Majid—Re-Examining The Finance Growth Nexus
Gadjah Mada International Journal of Business
May-August 2007, Vol. 9, No. 2, pp. 137–156

RE-EXAMINING
THE FINANCE-GROWTH NEXUS
Empirical Evidence from Indonesia
M. Shabri Abd. Majid
This paper empirically examines the short- and long-run relationships between financial development and economic growth
during the post-1997 financial crisis in Indonesia by employing a
battery of times-series techniques, such as Autoregressive Distributed Lag (ARDL) model, vector error correction model (VECM),
variance decompositions (VDCs), and impulse-response functions
(IRFs). Based on the ARDL (2, 0, 1, 2) model, the study finds that
there exists a long-run equilibrium between economic growth and
financial depth, share of investment, and inflation. In the long run,
inflation is found to be the only variable which significantly (negatively) affects economic growth, implying a crucial role of maintaining a low rate of inflation in promoting the economic growth in the
country. As for the dynamic causalities among the variables, the
study finds the bidirectional causation between economic growth
and investment, while the unidirectional causation is only found
running from financial depth to investment. The finding of independence between economic growth and financial development supports the view of “the independent hypothesis” of Lucas (1988).
Finally, based on VDCs and IRFs, the study documents that the
variations in the economic growth respond more to shocks in the
price stability (inflation), followed by investment and financial
development. Our findings indicate that if policy makers want to
promote growth, attention should be focused on long-run policies,
i.e., maintaining the low rate of inflation.
Keywords: ARDL; financial development; growth; impulse response function;
Indonesia; multivariate causality
JEL Classification: C32, O16.
137

Majid—Re-Examining The Finance Growth Nexus

adopting different methods, sets of data,
and sample may lead to inconsistent
findings. This study is, therefore, aimed
at empirically re-examining short- and
long-run relationships between financial development and economic growth
in the Indonesian economy during the
post-1997 financial turmoil by adopting the latest technique, which is
autoregressive distributed lag (ARDL)
bound testing approach to examine the
cointegration. It also attempts to investigate the finance-growth nexus using
multivariate causality tests within a
vector error correction model (VECM).
Finally, the paper also seeks to explore
the relative strength of the variables in
affecting economic growth using the
variance decompositions (VDCs) and
the impulse response function (IRFs)
based on the structural vector
autoregression (VAR) framework.
The rest of the paper is organized
as follows. Section 2 discusses the
theoretical issues on the relationship
between financial development and
economic growth. The empirical framework and data utilized in this study are
delineated in Section 3. The empirical
results and discussion on the findings
are presented in Section 4. Finally,
Section 5 summarizes the main findings and provides some policy implications.

Theoretical Underpinnings
The connection between financial
development and economic growth has
been a subject of considerable interest
in the development of economic and

finance literature in recent years. In
this framework, financial development
is considered to be the principal input
for economic growth. It is an important
element to affect the rate of economic
growth by altering productivity growth
and the efficiency of capital. It also
affects the accumulation of capital
through its impacts on savings rate by
altering the proportion of savings
(Pagano 1993; Levine 1997). The theoretical support can be traced back to
the work of Schumpeter (1911) where
he argues that financial intermediaries
sector alter the mobilizing of savings
for successful projects by managing
risk, monitoring managers, and then
facilitating transactions which will essentially improve technological innovation and economic development. In
their seminal works, McKinnon (1973)
and Shaw (1973) assert that the financial liberalization will increase savings and capital accumulation, which
is finally to be invested and therefore
enhances growth.
Heretofore, the development
theory of economic growth has been
widely used as literature in the study of
economic development, macroeconomics, and other related subjects.
Some of these theories were introduced by Rostow (1960), Harrod
(1939), Domar (1946), Lewis (1954),
and Solow (1956). However, only few
of these theories focus explicitly on the
role of financial development in promoting economic growth. On one hand,
Harrod (1939) and Domar (1946) opine
that in order to increase a growth rate,
new investments representing net ad139

Majid—Re-Examining The Finance Growth Nexus

units, would provide an efficient allocation of resources, thereby leading
other economic sectors in their growth
processes. This view has received considerable supports from recent empirical studies (Greenwood and Jovanovic
1990; Thakor 1996, to name a few).
The second one is “the growth-led
finance hypothesis” or “the demandfollowing view.” This view was advanced by Robinson (1952), and it
states that financial development follows economic growth, or where an
enterprise leads finance follows. Accordingly, as the real side of the
economy expands, its demand for certain financial instruments, arrangements, and markets increases, leading
to the growth of these services. Empirical supports for this second view
can be found, for instances, in the
studies of Friedman and Schwartz
(1963) and Demetrides and Hussein
(1996).
The third view is the “the feedback hypothesis” or “the bidirectional
causality view.” This view postulates
that finance and economic developments are mutually causal and have
bidirectional causality. In this hypothesis, it is asserted that a country with a
well-developed financial system could
promote high economic expansion
through technological changes, products and services innovation
(Schumpeter 1911). This will, in turn,
create a high demand for the financial
arrangements and services (Levine
1997). As the banking institutions effectively response to this demand, these
changes will stimulate higher economic

achievements. Both financial and economic developments therefore are positively interdependent, and their relationship could lead to bidirectional
causality (Choong et al. 2003). Empirical supports for this view can also
be found, for examples, in the works of
Greenwood and Smith (1997) and
Luintel and Khan (1999).
Eventually, the fourth view is “the
independent hypothesis.” This view
was originally put forward by Lucas
(1988), arguing that financial and economic development growth are not
causally related, or in the words of
Lucas (1988), “economics badly overstresses the role of financial factors in
economic growth.” Meanwhile,
Chandavarkar (1992) notes that “none
of the pioneers of the development
economics….even list finance as a factor of development.”
From the aforementioned exposition of different streams of thoughts on
the relationship between financial and
economic developments, it is obvious
that the literature on this issue is mixed
and inconclusive. Accordingly, it is
appropriate and timely to empirically
re-examine the finance-growth nexus
in the largest Moslem population country, Indonesia. Does the finance-growth
nexus in Indonesia support the first
view (the finance-led growth hypothesis or the supply-leading view), the
second view (the growth-led finance
hypothesis or the demand-following
view), the third view (the feedback
hypothesis or the bidirectional causality view), or the last view (the independent hypothesis)? The extent to which
141

Majid—Re-Examining The Finance Growth Nexus

provides more motivation for the study
to adopt this model.
The ARDL model used in this
study can be written as follows:
GDPt = 0 + 1FDt +2 SIt +
3INFt + et ................. (1)
where, GDPt = real output at time t,
FDt = a measure of financial
depth,
SIt = the share of investment,
INFt = inflation, and
et
= an error term.
The error correction version of
ARDL framework pertaining to the
variables in the Equation (1) can be
reproduced as follows:
p

GDPt = 0 + iGDPt-i +
i=1

p

p

iFDt-i + iSIt-i +

i=0
i=0

iINFt-i + 1GDPt-1 +
2FDt-1 + 3SIt-1 +
4INFt-1 u1t .............(2)
The terms with the summation
signs in the Equation (2) represent the
error correction dynamic while the second part (termed with s) corresponds
to the long-run relationship. The null
of no cointegration in the long-run
relationship, defined with H0: 1 = 2 =
3 = 4 = 0, is tested against the alter-

native of H0:1 2 3 4  0 by the
means of familiar F-test. However, the
asymptotic distribution of this F-statistic is nonstandard irrespective of
whether the variables are I(0) or I(1).
For a small sample size ranging from
30 to 80 observations, Narayan (2004)
has tabulated two sets of appropriate
critical values. One set assumes that all
variables are I(1) and the other assumes that they are all I(0). This provides a bound covering all possible
classifications of the variables into I(1)
and I(0) or even fractionally integrated.
If the F-statistic exceeds the upper
bound level, the null is rejected, which
indicates the existence of cointegration.
On the other hand, if the F-statistic
falls below the bound level, the null
cannot be rejected, which supports no
cointegration. If, however, it falls
within the band, the result is inconclusive.
Finally, in order to determine the
optimal lag-length incorporated into
the model and to select the ARDL
model to be estimated, the study employs the Akaike Information Criteria
(AIC). Since our study utilizes quarterly data with only 36 observations,
the possible optimal lag-length to be
considered is only 4.

Vector Error Correction Model
(VECM) Framework
To examine the multivariate causality relationship among the variables,
the study employs the vector error correction model (VECM) framework.
The VECM regresses the changes in
143

Majid—Re-Examining The Finance Growth Nexus

strength of the causality among the
variables beyond the sample period.
Variance decompositions (VDCs)
therefore allow us to examine the outof sample causality among the variables in the VAR system. It measures
the percentage of the forecast error of
variable that is explained by another
variable. Precisely, it indicates the relative impact that one variable has on
another variable. At the same time, it
provides information on how a variable of interest responds to shocks or
innovations in other variables. The
VAR model is a system of reduced
form dynamic linear equations in which
each variable is expressed as a function of a serially uncorrelated errors
and an equal number of lags of all
variables in the system (Abdullah 1998;
Enders 1995). This VAR model assumes that the contemporaneous correlations of errors across equations are
nonzero, and therefore there are no
contemporaneous explanatory variables in the model. The error terms
(also referred to as innovations) can
provide a potential source of new information on the movements in a variable during current period. In order to
interpret the economic implications
from the VAR model, Sim’s (1980)
innovation accounting procedure is
employed. This procedure involves the
decomposition of forecast error variance of each variable into components
attributable to its own innovations and
to shocks of other variables in the
system.

The above procedures of VDCs
together with the IRFs analysis (also
renowned as innovation accounting in
the literature) allow us to examine the
relationships among the variables. If
the correlations among the various innovations are large, the identification
problem is likely to be important. The
alternatives ordering should yield similar impulse responses and variance
decompositions.

Empirical Results
Before estimating the long-run
relationship between financial development and economic growth in Indonesia, we have to decide about the lagTable 1. F-statistics for Testing the
Existence of a Long-run Relationship
Lag-Length

F-Statistics

1
2
3
4

1.04320
4.5543 **
8.4077 ***
6.3412 ***

Note: The relevant critical value bounds are
taken from Case II with a restricted
intercept and no trend and number of
regressors = 3 (Narayan, 2004). They
are 4.480 – 5.700 at the 99% significance level, 3.170– 4.160 at the 95%
significance level and 2.618 – 3.502 at
the 90% significance level. **, and ***
denotes that F-Statistics falls above the
95%, and 99% upper bound, respectively.

145

Majid—Re-Examining The Finance Growth Nexus

mote economic development.4 Many
previous empirical studies document
that for countries with low inflation
rate (below 10 percent annually), their
economic growth will be accelerated
(Bekaert et al. 2005; Gylfason et al.
2001; and Hung 2003), while high
inflation (about 10 – 20 percent annually) could detriment the long-run economic growth (Gylfason et al. 2001;
and Andrés et al. 2004). Our findings
are in line with the later studies by
Gylfason et al. (2001); Andrés et al.
(2004) and Christopoulos and Tsionas
(2004). In their study, Christopoulos
and Tsionas (2004) find that during the

period from 1997 to 2000, a higher rate
of inflation in Peru has spoilt the economic growth of the country. In promoting the growth of economy in Indonesia, it is very important for the
government to maintain price stability
by reducing the rate of inflation below
two digits. A significant increase in the
prices of petroleum and cooking oil in
the early 2005 and mid-2006 respectively has hindered the growth of Indonesian economy, and it has also become one of the hindrances to the
government to totally recover its
economy.

Table 2: Multivariate ‘VECM’ Causality
Independent Variables
Dependent
Variables

GDP

FD

SI

INF

ECTt-1

-

1.3438
[0.2588]

2.6607*
[0.0923]

2.0512
[0.1360]

-1.4307***
(-3.5972)

FD

0.5405
[0.6596]

-

2.1997
[0.1346]

1.3082
[0.2968]

-0.1975*
(-2.0432)

SI

4.7974***
[0.0102]

3.6600*
[0.0688]

-

1.2395
[0.3193]

0.0385
(1.2656)

INF

1.6465
[0.2075]

0.1805
[0.6751]

0.6319
[0.5410]

-

-0.5203**
(-2.7004)

GDP

Note: ***, ** and * represent significance at the 1%, 5% and 10% levels, respectively. ECTt-1 is
derived by normalizing the cointegrating vectors on the GDP as proxy for economic growth,
producing residual r. By imposing restriction on the coefficients of each variable and conducting
Wald test, we obtain F-statistics for each coefficient in all equations. Figures in the parentheses and
squared parentheses represent t-statistics and probabilities for F-statistics, respectively.
4

See, for example, the IMF report for the year 2005. The average rate of inflation of Indonesia
was 10.45%, while the rest of ASEAN countries had rates of inflation of between 0.5 and 9.4%, i.e.,
Brunei Darussalam (1.22%), Malaysia (2.96%), Cambodia (5.56%), Laos PDR (7.17%), Myanmar
(9.37%), Philippines (7.64%), Singapore (0.47%), Thailand (4.54%), and Vietnam (8.25%).
147

Majid—Re-Examining The Finance Growth Nexus

economic shocks may exacerbate macroeconomic instability and reduce the
long-term growth. Thirdly, stock market development is likely to undermine the existing group-banking system in Indonesia which, despite their
many difficulties, has not been without merit in Indonesia as well as in
other highly successful East Asian
economies.
To further explore the dynamic
interaction between financial development and economic growth, the study
proceeds to test the variance decompositions (VDCs) and impulse-response
functions (IRFs). The results of VDCs
that are reported in Table 3 provide
detailed information on the relative
strength of the financial depth, share of
investment, and inflation in explaining
the changes in the economic growth.

From the VDCs and IRFs results, we
are also able to capture the relative
importance of various shocks and their
influences on the economic growth.
The VDCs and IRFs are simulated by
orthogonalizing the innovations in the
vector autoregression (VAR) equations
using the so-called Cholesky decomposition suggested by Sim (1980), with
the ordering of the variables: GDP,
FD, SI, INF.5 Based on VDCs results
for the horizon of 1 – 12 quarters as
reported in Table 3, we find that the
variations in the economic growth respond more to shocks in the price stability (inflation), and account for about
0 – 6.5 percent of economic growth
forecast error variance after three years.
Meanwhile, the variations in the economic growth responding to shocks in
the share of investment and financial

Table 3. Variance Decompositions
Horizon
(Quarterly)
1
2
4
6
8
10
12

Explained by shocks in:
GDP

FD

SI

INF

100
95.39
89.07
88.11
87.46
87.26
87.19

0.00
0.10
0.78
0.95
1.12
1.17
1.20

0.00
4.18
4.96
5.09
5.02
5.10
5.06

0.00
0.33
5.19
5.85
6.39
6.47
6.55

5
We also have tried to use different orderings of the variables such as GDP, FD, INF, SI; GDP,
INF, SI, FD; and GDP, INF, FD, SI, and to employ the generalized impulses which do not depend
on the VAR ordering, as described by Pesaran and Shin (1998). However, their results are very much
similar.

149

Majid—Re-Examining The Finance Growth Nexus

Continued from Figure 1

Response of GDP to INF

2
1
0
-1
-2
1

2

3

4

5

6

Figure 2. CUSUM and CUSUMSQ Plots
Plot of Cumulative Sum of Recursive Residuals
15
10
5
0
-5
-10
-15
1999Q1
2000Q2
2001Q3
2002Q4
2004Q1
2005Q2
The straight lines represent critical bounds at 5% significance level

2006Q3

2006Q4

Plot of Cumulative Sum of Squares of Recursive Residuals
1.5
1.0
0.5
0.0
-0.5
1999Q1

2000Q2

2001Q3

2002Q4

2004Q1

2005Q2

2006Q3

2006Q4

The straight lines represent critical bounds at 5% significance level

To complement our analysis on
the VDCs, we further generate the IRFs,
as described above. As reported in
Figure 1, the overall results seem to be
very much consistent with our earlier
findings. Economic growth seems to
have an immediate negative response

to shocks in the price stability and
share of investment, while no significant effect is found between the shocks
in the financial development to the
innovations in the economic growth.
This further implies that any policies
pertaining to the price stability and
151

Majid—Re-Examining The Finance Growth Nexus

variations in the economic growth responding to shocks in the share of
investment and financial development
only account for about 0 – 5 percent.
Economic growth seems to have an
immediate negative response to shocks
in the price stability, while no significant effect is found between the shocks
in the financial development and the
innovations in the economic growth.
The variations in the economic growth
are, however, much depending on its
own innovations.
The most important implication
of our findings is a policy recommendation: if policy makers want to promote growth, then attention should be
focused on maintaining the lower rate
of inflation in the long run. However,
any efforts put forward by the Indonesian government to promote the economic growth by further enhancing,
deregulating, and restructuring the financial sector (banking and stock market) are found to be insignificant
endeavours in promoting the growth
of economy in Indonesia. As for the
dynamic causalities among the variables, the study finds the bidirectional

causation between economic growth
and investment, while the unidirectional causation is only found running
from financial depth to investment. On
the other hand, to promote economic
growth in the short run, the conducive
investment environment should be
improved by enhancing the modern
the financial institutions, both in banking sector and in stock market.
Finally, this study has re-examined the finance-growth nexus financial turmoil in Indonesia during the
post-1997 period. Hence, to enrich the
findings, more robust analyses are
needed, e.g., a comparative analysis
between the pre- and post-1997 financial turmoil periods, thereby providing
a comprehensive policy recommendation. Additionally, the enrichment of
the findings could also be done by
including more countries into the analysis, such as by examining all ASEAN
countries. A comparative study between the ASEAN economies and developed markets would also provide
additional insights into the existing
empirical evidence.

References
Abdullah, D. A. 1998. Money growth variability and stock returns: An innovations
accounting analysis. International Economic Journal 12 (4): 89-104.
Al-Yousif, Y. K. 2002. Financial development and economic growth: Another look at the
evidence from developing countries. Review of Financial Economics 11: 131–150.
Andrés, J., I. Hernando, and J. D. Lo´pez-Salido. 2004. The role of the financial system
in the growth–inflation link: The OECD experience. European Journal of Political
Economy 20: 941–961.
Arestis, P., and P. Demetriades. 1997. Financial development and economic growth:
Assessing the evidence. The Economic Journal 107: 783-799.
153

Majid—Re-Examining The Finance Growth Nexus
Heinesen, E. 1995. A macroeconomic rationing model estimated by cointegration
techniques and generalized method of moments. Economic Modelling 12 (2): 97110.
Hung, F. S. 2003. Inflation, financial development, and economic growth. International
Review of Economics and Finance 12: 45–67.
International Monetary Fund. 2006. International Financial Statistic Online.
www.imfstatistics.org.
Kasa, K. 1992. Common stochastic trends in international stock markets. Journal of
Monetary Economics 29: 95-124.
Khan, S.M., and A. S. Senhaji. 2000. Financial and economic growth: An overview. IMF
Working Paper. Washington: International Monetary Fund.
King, R.G., and R. Levine. 1993. Finance and growth: Schumpeter might be right. The
Quarterly Journal of Economics: 717-737.
Laurenceson, J and J. C. H. Chai. 2003. Financial Reform and Economic Development
in China. Cheltenham, UK: Edward Elgar.
Levine, R. 1997. Financial development and economic growth: Views and agenda.
Journal of Economic Literature 35 (2): 688-726.
Lewis, W. A. 1954. Economic development with unlimited supplies of labour. The
Manchester School 22: 139-191.
Lucas, R.E. 1988. On the mechanics of economic development. Journal of Monetary
Economics 22 (1): 3-42.
Luintel, B.K., and M. Khan. 1999. A quantitative re-assessment of the finance-growth
nexus: evidence from multivariate VAR. Journal of Development Economics 60:
381-405.
McKinnon, R. I. 1973. Money and Capital in Economic Development. Washington, DC:
The Brooking Institution.
Mussa, M. 2006. Global economic prospects 2006/2007: Continued solid growth in 2006,
rising risks for inflation, financial markets, and growth for 2007. Paper. Presented
at the ninth semiannual meeting on Global Economic Prospects, Washington, DC,
April 11.
Narayan, P. K. 2004. Reformulating critical values for the bounds f-statistics approach to
cointegration: An application to the tourism demand model for Fiji. Discussion
Paper No. 02/04. Department of Economics, Monash University. Australia.
Pagano, M. 1993. Financial markets and growth: An overview. European Economic
Review 37: 613-622.
Pesaran, M. H., Y. Shin, and R. J. Smith. 1996. Testing for the existence of a long-run
relationship. DAE Working Paper No. 9622. Department of Applied Economics,
University of Cambridge.
Pesaran, M. H., and Y. Shin. 1998. Generelazied impulse response analysis in linear
multivariate models. Economic Letters 58: 17-29.

155