Description
The purpose of this article is to examine the inter-relationship and direction of causality
among three macroeconomic variables such as trade liberalization, financial development and economic
growth
Journal of Financial Economic Policy
Trade Openness, Financial Development Index and Economic Growth: Evidence from
India (1971-2012)
Dogga Satyanarayana Murthy Suresh Kumar Patra Amaresh Samantaraya
Article information:
To cite this document:
Dogga Satyanarayana Murthy Suresh Kumar Patra Amaresh Samantaraya , (2014),"Trade Openness,
Financial Development Index and Economic Growth", J ournal of Financial Economic Policy, Vol. 6 Iss 4 pp.
362 - 375
Permanent link to this document:
http://dx.doi.org/10.1108/J FEP-10-2013-0056
Downloaded on: 24 January 2016, At: 21:50 (PT)
References: this document contains references to 26 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 344 times since 2014*
Users who downloaded this article also downloaded:
Omotola Awojobi, (2013),"Does trade openness and financial liberalization foster growth: An empirical
study of Greek economy", International J ournal of Social Economics, Vol. 40 Iss 6 pp. 537-555 http://
dx.doi.org/10.1108/03068291311321848
Salih Turan Katircioglu, Neslihan Kahyalar, Hasret Benar, (2007),"Financial development, trade and growth
triangle: the case of India", International J ournal of Social Economics, Vol. 34 Iss 9 pp. 586-598 http://
dx.doi.org/10.1108/03068290710778615
Muhammad Tahir, Imran Khan, (2014),"Trade openness and economic growth in the Asian region", J ournal
of Chinese Economic and Foreign Trade Studies, Vol. 7 Iss 3 pp. 136-152 http://dx.doi.org/10.1108/
J CEFTS-05-2014-0006
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Trade Openness, Financial
Development Index and
Economic Growth
Evidence from India (1971-2012)
Dogga Satyanarayana Murthy, Suresh Kumar Patra and
Amaresh Samantaraya
Department of Economics, Pondicherry University, Puducherry, India
Abstract
Purpose – The purpose of this article is to examine the inter-relationship and direction of causality
among three macroeconomic variables such as trade liberalization, fnancial development and economic
growth.
Design/methodology/approach – The empirical analysis is based on the principal component
analysis as method to construct fnancial development index (FDI), augmented Dickey–Fuller and
Phillips–Perron tests as the unit root test, Johansen’s co-integration test and VECM for direction of
causality in the long run among TOP, FDI and economic growth.
Findings – The empirical results confrmed that there exists a long-run association among trade
openness, fnancial development and economic growth. This study has also found that there is
bidirectional causality between fnancial development and growth. However, the causality runs from
growth to fnance is stronger than that fromfnance to growth. This study also observed unidirectional
causality that runs from fnancial development and economic growth to trade openness.
Research limitations/implications – The policy implications that could be drawn fromthe present
study is that, initiation of fnancial reforms to improve the size of fnancial systemwould lead to higher
economic growth. Another key implication fromthis study is that because trade openness has no effect
on both domestic fnancial sector development and output growth, it would be better to deploy the
resources into creating a sustained domestic demand rather than concentrating more on the external
front in general and trade openness in particular.
Originality/value – The study constructs a summary IFDfor India by taking into account four broad
fnancial development indicators for the period 1971-2012. The present paper also suggests that it
would be better to deploy the resources to create a sustained domestic demand rather than
concentrating more on the external front in general and trade openness in particular.
Keywords Trade openness, Financial development index, Economic growth, VECM
Paper type Research paper
1. Introduction
The relationship between trade openness, fnancial development and economic growth
has been widely discussed and debated in both the theoretical and empirical literature.
It is argued that trade openness and fnancial development policies positively infuence
gross domestic product growth by reducing the distortion in the production process
(Atif et al., 2010). It is also widely evident fromthe fact that countries with high degree of
JEL classifcation – F13, G21, O1, B23
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JFEP
6,4
362
Journal of Financial Economic Policy
Vol. 6 No. 4, 2014
pp. 362-375
©Emerald Group Publishing Limited
1757-6385
DOI 10.1108/JFEP-10-2013-0056
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
trade openness and well-developed fnancial system have been registering higher GDP
growth as compared to the countries with low-fnancial sector development and restrictive
trade policies. Opening up of an economy for the cross border fows of goods and services
creates high competitive environment, which will drive down the revenue of existing frms
and diminish their profts, requiring them to search for external sources of fnance
(Quy-Toan and Levchenko, 2004). These sources of fnance will be available only if the
fnancial system is able to solve the problem of asymmetric information and other
market imperfections in the present fnancial system. As a result, present frms and their
rich, elite owners will now have incentives to support the compulsory institutional
reforms to make the fnancial system effcient and well-functioning. The resulting
reforms, in turn, will extend the size of the fnancial systemand thereby foster economic
growth. Thus, free trade can be an important ingredient that stimulates economic
growth through domestic fnancial sector development. However, recent global
fnancial crisis that cropped up in July 2007 in the USA and later enveloped the entire
world has again re-opened the debate over open economy policy, as it resulted in
macroeconomic instability of the national economies and also put entire world economy
in trouble with falling outputs and rising unemployment levels across the world.
Similarly, the role of fnancial sector development in economic growth is the long-run
debated issues among economists (Schumpeter, 1911; Patrick, 1966; Goldsmith, 1969:
McKinnon, 1973; Shaw, 1973). With the unleashing of endogenous growth theory,
several studies have been attempted to knowhowthe development of fnancial markets
and institutions may affect economic growth (Greenwood and Jovanovic, 1990;
Bencivenga and Smith, 1991; King and Levine, 1993; Arestis, 2005; Shahbaz et al., 2013).
Finance can affect growth in an endogenous growth model by allowing higher returns
on investment (Greenwood and Jovanovic, 1990), through increasing the rate of savings
(Bencivenga and Smith, 1991) and by rising human capital accumulation. It implies that
from the macroeconomic point of view, well-developed fnancial markets and
institutions of a country will able to convert a given amount of inputs, k, into larger
amount of output, Y. This is because the production function is a rising function of the
fnancial development of the economy (Roubini and Sala-I Martin, 1992).
The important question that rambles in the mind of many researchers whether there
exists any inter-relationships among trade openness, fnancial development and
economic growth. Hence, an attempt has been undertaken to know the inter-relation
between the above three variables in India for the period from 1970-1971 to 2011-2012.
India, like many developing countries, in 1991, as a part of the structural adjustment and
macroeconomics stabilization programmes, introduced various macroeconomic, trade
and fnancial sector reforms to promote economic growth. Thus, liberal fnancial policy
regime has been replaced in place of an old controlled regime. This resulted in
tremendous improvement in the fnancial systemin terms of total fnancial assets, bank
credit to private sector as percentage of GDP, increase in the quality of bank assets, etc.
In 1991, fnancial assets as percentage of GDP were 17.61 per cent and it increased to
116.93 in 2011. Similarly, trade reforms in the form of removal of license-permit raj,
abolition of quantitative restrictions and reducing import tariff resulted in signifcant
rise in Indian foreign trade. In 1991, India’s foreign trade as percentage of GDP has been
increased from 6.72 per cent in 1991 to 73.14 per cent in 2011. The average growth rate
during this period (i.e. from 1991-2011) was 6.7 per cent.
363
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The present study constructs a summary index of fnancial development (IFD) by
applying principal component analysis (PCA) and uses co-integration test for long-run
association and vector error correction mechanism (VECM)-based Granger causality
approach to knowthe causality between trade liberalization, fnancial development and
economic growth. Thus, the major advantages of the presents study are constructing a
broad measure for assessing fnancial development in India by taking into account a
large set of variables, which together represent the development of fnancial system in
terms of its size, activity and using appropriate econometric tools while taking into
account non-stationarity of the time series data to examine the causal link between trade
openness, fnancial development and economic growth. Our results confrmthe long-run
association between these three macroeconomic variables in India. In addition to this,
our results based on Granger causality approach used in VECM framework strongly
support the old Schumpeter (1911) and Patrick’s (1966)[1] fnance-led growth hypothesis
in the long run.
The rest of the paper is as follows. Section 2 describes theoretical and conceptual
framework of the study. Section 3 outlines the data and methodology. The empirical
results are reported in Section 4. The last section concludes.
2.Theoretical underpinning
2.1 Trade openness – economic growth
The concept of absolute and comparative cost advantage theories emphasized that like
inter-regional trade, the international trade is also benefcial for the trading countries
because each country will specialize in the production of particular commodities but not
in the all commodities, and also countries differ in the resource endowments. Therefore,
when a country enters into trade with other country, it can export those commodities in
which its production cost is less, and can import those commodities in which its
production cost is high. This results in greater output and consumer welfare in both the
trading countries, which in turn, will lead to higher employment and hence economic
growth.
Thus, the classical economists were in favour of free trade policy, as they assumed
that free trade among different nations maximizes the output and employment of all the
participating countries (Salvatore, 2010). Edwards (1998) noted that countries that are
more open to the rest of the world are better placed in capturing the advanced
technologies of leading nations. If the costs of technological imitation are less than the
costs of domestic-developed innovations, then a poorer country will grow faster than a
high developed one. This faster rate of growth will continue as long as that country
remains opened to attract newideas until, at some point, equilibriumis reached and the
rate of growth slows down.
However, economists like Prebish, Singer, Myint and Miyrdal argues that free trade
between developed and developing countries shifts the gains from developing to
developed nation because developing countries are largely dependent on the production
of primary goods, whereas the developed nations mostly hinge upon manufacturing
products. The demand and prices of the primary goods will often deteriorate as
compared to manufacturing product. Therefore, free trade leads to secular deterioration
in the terms of trade of developing countries.
JFEP
6,4
364
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.2 Trade openness – fnancial development
There are signifcant differences in fnancial sector development across countries and even
within the country over a period. This is mainly because fnancial development is also
infuencedbyother factors like legal systeminthe country(Shleifer andVishny, 1998), trade
and fnancial openness (Rajan and Zingales, 2003). It is argued that the trade openness
affects fnancial development because free trade will result in uncertainty and income
inconsistency of agents, which in turn raises the demand for insurance and other fnancial
services and thereby increases the size of the fnancial system(Newbery and Stiglitz, 1984).
Furthermore, free trade among different nations generally will increase the demand for
external fnance, as theyproduce more fnancial-dependent goods. Thus, free trade increases
the demand for external fnance and thereby size and quality of the fnancial system.
However, Quy-Toan and Levchenko (2004) noted that if there is free trade between
rich and poor countries, in rich countries, more trade would be associated with faster
fnancial development as they are specialized in fnancial-dependent good. Whereas
more trade lead to deterioration in the size of the fnancial system in poor countries, as
they import fnancial-dependent goods rather than produce them domestically. While,
Rajan and Zingales (2003) postulate that trade openness is linked with fnancial market
development, especially when cross-border capital fows are free, and that changes in
openness are associated with changes in the size of fnancial markets.
Further, free trade helps to develop the domestic fnancial markets and then the
economic growth. Because free trade expands the size of the market for domestic goods,
which in turn encourages the domestic production and thus production of more goods
and services, more capital is required. Therefore, allowing foreign capital into domestic
fnancial markets by fnancial openness increases the availability of funds, which in
turn lowers the cost of borrowing and thereby increases the investment and economic
growth (Peter, 2003). Thus, trade openness and fnancial openness are not substitute
rather they are complementary to each other as their co existence will result in domestic
fnancial sector development and hence higher economic growth. Similarly, Rajan and
Zingales (2003) hypothesis also suggest that simultaneous openness of both trade and
capital fows are preconditions for fnancial development.
2.3 Financial development – economic growth
Financial development affects economic growth mainly by increasing the marginal
productivity of capital through transferring funds from relatively less to relatively more
productive uses and also through increasing the rate of saving (Bencivenga and Smith,
1991). Financial institutions increase the rate of saving by pooling small savings from
different segments across country and making them into a huge amount of loans for
productive uses. Similarly, fnancial system can increase the marginal productivity of
capital because they are more effcient and experts in the collection and evaluation of
information on the different alternative investment projects and they can also reduce risk to
individual to undertake riskier and more productive investment by directing funds to
relative liquidity and higher-yield assets (Pagano, 1993). Thus, fnancial development frst
increases the rate of saving and then to increase the marginal productivity of capital,
fnancial institutions transferringthese funds toproductive uses andtherebyoutput growth.
Similarly, economic growth also encourages the development of fnancial markets and
institutions by creating demand for different types of fnancial services (Robinson, 1952;
Patrick, 1966). As economygrows, it creates additional demandfor fnancial services, which
365
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
bring about a supply response in the growth of the fnancial system. Similarly, as economy
grows, there will be a greater variance in the growth rates among different sectors in the
economy and thereby an increase in the demand for fnancial intermediation to transfer
funds from slow-growth sectors to fast-growing sector. Thus, fnancial development and
economic growth infuences each other. More specifcally, in initial stage of development
growth follows fnance and in advance stage, fnance follow economic growth (Patrick,
1966).
3. Data and methodology
This section reports the data and methods used to examine the causal link between
fnancial development, economic growth and trade openness, and it also summarizes
various indicators used in the present study for assessing fnancial sector development.
3.1 Data and methods used
To construct a summary IFD for India, present study takes into account various
fnancial development indicators such as the ratio of broad money (M3) to narrow
money (M1), which is used as a proxy fnancial innovation ratio (FIR), money supply as
percentage of GDP (M3) and the ratio of banking credit to private sector to GDP.
Similarly, as an indicator for economic activity, the study used GDP at factor cost at
constant prices (2004-2005 ?100) and the amount of total trade i.e. sum of exports and
imports as percentage of GDP used as proxy of trade openness (TOP). The annual time
series data on the above variables has been collected from the Reserve Bank of India
handbook of statistics, 2012-2013.
The empirical analysis mainly consists of three parts:
(1) construction of fnancial development index (FDI) by using PCA;
(2) subsequently using this index, estimating Johansen’s co-integration test to
analyse the long-run association between real GDP growth, fnancial
development (IFD) and trade openness (TOP); and
(3) we proceed for causality test based on VECM framework.
3.2 Principal component analysis
The present study used PCA for construction of a summary IFD. The PCA is a
multivariate technique used to transform the original data consisting of a set of
variables into a linear combination of a small set of variables known as principal
components (PCs) so that the bulk of the variation in the original data is explained.
These PCs are newentities and they are extracted fromthe original data set of variables
after taking into account the correlation matrix. Among extracted PCs, the frst PC
would be best component as it explains greater variance than the rest of the PCs.
Linear combinations of each component’s factor loading can be expressed as follows:
PC
k
? ?
1k
?
1
? ?
2k
?
2
? ......... ? ?
nk
?
n
(1)
Where,
PC
k
is the kth PC.
X
1
, X
2
[…], X
n
are the variables used in the PCA.
?
1k
, ?
2k
, […] ?
nk
are factor loadings of respective X
i
in kth PC.
JFEP
6,4
366
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The number of PCs extracted should be less than or equal to the number of variables
used in the analysis.
3.3 Unit root test
We used the augmented Dickey–Fuller (ADF) test as test of unit root which is based on
the assumption of serially correlated error terms. The ADF test consists of estimating
the following regression:
?Y
t
? ? ? ?
t
? ??
t?1
? ?
1
??
t?1
? ........ ? ?
p?1
??
t?p?1
? ?
t
(2)
Where ?
t
is serially correlated.
The Phillips–Perron (1988) test is mostly used to examine time series whose
differences may have mixed autoregressive–moving average (p, q) processes of
unknown order in that the test statistic includes a nonparametric allowance for
heteroscedasticity and serial correlation in testing the regression. It involves estimating
the following equation:
?
t
? ?
0
? ?
1
?
t?1
? ?
2
(
t ? T/2
)
? v
t
(3)
Where T indicates the number of observations and ?
t
represents the error term. There
will be no unit root if ?1
?1
?0, as in the ADF test, and thus we can test the stationarity
of a variable without the trend by dropping the trend term.
3.4 Co-integration test
To evaluate the long-run relationship between real GDP growth (LGDP), fnancial
development (LIFD) and trade openness (LTOP), we have used the maximumlikelihood
test procedure recognized by Johansen and Juselious (1990) and Johansen (1991).
Especially, the VAR (vector autoregression) with m-lag and Gaussian error can take the
following form (if X
t
is a vector of n stochastic variables):
?X
t
? ? ? ?
1
?X
t?1
? ··· ? ?
m?1
?X
t?m?1
? ?X
t?1
? ?
t
(4)
Where ?
1
, ?
p?1
and ?are coeffcient matrices, ?
t
is a vector of white noise process and
? contains all deterministic elements.
Determination of the rank (r) of Matrix ?? is the central point of conducting
co-integration procedure developed by Johansen. There are mainly three feasible
outcomes in the present application. Firstly, it can be of full rank (r ?n) (which would
imply that the variables are stationary processes), which contradicts the previous
fnding of non-stationarity. Secondly, the rank of ?can be zero (r ?0), which represents
no long-run relationship among the variables. It will be suitable to estimate the model in
either frst differences or levels, when ??is of either zero rank or full rank, respectively.
Finally, when there is at most r co-integrating vectors 0 ? r ? n (i.e. reduced rank), it
suggests that there are (n ? r) common stochastic trends. On the basis of Akaike
information criterion (AIC), the lag in the VAR is chosen. Johansen’s co-integration
procedure deals with two likelihood ratio test statistics such as trace test and the
maximum eigenvalue (?-max) test. But, in our study, we have taken into account the
trace test to determine which of the three as stated above is supported by the data.
367
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.5 Granger causality
If GDP growth (LGDP) has a long-run relationship with money fnancial development
(IFD) and trade openness (LTOP), the further step is to investigate the causal
relationship among these variables because if two or more variables are co-integrated,
there must exist causality in at least one direction (Engel and Granger, 1987). We,
therefore, move further to determine whether real GDP will Granger-cause fnancial
development and trade openness and vice-versa, using VECM which is given in the
following as form:
?z
t
? ? ? ?t ? ?z
t?1
?
?
i?1
p?i
?
t
?y
t?i
?
?
i?1
p?1
?
t
?x
t?i
? ?
t
(5)
Where, ? is the frst-difference operator. The long-run multiplier matrix ? as:
? ?
?
?
YY
?
YX
?
XY
?
XX
?
As the diagonal elements of the above matrix are unrestricted, the selected series can
followeither I(1) or I(0). If ?
YY
?0, then Yis I(0). On the contrary, if ?
YY
?0 then Yis I(1).
The VECM procedures described above are imperative in the testing of at most one
co-integrating vector between dependent variable y
t
and a set of explanatory
variables x
t
.
Hence, the VEC model is as follows:
?LGDP
t
? ?
g
(LGDP
t?1
? a
0
? a
1
LTOP
t?1
? a
2
LIFD) ?
?
?
11
(i)?LGDP
t?i
?
?
?
12
(i)?LIFD
t?i
?
?
?
13
(i)?LTOP
t?i
? ?
gt
(6)
?LIFD
t
? ?
m
(LGDP
t?1
? a
0
? a
1
LTOP
t?1
? a
2
LIFD) ?
?
?
21
(i)?LGDP
t?i
?
?
?
22
(i)?LIFD
t?i
?
?
?
23
(i)?LTOP
t?i
? ?
mt
(7)
?LTOP
t
? ?
P
(LGDP
t?1
? a
0
? a
1
LTOP
t?1
? a
2
LIFD) ?
?
?
31
(i)?LGDP
t?i
?
?
?
32
32(i)?LIFD
t?i
?
?
?
33
(i)?LTOP
t?i
? ?
PT
(8)
The above systemis a three variables VARmodel augmentedby e
t?1
. If ?s are zero, then
it is a simple VAR model and there is no co-integration. Hence, co-integration and error
correction are equivalent representation (Granger representation theorem).
In the view of Engle and Granger (1987), if there exists co-integration between two
variables, then a conventional causality test commonly known as error correction model
should be applied. The VECM specifcation restricts the long-run behaviour of the
endogenous variables to converge to their co-integrating relationships while allowing a
wide range of short-run dynamics (Granger causality). As the deviation from long-run
equilibrium is rectifed gradually through a series of partial short-run adjustments,
co-integrating term is otherwise known as the error correction term.
JFEP
6,4
368
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4. Report on empirical results
In this section, we have reported results on construction of a summary IFD for India by
using PCA followed by empirical results for co-integration and causality tests.
4.1 Constructing FDI
In PCA, loadings or scores refecting how each of the indicators included in each of the
index contributes to the PC were computed. The factor loadings fromthe best PC, which
explain more than 95 per cent of the total variance, are used for construction of the fnal
index.
Prior to using PCA, we test for factorability of the data used in the study. Bartlett’s
test of sphericity and Kaiser–Meyer–Olkin measure of sampling adequacy (KMO)[2] are
carried out to test the suitability of the data for PCA. For KMO, a value of 0.6 is required
for good PCA. Our data clearly supports the use of PCA for construction of index as
Bartlett’s test is highly signifcant (p ?0.00) and the value of KMO (i.e. 0.64) is greater
than 0.6 (see Table I).
The construction of FDI can be written in the following terms:
IFD ? 1 ? q
1
FIR ? q
2
M3 ? q
3
PRVT (9)
Where, FIR, M3 and PRVT are variables used and q
1
, q
2
, q
3
[…] […] q
i
are the factor
loadings of corresponding variable in each PC. The loadings or scores for the various
PCs estimated for construction of IFD are presented in Table II.
Because our frst PC(PC
1
) itself captures 96 per cent of the total variance[3], its values
are used as weights to construct a summary IFDin India as per the formula given below:
IFD
t
? (0.573) FIR
t
? (0.586) M3
t
? (0.573) PRVT
t
Where, the subscript “t” refers to the year from 1970-1971 to 2011-2012.
For example, IFDvalue for 1970-1971 is calculated by multiplying the factor loadings
as given in equation 10 with the respective values of FIR, M3 and PRVT for the year
1970-1971. Table III reports IFD values for the period 1971-1972 to 2011-2012.
Table I.
Results of Bartlett and
KMO test
K KMO and Bartlett’s test
KMO measure of sampling adequacy 0.64
Bartlett’s test of sphericity 318.3
(p-value) (0.00)
Table II.
Variables and its factor
loadings in each principal
component
Variables
Eigenvectors (PC
k
)
PC
1
PC
2
PC
3
FIR 0.573 ?0.699 0.427
M3 0.586 ?0.015 ?0.810
PRVT 0.573 0.715 0.401
Eigen values (E
k
) 2.886 0.099 0.015
369
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Subsequently, using the above index, we estimated long-run association and causal link
between economic growth (GDP), fnancial development (IFD) and trade openness
(TOP)[4] by using Johansen co-integration test and the Granger causality test used in the
VECM framework. Results of co-integration and causality tests are discussed in
Table IV.
The frst step in applying the co-integration technique is to determine the degree of
integration of each variable in model. The above table presents the ADF test as the test
of unit root. In Table IV, it is seen that all variables are found to be non-stationary in
levels but stationary in frst difference with intercept at 1 per cent level of signifcance.
Hence, all the variables are integrated of order 1 [I (1)]. Therefore, we proceed to apply
co-integration tests between the variables such as LGDP, LIFDand LTOP to detect any
possible long run equilibrium between the series.
After confrming the degree of integration of each variable, we proceed to estimate
long-run association among variables. To explore this relationship, the study
determines the optimal lag length of the model by using the VAR test and Lag 2 is
indentifed as the optimal lag based on the Schwarz information criterion and Hannan–
Quinn information criterion. The result is presented in Table V.
Fromthe above Table VI, it is concluded that the null hypothesis of no co-integration
is rejected at 1 per cent level as p-value is less than 5 per cent. But, the null of both at most
1 and at most 2 co-integration relationship among the variables cannot be rejected
because the trace statistics, i.e. 12.05 and 0.17, are less than the critical value, i.e. 15.49
and 3.84, respectively. Therefore, trace test indicates three co-integrating equations at
Table IV.
Unit root test results
Variables
ADF test Phillips–Perron test
Levels First Diff. Levels First Diff.
LGDP 0.096 (0.96) ?4.52 (0.00)* 0.40 (0.98) ?4.53 (0.00)*
LIFD ?0.58 (0.86) ?4.80 (0.00)* ?0.60 (0.85) ?4.86 (0.00)*
LTOP 0.41 (0.90) ?5.13 (0.00)* ?0.41 (0.89) ?5.13 (0.00)*
Notes: p-values are in the parentheses; *denotes signifcance at 1 per cent level
Table III.
Index of fnancial
development
Year IFD Year IFD
1970-1971 20.53 2001-2002 55.99
1971-1972 21.92 2002-2003 61.24
1972-1973 23.13 2003-2004 61.95
1973-1974 22.72 2004-2005 65.16
1974-1975 22.27 2005-2006 69.84
1975-1976 24.29 2006-2007 73.67
1976-1977 27.10 2007-2008 77.60
1977-1978 28.24 2008-2009 80.14
1978-1979 31.65 2009-2010 82.47
1979-1980 34.54 2010-2011 83.06
1980-1981 33.81 2011-2012 84.11
1981-1982 33.92
JFEP
6,4
370
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the 1 per cent level. Hence, there exists long-run association among all the variables.
Now, we can go for Granger causality approach using VECM.
The short-run error correction term, D [LGDP (?1)] shows the short-run adjustment
of GDP to its own deviation from long run. This is of the correct sign and statistically
signifcant, indicating that deviation from the long-run rate of GDP is corrected by 51
per cent in the next period. Similarly, D [LIFD (?2)] shows the short-run adjustment of
fnancial development to its own deviation fromlong run. This is of the correct sign, i.e.
negative and statistically signifcant, indicating that deviation fromthe long-run rate of
fnancial development is corrected by 28 per cent in the next period. Moreover, a
signifcant error correction confrms the existence of a stable long-run relationship
between the dependent variable, i.e. GDP, and the signifcant regressor, i.e. fnancial
sector development index, whereas the trade openness is not found to have any
signifcant relationship with LIFDand LGDP, both in the short run and long run. Hence,
the negative and signifcant error correction term also reveals that in the long-run
causality runs from fnancial development to economic growth but not vice versa (see
Table VII).
Since, we found long-run relationship between fnancial development, trade
openness and economic growth, the study proceeds to examine the short-run causality
link between these three key macroeconomic variables by applying the VEC Granger
causality/block exogeneity Wald tests. From Table VIII, it is clear that there is
bidirectional causality between fnancial development and economic growth measured
in terms of nominal GDP. However, the causality running from economic growth to
fnancial sector development is stronger than that of running from fnancial sector
development to economic growth. Similarly, it is also found that there is unidirectional
causality between GDP and trade openness, trade openness and fnancial development,
Table V.
VAR optimal lag selection
Lag Log L LR FPE AIC SC HQ
0 3.192895 NA 0.000199 ?0.010716 0.121244 0.035341
1 202.2382 353.8583 5.17e-09 ?10.56879 ?10.04095 ?10.38456
2 227.7135 41.04363* 2.10e-09 ?11.48409 ?10.56037* ?11.16168*
3 237.4825 14.11078 2.07e-09* ?11.52681* ?10.20721 ?11.06623
4 244.8779 9.449587 2.40e-09 ?11.43766 ?9.722181 ?10.83891
5 249.1337 4.728665 3.44e-09 ?11.17409 ?9.062734 ?10.43717
Notes: *Indicates lag order selected by the criterion; LR: sequential modifed LR test statistic (each
test at 5 per cent level); FPE: fnal prediction error; AIC: Akaike information criterion; SC: Schwarz
information criterion; HQ: Hannan–Quinn information criterion
Table VI.
Johansen’s maximum
likelihood estimates for
the long-run relationship
between LGDP, LIFD and
LTOP
Johansen test statistics
Testing hypothesis Critical value
H
0
H
A
?
trace
5 (%) Probability
r ?0 r ?0 42.09 29.80 0.00
r ?1 r ?1 12.05 15.49 0.15
r ?2 r ?2 0.17 3.84 0.67
371
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
which is running from economic growth and fnancial development to trade
liberalization.
5. Conclusion
This study empirically investigates the inter-relation between trade liberalization,
fnancial development and economic growth in India for the period 1970-2011. Firstly,
this study constructs a summary IFDby using PCAand subsequently using this index,
the long-run association and causal link between fnancial development, economic
growth and trade openness in India have been examined by using Johansen’s
co-integration test and VECM-based Granger causality approach, respectively.
The results obtained using Johansen’s co-integration test confrms the long-run
association between the above three variables in India. Further, the Granger causality
approach based on VECM confrms the unidirectional causality that is running from
fnancial development to economic growth in India. Similarly, VECGranger causality or
block exogeneity Wald test used for short-run causality also provides evidence of
bidirectional causality between fnancial development and economic growth. However,
it is found that in India, the demand following concept of fnancial development or
growth-led fnancial development is found to be more robust compared to the
supply-leading concept of fnancial development during 1971-2011. This is because
fnancial development measured by three parameters such as FIR, money supply (M3)
ratio and ratio of banking credit to private sector to GDP is more sensitive to growth.
Table VII.
Vector error correction
mechanism
Error correction D(LGDP) D(LIFD) D(LTOP)
CointEq1 ?0.07 (?2.32)** ?0.01 (?0.50) 0.33 (3.70)
D(LGDP(?1)) 0.52 (3.50)* ?0.89 (?8.30) 0.74 (1.82)
D(LGDP(?2)) ?0.37 (?1.30) 0.11 (0.51) 1.30 (1.65)
D(LIFD(?1)) ?0.20 (?0.77) 0.17 (0.91) 1.50 (2.06)
D(LIFD(?2)) ?0.29 (?2.15)** 0.10 (1.06) 0.41 (1.13)
D(LTOP(?1)) ?0.04 (?0.77) ?0.05 (?1.11) 0.00 (0.03)
D(LTOP(?2)) 0.00 (0.01) 0.08 (1.65) 0.26 (1.47)
C 0.13 (2.93) 0.12 (3.77) ?0.24 (?1.99)
Notes: t-values in parenthesis. *Indicates signifcance at the 1 per cent level; **5 per cent level of
signifcance
Table VIII.
VEC Granger causality/
block exogeneity Wald
tests
Null hypothesis ?
2
p-value
D(LIFD) does not Granger-cause D(LGDP) 4.79*** 0.07
D(LTOP) does not Granger-cause D(LGDP) 0.60 0.73
D(LGDP) does not Granger-cause D(LIFD) 72.21* 0.00
D(LTOP) does not Granger-cause D(LIFD) 3.47 0.17
D(LGDP) does not Granger-cause D(LTOP) 8.34** 0.01
D(LIFD) does not Granger-cause D(LTOP) 5.47*** 0.06
Notes: *1 per cent level of signifcance; **5 per cent level of signifcance; ***10 per cent level of
signifcance
JFEP
6,4
372
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Thus, the empirical fnds of the present study clearly associates with the old
Schumpeterian hypothesis (1911)[5] and Patrick’s (1966) supply-leading hypothesis of
fnancial development for India. The stylized empirical evidence for the steady state
effects of fnancial development on economic growth suggests the need to continue
economic reforms to further develop and strengthen the fnancial sector and supplement
the efforts aimed at achieving the national objective of high and sustained economic
growth.
The empirical evidence of the study, on the other hand, also reveals that trade
openness has no effect on both fnancial development and economic growth, rather it is
observed that both fnancial development and economic growth signifcantly affects
trade openness, and that fnancial development and economic growth are more sensitive
to each other. Hence, the policy implication that could be drawn from the above
conclusion is that because trade openness has no effect on both domestic fnancial sector
development and output growth, it would be better to deploy the resources into creating
a sustained domestic demand rather than concentrating more on the external front in
general and trade openness in particular.
Finally, the empirical evidence of the present study provides certain policy lessons to
many countries, particularly to those countries like China and Germany that are so far
concentrated on boosting their export growth by maintaining competitive exchange rate
through costly intervention policies. Its time China and the countries resorting to
export-led growth models need to realize that the costs of this kind of policies would
outweigh the benefts and in the mediumand long run, this would be counterproductive.
Hence, they should rather focus on improving the purchasing power of their domestic
populations by running defcits and spending their surpluses, instead of dumping them
in forex reserve accumulation.
Notes
1. According to Patrick’s (1966) supply-leading hypothesis, deliberative creation of new
fnancial institutions, instruments and services in advance of demand for them, leads to more
capital accumulation through increase the volume of savings, and hence economic growth.
2. The KMO test measures the sampling adequacy and used to test whether the partial
correlation among the variables are small. Bartlett’s test of sphericity is used to test the
correlation matrix is an identity matrix or not.
3. ?E
k
?2.886 ?0.099 ?0.015 ?3.00. PC
1
?2.886/3.00 ?0.962
4. All the variables such as IFD, GDP and TOP used in the present study are calculated in log
form.
5. According to the Schumpeterian hypothesis, active and well-functioning fnancial system,
particularly banking system, promotes economic growth by identifying and funding the
entrepreneurs with fnest chances of success as also implementing innovative products and
production methods.
References
Arestis, P. (2005), “Financial liberalization and the relationship between fnance and growth”,
CEPP Working Paper no. 05/05, University of Cambridge, 20 June.
373
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Atif, R.M., Jadoon, A., Zaman, K., Ismail, A. and Seemab, R. (2010), “Trade liberalization, fnancial
development and economic growth: evidence from Pakistan (1980-2009)”, Journal of
International Academic Research, Vol. 10 No. 2, pp. 30-37.
Bencivenga, V.R. and Smith, B.D. (1991), “Financial intermediation and endogenous growth”,
Review of Economic Studies, Vol. 58 No. 2, pp. 195-209.
Edwards, S. (1998), “Openness, Productivity and growth: what do we really know?”, Economic
Journal, Royal Economic Society, Vol. 108 No. 447, pp. 383-398.
Engle, R.F. and Granger, C.W.J. (1987), “Co-integration and error correction: representation,
estimation, and testing”, Econometrica, Vol. 55 No. 2, pp. 251-276.
Goldsmith, R.W. (1969), Financial Structure and Development, Yale University Press, NewHaven,
CT.
Greenwood, J. and Jovanovic, B. (1990), “Financial development, growth and the distribution of
income”, Journal of Political Economy, Vol. 98 No. 5, pp. 1076-1108.
Johansen, S. (1991), “Estimation and hypothesis testing of co-integration vectors in Gaussian
vector autoregressive models”, Econometrica, Vol. 59 No. 6, pp. 1551-1580.
Johansen, S. and Juselious, K. (1990), “Maximum likelihood estimation and inference on
cointegration with applications to the demand for the money”, Oxford Bulletin of Economics
and Statistics, Vol. 52 No. 2, pp. 169-210.
King, R.G. and Levine, R. (1993), “Finance and growth: Schumpeter might be right”, Quarterly
Journal of Economics, Vol. 108 No. 3, pp. 7171-7737.
Mckinnon, R.I. (1973), Money and Capital in Economic Development, Brooking Institution,
Washington, DC.
Newbery, D.M.G. and Stiglitz, J. (1984), “Pareto inferior trade”, Reviewof Economic Studies, Vol. 51
No. 1, pp. 1-12.
Patrick, H.T. (1966), “Financial development and economic growth in underdeveloped countries”,
Economic Development and Cultural Change, Vol. 14 No. 2, pp. 174-189.
Pagno, M. (1993), “Financial markets and growth: an overview”, European Economic Review,
Vol. 37 Nos 2/3, pp. 613-622.
Peter, B.H. (2003), “Capital account liberalization, the cost of capital, and economic growth”, NBER
Working Paper No. 9488, Cambridge, 21 February.
Philips, P.C.B. and Perron, P. (1988), “Testing for a Unitroot in Time series Regrreson”, Biometrica,
Vol. 75 No. 2, pp. 335-346.
Quy-Toan, D. and Levchenko, A.A. (2004), “Trade and Financial Development”, World Bank
Policy Research Working Paper No. 3347.
Rajan, R. and Zingales, L. (2003), “The great reversals: the politics of fnancial development in the
20th century”, Journal of Financial Economics, Vol. 69 No. 1, pp. 5-50.
Robinson, J. (1952), The Generalization of the General Theory, The Rate of Interests and Other
Essay, Macmillan, London, pp. 67-142.
Roubini, N. and Sala-I-Martin, X. (1991), “Financial development, trade regimes and economic
growth”, NBER Working Paper No, MA, No. 3876.
Salvatore, D. (2010), International Economics, Johan Wiley Publication, Hoboken, NJ.
Schumpeter, J.A. (1911), The Theory of Economic Development, Harvard University Press,
Cambridge.
Shahbaz, M., Shabbir, M.S. and Butt, M.S. (2013), “Effect of fnancial development on agricultural
growth in Pakistan”, International Journal of Social Economics, Vol. 40 No. 8, pp. 707-728.
JFEP
6,4
374
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Shaw, E.S. (1973), Financial Deepening in Economic Development, Harvard University Press,
Cambridge, M.A.
Shleifer, A. and Vishny, R.W. (1998), “The quality of government”, Harvard Institute of Economic
Research Working Paper 1847, Harvard.
Further reading
Gurley, J. and Shaw, E. (1955), “Financial aspects of economic development”, American Economic
Review, Vol. 45 No. 4, pp. 515-538.
About the authors
Dogga Satyanarayana Murthy is currently pursuing PhD in Economics in Pondicherry
University, India. I have qualifed UGC NET and AP SLET and received PRATIBA AWARD in
BEd. I have also attended and presented papers in many national workshops, national and
international seminars and a paper also accepted and to be published in JHM, SAGE Publication,
Jaipur, India, and another paper to be published in Asian Economic Review, Indian Institute of
Economic Society, Hyderabad, India. I have also published two more papers in Global Journal of
Finance and Management as seminar proceedings. Dogga Satyanarayana Murthy is a
corresponding author and can be contacted at: [email protected]
Suresh Kumar Patra is currently pursuing PhDin Economics in Pondicherry University, India.
I received a Gold Medal as the toper in MPhil in Economics from Pondicherry University and
qualifed for the Tamil Nadu SETin 2013, and I amalso receiving the ICSSRFellowship. I do have
special interest in Mathematical Economics, Econometrics, Economic Growth and Development;
and also in Health Economics. I have also published one book on “An Evaluation of National Rural
Health Mission In Odisha: With special reference to Puri District”. I have also attended and
presented papers in many national workshops, national and international seminars and a paper
also accepted and to be published in JHM, SAGE Publication, Jaipur, India.I have also published
two more papers in Global Journal of Finance and Management as seminar proceedings.
Amaresh Samantaraya is presently working as Associate Professor in the Department of
Economics, Pondicherry University. He teaches courses on Macroeconomics, Economics of
Money &Banking and Applied Econometrics. His research interest is diverse, covering empirical
issues related to conduct of monetary policy in India, fnancial sector developments and
inter-regional disparity in Indian growth experience. He has published numerous research studies
in various national and international professional journals in economics and fnance, and has
published a book on the conduct of monetary policy in India. Prior to joining academics, he worked
for over a decade in various capacities, undertaking economic research in the Reserve Bank of
India (RBI) – the central bank and the monetary authority for India.
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
375
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_297673168.pdf
The purpose of this article is to examine the inter-relationship and direction of causality
among three macroeconomic variables such as trade liberalization, financial development and economic
growth
Journal of Financial Economic Policy
Trade Openness, Financial Development Index and Economic Growth: Evidence from
India (1971-2012)
Dogga Satyanarayana Murthy Suresh Kumar Patra Amaresh Samantaraya
Article information:
To cite this document:
Dogga Satyanarayana Murthy Suresh Kumar Patra Amaresh Samantaraya , (2014),"Trade Openness,
Financial Development Index and Economic Growth", J ournal of Financial Economic Policy, Vol. 6 Iss 4 pp.
362 - 375
Permanent link to this document:
http://dx.doi.org/10.1108/J FEP-10-2013-0056
Downloaded on: 24 January 2016, At: 21:50 (PT)
References: this document contains references to 26 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 344 times since 2014*
Users who downloaded this article also downloaded:
Omotola Awojobi, (2013),"Does trade openness and financial liberalization foster growth: An empirical
study of Greek economy", International J ournal of Social Economics, Vol. 40 Iss 6 pp. 537-555 http://
dx.doi.org/10.1108/03068291311321848
Salih Turan Katircioglu, Neslihan Kahyalar, Hasret Benar, (2007),"Financial development, trade and growth
triangle: the case of India", International J ournal of Social Economics, Vol. 34 Iss 9 pp. 586-598 http://
dx.doi.org/10.1108/03068290710778615
Muhammad Tahir, Imran Khan, (2014),"Trade openness and economic growth in the Asian region", J ournal
of Chinese Economic and Foreign Trade Studies, Vol. 7 Iss 3 pp. 136-152 http://dx.doi.org/10.1108/
J CEFTS-05-2014-0006
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Trade Openness, Financial
Development Index and
Economic Growth
Evidence from India (1971-2012)
Dogga Satyanarayana Murthy, Suresh Kumar Patra and
Amaresh Samantaraya
Department of Economics, Pondicherry University, Puducherry, India
Abstract
Purpose – The purpose of this article is to examine the inter-relationship and direction of causality
among three macroeconomic variables such as trade liberalization, fnancial development and economic
growth.
Design/methodology/approach – The empirical analysis is based on the principal component
analysis as method to construct fnancial development index (FDI), augmented Dickey–Fuller and
Phillips–Perron tests as the unit root test, Johansen’s co-integration test and VECM for direction of
causality in the long run among TOP, FDI and economic growth.
Findings – The empirical results confrmed that there exists a long-run association among trade
openness, fnancial development and economic growth. This study has also found that there is
bidirectional causality between fnancial development and growth. However, the causality runs from
growth to fnance is stronger than that fromfnance to growth. This study also observed unidirectional
causality that runs from fnancial development and economic growth to trade openness.
Research limitations/implications – The policy implications that could be drawn fromthe present
study is that, initiation of fnancial reforms to improve the size of fnancial systemwould lead to higher
economic growth. Another key implication fromthis study is that because trade openness has no effect
on both domestic fnancial sector development and output growth, it would be better to deploy the
resources into creating a sustained domestic demand rather than concentrating more on the external
front in general and trade openness in particular.
Originality/value – The study constructs a summary IFDfor India by taking into account four broad
fnancial development indicators for the period 1971-2012. The present paper also suggests that it
would be better to deploy the resources to create a sustained domestic demand rather than
concentrating more on the external front in general and trade openness in particular.
Keywords Trade openness, Financial development index, Economic growth, VECM
Paper type Research paper
1. Introduction
The relationship between trade openness, fnancial development and economic growth
has been widely discussed and debated in both the theoretical and empirical literature.
It is argued that trade openness and fnancial development policies positively infuence
gross domestic product growth by reducing the distortion in the production process
(Atif et al., 2010). It is also widely evident fromthe fact that countries with high degree of
JEL classifcation – F13, G21, O1, B23
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1757-6385.htm
JFEP
6,4
362
Journal of Financial Economic Policy
Vol. 6 No. 4, 2014
pp. 362-375
©Emerald Group Publishing Limited
1757-6385
DOI 10.1108/JFEP-10-2013-0056
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
trade openness and well-developed fnancial system have been registering higher GDP
growth as compared to the countries with low-fnancial sector development and restrictive
trade policies. Opening up of an economy for the cross border fows of goods and services
creates high competitive environment, which will drive down the revenue of existing frms
and diminish their profts, requiring them to search for external sources of fnance
(Quy-Toan and Levchenko, 2004). These sources of fnance will be available only if the
fnancial system is able to solve the problem of asymmetric information and other
market imperfections in the present fnancial system. As a result, present frms and their
rich, elite owners will now have incentives to support the compulsory institutional
reforms to make the fnancial system effcient and well-functioning. The resulting
reforms, in turn, will extend the size of the fnancial systemand thereby foster economic
growth. Thus, free trade can be an important ingredient that stimulates economic
growth through domestic fnancial sector development. However, recent global
fnancial crisis that cropped up in July 2007 in the USA and later enveloped the entire
world has again re-opened the debate over open economy policy, as it resulted in
macroeconomic instability of the national economies and also put entire world economy
in trouble with falling outputs and rising unemployment levels across the world.
Similarly, the role of fnancial sector development in economic growth is the long-run
debated issues among economists (Schumpeter, 1911; Patrick, 1966; Goldsmith, 1969:
McKinnon, 1973; Shaw, 1973). With the unleashing of endogenous growth theory,
several studies have been attempted to knowhowthe development of fnancial markets
and institutions may affect economic growth (Greenwood and Jovanovic, 1990;
Bencivenga and Smith, 1991; King and Levine, 1993; Arestis, 2005; Shahbaz et al., 2013).
Finance can affect growth in an endogenous growth model by allowing higher returns
on investment (Greenwood and Jovanovic, 1990), through increasing the rate of savings
(Bencivenga and Smith, 1991) and by rising human capital accumulation. It implies that
from the macroeconomic point of view, well-developed fnancial markets and
institutions of a country will able to convert a given amount of inputs, k, into larger
amount of output, Y. This is because the production function is a rising function of the
fnancial development of the economy (Roubini and Sala-I Martin, 1992).
The important question that rambles in the mind of many researchers whether there
exists any inter-relationships among trade openness, fnancial development and
economic growth. Hence, an attempt has been undertaken to know the inter-relation
between the above three variables in India for the period from 1970-1971 to 2011-2012.
India, like many developing countries, in 1991, as a part of the structural adjustment and
macroeconomics stabilization programmes, introduced various macroeconomic, trade
and fnancial sector reforms to promote economic growth. Thus, liberal fnancial policy
regime has been replaced in place of an old controlled regime. This resulted in
tremendous improvement in the fnancial systemin terms of total fnancial assets, bank
credit to private sector as percentage of GDP, increase in the quality of bank assets, etc.
In 1991, fnancial assets as percentage of GDP were 17.61 per cent and it increased to
116.93 in 2011. Similarly, trade reforms in the form of removal of license-permit raj,
abolition of quantitative restrictions and reducing import tariff resulted in signifcant
rise in Indian foreign trade. In 1991, India’s foreign trade as percentage of GDP has been
increased from 6.72 per cent in 1991 to 73.14 per cent in 2011. The average growth rate
during this period (i.e. from 1991-2011) was 6.7 per cent.
363
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The present study constructs a summary index of fnancial development (IFD) by
applying principal component analysis (PCA) and uses co-integration test for long-run
association and vector error correction mechanism (VECM)-based Granger causality
approach to knowthe causality between trade liberalization, fnancial development and
economic growth. Thus, the major advantages of the presents study are constructing a
broad measure for assessing fnancial development in India by taking into account a
large set of variables, which together represent the development of fnancial system in
terms of its size, activity and using appropriate econometric tools while taking into
account non-stationarity of the time series data to examine the causal link between trade
openness, fnancial development and economic growth. Our results confrmthe long-run
association between these three macroeconomic variables in India. In addition to this,
our results based on Granger causality approach used in VECM framework strongly
support the old Schumpeter (1911) and Patrick’s (1966)[1] fnance-led growth hypothesis
in the long run.
The rest of the paper is as follows. Section 2 describes theoretical and conceptual
framework of the study. Section 3 outlines the data and methodology. The empirical
results are reported in Section 4. The last section concludes.
2.Theoretical underpinning
2.1 Trade openness – economic growth
The concept of absolute and comparative cost advantage theories emphasized that like
inter-regional trade, the international trade is also benefcial for the trading countries
because each country will specialize in the production of particular commodities but not
in the all commodities, and also countries differ in the resource endowments. Therefore,
when a country enters into trade with other country, it can export those commodities in
which its production cost is less, and can import those commodities in which its
production cost is high. This results in greater output and consumer welfare in both the
trading countries, which in turn, will lead to higher employment and hence economic
growth.
Thus, the classical economists were in favour of free trade policy, as they assumed
that free trade among different nations maximizes the output and employment of all the
participating countries (Salvatore, 2010). Edwards (1998) noted that countries that are
more open to the rest of the world are better placed in capturing the advanced
technologies of leading nations. If the costs of technological imitation are less than the
costs of domestic-developed innovations, then a poorer country will grow faster than a
high developed one. This faster rate of growth will continue as long as that country
remains opened to attract newideas until, at some point, equilibriumis reached and the
rate of growth slows down.
However, economists like Prebish, Singer, Myint and Miyrdal argues that free trade
between developed and developing countries shifts the gains from developing to
developed nation because developing countries are largely dependent on the production
of primary goods, whereas the developed nations mostly hinge upon manufacturing
products. The demand and prices of the primary goods will often deteriorate as
compared to manufacturing product. Therefore, free trade leads to secular deterioration
in the terms of trade of developing countries.
JFEP
6,4
364
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.2 Trade openness – fnancial development
There are signifcant differences in fnancial sector development across countries and even
within the country over a period. This is mainly because fnancial development is also
infuencedbyother factors like legal systeminthe country(Shleifer andVishny, 1998), trade
and fnancial openness (Rajan and Zingales, 2003). It is argued that the trade openness
affects fnancial development because free trade will result in uncertainty and income
inconsistency of agents, which in turn raises the demand for insurance and other fnancial
services and thereby increases the size of the fnancial system(Newbery and Stiglitz, 1984).
Furthermore, free trade among different nations generally will increase the demand for
external fnance, as theyproduce more fnancial-dependent goods. Thus, free trade increases
the demand for external fnance and thereby size and quality of the fnancial system.
However, Quy-Toan and Levchenko (2004) noted that if there is free trade between
rich and poor countries, in rich countries, more trade would be associated with faster
fnancial development as they are specialized in fnancial-dependent good. Whereas
more trade lead to deterioration in the size of the fnancial system in poor countries, as
they import fnancial-dependent goods rather than produce them domestically. While,
Rajan and Zingales (2003) postulate that trade openness is linked with fnancial market
development, especially when cross-border capital fows are free, and that changes in
openness are associated with changes in the size of fnancial markets.
Further, free trade helps to develop the domestic fnancial markets and then the
economic growth. Because free trade expands the size of the market for domestic goods,
which in turn encourages the domestic production and thus production of more goods
and services, more capital is required. Therefore, allowing foreign capital into domestic
fnancial markets by fnancial openness increases the availability of funds, which in
turn lowers the cost of borrowing and thereby increases the investment and economic
growth (Peter, 2003). Thus, trade openness and fnancial openness are not substitute
rather they are complementary to each other as their co existence will result in domestic
fnancial sector development and hence higher economic growth. Similarly, Rajan and
Zingales (2003) hypothesis also suggest that simultaneous openness of both trade and
capital fows are preconditions for fnancial development.
2.3 Financial development – economic growth
Financial development affects economic growth mainly by increasing the marginal
productivity of capital through transferring funds from relatively less to relatively more
productive uses and also through increasing the rate of saving (Bencivenga and Smith,
1991). Financial institutions increase the rate of saving by pooling small savings from
different segments across country and making them into a huge amount of loans for
productive uses. Similarly, fnancial system can increase the marginal productivity of
capital because they are more effcient and experts in the collection and evaluation of
information on the different alternative investment projects and they can also reduce risk to
individual to undertake riskier and more productive investment by directing funds to
relative liquidity and higher-yield assets (Pagano, 1993). Thus, fnancial development frst
increases the rate of saving and then to increase the marginal productivity of capital,
fnancial institutions transferringthese funds toproductive uses andtherebyoutput growth.
Similarly, economic growth also encourages the development of fnancial markets and
institutions by creating demand for different types of fnancial services (Robinson, 1952;
Patrick, 1966). As economygrows, it creates additional demandfor fnancial services, which
365
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
bring about a supply response in the growth of the fnancial system. Similarly, as economy
grows, there will be a greater variance in the growth rates among different sectors in the
economy and thereby an increase in the demand for fnancial intermediation to transfer
funds from slow-growth sectors to fast-growing sector. Thus, fnancial development and
economic growth infuences each other. More specifcally, in initial stage of development
growth follows fnance and in advance stage, fnance follow economic growth (Patrick,
1966).
3. Data and methodology
This section reports the data and methods used to examine the causal link between
fnancial development, economic growth and trade openness, and it also summarizes
various indicators used in the present study for assessing fnancial sector development.
3.1 Data and methods used
To construct a summary IFD for India, present study takes into account various
fnancial development indicators such as the ratio of broad money (M3) to narrow
money (M1), which is used as a proxy fnancial innovation ratio (FIR), money supply as
percentage of GDP (M3) and the ratio of banking credit to private sector to GDP.
Similarly, as an indicator for economic activity, the study used GDP at factor cost at
constant prices (2004-2005 ?100) and the amount of total trade i.e. sum of exports and
imports as percentage of GDP used as proxy of trade openness (TOP). The annual time
series data on the above variables has been collected from the Reserve Bank of India
handbook of statistics, 2012-2013.
The empirical analysis mainly consists of three parts:
(1) construction of fnancial development index (FDI) by using PCA;
(2) subsequently using this index, estimating Johansen’s co-integration test to
analyse the long-run association between real GDP growth, fnancial
development (IFD) and trade openness (TOP); and
(3) we proceed for causality test based on VECM framework.
3.2 Principal component analysis
The present study used PCA for construction of a summary IFD. The PCA is a
multivariate technique used to transform the original data consisting of a set of
variables into a linear combination of a small set of variables known as principal
components (PCs) so that the bulk of the variation in the original data is explained.
These PCs are newentities and they are extracted fromthe original data set of variables
after taking into account the correlation matrix. Among extracted PCs, the frst PC
would be best component as it explains greater variance than the rest of the PCs.
Linear combinations of each component’s factor loading can be expressed as follows:
PC
k
? ?
1k
?
1
? ?
2k
?
2
? ......... ? ?
nk
?
n
(1)
Where,
PC
k
is the kth PC.
X
1
, X
2
[…], X
n
are the variables used in the PCA.
?
1k
, ?
2k
, […] ?
nk
are factor loadings of respective X
i
in kth PC.
JFEP
6,4
366
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The number of PCs extracted should be less than or equal to the number of variables
used in the analysis.
3.3 Unit root test
We used the augmented Dickey–Fuller (ADF) test as test of unit root which is based on
the assumption of serially correlated error terms. The ADF test consists of estimating
the following regression:
?Y
t
? ? ? ?
t
? ??
t?1
? ?
1
??
t?1
? ........ ? ?
p?1
??
t?p?1
? ?
t
(2)
Where ?
t
is serially correlated.
The Phillips–Perron (1988) test is mostly used to examine time series whose
differences may have mixed autoregressive–moving average (p, q) processes of
unknown order in that the test statistic includes a nonparametric allowance for
heteroscedasticity and serial correlation in testing the regression. It involves estimating
the following equation:
?
t
? ?
0
? ?
1
?
t?1
? ?
2
(
t ? T/2
)
? v
t
(3)
Where T indicates the number of observations and ?
t
represents the error term. There
will be no unit root if ?1
?1
?0, as in the ADF test, and thus we can test the stationarity
of a variable without the trend by dropping the trend term.
3.4 Co-integration test
To evaluate the long-run relationship between real GDP growth (LGDP), fnancial
development (LIFD) and trade openness (LTOP), we have used the maximumlikelihood
test procedure recognized by Johansen and Juselious (1990) and Johansen (1991).
Especially, the VAR (vector autoregression) with m-lag and Gaussian error can take the
following form (if X
t
is a vector of n stochastic variables):
?X
t
? ? ? ?
1
?X
t?1
? ··· ? ?
m?1
?X
t?m?1
? ?X
t?1
? ?
t
(4)
Where ?
1
, ?
p?1
and ?are coeffcient matrices, ?
t
is a vector of white noise process and
? contains all deterministic elements.
Determination of the rank (r) of Matrix ?? is the central point of conducting
co-integration procedure developed by Johansen. There are mainly three feasible
outcomes in the present application. Firstly, it can be of full rank (r ?n) (which would
imply that the variables are stationary processes), which contradicts the previous
fnding of non-stationarity. Secondly, the rank of ?can be zero (r ?0), which represents
no long-run relationship among the variables. It will be suitable to estimate the model in
either frst differences or levels, when ??is of either zero rank or full rank, respectively.
Finally, when there is at most r co-integrating vectors 0 ? r ? n (i.e. reduced rank), it
suggests that there are (n ? r) common stochastic trends. On the basis of Akaike
information criterion (AIC), the lag in the VAR is chosen. Johansen’s co-integration
procedure deals with two likelihood ratio test statistics such as trace test and the
maximum eigenvalue (?-max) test. But, in our study, we have taken into account the
trace test to determine which of the three as stated above is supported by the data.
367
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.5 Granger causality
If GDP growth (LGDP) has a long-run relationship with money fnancial development
(IFD) and trade openness (LTOP), the further step is to investigate the causal
relationship among these variables because if two or more variables are co-integrated,
there must exist causality in at least one direction (Engel and Granger, 1987). We,
therefore, move further to determine whether real GDP will Granger-cause fnancial
development and trade openness and vice-versa, using VECM which is given in the
following as form:
?z
t
? ? ? ?t ? ?z
t?1
?
?
i?1
p?i
?
t
?y
t?i
?
?
i?1
p?1
?
t
?x
t?i
? ?
t
(5)
Where, ? is the frst-difference operator. The long-run multiplier matrix ? as:
? ?
?
?
YY
?
YX
?
XY
?
XX
?
As the diagonal elements of the above matrix are unrestricted, the selected series can
followeither I(1) or I(0). If ?
YY
?0, then Yis I(0). On the contrary, if ?
YY
?0 then Yis I(1).
The VECM procedures described above are imperative in the testing of at most one
co-integrating vector between dependent variable y
t
and a set of explanatory
variables x
t
.
Hence, the VEC model is as follows:
?LGDP
t
? ?
g
(LGDP
t?1
? a
0
? a
1
LTOP
t?1
? a
2
LIFD) ?
?
?
11
(i)?LGDP
t?i
?
?
?
12
(i)?LIFD
t?i
?
?
?
13
(i)?LTOP
t?i
? ?
gt
(6)
?LIFD
t
? ?
m
(LGDP
t?1
? a
0
? a
1
LTOP
t?1
? a
2
LIFD) ?
?
?
21
(i)?LGDP
t?i
?
?
?
22
(i)?LIFD
t?i
?
?
?
23
(i)?LTOP
t?i
? ?
mt
(7)
?LTOP
t
? ?
P
(LGDP
t?1
? a
0
? a
1
LTOP
t?1
? a
2
LIFD) ?
?
?
31
(i)?LGDP
t?i
?
?
?
32
32(i)?LIFD
t?i
?
?
?
33
(i)?LTOP
t?i
? ?
PT
(8)
The above systemis a three variables VARmodel augmentedby e
t?1
. If ?s are zero, then
it is a simple VAR model and there is no co-integration. Hence, co-integration and error
correction are equivalent representation (Granger representation theorem).
In the view of Engle and Granger (1987), if there exists co-integration between two
variables, then a conventional causality test commonly known as error correction model
should be applied. The VECM specifcation restricts the long-run behaviour of the
endogenous variables to converge to their co-integrating relationships while allowing a
wide range of short-run dynamics (Granger causality). As the deviation from long-run
equilibrium is rectifed gradually through a series of partial short-run adjustments,
co-integrating term is otherwise known as the error correction term.
JFEP
6,4
368
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4. Report on empirical results
In this section, we have reported results on construction of a summary IFD for India by
using PCA followed by empirical results for co-integration and causality tests.
4.1 Constructing FDI
In PCA, loadings or scores refecting how each of the indicators included in each of the
index contributes to the PC were computed. The factor loadings fromthe best PC, which
explain more than 95 per cent of the total variance, are used for construction of the fnal
index.
Prior to using PCA, we test for factorability of the data used in the study. Bartlett’s
test of sphericity and Kaiser–Meyer–Olkin measure of sampling adequacy (KMO)[2] are
carried out to test the suitability of the data for PCA. For KMO, a value of 0.6 is required
for good PCA. Our data clearly supports the use of PCA for construction of index as
Bartlett’s test is highly signifcant (p ?0.00) and the value of KMO (i.e. 0.64) is greater
than 0.6 (see Table I).
The construction of FDI can be written in the following terms:
IFD ? 1 ? q
1
FIR ? q
2
M3 ? q
3
PRVT (9)
Where, FIR, M3 and PRVT are variables used and q
1
, q
2
, q
3
[…] […] q
i
are the factor
loadings of corresponding variable in each PC. The loadings or scores for the various
PCs estimated for construction of IFD are presented in Table II.
Because our frst PC(PC
1
) itself captures 96 per cent of the total variance[3], its values
are used as weights to construct a summary IFDin India as per the formula given below:
IFD
t
? (0.573) FIR
t
? (0.586) M3
t
? (0.573) PRVT
t
Where, the subscript “t” refers to the year from 1970-1971 to 2011-2012.
For example, IFDvalue for 1970-1971 is calculated by multiplying the factor loadings
as given in equation 10 with the respective values of FIR, M3 and PRVT for the year
1970-1971. Table III reports IFD values for the period 1971-1972 to 2011-2012.
Table I.
Results of Bartlett and
KMO test
K KMO and Bartlett’s test
KMO measure of sampling adequacy 0.64
Bartlett’s test of sphericity 318.3
(p-value) (0.00)
Table II.
Variables and its factor
loadings in each principal
component
Variables
Eigenvectors (PC
k
)
PC
1
PC
2
PC
3
FIR 0.573 ?0.699 0.427
M3 0.586 ?0.015 ?0.810
PRVT 0.573 0.715 0.401
Eigen values (E
k
) 2.886 0.099 0.015
369
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Subsequently, using the above index, we estimated long-run association and causal link
between economic growth (GDP), fnancial development (IFD) and trade openness
(TOP)[4] by using Johansen co-integration test and the Granger causality test used in the
VECM framework. Results of co-integration and causality tests are discussed in
Table IV.
The frst step in applying the co-integration technique is to determine the degree of
integration of each variable in model. The above table presents the ADF test as the test
of unit root. In Table IV, it is seen that all variables are found to be non-stationary in
levels but stationary in frst difference with intercept at 1 per cent level of signifcance.
Hence, all the variables are integrated of order 1 [I (1)]. Therefore, we proceed to apply
co-integration tests between the variables such as LGDP, LIFDand LTOP to detect any
possible long run equilibrium between the series.
After confrming the degree of integration of each variable, we proceed to estimate
long-run association among variables. To explore this relationship, the study
determines the optimal lag length of the model by using the VAR test and Lag 2 is
indentifed as the optimal lag based on the Schwarz information criterion and Hannan–
Quinn information criterion. The result is presented in Table V.
Fromthe above Table VI, it is concluded that the null hypothesis of no co-integration
is rejected at 1 per cent level as p-value is less than 5 per cent. But, the null of both at most
1 and at most 2 co-integration relationship among the variables cannot be rejected
because the trace statistics, i.e. 12.05 and 0.17, are less than the critical value, i.e. 15.49
and 3.84, respectively. Therefore, trace test indicates three co-integrating equations at
Table IV.
Unit root test results
Variables
ADF test Phillips–Perron test
Levels First Diff. Levels First Diff.
LGDP 0.096 (0.96) ?4.52 (0.00)* 0.40 (0.98) ?4.53 (0.00)*
LIFD ?0.58 (0.86) ?4.80 (0.00)* ?0.60 (0.85) ?4.86 (0.00)*
LTOP 0.41 (0.90) ?5.13 (0.00)* ?0.41 (0.89) ?5.13 (0.00)*
Notes: p-values are in the parentheses; *denotes signifcance at 1 per cent level
Table III.
Index of fnancial
development
Year IFD Year IFD
1970-1971 20.53 2001-2002 55.99
1971-1972 21.92 2002-2003 61.24
1972-1973 23.13 2003-2004 61.95
1973-1974 22.72 2004-2005 65.16
1974-1975 22.27 2005-2006 69.84
1975-1976 24.29 2006-2007 73.67
1976-1977 27.10 2007-2008 77.60
1977-1978 28.24 2008-2009 80.14
1978-1979 31.65 2009-2010 82.47
1979-1980 34.54 2010-2011 83.06
1980-1981 33.81 2011-2012 84.11
1981-1982 33.92
JFEP
6,4
370
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the 1 per cent level. Hence, there exists long-run association among all the variables.
Now, we can go for Granger causality approach using VECM.
The short-run error correction term, D [LGDP (?1)] shows the short-run adjustment
of GDP to its own deviation from long run. This is of the correct sign and statistically
signifcant, indicating that deviation from the long-run rate of GDP is corrected by 51
per cent in the next period. Similarly, D [LIFD (?2)] shows the short-run adjustment of
fnancial development to its own deviation fromlong run. This is of the correct sign, i.e.
negative and statistically signifcant, indicating that deviation fromthe long-run rate of
fnancial development is corrected by 28 per cent in the next period. Moreover, a
signifcant error correction confrms the existence of a stable long-run relationship
between the dependent variable, i.e. GDP, and the signifcant regressor, i.e. fnancial
sector development index, whereas the trade openness is not found to have any
signifcant relationship with LIFDand LGDP, both in the short run and long run. Hence,
the negative and signifcant error correction term also reveals that in the long-run
causality runs from fnancial development to economic growth but not vice versa (see
Table VII).
Since, we found long-run relationship between fnancial development, trade
openness and economic growth, the study proceeds to examine the short-run causality
link between these three key macroeconomic variables by applying the VEC Granger
causality/block exogeneity Wald tests. From Table VIII, it is clear that there is
bidirectional causality between fnancial development and economic growth measured
in terms of nominal GDP. However, the causality running from economic growth to
fnancial sector development is stronger than that of running from fnancial sector
development to economic growth. Similarly, it is also found that there is unidirectional
causality between GDP and trade openness, trade openness and fnancial development,
Table V.
VAR optimal lag selection
Lag Log L LR FPE AIC SC HQ
0 3.192895 NA 0.000199 ?0.010716 0.121244 0.035341
1 202.2382 353.8583 5.17e-09 ?10.56879 ?10.04095 ?10.38456
2 227.7135 41.04363* 2.10e-09 ?11.48409 ?10.56037* ?11.16168*
3 237.4825 14.11078 2.07e-09* ?11.52681* ?10.20721 ?11.06623
4 244.8779 9.449587 2.40e-09 ?11.43766 ?9.722181 ?10.83891
5 249.1337 4.728665 3.44e-09 ?11.17409 ?9.062734 ?10.43717
Notes: *Indicates lag order selected by the criterion; LR: sequential modifed LR test statistic (each
test at 5 per cent level); FPE: fnal prediction error; AIC: Akaike information criterion; SC: Schwarz
information criterion; HQ: Hannan–Quinn information criterion
Table VI.
Johansen’s maximum
likelihood estimates for
the long-run relationship
between LGDP, LIFD and
LTOP
Johansen test statistics
Testing hypothesis Critical value
H
0
H
A
?
trace
5 (%) Probability
r ?0 r ?0 42.09 29.80 0.00
r ?1 r ?1 12.05 15.49 0.15
r ?2 r ?2 0.17 3.84 0.67
371
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
which is running from economic growth and fnancial development to trade
liberalization.
5. Conclusion
This study empirically investigates the inter-relation between trade liberalization,
fnancial development and economic growth in India for the period 1970-2011. Firstly,
this study constructs a summary IFDby using PCAand subsequently using this index,
the long-run association and causal link between fnancial development, economic
growth and trade openness in India have been examined by using Johansen’s
co-integration test and VECM-based Granger causality approach, respectively.
The results obtained using Johansen’s co-integration test confrms the long-run
association between the above three variables in India. Further, the Granger causality
approach based on VECM confrms the unidirectional causality that is running from
fnancial development to economic growth in India. Similarly, VECGranger causality or
block exogeneity Wald test used for short-run causality also provides evidence of
bidirectional causality between fnancial development and economic growth. However,
it is found that in India, the demand following concept of fnancial development or
growth-led fnancial development is found to be more robust compared to the
supply-leading concept of fnancial development during 1971-2011. This is because
fnancial development measured by three parameters such as FIR, money supply (M3)
ratio and ratio of banking credit to private sector to GDP is more sensitive to growth.
Table VII.
Vector error correction
mechanism
Error correction D(LGDP) D(LIFD) D(LTOP)
CointEq1 ?0.07 (?2.32)** ?0.01 (?0.50) 0.33 (3.70)
D(LGDP(?1)) 0.52 (3.50)* ?0.89 (?8.30) 0.74 (1.82)
D(LGDP(?2)) ?0.37 (?1.30) 0.11 (0.51) 1.30 (1.65)
D(LIFD(?1)) ?0.20 (?0.77) 0.17 (0.91) 1.50 (2.06)
D(LIFD(?2)) ?0.29 (?2.15)** 0.10 (1.06) 0.41 (1.13)
D(LTOP(?1)) ?0.04 (?0.77) ?0.05 (?1.11) 0.00 (0.03)
D(LTOP(?2)) 0.00 (0.01) 0.08 (1.65) 0.26 (1.47)
C 0.13 (2.93) 0.12 (3.77) ?0.24 (?1.99)
Notes: t-values in parenthesis. *Indicates signifcance at the 1 per cent level; **5 per cent level of
signifcance
Table VIII.
VEC Granger causality/
block exogeneity Wald
tests
Null hypothesis ?
2
p-value
D(LIFD) does not Granger-cause D(LGDP) 4.79*** 0.07
D(LTOP) does not Granger-cause D(LGDP) 0.60 0.73
D(LGDP) does not Granger-cause D(LIFD) 72.21* 0.00
D(LTOP) does not Granger-cause D(LIFD) 3.47 0.17
D(LGDP) does not Granger-cause D(LTOP) 8.34** 0.01
D(LIFD) does not Granger-cause D(LTOP) 5.47*** 0.06
Notes: *1 per cent level of signifcance; **5 per cent level of signifcance; ***10 per cent level of
signifcance
JFEP
6,4
372
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Thus, the empirical fnds of the present study clearly associates with the old
Schumpeterian hypothesis (1911)[5] and Patrick’s (1966) supply-leading hypothesis of
fnancial development for India. The stylized empirical evidence for the steady state
effects of fnancial development on economic growth suggests the need to continue
economic reforms to further develop and strengthen the fnancial sector and supplement
the efforts aimed at achieving the national objective of high and sustained economic
growth.
The empirical evidence of the study, on the other hand, also reveals that trade
openness has no effect on both fnancial development and economic growth, rather it is
observed that both fnancial development and economic growth signifcantly affects
trade openness, and that fnancial development and economic growth are more sensitive
to each other. Hence, the policy implication that could be drawn from the above
conclusion is that because trade openness has no effect on both domestic fnancial sector
development and output growth, it would be better to deploy the resources into creating
a sustained domestic demand rather than concentrating more on the external front in
general and trade openness in particular.
Finally, the empirical evidence of the present study provides certain policy lessons to
many countries, particularly to those countries like China and Germany that are so far
concentrated on boosting their export growth by maintaining competitive exchange rate
through costly intervention policies. Its time China and the countries resorting to
export-led growth models need to realize that the costs of this kind of policies would
outweigh the benefts and in the mediumand long run, this would be counterproductive.
Hence, they should rather focus on improving the purchasing power of their domestic
populations by running defcits and spending their surpluses, instead of dumping them
in forex reserve accumulation.
Notes
1. According to Patrick’s (1966) supply-leading hypothesis, deliberative creation of new
fnancial institutions, instruments and services in advance of demand for them, leads to more
capital accumulation through increase the volume of savings, and hence economic growth.
2. The KMO test measures the sampling adequacy and used to test whether the partial
correlation among the variables are small. Bartlett’s test of sphericity is used to test the
correlation matrix is an identity matrix or not.
3. ?E
k
?2.886 ?0.099 ?0.015 ?3.00. PC
1
?2.886/3.00 ?0.962
4. All the variables such as IFD, GDP and TOP used in the present study are calculated in log
form.
5. According to the Schumpeterian hypothesis, active and well-functioning fnancial system,
particularly banking system, promotes economic growth by identifying and funding the
entrepreneurs with fnest chances of success as also implementing innovative products and
production methods.
References
Arestis, P. (2005), “Financial liberalization and the relationship between fnance and growth”,
CEPP Working Paper no. 05/05, University of Cambridge, 20 June.
373
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Atif, R.M., Jadoon, A., Zaman, K., Ismail, A. and Seemab, R. (2010), “Trade liberalization, fnancial
development and economic growth: evidence from Pakistan (1980-2009)”, Journal of
International Academic Research, Vol. 10 No. 2, pp. 30-37.
Bencivenga, V.R. and Smith, B.D. (1991), “Financial intermediation and endogenous growth”,
Review of Economic Studies, Vol. 58 No. 2, pp. 195-209.
Edwards, S. (1998), “Openness, Productivity and growth: what do we really know?”, Economic
Journal, Royal Economic Society, Vol. 108 No. 447, pp. 383-398.
Engle, R.F. and Granger, C.W.J. (1987), “Co-integration and error correction: representation,
estimation, and testing”, Econometrica, Vol. 55 No. 2, pp. 251-276.
Goldsmith, R.W. (1969), Financial Structure and Development, Yale University Press, NewHaven,
CT.
Greenwood, J. and Jovanovic, B. (1990), “Financial development, growth and the distribution of
income”, Journal of Political Economy, Vol. 98 No. 5, pp. 1076-1108.
Johansen, S. (1991), “Estimation and hypothesis testing of co-integration vectors in Gaussian
vector autoregressive models”, Econometrica, Vol. 59 No. 6, pp. 1551-1580.
Johansen, S. and Juselious, K. (1990), “Maximum likelihood estimation and inference on
cointegration with applications to the demand for the money”, Oxford Bulletin of Economics
and Statistics, Vol. 52 No. 2, pp. 169-210.
King, R.G. and Levine, R. (1993), “Finance and growth: Schumpeter might be right”, Quarterly
Journal of Economics, Vol. 108 No. 3, pp. 7171-7737.
Mckinnon, R.I. (1973), Money and Capital in Economic Development, Brooking Institution,
Washington, DC.
Newbery, D.M.G. and Stiglitz, J. (1984), “Pareto inferior trade”, Reviewof Economic Studies, Vol. 51
No. 1, pp. 1-12.
Patrick, H.T. (1966), “Financial development and economic growth in underdeveloped countries”,
Economic Development and Cultural Change, Vol. 14 No. 2, pp. 174-189.
Pagno, M. (1993), “Financial markets and growth: an overview”, European Economic Review,
Vol. 37 Nos 2/3, pp. 613-622.
Peter, B.H. (2003), “Capital account liberalization, the cost of capital, and economic growth”, NBER
Working Paper No. 9488, Cambridge, 21 February.
Philips, P.C.B. and Perron, P. (1988), “Testing for a Unitroot in Time series Regrreson”, Biometrica,
Vol. 75 No. 2, pp. 335-346.
Quy-Toan, D. and Levchenko, A.A. (2004), “Trade and Financial Development”, World Bank
Policy Research Working Paper No. 3347.
Rajan, R. and Zingales, L. (2003), “The great reversals: the politics of fnancial development in the
20th century”, Journal of Financial Economics, Vol. 69 No. 1, pp. 5-50.
Robinson, J. (1952), The Generalization of the General Theory, The Rate of Interests and Other
Essay, Macmillan, London, pp. 67-142.
Roubini, N. and Sala-I-Martin, X. (1991), “Financial development, trade regimes and economic
growth”, NBER Working Paper No, MA, No. 3876.
Salvatore, D. (2010), International Economics, Johan Wiley Publication, Hoboken, NJ.
Schumpeter, J.A. (1911), The Theory of Economic Development, Harvard University Press,
Cambridge.
Shahbaz, M., Shabbir, M.S. and Butt, M.S. (2013), “Effect of fnancial development on agricultural
growth in Pakistan”, International Journal of Social Economics, Vol. 40 No. 8, pp. 707-728.
JFEP
6,4
374
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Shaw, E.S. (1973), Financial Deepening in Economic Development, Harvard University Press,
Cambridge, M.A.
Shleifer, A. and Vishny, R.W. (1998), “The quality of government”, Harvard Institute of Economic
Research Working Paper 1847, Harvard.
Further reading
Gurley, J. and Shaw, E. (1955), “Financial aspects of economic development”, American Economic
Review, Vol. 45 No. 4, pp. 515-538.
About the authors
Dogga Satyanarayana Murthy is currently pursuing PhD in Economics in Pondicherry
University, India. I have qualifed UGC NET and AP SLET and received PRATIBA AWARD in
BEd. I have also attended and presented papers in many national workshops, national and
international seminars and a paper also accepted and to be published in JHM, SAGE Publication,
Jaipur, India, and another paper to be published in Asian Economic Review, Indian Institute of
Economic Society, Hyderabad, India. I have also published two more papers in Global Journal of
Finance and Management as seminar proceedings. Dogga Satyanarayana Murthy is a
corresponding author and can be contacted at: [email protected]
Suresh Kumar Patra is currently pursuing PhDin Economics in Pondicherry University, India.
I received a Gold Medal as the toper in MPhil in Economics from Pondicherry University and
qualifed for the Tamil Nadu SETin 2013, and I amalso receiving the ICSSRFellowship. I do have
special interest in Mathematical Economics, Econometrics, Economic Growth and Development;
and also in Health Economics. I have also published one book on “An Evaluation of National Rural
Health Mission In Odisha: With special reference to Puri District”. I have also attended and
presented papers in many national workshops, national and international seminars and a paper
also accepted and to be published in JHM, SAGE Publication, Jaipur, India.I have also published
two more papers in Global Journal of Finance and Management as seminar proceedings.
Amaresh Samantaraya is presently working as Associate Professor in the Department of
Economics, Pondicherry University. He teaches courses on Macroeconomics, Economics of
Money &Banking and Applied Econometrics. His research interest is diverse, covering empirical
issues related to conduct of monetary policy in India, fnancial sector developments and
inter-regional disparity in Indian growth experience. He has published numerous research studies
in various national and international professional journals in economics and fnance, and has
published a book on the conduct of monetary policy in India. Prior to joining academics, he worked
for over a decade in various capacities, undertaking economic research in the Reserve Bank of
India (RBI) – the central bank and the monetary authority for India.
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
375
Trade Openess,
FDI and
Economic
Growth
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
5
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_297673168.pdf