rowth Revisited, Porld =evelopment, 1,, ,, pp. 11-'D/-.
=omar, 8 =. )1,+/*, H#apital 8xpansion, Rate of "rowth and 8mplo!ment,
Econo"etrica, 1+, 2, pp. 1'JD+J.
1/
8ngle, R ; and # P % "ranger. )1,3J*, H#oDintegration and 8rror #orrectionA
Representation, 8stimation and Testing, Econo"etrica, --, pp. 2-1DJ/.
8sfahani, > $. )1,,1*, H8xports, imports, and economic growth in semiD
industriali1ed countries, %ournal of =evelopment 8conomics, '-, pp. ,'D11/.
;eder, "ershon. )1,32*, H:n exports and economic growth, !ornal of
&evelop"ent Econo"ics, 12, pp. -,DJ'.
;rank, (ndre "under. )1,/,*, Batin (mericaA 4nderdevelopment or RevolutionW,
New ForkA <onthl! Review &ress.
"hatak, $ and $ P &rice. )1,,J*, H8xport #omposition and 8conomic "rowthA
#ointegration and #ausalit! 8vidence for India, Peltwirtschaftliches (rchiv,
1'', ', pp. -'3D-'.
"ranger, # P %. )1,/,*, HInvestigating #ausal Relationships b! 8conometric
<odels and #rossDspectral <ethods, Econo"etrica, 'J, ', pp. +2+D'3.
"ranger, # P %. )1,33*, H$ome Recent =evelopments in a #oncept of #ausalit!,
!ornal of Econo"etrics, ',, pp. 1,,D211.
"reenawa!, = and = $apsford. )1,,+*, HPhat does Biberalisation do for 8xports
and "rowth, Peltwirtschaftliches (rchiv, 1'., 1, pp. 1-2DJ+.
>arrod, R ;. )1,',*, H(n 8ssa! in =!namic Theor!, Econo"ic !ornal, +,, 1,',
pp. 1D''.
>eller, & $ and R # &orter. )1,J3*, H8xports and "rowthA (n empirical reD
investigation, %ournal of =evelopment 8conomics, -, 2, pp. 1,1D,'.
>irschmann, (. :. )1,-3*, The $trateg! of 8conomic =evelopment, New >avenA
Fale 4niversit! &ress.
>siao, < # P. )1,3J*, HTests of #ausalit! and 8xogeneit! between 8xports and
8conomic "rowthA The #ase of (sian NI#s, %ournal of 8conomic =evelopment,
12, 2, pp. 1+'D-,.
%ohansen, $. )1,33*, H$tatistical (nal!sis of #ointegration 0ectors, !ornal of
Econo"ic &yna"ics and (ontrol, 12, pp. 2'1D-+.
%ohansen, $. )1,,-*, BikelihoodD2ased Inference in #ointegrated 0ector
(utoregressive <odels, :xford 4niversit! &ress, :xford.
%ohansen, $ and L %uselius. )1,,.*, H<aximum Bikelihood 8stimation and
Inference on #ointegration D Pith (pplications for the =emand for <one!,
3xford 4lletin of Econo"ics and Statistics, -2, 2, pp. 1/,D21..
%ung, P. $. and &. %. <arshall. )1,3-*, H8xports, "rowth and #ausalit! in
=eveloping #ountries, !ornal of &evelop"ent Econo"ics, 13, pp. 1D12.
Lavoussi, R <. )1,3+*, H8xport 8xpansion and 8conomic "rowthA ;urther
8mpirical 8vidence, !ornal of &evelop"ent Econo"ics, 1+, pp. 2+1D-..
Lohli, I and N $ingh. )1,3,*, H8xports and "rowthA #ritical <inimum 8ffort and
=iminishing Returns, !ornal of &evelop"ent Econo"ics, '., pp. ',1D+...
1J
Lrueger, ( :. )1,J-*, The 2enefits and #osts of Import $ubstitution in IndiaA (
<icroeconomic $tud!, <inneapolis A 4niversit! of <innesota &ress.
Lrueger, (nne :. )1,,.*A ?(sian Trade and "rowth Bessons@. The 0"erican
Econo"ic +eview. )&apers and &roceedings*, 0ol. 3.A2, pp. 1.3D112.
Ball, $an6a!a. )2..1*, H#ompetitiveness, Technolog! and $kills, #heltanhamA
8dward 8lgar.
Bove, %. )1,,2*, H8xport Instabilit! and the =omestic 8conom!A Zuestions of
#ausalit!, %ournal of =evelopment $tudies, 23, +, pp. J'-D+2.
<ichael!, <ichael. )1,JJ*, H8xports and "rowthA (n 8mpirical Investigation,
%ournal of =evelopment 8conomics, +, pp. +,D-'
<ichael!, <. )1,J,*, H8xports and "rowthA ( Repl!, %ournal of =evelopment
8conomics, /, pp. 1+1D+'.
<oschos, =. )1,3,*, H8xport 8xpansion, "rowth and the Bevel of 8conomic
=evelopmentA (n 8mpirical (nal!sis, !ornal of &evelop"ent Econo"ics, '.,
pp. ,'D1.2.
Nidugala, " L. )2..1*, H8xports and 8conomic "rowth in IndiaA (n 8mpirical
Investigation, Indian 8conomic %ournal, +J, ', pp. /JDJ3.
&rebisch, R. )1,J.*, HTowards a New Trade &olic! for =evelopment, report b!
the $ecretar! "eneral of 4N#T(=, 4nited Nations, 1,/+. Reprinted in ".<.
<eier, Beading Issues in 8conomic =evelopmentA $tudies in international
povert!, $econd 8dition, New ForkA :xford 4niversit! &ress.
Rie1man, R, Phiteman # > and & < $ummers. )1,,/*, HThe 8ngine of "rowth or
Its >andmaidenW ( TimeD$eries (ssessment of 8xportDBed "rowth, 8mpirical
8conomics, 21, pp. JJD11..
Rodrik, =ani. )1,,/*A ?4nderstanding economic polic! reform. !ornal of
Econo"ic Literatre, '+, pp. ,D+1.
RosensteinDRodan, & N. )1,+'*, H&roblems of Industrialisation of 8astern and
$outhD8astern 8urope, Econo"ic !ornal, -', %uneD$ept, pp. 2.2D11.
$erletis, (postolos. )1,,2*, H8xport growth and #anadian economic
development, %ournal of =evelopment 8conomics, '3, pp. 1''D+-.
$heehe!, 8 %. )1,,.*, H8xports and "rowthA ( ;lawed ;ramework, %ournal of
=evelopment $tudies, 2J, pp. 111D1/.
$ims, # (. )1,J2*, H<one!, Income and #ausalit!, (merican 8conomic Review, /2, +,
pp. -+.D-2.
$ingh, <anmohan, )1,/+*A ?India1s Export Trends@, :xfordA #larendon &ress.
$rinivasan, T N. )1,3-*, HNeoclassical &olitical 8conom!, the $tate and 8conomic
=evelopment, (sian =evelopment Review, ', +, pp. '3D-3.
Porld 2ank, )1,,/*A ?India$ #ive 5ears of Stabili6ation and +efor"s and the (hallenges
0head7% Porld 2ank #ountr! $tud!A Pashington, =.#.
Porld Trade :rganisation, )2..-*A ?International Trade Statistics7, "eneva.
13
[[[[[[ )2..2*A ?Trade 'olicy +eview$ India7, )1,,3 and 2..2*, "eneva.
[[[[[[ )2..-*A 8*orld Trade report1, "eneva.
1,
Appendix 1able /: IndiaAs (hare of 1otal Borld Exports
2 In =( C *illion3
Dear Borld Exports Indian Exports (hare 2E3
1,+3 -3.,+ 1.'. 2.2.
1,+, /..++ 1.2. 1.,3
1,-. /2.-- 1.1+ 1.3'
1,-1 J,.1+ 1./. 2..2
1,-2 J'.,+ 1.2+ 1./3
1,-' J'.3. 1.1. 1.+,
1,-+ JJ..3 1.1J 1.-2
1,-- ,../- 1.2/ 1.',
1,-/ 1...+J 1.2/ 1.2-
1,-J 1.,.'1 1.'3 1.2/
1,-3 1.-.-' 1.22 1.1-
1,-, 112.22 1.'2 1.13
1,/. 12+.1. 1.'' 1..J
1,/1 12,.2+ 1.', 1..J
1,/2 1'-.+/ 1.+. 1..+
1,/' 1+3.+' 1./' 1.1.
1,/+ 1//.'. 1.J. 1..'
1,/- 13...1 1./, ..,+
1,// 1,J.1J 1.,- ..,,
1,/J 2.J.2+ 1./1 ..J3
1,/3 22J.2- 1.J/ ..J3
1,/, 2-,.1/ 1.3+ ..J1
1,J. 2,,.+- 2..' ../3
1,J1 ''-..2 2..+ ../1
1,J2 ',+.-- 2.+- ../2
1,J' --..J, 2.,2 ..-'
1,J+ 31-./, '.,' ..+3
1,J- 322.13 +.'/ ..-'
1,J/ ,2J.-2 -.-- ../.
1,JJ 1.-1..' /.'3 ../1
1,J3 121-.13 /./J ..--
1,J, 1-J+.11 J.31 ..-.
1,3. 1332.'2 3.-, ..+/
1,31 13,-.2+ 3.'. ..++
1,32 1J+,.1, ,.'/ ..-'
1,3' 1J.-.J' ,.1- ..-+
1,3+ 131'.23 ,.+- ..-2
1,3- 13'J.+3 ,.1+ ..-.
1,3/ 1,,..3- ,.+. ..+J
1,3J 2'-..32 11.'. ..+3
1,33 2/,1./J 1'.2' ..+,
1,3, 2,2,.,J 1-.3J ..-+
2.
1,,. ''23..1 1J.,J ..-+
1,,1 '+1J.,. 1J.J' ..-2
1,,2 '/+..J' 1,./' ..-+
1,,' '/'-.,' 21.-J ..-,
1,,+ +1',.3- 2-..2 ../.
1,,- +,+/.'+ '../' ../2
1,,/ -1/3.2, ''.11 ../+
1,,J --'3..3 '-..1 ../'
1,,3 -+-../J ''.++ ../1
1,,, -/++./J '-./J ../'
2... /'JJ.// +2.'3 ..//
2..1 /1'2..J +'.'/ ..J1
2..2 /+23.-, +,.2- ..JJ
2..' J+/-.'/ -J.., ..J/
2..+ ,./J.+' J1.J, ..J,
2..- 1.1,/.J. 3-.,' ..3+
21
3.0
3.2
3.4
3.6
3.8
4.0
4.2
4.4
1975 1980 1985 1990 1995 2000 2005
LEMP
12.4
12.8
13.2
13.6
14.0
14.4
1975 1980 1985 1990 1995 2000 2005
LGDP
12.6
12.8
13.0
13.2
13.4
13.6
13.8
14.0
1975 1980 1985 1990 1995 2000 2005
LGDPNETEXP
10.8
11.2
11.6
12.0
12.4
12.8
13.2
1975 1980 1985 1990 1995 2000 2005
LINVEST
8.0
8.5
9.0
9.5
10.0
10.5
11.0
11.5
12.0
1975 1980 1985 1990 1995 2000 2005
LREALEXP
8.5
9.0
9.5
10.0
10.5
11.0
11.5
12.0
1975 1980 1985 1990 1995 2000 2005
LREALIMP
22
doc_617151740.doc
WEATHER MAKERS PRESENTS RAINWATER MINERAL WATER WANTED FRENDS TO JOIN THIS VENTURE
Export and Economic Growth of India -
A Cointegration and Causality Evidence
Prepared by:
! C! Pradhan
"esearch #fficer
$inancial %ar&ets 'epartment
"E(E")E *A+ #$ I'IA
Central #ffice, $ort
%umbai - -..../
for presentation at the -0rd Annual Conference of 1he Indian Econometric
(ociety 21IE(3 to be held during 4anuary 5-6, 7..6 at the 'epartment of
8umanities and (ocial (ciences, II1 *ombay, Powai, %umbai - -....69 !
Abstract
The export led growth (ELG) hypothesis for India is tested with the data at
aggregate level covering post-liberalisation period. Stdy presents a test of the
ELG hypothesis by adopting co-integration hypothesis (!ohansen "ethod) for the
case of India. #ollowing hypotheses are tested$ (i) whether exports% i"ports and
G&' are cointegrated% (ii) whether export growth Granger cases G&' growth%
and (iii) and whether export growth Granger cases invest"ent. The finding
strengthens the arg"ent against the ELG hypothesis for the case of India and
strengthens the arg"ent that inspite of refor"s% it still retains so"e
characteristics of an i"port sbstitting econo"y. Since aggregate export data
for India incldes sectors% sch as% software exports% the ephoria abot
perceived sccesses in the I(T (infor"ation and co""nication technology)
sector for India see" so"ewhat pre"atre% given that at an aggregated level
there is little evidence to spport the export led growth hypothesis% which brings
into dobt the i"plicitly ass"ed prodctivity differentials and reslting positive
spillovers into the rest of the econo"y.
2
Export and Economic Growth of India -
A Cointegration and Causality Evidence
?
! C! Pradhan
:
The independent India had faced foreign exchange problem that leads to
export stagnation, as was happened in other underdeveloped countries. In this
context, Ragnar Nurkse
1
observed that, stagnation in underdeveloped countries
export was mainl! attributed to slowl! expanding world markets for traditional
products. The same argument was also corroborated b! Indian "overnment
#ommittee
2
and $. %. &atel
'
. The "eneral (greement on Tariff and Trade )"(TT*
report
+
illustrated the phenomenon of stagnant world export during 1,-.s mainl!
due to relativel! inelastic suppl! of exports as a result of slowl! expanding
domestic output and rapidl! expanding domestic demand in underdeveloped
countries, including India. In 1,/1, India was third onl! to 0ene1uela and 2ra1il
as an exporter among the underdeveloped countries )#ohen, 1,/+*. 2etween 1,+3
and 1,/1, while annual value of Indian exports had fluctuated around 4$ 5 1.'
billion, the world export had more than doubled from 4$ 5 -, billion to 4$ 5 12,
billion. The share of exports declined continuousl! from 2.2 per cent to 1.1 per
cent during the period )(ppendix Table 1*. 2en6amin #ohen )1,/+* had examined
the situation to conclude that declining market share was caused b! an increase in
the price of Indian exports relative to her competitors prices )deterioration of
terms of trade*.
7
8arlier version of the article was presented as a progress seminar at IIT 2omba! in $eptember 2../.
9
Research :fficer, ;inancial <arkets =epartment, R2I, <umbai and Research $cholar, =epartment of
>umanities and $ocial $ciences, IIT 2omba!, &owai, <umbai.
1
Ragnar Nurkse, ?&atterns of Trade and =evelopment@, E)ilibri" and Growth in the *orld Econo"y,
eds. "ottfried >aberler and Robert <. $tern )#ambridgeA >arvard 4niversit! &ress, 1,/1*.
2
+eport of the I"port and Export 'olicy (o""ittee )New =elhiA <inistr! of #ommerce and Industr!,
1,/2*.
'
$. %. &atel, ?8xport &rospects and 8conomic "rowthA India@, The Econo"ic !ornal % BCIC )$eptember
1,-,*.
+
Trends in International Trade% ( Report b! a &anel of 8xperts )"enevaA "(TT, 1,-3*.
'
The "overnment planners might have felt that a dramatic decline in Indias
foreign exchange reserves was a preDreEuisite for obtaining foreign assistance
called for the $econd ;ive Fear &lan. 8ven if a polic! of export promotion had
been considered generall! desirable, its implementation for traditional Indian
exports conflicted with other economic goals of the Indian "overnment G
increasing the flow of foreign aid, controlling imports, increasing government
revenues, maintaining shortDrun labour emplo!ment, stabilising the domestic price
of important consumer goods, encouraging smallDscale producers who have high
production costs and conserving natural resources. The policies of the Indian
government freEuentl! affected production costs and domestic demand so as to
raise the suppl! price for Indian exports. <oreover, Indias bitter experience of
colonial rule and s!mpath! for socialist beliefs resulted in a cautious polic!
environment where selfDreliance and indigenous efforts were vigorousl!
encouraged b! the government.
In addition, the grand economic theories )H2ig &ush theories and
unbalancedIstrategic growth models* attributed variousl! to RosensteinDRodan
)1,+'*, >arrod )1,',* and =omar )1,+/* and >irschmann )1,-3* led to a
dominant role for state in most areas of industrial activit!. The need to correct the
HantiDexport bias was documented in the Third &lan and export subsidisation was
introduced in 1,/2. <an! economic and political happenings in the late 1,/.s,
however, subseEuentl! stalled the process of opening up of the econom!. 8xport
subsidies were reinstated and even augmented and industrial licensing reverted to
more restrictive mode. (s a result of tightness in the export and import policies of
the government, Indias export share slips down to a level of ..J per cent in 1,J..
=uring the 1,J.s and later, influenced b! the ideas forwarded b! &rebisch )1,J.*
and ;rank )1,/,*, the Indian state eventuall! developed an intricate bod! of rules
and regulations which led to a highl! protected econom! where government
departments displa!ed increasing levels of interventionism in the basic
functioning of the econom!. The state sector grew at the cost of a large private
sector. The ke! outcome was that private industr! lobbied for and received
protection behind tariffs and Euota walls, which ultimatel! undermined the
+
competitiveness of Indian industr! in general and led to highDcost inefficient
production. This was accompanied b! rent seeking behaviour b! agents of state
K2hagwati )1,32*, Lrueger )1,J-* and $rinivasan )1,3-*M. Inspite of this, India
has managed to create a highl! diversified industrial base and it has managed to
develop competencies in a wide range of industrial activities KBall )2..1*M.
India has been described as an Himport substituting countr! par excellence
KRodrik )1,,/A1-*M. ( balance of pa!ments crisis in 1,,1 led to the initiation of
an ongoing process of trade liberalisation. These events corrected the inDbuilt
s!stemic bias against exports and the! have led to a degree of correction of the
price distortions in the Indian econom! through the creation of a more open
econom!. <ore importantl!, increased competition and the presence of firms
from foreign markets has in6ected a greater degree of Eualit! consciousness and
customer orientation, which had hitherto been largel! absent due to the lack of
competitive pressures. In the past there were few foreign firms present in the
protected domestic Indian market. These changes have reduced the tendenc! of
Indian firms to seek and obtain protection from foreign imports. &olic! reform has
also reduced the effectiveness of attempts b! Indian firms to hide behind high
tariff barriers and it has challenged interests that have attempted to perpetuate
inefficient production.
In recent !ears, Indias percentage share in world exports has been
increasing steadil!, though at a slow pace. ;urther, India is building up new areas
of strength in export markets b! moving to computer software exports, exports of
pharmaceuticals and engineering manufactures in addition to traditional export
strengths in gems N 6eweller!, textiles and primar! products KN($$#:<
-
and
="#I$
/
)various issues*M. These events have succeeded in reducing the
ideological opposition to trade which derived in part from Indias colonial
experience combined with an obsession with selfDsufficienc! at an! cost. The
purpose of this paper is to investigate the following h!pothesesA )1* whether
exports, imports and "=& are cointegrated using the %ohansens approachO )2*
-
N($$#:<, $tatistics of the Indian IT Industr! )online*, )downloaded from www.nasscom.org*,
New =elhiA National (ssociation of $oftware and $ervice #ompanies.
/
=irectorate "eneral of #ommercial Intelligence and $tatisticsO :fficial Trade statistics.
-
whether export growth "ranger causes "=& growthO and )'* whether export
growth "ranger causes investment. This paper is structured as follows. $ection 2
provides a review of previous studies as well as a surve! of the work done for the
case of India. $ection ' outlines the data sources and provides a description of the
specific time series investigated in this stud!. It also presents in detail the
methodolog! and formal techniEues emplo!ed in the empirical anal!sis, as well as
the results obtained. $ection + summarises the main findings.
7! ;iterature "eview
7!/ 1he Export-;ed Growth 2E;G3 8ypothesis
There are fair amount of literature on the empirical investigation of the
export led growth )8B"* h!pothesis as well as investigations using "ranger
)1,/,* causalit!. There is the well known argument about the greater
effectiveness of export oriented industrialisation )8:I* K2hagwati )1,32*, Lrueger
)1,J-*, and $rinivasan )1,3-*M as compared to import substituting
industrialisation )I$I* K&rebisch )1,J.*O ;rank )1,/,*O <!rdal
J
)1,-J*M. The
opposing views on trade as an Hengine of growth KBewis )1,3.*M or a
Hhandmaiden of growth KLravis )1,J.*O Riedel )1,3+*M are also well known.
There have been several studies that have found some association between
exports )or export growth* and output )"=&* levels )or output growth*. ;or the
case of developing countries anal!tical work originall! focused on correlations
between exports and income K8mer! )1,/J*, <ai1els )1,/3*, Lravis )1,J.*M,
moving on to studies with limited samples K2alassa )1,J3*M, followed b! studies
focusing on aggregate production functions that included exports as an
explanator! variable K;eder )1,32*M. There have been studies on the existence of a
threshold effect as well KLavoussi )1,3+*, <oschos )1,3,*, Lohli and $ingh
)1,3,*M. These have been supplemented b! causalit! tests K%ung and <arshall
)1,3-*O #how )1,3J*M. The econometric methods emplo!ed in this anal!sis have
been significantl! influenced b! the work of "ranger )1,/,, 1,33*, 8ngle and
J
Econo"ic Theory and ,nderdeveloped +egions, 1,-J.
/
"ranger )1,3J*, %ohansen )1,33, 1,,-*, and %ohansen and %uselius )1,,.*, among
others.
The idea that export growth is one of the ma6or determinants of output
growth Kvi1. the export led growth )8B"* h!pothesisM is a recurrent one. 8xport
growth ma! effect output growth through positive externalities on nonDexports,
through the creation of more efficient management st!les, improved production
techniEues, increased scale economies, improved allocative efficienc!, and better
abilit! to generate d!namic comparative advantage. If there are incentives to
increase investment and improve technolog! this would impl! a productivit!
differential in favour of the export sector )in other words, marginal factor
productivities are expected to be higher in the export sector than in the other
sectors of the econom!*. It is thus argued that an expansion of exports, even at the
cost of other sectors, will have a net positive effect on the rest of the econom!. It
ma! also ease the foreign exchange constraint. There could also be positive
spillover effects on the rest of the econom!. These factors notwithstanding, the
empirical evidence for the 8B" h!pothesis is mixed. Time series evidence fails to
provide uniform support to the 8B" h!pothesis whereas a wide bod! of literature
appl!ing a range of cross section t!pe methodologies strongl! supports an
association between exports and growth. In other words, cross section results
appear to find a close and robust relationship, while time series results are less
conclusive.
$tudies such as %ung and <arshall )1,3-*, #how )1,3J*, >siao )1,3J*,
=arrat )1,3J*, (fxentiou and $erletis )1,,1*, 2ahmaniD:skooee et al )1,,1*,
=odaro )1,,1*, "reenawa! and $apsford )1,,+* and Bove )1,,2* have cast some
doubt on the validit! of the 8B" h!pothesis. :thers such as $erletis )1,,2*,
>enriEue and $adorsk! )1,,/*, 2ahmaniD:skooee and (lse )1,,'*, "hatak et al
)1,,-* and Nidugala )2..1* provide fairl! robust evidence in favour of the 8B"
h!pothesis. <ost of the time series studies emplo! the "ranger method, while
onl! a few studies combine "rangers test with the (kaikes Information
#riterion )(I#* to determine the optimal lag length in the "ranger causalit! test.
J
The latter approach removes the ambiguit! involved in the arbitrar! choice of lag
lengths. ;urther, most studies Kwith exceptions like (fxentiou and $erletis )1,,1*
and 2ahmaniD:skooee and (lse )1,,'*M do not consider whether exports and
income are themselves cointegrated. Thus there ma! not exist a genuine long term
relationship between exports and outputA the results ma! indicate a pure short run
relationship.
7!7 India<s Case
There are a few studies on this sub6ect for the case of India as well.
=hawan and 2iswal )1,,,* investigate the 8B" h!pothesis using a vector
autoregressive )0(R* model b! considering the relationship between real "=&,
real export and terms of trade for India for the period 1,/1 to 1,,'. The! emplo!
a multivariate framework using %ohansens cointegration procedure and find a
longDrun eEuilibrium relationship between the three variables and the causal
relationship flows from the growth in "=& and terms of trade to the growth in
exports. >owever, the! conclude that the causalit! from exports to "=& appears
to be a short run phenomenon. In a similar framework, (safuD(d6a!e et al )1,,,*
consider three variablesA exports, real output and imports )for the period 1,/.D
1,,+*. The! do not find an! evidence of the existence of a causal relationship
between these variables for the case of India and no support for the 8B"
h!pothesis, which is not too surprising given Indias economic histor! and trade
policies. (nwer and $ampath )2..1*, also find evidence against the 8B"
h!pothesis for India. In contrast, Nidugala )2..1* builds on 8sfahanis )1,,1*
model and uses an augmented production function with exports as a regressor.
Nidugala finds evidence in support of the 8B" h!pothesis for the case of India,
particularl! in the 1,3.s. >e finds that export growth had a significant impact on
"=& growth. ;urther, his stud! reveals that growth of manufactured exports had a
significant positive relationship with "=& growth, while the growth of primar!
exports had no such influence. "hatak and &rice )1,,J* test the 8B" h!pothesis
for India for the period 1,/.D1,,2, using exports as regressors and measure of
"=& that nets out exports, along with exports and imports as additional variables.
Their results indicate that real )aggregate* export growth is "rangerDcaused b!
3
nonDexport real "=& growth in India over 1,/.D,2. Their cointegration tests
confirm the long run nature of this relationship. >owever, imports do not appear
to be important for the case of India. (s corroborated subseEuentl! b! Nidugala
)2..1*, their disaggregated anal!sis shows that nonDtraditional manufactured
exports )such as machiner! and transport eEuipment* are found to "ranger cause
output growth, while traditional manufactures )such as textiles, wood, paper* have
little effect.
7!0 Excluding growth accounting effects
In empirical anal!sis of trade data a ma6or problem arises from the fact that
exports are themselves a component of output, via the national income accounting
identit! Ksee <ichael! )1,JJ, 1,J,*, >eller and &orter )1,J3*, ;eder )1,32*,
(fxentiou and $erletis )1,,1*, Bove )1,,2*, 8sfahani )1,,1*, "reenawa! )1,,+*,
"hatak and &rice )1,,J* and $heehe! )1,,.*M. The results of such a model are
likel! to suffer from a simultaneit! bias since export growth ma! itself be a
function of the increase in output. ;ollowing ;eder )1,32*, the framework of the
econom! can be divided into two sectors D export and nonDexport. In this anal!sis,
there has been a separation of economic influence of exports on output from that
incorporated in the HNational Income (ccounting relationship b! using a measure
of "=& )F* that netted out exports )C*.
)1* ..... u C a a F 1 . + + =
Phere Fstands for rates of change of "=& while C stands for export growth.
0 Empirical Analysis
0!/ 'ata (ource
There are two basic sources for data on Indias exports. :ne set is
compiled b! the =irectorate "eneral of #ommercial Intelligence and $tatistics
)="#IN$*, <inistr! of #ommerce, "overnment of India and the other is
compiled b! the Reserve 2ank of India )R2I*. The ="#I$ compiles information
on real transactions, reporting EuantitiesIvolumes of exports as well as export
earnings in Indian rupees. 8xports are decomposed into headings matching with
,
the IT# )>$*
3
$tandard Industrial #lassification )$I#* codes. Thus exports are
broken down b! $I# categories and b! destination. R2I compiles export data on
its balance of pa!ment statistics component b! aggregating the econom! wide
financial transactions related to exports. 8xporters and financial intermediaries
have to provide this information to the R2I b! statute. ="#I$ data has been used
much more freEuentl! in the literature and the R2Is data has been relativel! less
freEuentl! referred to. In this stud!, R2Is data sets used for anal!sis, in part to
correct the above mentioned lacuna and to include invisible receipts and pa!ment
along with current account to capture recent upsurge in services export.
(ccordingl!, the data used in this exercise has been compiled from the Reserve
2ank of Indias H>andbook of $tatistics on the Indian 8conom!, 2..+D.- starting
from 1,J.DJ1 )referred to as 1,J1* to 2..+D.- )referred to as 2..-*. The national
income and trade data at aggregate level have been used in order to facilitate
further disaggregation subseEuentl!. The following time series data at aggregate
level are adapted for the period 1,J1D2..-A
1. "=& at constant market prices )gross domestic product*
2. "=& net of exports
'. Real exports )exports deflated b! the time series of unit price index of exports*
+. Real imports
-. "ross domestic capital ;ormation as prox! for Investment
/. 8mplo!ment in the formal sector
The constant "=& estimates are used in this stud! to permit intertemporal
comparisons. The time series of unit price index of exports is used to deflate the
export series. The prefix HB stands for the natural logarithm of the concerned time
series. (ll econometric estimations in this paper have been carried out using
Eviews -.
The data emplo!ed in this stud! are graphicall! displa!ed in (ppendixA
logarithmic transformations of time series data. In all the cases except "=& and
3
International Trade #lassification )>armonised $!stem*.
1.
"=& without exports, the probabilit! of the %arEueD2erra test statistic provides
evidence in favour of the null h!pothesis of a normal distribution. (dditionall!,
simple correlations are estimated. It is pertinent to note the negative correlations
between emplo!ment and all economic variables )income, income without
exports, real export and investment*.
0!7 =nit "oots and Cointegration
In investigating the export led growth )8B"* h!pothesis, the traditional
approach of first differencing disregards potentiall! important eEuilibrium
relationships among the levels of the series to which the h!potheses of economic
theor! usuall! appl! K8ngle and "ranger )1,3J*M. The first test for an ordinar!
unit root D single series unit root tests include (ugmented =icke!D;uller )(=;*.
Table 1 summarises the results for unit root tests on levels and in first differences
of the data. $trong evidence emerges that all the time series are I )1*. In Table 1,
for the (ugmented =ick! =icke!D;uller test, the lag length is based on the
$chwar1 Information #riterion, while for the &hillipsD&erron 4nit Root Test
bandwidth selection is based on Newe!DPest.
1able /: =nit "oot 1ests
Bevel (=;
test
$tatistics
Bevel &&
Test
$tatistics
;irst =ifference
(=; Test
$tatistics
;irst
=ifference &&
Test $tatistics
"=& without
8xports
D..J.+ D....1 D2.-., D/.''J
8xports J.,+/ 2,.'/+ 1.//+ D1.'31
Imports 1..21/ +..'3 1.+22 D..'-3
Investment +.11+ 1-.'3/ ..J1, D+.+2-
8mplo!ment D'.+.J D2.J1J D'..2- D2.,+,
1Q critical value D'./', D+.2-2 D+.2/2 D2.1'.
1Q critical value
)Trend and
Intercept*
D2..+, D2..+, D+.2J' D+.2/'
;ollowing a multivariate approach the estimation proceed with considering
the cointegration h!pothesis between output )"=&*, real exports and real imports.
11
These variables have been chosen for anal!sis for three reasons. ;irst, Rie1mann
et al )1,,/* have suggested that imports are an important variable while
considering causalit! between exports and growth, and omission of imports could
lead to biased results
,
. $econdl!, testing the 8B" h!pothesis is an explicit
ob6ective and the chosen variables seem to be appropriate for such an exercise.
;inall!, given the set of variables for which time series data is available for India,
both investments and emplo!ment seem less appropriate. In the case of
investments, foreign direct investment is excluded and the series are thus
underestimates for total investment in India, especiall! during the 1,,.s.
8mplo!ment data is ambiguous because of definitional and reporting issues. <ore
seriousl!, with a large informal, unorganised sector emplo!ment data also suffers
from underestimation problems. In this anal!sis, two cases are considered. ;irst,
using the %ohansen method, the test has been done whether there is a cointegrating
relationship between exports, imports and "=&. $econdl!, estimation used b!
considering the case of exports, imports and "=& net of exports in order to avoid
the Haccounting effect.
1able 7: 4ohansen Cointegration 1est >lnG'P, lnExport, lnImport?
r 8igen
0alue
Trace
$tatistics
-Q #0
&rob
?
<axD
8igen
$tatistic
-Q #0 &rob
None ..',+''J 2/.3323. 2,.J,J.J 1/.-+J2+ 21.1'1/2 ..1,+-
(t most 1 ..2.2-32 1..''--- 1-.+,+J1 J.+J.+2J 1+.2/+/. ..+'-1
(t most 2 ...3'1/. 2.3/-12+ '.3+1+// 2.3/-12+ '.3+1+// ...,.-
The results for the first case are presented in Table 2. Then repeat the test
replacing "=& with "=& less exports )Table '*. (s Tables 2 and ' shows, the null
h!pothesis can not be re6ected at the -Q significance level. The %ohansen procedure, like
man! others, reEuires estimation of various structural and nuisance parameters. ;or
,
Inclusion of imports )along with exports and output* in this anal!sis allows to examine the notion that
imports relieve the foreign exchange constraint that developing countries often face. This referred to as
import compression K)8sfahani)1,,1*M.
7
<ackinnonD>augD<ichelis )1,,,* pDvalues.
12
example, a vector autoregressive )0(R* log order must be specified and then the lag
parameters are estimated.
1able 0: 4ohansen Cointegration 1est >lnG'Pless export, lnExport, lnImport?
r 8igen
0alue
Trace
$tatistics
-Q #0
&rob
<axD
8igen
$tatistic
-Q #0 &rob
None ..'//,/- 1,.++J,J 2,.J,J.J 1-..33-, 21.1'1/2 ..2323
(t most 1 ...JJ,,, +.'-,'J- 1-.+,+J1 2./J,,1+ 1+.2/+/. ..,/-3
(t most 2 ...+,/1, 1./J,+/1 '.3+1+// 1./J,+/1 '.3+1+// ..1,-.
0!0 Granger causality
#orrelation does not necessaril! impl! causation in an! meaningful sense of the
word. The econometric anal!sis has full of magnificent correlations, which are simpl!
spurious or meaningless. To investigate the causalit! between "=& )and "=& less
exports* on the one hand and exports on the other, simple "ranger causalit! test has been
performed b! estimating the bivariate autoregressive processes for "=& )and "=& less
exports* and exports. The ob6ective of this exercise is to test the export led growth )8B"*
h!pothesis for India empiricall!. ;urthermore, building on preceding anal!sis, it can be
argued that export growth can stimulate investments )gross domestic capital formation*,
especiall! if there exists a productivit! differential between the export sector and the
nonDexport sector. In such cases, investment would be expected to increase in those
sectors of the econom! where productivit! and returns are higher )the export sector*.
8Euall! well, it is theoreticall! plausible to expect the reverseA the case where increased
investment would also stimulate export growth. Phether investments are in social
overhead capital )infrastructure* or in specific industries, there could be an overall
beneficial effect of investments on exports. 8mpirical anal!sis of the h!pothesis has been
presented in Table 3.
Thus the estimated eEuationA
Ryt S a. T a1Ryt.1 T ... T a1Ryt.1 T b1Rxt.1 T ... T blxt.l ///..)-*
Rxt S a. T a1Rxt.1 T ... T a1Rxt.1 T b1Ryt.1 T ... T blyt.l ///.. )/*
1'
The reported ;Dstatistics are the Pald statistics for the 6oint h!pothesisA
b1 S ... S bl S . UUU )J*
The null h!pothesis is therefore that C does not "rangerDcause F in the first regression
and that F does not "rangerDcause C in the second regression.
1able -: Granger Causality 2"eal Export: G'P net Export3
Null >!pothesisA :bs ;D$tatistic &robabilit!
R8(B8C& does not "ranger #ause "=&N8T8C& '+ -.J'+// .!.77@9
"=&N8T8C& does not "ranger #ause R8(B8C& '+ ....1,' ..,/-2.
1able 5: Granger Causality 2"eal Export: G'P3
Null >!pothesisA :bs ;D$tatistic &robabilit!
R8(B8C& does not "ranger #ause "=& '+ ../+/'3 ..+2J-'
"=& does not "ranger #ause R8(B8C& '+ ..12+22 ..J2/3,
1able 9: Granger Causality 2"eal Export: "eal Investment3
Null >!pothesisA :bs ;D$tatistic &robabilit!
IN08$T does not "ranger #ause R8(B8C& '+ ..-+21. ..+/J1.
R8(B8C& does not "ranger #ause IN08$T '+ +.32-3/ .!.059-
In all the cases of Tables +, - and /, the reported probabilities are greater than
...- except for real export and "=& net of export and Investment. $ince annual
observations have been used, onl! one lag is emplo!ed. The h!pothesis that exports
"ranger cause investment )or vice versa* can not be full! re6ected. (t this significance
level, we could marginall! accept the h!pothesis that growth in investment and income
would "ranger causes growth in real exports. The evidence in this section does not
provide an! support for the causalit! relationship between exports and output )"=&*.
There is weak evidence suggesting that the direction of causalit! runs from "=& to
1+
exports, which further strengthens the case against the 8B" h!pothesis for the case of
India.
-! Conclusion
This stud! has examined export led growth )8B"* h!pothesis for India
using different approaches b! emplo!ing data at the aggregate level covering the
postDliberalisation period. It presents a more upDtoDdate test of the 8B" h!pothesis
for India and able to capture the effects of liberalisation on exports and output
growth. 2! emplo!ing cointegration h!pothesis )following the %ohansen method*
for the case of India, the stud! investigate the following h!pothesesA )i* whether
exports, imports and "=& are cointegrated using the %ohansen approach, )ii*
whether export growth "ranger causes "=& growth, and )iii* and whether export
growth "ranger causes investment. ;or the first two cases, strong evidence is
found against the cointregration h!pothesis. In this anal!sis, it also fail to find
support for the h!pothesis that exports "ranger cause "=&, using two measures
for "=& )"=& with exports and "=& without exports*. The same holds for the
relationship between exports and investment. This strengthens the argument
against the 8B" h!pothesis for the case of India and strengthens the argument that
inspite of reforms, it still retains some characteristics of an import substituting
econom!. $ince aggregate export data for India includes sectors such as software
exports, the euphoria about perceived successes in the I#T )information and
communication technolog!* sector for India seem somewhat premature, given that
at an aggregated level there is little evidence to support the export led growth
h!pothesis, which brings into doubt the implicitl! assumed productivit!
differentials and resulting positive spillovers into the rest of the econom!.
VVVVVVVVVVV
*ibliography and "eferences
1-
(fxentiou, & # and ( $erletis. )1,,1*, H8xports and "N& #ausalit! in the
Industrial #ountriesA 1,-.D1,3-, L!klos, ++, 2, pp. 1/JDJ,.
(hluwalia, <ontek $. )2..2*A ?8conomic Reforms in India since 1,,1A >as
"radualism PorkedW@, !ornal of Econo"ic 'erspectives, 0ol. 1/, No J, pp. /JD
33.
(lston, Richard <.O %. R. Learl, and <ichael 2. 0aughan. )1,,2*A XIs There a
#onsensus (mong 8conomists in the 1,,.YsWX 0"er. Econ. +ev. )&apers and
&roceedings*, 32A2, pp. 1-3DJJ.
(nwer, < $ and R L $ampath. )2..1*, H8xports and 8conomic "rowth, Indian
8conomic %ournal, +J, ', pp. J,D33.
(safuD(d6a!e, % and = #hakrabort!. )1,,,*, H8xportDled "rowth and Import
#ompressionA ;urther Time $eries 8vidence ;rom B=#s, (ustralian 8conomic
&apers, '3, pp. 1/+DJ-.
2ahmaniD:skoee, < and % (lse. )1,,'*, H8xport "rowth and 8conomic "rowthA
(n (pplication of #ointegration and 8rrorD#orrection <odelling, %ournal of
=evelopment (reas, 2J, %ul, pp. -'-D+2.
2ahmaniD:skoee, <, > <ohtadi and " $habsigh. )1,,1*, H8xports, "rowth and
#ausalit! in B=#sA ( ReDexamination, %ournal of =evelopment 8conomics, '/,
pp. +.-D1-.
2alassa, 2. )1,J3*, H8xports and 8conomic "rowthA ;urther 8vidence, !ornal
of &evelop"ent Econo"ics, -, pp. 131D3,.
2haduri, (mit and =eepak Na!!ar )1,,/*A ?The Intelligent 'erson1s Gide to
Liberalisation@, &enguin 2ooks, New =elhi.
2hagwati, %. )1,32*, H=irectl! unproductive profit seeking )=4&* activities,
%ournal of &olitical 8conom!, ,., -, pp. ,33D1..2.
2reitung, %. )2..2*A HNonparametric Tests for 4nit Roots and #ointegration,
!ornal of Econo"etrics, 1.3, '+'D'/+.
#how, & # F. )1,3J*, H#ausalit! 2etween 8xport "rowth and Industrial
=evelopmentA 8mpirical 8vidence from the NI#s, !ornal of &evelop"ent
Econo"ics, 2/, pp. --D/'.
#ohen, 2en6amin I. )1,/+*A ?The $tagnation of Indian 8xports, 1,-1D/1@, The
2arterly !ornal of Econo"ics, 0ol. J3, No. +, &&. /.+D/2..
=arrat, (li ;. )1,3J*, H(re exports engine of growthW (nother look at the
evidence, (pplied 8conomics, 1,, pp. 2JJD3'.
=hawan, 4 and 2 2iswal. )1,,,*, HReDexamining export led growth h!pothesisA a
multivariate cointegration anal!sis for India, (pplied 8conomics, '1, pp. -2-D'..
=odaro, $anto. )1,,1*, H#omparative (dvantage, Trade and "rowthA 8xportDBed
"rowth Revisited, Porld =evelopment, 1,, ,, pp. 11-'D/-.
=omar, 8 =. )1,+/*, H#apital 8xpansion, Rate of "rowth and 8mplo!ment,
Econo"etrica, 1+, 2, pp. 1'JD+J.
1/
8ngle, R ; and # P % "ranger. )1,3J*, H#oDintegration and 8rror #orrectionA
Representation, 8stimation and Testing, Econo"etrica, --, pp. 2-1DJ/.
8sfahani, > $. )1,,1*, H8xports, imports, and economic growth in semiD
industriali1ed countries, %ournal of =evelopment 8conomics, '-, pp. ,'D11/.
;eder, "ershon. )1,32*, H:n exports and economic growth, !ornal of
&evelop"ent Econo"ics, 12, pp. -,DJ'.
;rank, (ndre "under. )1,/,*, Batin (mericaA 4nderdevelopment or RevolutionW,
New ForkA <onthl! Review &ress.
"hatak, $ and $ P &rice. )1,,J*, H8xport #omposition and 8conomic "rowthA
#ointegration and #ausalit! 8vidence for India, Peltwirtschaftliches (rchiv,
1'', ', pp. -'3D-'.
"ranger, # P %. )1,/,*, HInvestigating #ausal Relationships b! 8conometric
<odels and #rossDspectral <ethods, Econo"etrica, 'J, ', pp. +2+D'3.
"ranger, # P %. )1,33*, H$ome Recent =evelopments in a #oncept of #ausalit!,
!ornal of Econo"etrics, ',, pp. 1,,D211.
"reenawa!, = and = $apsford. )1,,+*, HPhat does Biberalisation do for 8xports
and "rowth, Peltwirtschaftliches (rchiv, 1'., 1, pp. 1-2DJ+.
>arrod, R ;. )1,',*, H(n 8ssa! in =!namic Theor!, Econo"ic !ornal, +,, 1,',
pp. 1D''.
>eller, & $ and R # &orter. )1,J3*, H8xports and "rowthA (n empirical reD
investigation, %ournal of =evelopment 8conomics, -, 2, pp. 1,1D,'.
>irschmann, (. :. )1,-3*, The $trateg! of 8conomic =evelopment, New >avenA
Fale 4niversit! &ress.
>siao, < # P. )1,3J*, HTests of #ausalit! and 8xogeneit! between 8xports and
8conomic "rowthA The #ase of (sian NI#s, %ournal of 8conomic =evelopment,
12, 2, pp. 1+'D-,.
%ohansen, $. )1,33*, H$tatistical (nal!sis of #ointegration 0ectors, !ornal of
Econo"ic &yna"ics and (ontrol, 12, pp. 2'1D-+.
%ohansen, $. )1,,-*, BikelihoodD2ased Inference in #ointegrated 0ector
(utoregressive <odels, :xford 4niversit! &ress, :xford.
%ohansen, $ and L %uselius. )1,,.*, H<aximum Bikelihood 8stimation and
Inference on #ointegration D Pith (pplications for the =emand for <one!,
3xford 4lletin of Econo"ics and Statistics, -2, 2, pp. 1/,D21..
%ung, P. $. and &. %. <arshall. )1,3-*, H8xports, "rowth and #ausalit! in
=eveloping #ountries, !ornal of &evelop"ent Econo"ics, 13, pp. 1D12.
Lavoussi, R <. )1,3+*, H8xport 8xpansion and 8conomic "rowthA ;urther
8mpirical 8vidence, !ornal of &evelop"ent Econo"ics, 1+, pp. 2+1D-..
Lohli, I and N $ingh. )1,3,*, H8xports and "rowthA #ritical <inimum 8ffort and
=iminishing Returns, !ornal of &evelop"ent Econo"ics, '., pp. ',1D+...
1J
Lrueger, ( :. )1,J-*, The 2enefits and #osts of Import $ubstitution in IndiaA (
<icroeconomic $tud!, <inneapolis A 4niversit! of <innesota &ress.
Lrueger, (nne :. )1,,.*A ?(sian Trade and "rowth Bessons@. The 0"erican
Econo"ic +eview. )&apers and &roceedings*, 0ol. 3.A2, pp. 1.3D112.
Ball, $an6a!a. )2..1*, H#ompetitiveness, Technolog! and $kills, #heltanhamA
8dward 8lgar.
Bove, %. )1,,2*, H8xport Instabilit! and the =omestic 8conom!A Zuestions of
#ausalit!, %ournal of =evelopment $tudies, 23, +, pp. J'-D+2.
<ichael!, <ichael. )1,JJ*, H8xports and "rowthA (n 8mpirical Investigation,
%ournal of =evelopment 8conomics, +, pp. +,D-'
<ichael!, <. )1,J,*, H8xports and "rowthA ( Repl!, %ournal of =evelopment
8conomics, /, pp. 1+1D+'.
<oschos, =. )1,3,*, H8xport 8xpansion, "rowth and the Bevel of 8conomic
=evelopmentA (n 8mpirical (nal!sis, !ornal of &evelop"ent Econo"ics, '.,
pp. ,'D1.2.
Nidugala, " L. )2..1*, H8xports and 8conomic "rowth in IndiaA (n 8mpirical
Investigation, Indian 8conomic %ournal, +J, ', pp. /JDJ3.
&rebisch, R. )1,J.*, HTowards a New Trade &olic! for =evelopment, report b!
the $ecretar! "eneral of 4N#T(=, 4nited Nations, 1,/+. Reprinted in ".<.
<eier, Beading Issues in 8conomic =evelopmentA $tudies in international
povert!, $econd 8dition, New ForkA :xford 4niversit! &ress.
Rie1man, R, Phiteman # > and & < $ummers. )1,,/*, HThe 8ngine of "rowth or
Its >andmaidenW ( TimeD$eries (ssessment of 8xportDBed "rowth, 8mpirical
8conomics, 21, pp. JJD11..
Rodrik, =ani. )1,,/*A ?4nderstanding economic polic! reform. !ornal of
Econo"ic Literatre, '+, pp. ,D+1.
RosensteinDRodan, & N. )1,+'*, H&roblems of Industrialisation of 8astern and
$outhD8astern 8urope, Econo"ic !ornal, -', %uneD$ept, pp. 2.2D11.
$erletis, (postolos. )1,,2*, H8xport growth and #anadian economic
development, %ournal of =evelopment 8conomics, '3, pp. 1''D+-.
$heehe!, 8 %. )1,,.*, H8xports and "rowthA ( ;lawed ;ramework, %ournal of
=evelopment $tudies, 2J, pp. 111D1/.
$ims, # (. )1,J2*, H<one!, Income and #ausalit!, (merican 8conomic Review, /2, +,
pp. -+.D-2.
$ingh, <anmohan, )1,/+*A ?India1s Export Trends@, :xfordA #larendon &ress.
$rinivasan, T N. )1,3-*, HNeoclassical &olitical 8conom!, the $tate and 8conomic
=evelopment, (sian =evelopment Review, ', +, pp. '3D-3.
Porld 2ank, )1,,/*A ?India$ #ive 5ears of Stabili6ation and +efor"s and the (hallenges
0head7% Porld 2ank #ountr! $tud!A Pashington, =.#.
Porld Trade :rganisation, )2..-*A ?International Trade Statistics7, "eneva.
13
[[[[[[ )2..2*A ?Trade 'olicy +eview$ India7, )1,,3 and 2..2*, "eneva.
[[[[[[ )2..-*A 8*orld Trade report1, "eneva.
1,
Appendix 1able /: IndiaAs (hare of 1otal Borld Exports
2 In =( C *illion3
Dear Borld Exports Indian Exports (hare 2E3
1,+3 -3.,+ 1.'. 2.2.
1,+, /..++ 1.2. 1.,3
1,-. /2.-- 1.1+ 1.3'
1,-1 J,.1+ 1./. 2..2
1,-2 J'.,+ 1.2+ 1./3
1,-' J'.3. 1.1. 1.+,
1,-+ JJ..3 1.1J 1.-2
1,-- ,../- 1.2/ 1.',
1,-/ 1...+J 1.2/ 1.2-
1,-J 1.,.'1 1.'3 1.2/
1,-3 1.-.-' 1.22 1.1-
1,-, 112.22 1.'2 1.13
1,/. 12+.1. 1.'' 1..J
1,/1 12,.2+ 1.', 1..J
1,/2 1'-.+/ 1.+. 1..+
1,/' 1+3.+' 1./' 1.1.
1,/+ 1//.'. 1.J. 1..'
1,/- 13...1 1./, ..,+
1,// 1,J.1J 1.,- ..,,
1,/J 2.J.2+ 1./1 ..J3
1,/3 22J.2- 1.J/ ..J3
1,/, 2-,.1/ 1.3+ ..J1
1,J. 2,,.+- 2..' ../3
1,J1 ''-..2 2..+ ../1
1,J2 ',+.-- 2.+- ../2
1,J' --..J, 2.,2 ..-'
1,J+ 31-./, '.,' ..+3
1,J- 322.13 +.'/ ..-'
1,J/ ,2J.-2 -.-- ../.
1,JJ 1.-1..' /.'3 ../1
1,J3 121-.13 /./J ..--
1,J, 1-J+.11 J.31 ..-.
1,3. 1332.'2 3.-, ..+/
1,31 13,-.2+ 3.'. ..++
1,32 1J+,.1, ,.'/ ..-'
1,3' 1J.-.J' ,.1- ..-+
1,3+ 131'.23 ,.+- ..-2
1,3- 13'J.+3 ,.1+ ..-.
1,3/ 1,,..3- ,.+. ..+J
1,3J 2'-..32 11.'. ..+3
1,33 2/,1./J 1'.2' ..+,
1,3, 2,2,.,J 1-.3J ..-+
2.
1,,. ''23..1 1J.,J ..-+
1,,1 '+1J.,. 1J.J' ..-2
1,,2 '/+..J' 1,./' ..-+
1,,' '/'-.,' 21.-J ..-,
1,,+ +1',.3- 2-..2 ../.
1,,- +,+/.'+ '../' ../2
1,,/ -1/3.2, ''.11 ../+
1,,J --'3..3 '-..1 ../'
1,,3 -+-../J ''.++ ../1
1,,, -/++./J '-./J ../'
2... /'JJ.// +2.'3 ..//
2..1 /1'2..J +'.'/ ..J1
2..2 /+23.-, +,.2- ..JJ
2..' J+/-.'/ -J.., ..J/
2..+ ,./J.+' J1.J, ..J,
2..- 1.1,/.J. 3-.,' ..3+
21
3.0
3.2
3.4
3.6
3.8
4.0
4.2
4.4
1975 1980 1985 1990 1995 2000 2005
LEMP
12.4
12.8
13.2
13.6
14.0
14.4
1975 1980 1985 1990 1995 2000 2005
LGDP
12.6
12.8
13.0
13.2
13.4
13.6
13.8
14.0
1975 1980 1985 1990 1995 2000 2005
LGDPNETEXP
10.8
11.2
11.6
12.0
12.4
12.8
13.2
1975 1980 1985 1990 1995 2000 2005
LINVEST
8.0
8.5
9.0
9.5
10.0
10.5
11.0
11.5
12.0
1975 1980 1985 1990 1995 2000 2005
LREALEXP
8.5
9.0
9.5
10.0
10.5
11.0
11.5
12.0
1975 1980 1985 1990 1995 2000 2005
LREALIMP
22
doc_617151740.doc