Int J Fin Econ - 2006 - Demetriades - Finance Institutions and Economic Development

INTERNATIONAL JOURNAL OF FINANCE AND ECONOMICS
Int. J. Fin. Econ. 11: 245–260 (2006)

Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ijfe.296
FINANCE, INSTITUTIONS AND ECONOMIC DEVELOPMENT

PANICOS DEMETRIADESa,*,y and SIONG HOOK LAWb
a
Department of Economics, University of Leicester, UK
b
Department of Economics, University Putra Malaysia, Malaysia
ABSTRACT
Using data from 72 countries for the period 1978–2000, we find that financial development has larger effects on GDP
per capita when the financial system is embedded within a sound institutional framework. Moreover, we find that
financial development is most potent in middle-income countries, where its effects are particularly large when
institutional quality is high. Importantly, we also find that in low-income countries the influence of financial
development is at its weakest; in these countries, more finance without sound institutions may not succeed in delivering
long-run economic benefits. Copyright # 2006 John Wiley & Sons, Ltd.
JEL CODE: G1; O1; O4
KEY WORDS: Financial development; institutional quality; economic development; pooled mean group estimation
1. INTRODUCTION
It is now widely accepted that factor accumulation and technological change alone cannot adequately
explain differences in economic growth across countries. Institutions and finance are separately emerging as
the key fundamental determinants of economic growth in recent literature.
Institutions are the rules of the game in a society by which the members of a society interact and shape
the economic behaviour of agents. They may be treated as ‘social technologies’ in the operation of
productive economic activities (Nelson and Sampat, 2001). When the rules change frequently or are not
respected, when corruption is widespread or when property rights are not well defined or enforced, markets
will not function well, uncertainty would be high, and, as a result, the allocation of resources would be
adversely affected. A number of recent papers provide empirical evidence that confirms the importance of
institutional quality for economic performance.1 Rodrik et al. (2002) find that quality of institutions
overrides geography and integration (international trade) in explaining cross-country income levels. Hall
and Jones (1999) find that differences in capital accumulation, productivity and output per worker across
countries are driven by differences in institutions and government policies. Knack and Keefer (1995) find a
positive and significant relationship between institutional indicators, such as quality of bureaucracy, and
economic growth utilizing cross-country data. Mauro (1995) demonstrates that the countries that have a
higher corruption index tend to have persistently lower growth. Rodrik (1997) finds that institutional
quality does exceptionally well in rank-ordering East Asian countries according to their growth
performance while Pistor et al. (1998) highlight the role of law and legal systems in promoting Asian
economic growth.
*Correspondence to: P. Demetriades, Department of Economics, University of Leicester, University Road, Leicester LE1 7RH, UK.
y
E-mail: [email protected]
Copyright # 2006 John Wiley & Sons, Ltd.

10991158, 2006, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ijfe.296 by National University of Science and Technology, Wiley Online Library on [31/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
246 P. DEMETRIADES AND S. HOOK LAW
Financial intermediaries perform an important function in the development process, particularly through
their role in allocating resources to productive uses. The increased availability of financial instruments
reduces transaction and information costs while larger and more efficient financial markets help economic
agents hedge, trade and pool risk, thereby raising investment and economic growth (Goodhart, 2004).
Levine (2003) provides an excellent overview of a large body of empirical literature that suggests that
financial development can robustly explain differences in economic growth across countries. Zingales
(2003), however, questions the extent to which cross-country relationships can be utilized for policy
purposes, especially since there is a bunch of variables, all positively correlated with growth, which are also
highly correlated among themselves. These difficulties have prompted a number of authors to examine the
relationship using time-series data for individual countries in the hope of a better understanding of the
causality between finance and growth. The evidence from these studies suggests that while the relationship
between financial development and growth within individual countries is typically a positive one, causality
tends to vary considerably from country to country (e.g. Demetriades and Hussein, 1996). It is, therefore,
not sensible to draw out any policy implications from the positive association obtained between finance and
growth obtained from cross-country studies that would be applicable to every country in the world. More
finance may mean more growth in some cases but not in others; knowing where it does and where it does
not is critical for policy makers. Understanding why there is such variation across countries is, therefore, an
important next step for both policy makers and academics.
The variation in causality between finance and growth detected in time-series studies suggests that there
are important differences in the way in which finance influences economic growth across countries. Arestis
and Demetriades (1997) suggest that it may reflect institutional differences across countries. This idea is
developed further in Demetriades and Andrianova (2004), who argue that varying causal patterns may
reflect differences in the quality of finance, which are, in turn, determined by the quality of financial
regulation and the rule of law. For example, an increase in financial deepening, as captured by standard
indicators of financial development, may not result in increased growth because of corruption in the
banking system or political interference, which may divert credit to unproductive or even wasteful activities.
While this is a plausible conjecture, there is as yet little, if any, direct evidence to confirm that institutions
make a difference to the way in which finance affects economic growth. This paper represents a first step in
providing such evidence, by testing the hypothesis that the interaction between institutional quality and
financial development has a separate positive influence on economic growth, over and above the effect of
the levels of financial development and institutional quality. Testing this hypothesis within individual
countries requires data on institutional quality that span many decades, since institutions usually change
very slowly. Such data are only available for the last twenty years or so, which makes time-series analysis of
this issue not possible. However, we have been able to obtain institutional quality indicators for 72
countries for the period 1978–2000. We, therefore, utilize both cross-section and panel econometric
methods to test our hypothesis. Additionally, we also examine whether the estimated relationship varies in
accordance to the stage of economic development, as has recently been suggested by Rioja and Valev (2004)
in their analysis of the finance and growth nexus.
The paper is organized as follows. Section 2 lays down the empirical model, introduces the econometric
methodology and outlines the data. Section 3 presents and discusses the empirical findings. Section 4
summarizes and concludes.
2. EMPIRICAL MODEL, METHODOLOGY AND DATA
2.1. Empirical model

We assume that output in each country is determined by the following Cobb–Douglas production
function:
Yit ¼ Kita ðAit Lit Þ1a ð1Þ
Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Fin. Econ. 11: 245–260 (2006)
FINANCE, INSTITUTIONS AND ECONOMIC DEVELOPMENT 247
where Yit is real output in country i at time t; Kit is the stock of physical capital in country i at time t; Lit is
the stock of raw labour in country i at time t; Ait is a labour-augmenting factor reflecting the level of
technology and efficiency in country i at time t:2
It is assumed that a51; i.e. there are decreasing returns to capital. Raw labour and labour-augmenting
technology are assumed to evolve according to the following functions:
Lit ¼ Li0 eni t ð2Þ
Ait ¼ Ai0 e gi tþPit yi ð3Þ
where ni is the exogenous rate of growth of the labour force in country i; gi is the exogenous rate of
technological progress in country i; Pit is a vector of financial development, institutions and other factors
that may affect the level of technology and efficiency in country i at time t; and yi is a vector of coefficients
related to these variables.
In this framework, the state of labour-augmenting technology (variable A) depends not only on
exogenous technological improvements, determined by g; but also on the level of financial development and
the quality of institutions, such as the rule of law. Financial innovation may be critical in facilitating
technological breakthroughs, which may not occur without appropriate forms of financing. Moreover, the
routine monitoring and screening functions of the banking system help to ensure that only productive
investments are undertaken and that the capital stock remains productively employed (Pagano, 1993). The
presence of efficient, effective and uncorrupt institutions ensures that labour can be used for productive
purposes, instead of being wasted in dealing with red tape or rent-seeking activities (North, 1990; Nelson
and Sampat, 2001).
In a neoclassical growth framework, such as this one, the effects of financial development and institutions
on growth can only be transient, i.e. dPit =dt is assumed to be zero in the steady state (but can be positive or
negative in transition). However, the level of Pi can vary across countries in the steady state, which means
that the level of per-capita income can also vary across countries. Thus, different countries may well
converge to different steady states, depending on their steady-state level of financial and institutional
development.
In the steady state, output per effective worker ½Y=AL is constant while output per worker ½Y=L grows
at the exogenous rate g: In general, output in effective worker terms evolves as follows:
Yit
¼ ðkit Þa
Ait Lit
In (raw) worker terms, output evolves according to
Yit
¼ Ait ðkit Þa ð4Þ
Lit
Let
yit ¼ ðYit =Lit Þ
Taking logs of both sides of Equation (4), we get
ln yit ¼ ln Ait þ a ln kit
Utilizing Equation (3) we obtain
ln yit ¼ ln A0 þ ð1 aÞgi t þ ð1 aÞyi Pit þ a ln kit ð5Þ
Equation (5) describes the evolution of output per worker or labour productivity, as a function of a vector
of financial and institutional variables, which may change over time, the level of physical capital and the
exogenous growth rate of output.
As essentially the production function relationship, it could be argued that Equation (5) is valid both
within and outside of the steady state. This is an advantage for estimation purposes, particularly when
using static panel data techniques.3 Moreover, Equation (5) is not dependent on assumptions regarding the
behaviour of saving, which may introduce further departures from reality. Equation (5), therefore, offers a
reasonable basis for estimation.
In order to estimate (5), we need to specify a functional form for the vector P and add an appropriate
error term. The simplest form we utilize is linear, as follows:
ln yit ¼ ln A0 þ ð1 aÞgi t þ ð1 aÞy1i P1it þ ð1 aÞy2i P2it þ a ln kit þ eit ð6Þ
where P1 is a financial development indicator, P2 is a variable summarizing the quality of institutions and eit
is an error term, the properties of which may be model specific.
The second functional form for P is a non-linear one in that in addition to the linear terms it also
contains a multiplicative term, representing the interaction between financial development and institutions
on growth, as follows:
ln yit ¼ ln A0 þ ð1 aÞgi t þ ð1 aÞy1i P1it þ ð1 aÞy2i P2it þ ð1 aÞy3i ðP1it P2it Þ þ a ln kit þ Zit ð7Þ
where Zit is a new error term.
Equations (6) and (7) provide the basis for the empirical models. In reduced form, they can, respectively,
be re-written as follows:
ln yit ¼ b0i þ b1i t þ b2i FD1it þ b3i INSi þ b4i ln kit þ eit ð8Þ
ln yit ¼ b0i þ b1i t þ b2i FD1it þ b3i INSi þ b4i ln kit þ b5i FDit INSit þ Zit ð9Þ
where the b’s are parameters to be estimated and FD and INS denote financial development and
institutional quality, respectively.
2.2. Econometric approach

Cross-sectional estimation. Numerous studies have examined the determinants of economic growth
using cross-section data, including classic papers such as Barro and Sala-i-Martin (1992) and
Mankiw et al. (1992). In these studies the dependent and independent variables are averaged over a
fairly long period (usually 20 or move years), which is meant to capture the steady-state relationship
between the variables concerned. Our first set of estimations utilizes cross-sectional estimations, which
enables us to gauge our results against literature benchmarks. In order to estimate Equations (8) and (9)
using cross-section analysis, we use country averages for each variable over the full 23-year period
(1978–2000). Regional dummies for Latin America, East Asia and sub-Saharan Africa are also included in
both equations.
Panel data estimation. Equations (8) and (9) can be estimated directly using panel data techniques, which
allow both cross-section and time-series variation in all variables. The empirical analysis of the model in
Equations (8) and (9) above generally involve a system of N T equations (N countries and T time
observations) that can be examined in different ways. In this study, the main econometric approaches
employed include different forms of pooled cross-section time-series regressions, which are discussed below.
While cross-sectional estimation methods may, in principle, capture the long-run relationship between
the variables concerned, they do not take advantage of the time-series variation in the data, which could
increase the efficiency of estimation. It is, therefore, preferable to estimate the model using panel data
techniques, which, however, require careful econometric modelling of dynamic adjustment. The static panel
data technique based on either pooling or fixed effects, which could be applied to Equation (8) or (9), makes
no attempt to accommodate possibly heterogeneous dynamic adjustment around the long-run equilibrium
relationship (Pesaran and Smith, 1995; Pesaran et al. 1999). Careful modelling of short-run dynamics
requires a slightly different econometric modelling approach. We, therefore, assume that Equation (8) [or
(9)] represents the equilibrium relationship, which holds in the long run, but that the dependent variable
may deviate from its equilibrium path in the short run. This may be because output shocks may be
persistent over time.
The parameter estimates of Equations (8) and (9) can then be obtained using two recently developed
methods for the statistical analysis of dynamic panel data, namely the mean group (MG) and pooled mean
group (PMG) estimators proposed by Pesaran and Smith (1995) and Pesaran et al. (1999), respectively.
These methods are well suited to the analysis of dynamic panels that have both large time and cross-section
data fields. In addition, both estimators have the advantage of being able to accommodate both the long-
run equilibrium and the possibly heterogeneous dynamic adjustment process. So far these methods have
been applied to studies of money demand, energy demand, economic growth and convergence.
Following Pesaran et al. (1999), we base our panel analysis on the unrestricted error correction ARDL
ðp; qÞ representation:
X
p1 X
q1
Dyit ¼ fi yi;t1 þ b0i xi;t1 þ lij Dyi;tj þ g0ij Dxi;tj þ mi þ uit ; i ¼ 1; 2; . . . ; N; t ¼ 1; 2; . . . ; T
j¼1 j¼0
ð10Þ
where yit is a scalar dependent variable, xit is the k 1 vector of regressors for group i; mi represent the fixed
effects, fi is a scalar coefficient on the lagged dependent variable, b0i ’s is the k 1 vector of coefficients on
explanatory variables, lij ’s are scalar coefficients on lagged first-differences of dependent variables, and gij ’s
are k 1 coefficient vectors on first-difference of explanatory variables and their lagged values. We assume
that the disturbances uit ’s in the ARDL model are independently distributed across i and t; with zero means
and variances s2i 40: Further, assuming that fi 50 for all i; and therefore there exists a long-run
relationship between yit and xit defined by
yit ¼ y0i xit þ Zit ; i ¼ 1; 2; . . . ; N; t ¼ 1; 2; . . . ; T ð11Þ
where ¼y0i b0i =fi
is the k 1 vector of the long-run coefficients, and Zit ’s are stationary with possibly non-
zero means (including fixed effects). Equation (10) can be re-written as
X
p1 X
q1
Dyit ¼ fi Zi;t1 þ lij Dyi;tj þ g0ij Dxi;tj þ mi þ uit ð12Þ
j¼1 j¼0
where Zi;t1 is the error correction term given by (11), hence fi is the error correction coefficient measuring
the speed of adjustment towards the long-run equilibrium.
Under this general framework, Pesaran et al. (1999) propose the pooled mean group (PMG) estimator.
This estimator allows the intercepts, short-run coefficients and error variances to differ freely across groups,
but the long-run coefficients are constrained to be the same; that is, yi ¼ y for all i: The group-specific short-
run coefficients and the common long-run coefficients are computed by the pooled maximum likelihood
estimation. These estimators are denoted by
PN * PN * PN *
fi bi lij
f# PMG ¼ i¼1 ; b# PMG ¼ i¼1 ; l# jPMG ¼ i¼1 ; j ¼ 1; . . . ; p 1
N N N
PN
d* ij
d# jPMG ¼ i¼1
; j ¼ 0; . . . ; q 1; y# PMG ¼ y* ð13Þ
N
On the other hand, the MG estimator proposed by Pesaran and Smith (1995) allows for heterogeneity of
all the parameters and gives the following estimates of short-run and long-run parameters:
PN # PN # PN #
fi bi lij
f# MG ¼ i¼1 ; b# MG ¼ i¼1 ; l# jMG ¼ i¼1 ; j ¼ 1; . . . ; p 1
N N N
PN
d# ij 1X N
d# jMG ¼ i¼1
; j ¼ 0; . . . ; q 1; y# MG ¼ ðb# i =f# i Þ ð14Þ
N N i¼1
where f # i ; b# i ; l# ij and g# ij are the OLS estimates obtained individually from Equation (10). In other words, the
MG approach consists of estimating separate regressions for each country and computing averages of the
country-specific coefficients (e.g. Evans, 1997; Lee et al., 1997). This estimator is likely to be inefficient in
small country samples, where any country outlier could severely influence the averages of the country
coefficients.
The MG estimator provides consistent estimates of the mean of the long-run coefficients, though these
will be inefficient if slope homogeneity holds. Under long-run slope homogeneity, the pooled estimators are
consistent and efficient. The hypothesis of homogeneity of the long-run policy parameters cannot be
assumed a priori and is tested empirically in all specifications. The presence of heterogeneity in the means of
the coefficients is examined by a Hausman-type test (Hausman, 1978) applied to the difference between the
MG and the PMG. Under the null hypothesis the difference in the estimated coefficients between the MG
and PMG estimators is not significant and PMG is more efficient.
2.3. Data
The data set consists of a panel of observations for 72 countries4 for the period 1978–2000. The sample
countries are grouped into three groups: high-, middle- and low-income based on the World Bank
classification.5 Annual data on real GDP per capita, gross fixed capital formation, and three alternative
financial development indicators (liquid liabilities, private sector credit and domestic credit provided by the
banking sector, all expressed as ratios to GDP) are collected from World Development Indicators (World
Bank CD-ROM 2002). All these data are converted to US dollars based on 1995 constant prices. The
capital stock is constructed from the gross investment figures following the perpetual inventory method.
Initial capital stocks are calculated using the assumption that over long periods of time capital and output
grow at the same rate. A depreciation rate of 6% and the average growth rate of the initial 5 years are used
to generate the initial level of capital stock (see Hall and Jones, 1999). Capital stock per capita is derived as
a ratio of the total capital stock to total population.
The data set on institutional quality indicators employed in this study was assembled by the IRIS
Center of the University of Maryland from the International Country Risk Guide (ICRG)}a monthly
publication of Political Risk Services (PRS). Following Knack and Keefer (1995), five PRS indicators are
used to measure the overall institutional environment, namely: (i) Corruption, which reflects the
likelihood that officials will demand illegal payment or use their position or power to their own
advantage; (ii) Rule of Law, which reveals the degree to which citizens are willing to accept established
institutions to make and implement laws and to adjudicate dispute. It can also be interpreted as a measure
of ‘rule obedience’ (Clague, 1993) or government credibility; (iii) Bureaucratic Quality, which represents
autonomy from political pressure, strength, and expertise to govern without drastic changes in policy or
interruptions in government services, as well as the existence of an established mechanism for recruitment
and training of bureaucrats; (iv) Government Repudiation of Contracts, which describes the risk of a
modification in a contract taking due to change in government priorities; and (v) Risk of Expropriation,
which reflects the risk that the rules of the game may be abruptly changed. The above first three variables
are scaled from 0 to 6, whereas the last two variables are scaled from 0 to 10. Higher values imply
better institutional quality and vice versa. The institutions indicator is obtained by summing the above five
indicators.6
3. ESTIMATION RESULTS
Equations (8) and (9) are first estimated on the full sample of countries using the OLS cross-country
estimator.7 The results are reported in Table 1, where Models 1–6 are estimates of Equation (8), utilizing
alternative proxies for financial development. Similarly, Models 7–12 are estimates of Equation (9), which
includes the interaction term between financial development and institutions.
Table 1. Results of OLS estimation. Dependent variable: ln real GDP per capita (72 countries, 1978–2000)
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12
Constant 9.38 7.45 3.94 2.16 2.04 1.60 1.55 2.57 3.22 3.43 2.69 1.37
(1.49) (1.64) (2.27) (1.33) (1.39) (1.43) (1.59) (1.43) (1.46) (1.68) (1.34) (1.08)
T 0.02 0.04 0.04 0.06 0.07 0.08 0.06 0.07 0.36 0.40 0.43 0.40
(2.13)nn (2.04)nn (2.23)nn (2.12)nn (2.30)nn (2.24)nn (2.16)nn (2.43)nn (2.31)nn (1.74)n (1.86)n (1.67)n
ln K 0.18 0.20 0.24 0.23 0.21 0.23 0.20 0.21 0.23 0.24 0.26 0.25
(2.37)nn (2.40)nn (2.33)nn (2.51)nn (2.45)nn (2.36)nn (1.98)n (1.87)n (2.36)nn (2.37)nn (2.51)nn (2.08)nn
INS 0.25 0.23 0.24 0.20 0.22 0.21 0.31 0.33 0.37 0.26 0.22 0.23
Copyright # 2006 John Wiley & Sons, Ltd.

(2.54)nn (2.52)nn (2.48)nn (2.65)nnn (2.76)nnn (3.44)nnn (1.78)n (1.96)n (1.79)n (1.65)n (0.75) (0.91)
LIA 0.28 } } 0.20 } } 0.30 } } 0.25 } }
(2.35)nn (1.85)n (2.41)nn (2.35)nn
PRI } 0.35 } } 0.24 } } 0.38 } } 0.32 }
(3.32)nnn (2.34)nn (2.41)nn (2.48)nn
DOC } } 0.22 } } 0.20 } } 0.27 } } 0.24
(2.38)nn (1.75)n (1.79)n (2.13)nn
LIA INS } } } } } } 0.40 } } 0.38 } }
(2.36)nn (2.55)nn
PRI INS } } } } } } } 0.46 } } 0.42 }
(2.47)nn (2.23)nn
DOC INS } } } } } } } } 0.38 } } 0.35
(2.35)nn (2.28)nn
Latin America } } } 0.72 0.64 0.68 } } } 0.60 0.64 0.62
(1.72)n (1.92)n (1.87)n (2.12)nn (1.87)n (1.95)n
East Asia } } } 0.30 0.49 0.30 } } } 0.49 0.50 0.41
(1.88)n (1.53) (0.93) (1.45) (1.56) (1.34)
Sub-Saharan } } } 0.62 0.50 0.68 } } } 0.80 0.50 0.82
Africa (2.03)nn (1.91)n (2.32)nn (2.53)nn (2.07)nn (2.26)nn
Adj R2 0.63 0.64 0.63 0.72 0.73 0.72 0.62 0.64 0.62 0.70 0.74 0.74
FINANCE, INSTITUTIONS AND ECONOMIC DEVELOPMENT
Jarque–Bera 5.29 3.88 3.65 3.58 3.44 3.39 5.06 3.94 3.67 5.07 4.77 3.67
(w2-stat) (0.07) (0.14) (0.16) (0.17) (0.18) (0.18) (0.08) (0.14) (0.16) (0.08) (0.09) (0.16)
White test 11.32 13.03 14.72 11.35 16.78 12.04 11.03 16.62 14.99 12.05 17.22 11.39
(w2-stat) (0.18) (0.11) (0.07) (0.18) (0.08) (0.15) (0.35) (0.08) (0.13) (0.15) (0.07) (0.19)
Ramsey RESET 1.09 0.37 0.92 2.24 1.43 1.76 2.00 0.87 0.95 1.06 0.27 0.54
(0.36) (0.77) (0.44) (0.09) (0.24) (0.16) (0.12) (0.46) (0.42) (0.37) (0.85) (0.65)
Notes: Figures in parentheses are t-statistics except for Jarque–Bera normality, White heteroscedasticity and Ramsey RESET tests, which are p-values. Significance at 1%, 5% and
10% denoted by nnn, nn and n, respectively.
251
Int. J. Fin. Econ. 11: 245–260 (2006)

The estimated coefficients on physical capital are consistent with theory. The results of the diagnostic
tests suggest that the models are relatively well specified. In addition, the adjusted R2 suggests that these
models explain about 62–74% of the variation in real GDP per capita. In Models 1–3, all three financial
development indicators, as well as the institutions variable are positive and statistically significant, as
expected. When the regional dummies are included, as shown in Models 4–6, the institutions variable
remains highly significant. However, of the three financial development indicators only the private sector
credit remains statistically significant at the conventional level. In Models 7–9, the newly included
interaction term is highly significant, while the significance of the institutions variable is reduced to the 10%
level. All three financial development indicators remain significant at conventional levels except for
domestic credit. In Models 10–12 where the regional dummies are also included, both the interaction term
and the financial development indicator remain highly significant, and the coefficients of the interaction
term are larger than those of the financial development indicators. However, the linear institutions term is
no longer significant. These findings seem to indicate that both the quantity and the quality of finance
matter for economic development, while institutions matter only in so far as they can improve the quality of
finance.
We now turn to examine the extent to which the above findings vary with the stage of economic
development, by re-estimating the models utilizing panels of high-, middle- and low-income countries.
Tables 2, 4 and 6 report estimates of Equation (8), whereas Tables 3, 5 and 7 report estimates of
Equation (9). These tables utilize three alternative panel data estimators: static fixed effect; PMG,
which imposes common long-run effects, and MG, which imposes no restrictions. The tables present
estimates of the long-run coefficients, the adjustment coefficient, and joint Hausman test statistics.8 The
comparison between MG and PMG is based on the Hausman test. The lag order is first chosen in each
country on the unrestricted model with one lag for the independent variable.9 Because the time span of the
panel data is only 23 years, the MG estimator suffers from too few degrees of freedom. The Hausman test
statistic fails to reject the restriction of common long-run coefficients. Hence, the MG estimator is not as
informative as the PMG estimator and we therefore focus our commentary on the PMG and static fixed
effect results.
Table 2 reports the three alternative panel data estimators of Equation (8) for high-income countries
when the interaction term is not included. These results reveal that the signs of the long-run coefficients
remain similar to those obtained by OLS and are consistent with theory. The static fixed effect estimator
and PMG estimator demonstrate that the coefficients of liquid liabilities and private sector credit are
positive and statistically significant, while domestic credit is not statistically significant. The institutional
quality indicator, however, is statistically significant only at the 10% level based on the static fixed effect
and PMG results. Table 3, which includes the interaction term, also suggests that institutional quality, on
its own, is insignificant for high-income countries. The interaction term, however, is significant at
conventional levels when using liquid liabilities and private sector credit as the financial development
indicators. However, it is significant only at the 10% level when financial development is proxied by
domestic credit. These findings seem to suggest that even within high-income countries financial
development has positive direct and indirect effects on long- run output per capita. The same cannot,
however, be said for institutional quality, the effects of which seem to be working largely through the
financial system.
The panel data estimations results for middle-income countries without the interaction terms are
reported in Table 4. The estimated coefficients of the financial development indicators, all of which are
strongly significant, are larger than the corresponding ones for the high-income group. This finding is
consistent with Rioja and Valev (2004), who also find a much stronger growth-enhancing effect of financial
development in middle-income countries compared to high-income countries. Institutional quality also has
a positive and highly significant effect in all models. Thus, our findings suggest that good institutions are
more important for economic development in middle-income countries than in high-income countries. This
conclusion is also supported by the results reported in Table 5, where the institutional quality indicator is
positive and significant in all three models, irrespective of estimator used. Moreover, the estimated
coefficients of the interaction term are not only positive and highly significant in all three models,
Table 2. Alternative panel data estimations of high-income countries. Dependent variable: ln real GDP per capita (24
countries, 1978–2000; without interaction term)
Static fixed-effects PMG estimators MG estimators
estimators
Liquid liabilities/GDP ðLIAÞ
T 0.01 0.01 0.07
(0.70) (1.50) (2.53)nnn
ln K 0.03 0.40 0.43
(1.76)n (10.32)nnn (5.87)nnn
LIA 0.15 0.14 0.30
(6.24)nnn (2.76)nnn (1.13)nnn
INS 0.12 0.12 0.20
(1.79)n (1.68)n (2.11)nn
Error correction adjustment 1 0.37 0.42
(N/A) (4.15)nnn (6.54)nnn
H test for long-run homogeneity } 8.11 (0.08)
Private sector credit/GDP ðPRIÞ

T 0.81 0.89 0.85
(14.35)nnn (21.84)nnn (22.85)nnn
ln K 0.09 0.06 0.06
(2.41)nn (2.06)nn (2.06)nn
PRI 0.10 0.12 0.16
(2.39)nn (2.39)nn (1.89)n
INS 0.11 0.17 0.17
(1.79)n (1.92)n (2.08)nn
(N/A) (3.36)nnn (5.33)nnn
Domestic credit/GDP ðDOCÞ

T 0.81 0.89 0.86
(14.33)nnn (20.26)nnn (24.23)nnn
ln K 0.09 0.11 0.07
(2.47)nn (3.68)nnn (2.33)nn
DOC 0.11 0.12 0.12
(1.14) (1.67)n (2.07)nn
INS 0.11 0.12 0.07
(1.67)n (1.88)n (2.33)nn
(N/A) (4.58)nnn (6.58)
Notes: All equations include a constant country-specific term. Figures in parentheses are t-statistics except for Hausman tests (H),
which are p-values. nnn, nn and n indicate significance at the 1%, 5% and 10% levels, respectively. N T ¼ 552:
again irrespective of estimator, but are also uniformly higher than the corresponding ones for high-income
countries. Finally, a similar picture emerges when comparing the estimated coefficients on the
financial development indicators with the corresponding ones in Table 3. To conclude, these findings
suggest that both finance and institutional quality have large direct and indirect effects on economic
development in middle-income countries, which are greater than the corresponding ones for high-income
countries.
Tables 6 and 7 report the results for low-income countries. In Table 6, which excludes the interaction
term, the estimated coefficients on the financial development indicator are uniformly lower than the
corresponding ones for middle-income countries and uniformly higher that those for high-income
countries. However, these coefficients are less precisely estimated when using the fixed effects or MG
Table 3. Alternative panel data estimations of high-income countries. Dependent variable: ln real GDP per capita (24
countries, 1978–2000; with interaction term)
estimators
T 0.01 0.01 0.01
(1.67) (0.11) (1.32)
ln K 0.03 0.04 0.02
(1.81)n (1.18) (1.58)
LIA 0.26 0.16 0.32
(4.22)nnn (2.87)nnn (2.15)nn
INS 0.12 0.12 0.21
(1.53) (1.61)n (2.05)nn
LIA INS 0.31 0.28 0.11
(5.03)nnn (3.20)nnn (1.05)
(N/A) (4.21)nnn (6.54)nnn

T 0.80 0.90 0.84
(79.90)nnn (59.35)nnn (22.85)nnn
ln K 0.08 0.10 0.10
(2.02)nn (3.54)nnn (3.25)nnn
PRI 0.24 0.20 0.15
(1.90)n (2.20)nn (2.31)nn
INS 0.11 0.11 0.06
(1.48) (1.38) (2.02)nn
PRI INS 0.26 0.28 0.20
(2.50)nn (2.35)nn (1.01)
(N/A) (4.34)nnn (3.74)nnn

T 0.81 0.92 0.82
(76.04) (55.99) (4.34)nnn
ln K 0.09 0.10 0.09
(2.38)nn (2.54)nn (2.32)nn
DOC 0.05 0.14 0.12
(0.82) (2.25)nn (2.02)nn
INS 0.11 0.12 0.08
(1.69)n (1.78)n (2.30)nn
DOC INS 0.16 0.18 0.24
(1.47) (1.92)n (2.08)nn
(N/A) (4.40)nnn (4.10)
estimator, which renders them insignificant at the conventional 5% level in two out of three cases.
They are, however, significant when using the PMG estimator in all three cases. Institutional
quality, however, enters significantly throughout the three models when using the fixed effect and PMG
estimators. The estimated coefficients on institutional quality are larger than the corresponding ones
obtained for middle or high-income countries. Table 7 reveals that the estimated coefficients of the
Table 4. Alternative panel data estimations of middle-income countries. Dependent variable: ln real GDP per capita
(24 countries, 1978–2000; without interaction term)
estimators
T 0.01 0.02 0.04
(7.24)nnn (11.52)nnn (1.54)
ln K 0.15 0.25 0.30
(5.03)nnn (9.04)nnn (3.39)nnn
LIA 0.27 0.24 0.20
(6.04)nnn (5.51)nnn (2.49)nn
INS 0.12 0.10 0.13
(2.41)nn (2.16)nn (2.08)nn
(N/A) (5.04)nnn (7.07)nnn

T 0.01 0.01 0.01
(7.02)nnn (10.08)nnn (0.46)
ln K 0.15 0.25 0.29
(4.72)nnn (8.40)nnn (4.80)nnn
PRI 0.30 0.28 0.26
(3.61)nnn (3.64)nnn (1.98)n
INS 0.13 0.12 0.09
(2.39)nn (3.82)nnn (2.02)nn
(N/A) (5.89)nnn (7.92)nnn

T 0.01 0.02 0.03
(7.71)nnn (13.61)nnn (1.56)
ln K 0.13 0.23 0.18
(4.23)nnn (7.89)nnn (3.55)nnn
DOC 0.22 0.24 0.14
(2.35)nn (2.47)nn (1.98)n
INS 0.12 0.14 0.19
(2.46)nn (2.35)nn (1.86)n
(N/A) (5.91)nnn (7.28)
interaction terms are positive and significant throughout, irrespective of estimator used. The same
is true of institutional quality. However, the financial development indicator itself loses its
significance in the three models when using the fixed effects estimator, but remains significant
when using the PMG estimator. These ambiguous findings suggest that in low-income countries
institutional quality is a more robust determinant of long-term economic development than financial
development. It seems that without good institutions any positive effects of financial development
are weakened substantially, if they are to be found at all. It is also worth noting that in both
Tables 6 and 7, of the three financial development indicators, private sector credit enters with the largest
coefficients.
Table 5. Alternative panel data estimations of middle-income countries. Dependent variable: ln real GDP per capita
(24 countries, 1978–2000; with interaction term)
estimators
T 0.01 0.02 0.04
(8.68)nnn (11.63)nnn (1.23)
ln K 0.21 0.25 0.17
(7.12)nnn (9.12)nnn (1.11)
LIA 0.28 0.26 0.18
(249)nn (2.43)nn (2.35)nn
INS 0.13 0.12 0.15
(2.07)nn (2.35)nn (2.26)nn
LIA INS 0.35 0.38 0.22
(7.76)nnn (2.86)nnn (2.32)nn
(N/A) (5.00)nnn (4.09)nnn

T 0.01 0.01 0.03
(7.07)nnn (9.47)nnn (0.23)
ln K 0.19 0.26 0.35
(6.73)nnn (8.38)nnn (4.34)nnn
PRI 0.32 0.34 0.25
(3.71)nnn (3.07)nnn (2.16)nn
INS 0.12 0.12 0.15
(2.51)nn (2.38)nn (2.30)nn
PRI INS 0.38 0.40 0.30
(4.73)nnn (3.76)nnn (2.94)nnn
(N/A) (5.99)nnn (4.54)

T 0.01 0.02 0.05
(7.69)nnn (14.06)nnn (1.35)
ln K 0.16 0.23 0.32
(5.06)nnn (7.99)nnn (2.75)nnn
DOC 0.24 0.27 0.21
(2.44)nn (2.88)nnn (1.88)n
INS 0.12 0.11 0.16
(2.48)nn (2.06)nn (2.27)nn
DOC INS 0.33 0.35 0.28
(5.30)nnn (2.82)nnn (2.39)nn
(N/A) (5.94)nnn (2.85)nnn
4. CONCLUSION
Our findings suggest that financial development has larger effects on long-run economic development when
the financial system is embedded within a sound institutional framework. We found this to be particularly
true for poor countries, where more finance may or may not deliver significant benefits, if institutional
Table 6. Alternative panel data estimations of low-income countries. Dependent variable: ln real GDP per capita (24
countries, 1978–2000; without interaction term)
estimators
T 0.02 0.01 0.02
(1.77) (0.95) (0.23)
ln K 0.10 0.45 0.26
(4.02)nnn (11.62)nnn (2.81)nnn
LIA 0.20 0.21 0.18
(1.94)n (2.27)nn (1.96)n
INS 0.14 0.15 0.13
(6.39)nnn (2.26)nn (2.38)nn
(N/A) (4.22)nnn (6.96)nnn

T 0.03 0.01 0.01
(2.25)nn (4.61)nnn (0.14)
ln K 0.08 0.31 0.19
(3.32)nnn (10.62)nnn (3.53)nnn
PRI 0.26 0.25 0.31
(2.49)nn (2.27)nn (2.44)nn
INS 0.14 0.15 0.17
(7.07)nnn (2.46)nn (2.03)nn
(N/A) (3.95)nnn (4.04)nnn
Domestic credit/GDP (DOC)

T 0.03 0.04 0.01
(2.06)nn (5.11)nnn (0.11)
ln K 0.08 0.31 0.08
(3.32)nnn (10.95)nnn (0.65)
DOC 0.18 0.21 0.12
(1.94)n (2.02)nn (1.36)
INS 0.14 0.15 0.16
(2.65)nnn (2.37)nnn (1.69)n
(N/A) (3.89)nnn (3.53)nnn
quality is low. For poor countries, improvements in institutions are likely to deliver much larger direct
effects on economic development than finance on its own. They are also likely to have positive indirect
effects, through the financial system, particularly when the latter is providing large amounts of credit to the
private sector.
Our findings also suggest that financial development is most potent in delivering real economic
benefits in middle-income countries. Its effects are particularly large when institutional quality is high.
Institutional improvements can also deliver more economic development, especially when the financial
system is developed. The effects of financial development in high-income countries are smaller than in
middle-income countries. However, even in these countries its effects appear to be much larger when
institutional quality is high.
Table 7. Alternative panel data estimations of low-income countries. Dependent variable: ln real GDP per capita (24
countries, 1978–2000; with interaction term)
estimators
T 0.03 0.03 0.05
(2.32)nn (1.27) (0.39)
ln K 0.16 0.42 0.16
(6.18)nnn (10.29)nnn (3.18)nnn
LIA 0.18 0.20 0.17
(1.51) (2.41)nn (1.94)n
INS 0.14 0.13 0.14
(2.32)nn (2.37)nn (2.18)nn
LIA INS 0.28 0.25 0.23
(2.36)nn (2.34)nn (2.18)nn
(N/A) (4.52)nnn (4.48)nnn

T 0.05 0.02 0.05
(3.53)nnn (0.74) (0.40)
ln K 0.11 0.46 0.15
(4.60)nnn (10.15)nnn (2.83)nnn
PRI 0.23 0.28 0.25
(1.83)n (2.35)nn (2.19)nn
INS 0.14 0.15 0.11
(2.36)nn (2.17)nn (2.14)nn
PRI INS 0.30 0.32 0.28
(2.51)nn (3.55)nnn (2.41)nn
(N/A) (5.12)nnn (4.62)nnn

T 0.04 0.01 0.06
(3.11)nnn (0.31) (0.52)
ln K 0.14 0.52 0.18
(5.61)nnn (10.15)nnn (3.19)nnn
DOC 0.20 0.24 0.22
(1.88)n (2.47)nn (1.68)n
INS 0.13 0.14 0.15
(2.15)nn (2.28)nn (2.11)nn
DOC INS 0.26 0.30 0.24
(2.34)nn (2.63)nnn (2.35)nn
(N/A) (4.02)nnn (2.94)nnn
To conclude, it does appear to be the case that the interaction between financial development and
institutional quality, a variable that has been neglected by previous studies, is very important in terms of
economic development at all stages of development. Thus, ‘better finance, more growth’ seems to be a much
more widely applicable proposition than ‘more finance, more growth’.
ACKNOWLEDGEMENTS
We would like to thank two anonymous referees for their constructive suggestions that have helped to
improve the paper. We would also like to thank Gary Koop and Giovanni Urga for helpful comments.
Demetriades acknowledges financial support from the Nuffield Foundation (Grant SGS/0667/G).
APPENDIX A: LIST OF COUNTRIES USED IN THE ESTIMATION CATEGORIZED

THE WORLD BANK
High-income Middle-income Low-income

Australia AUS Argentina ARG Algeria DZA
Austria AUT Bolivia BOL Bangladesh BGD
Belgium BEL Brazil BRA Cameroon CMR
Canada CAN Chile CHL Egypt EGY
Denmark DNK Colombia COL Gambia GMB
Finland FIN Costa Rica CRI Ghana CHA
France FRA Cyprus CYP Haiti HAI
Germany GER Dominican Republic DOM Honduras HND
Greece GRC Ecuador ECU India IND
Iceland ICE El Salvador ESL Indonesia IDN
Ireland IRE Guatemala GTM Kenya KEN
Israel ISL Iran IRN Malawi MWI
Italy ITA Jamaica JAM Niger NIG
Japan JPN Korea KOR Pakistan PAK
Netherlands NEL Malaysia MYS Philippines PHL
New Zealand NZL Mexico MEX Senegal SEN
Norway NOR Nicaragua NIC Sierra Leone SIL
Portugal POR Panama PAN Sri Lanka LKA
Singapore SIN Papua New Guinea PNG Tanzania TAN
Spain ESP Paraguay PRY Thailand THA
Sweden SWE Peru PER Togo TOG
Switzerland CHE South Africa ZAF Tunisia TUN
United Kingdom GBE Syria SYR Zimbabwe ZWE
United States USA Uruguay URY Zambia ZMB
NOTES
1. Aron (2000) provides an excellent review of a large body of empirical literature that tries to link quantitative measures of
institutions with economic growth across countries and over time.
2. In the neoclassical growth model only labour-augmenting technological change is consistent with the existence of the steady state.
See Barro and Sala-i-Martin (1995).
3. Note, however, that an argument could also be made that the aggregate production function is only a long-run relationship
between inputs and output, because, say, the stocks of inputs may be over or under-utilized in the short run. In such a case
dynamic panel data techniques are called for}see below.
4. The list of countries is presented in Appendix A.
5. The World Bank classifies economies as low-income if the GDP per capita is less than US$755; middle-income if the GDP per
capita is between US$755 until US$9265 and high-income economies if the GDP per capita is more than US$9265.
6. The scale of corruption, bureaucratic quality and rule of law was first converted to 0 to 10 (multiplying them by 5/3) to make them
comparable to the other indicators. For robustness checks, we also used different weights for each indicator to construct the
aggregate index. The estimates are similar and are available on request.
7. The equations were also estimated using 2-stage least squares (2SLS), to check for possible endogeneity of the financial
development, capital stock and/or the institutional quality indicators. The results are not reported but are available on request.
8. We are grateful to Y. Shin for providing us with an appropriate GAUSS program.
9. The lag structure is ð1; 0; 0; 0; 0Þ for Equation (9) and the order of variables is as follows: RGDPC, Time Trend, financial
development, INS and K.
REFERENCES
Arestis P, Demetriades P. 1997. Financial development and economic growth: assessing the evidence. Economic Journal 107: 783–799.
Aron J. 2000. Growth and institutions: a review of the evidence. The World Bank Research Observer 15: 99–135.
Barro RJ, Sala-i-Martin X. 1992. Convergence. Journal of Political Economy 100: 223–251.
Barro RJ, Sala-i-Martin X. 1995. Economic Growth. McGraw-Hill: New York.
Clague C. 1993. Rule obedience, organizational loyalty, and economic development. Journal of Institutional and Theoretical Economics
149: 393–414.
Demetriades P, Andrianova S. 2004. Finance and growth: what we know and what we need to know. In Financial Development and
Growth: Explaining the Links, Goodhart CAE (ed.). Palgrave Macmillan: Basingstoke & New York; 38–65.
Demetriades P, Hussein K. 1996. Does financial development cause economic growth? Time series evidence from 16 countries. Journal
of Development Economics 51: 387–411.
Evans P. 1997. How fast do economies converge. The Review of Economics and Statistics 79: 219–225.
Goodhart CAE. 2004. Financial Development and Growth: Explaining the Links. Palgrave Macmillan: Basingstoke & New York.
Hall R, Jones C. 1999. Why do some countries produce so much more output per worker than others? Quarterly Journal of Economics
114: 83–116.
Hausman J. 1978. Specification tests in econometrics. Econometrica 46: 1251–1271.
Knack S, Keefer P. 1995. Institutions and economic performance: cross-country tests using alternative institutional measures.
Economics and Politics 7: 207–227.
Lee K, Pesaran MH, Smith RP. 1997. Growth and convergence in a multi-country empirical stochastic Solow model. Journal of
Applied Econometrics 12: 357–392.
Levine R. 2003. More on finance and growth: more finance, more growth? Federal Reserve Bank of St. Louis Review 85: 31–46.
Mankiw NG, Romer D, Weil DN. 1992. A contribution to the empirics of economic growth. Quarterly Journal of Economics 107:
407–437.
Mauro P. 1995. Corruption and growth. Quarterly Journal of Economics 110: 681–712.
Nelson RR, Sampat BN. 2001. Making sense of institutions as a factor shaping economic performance. Journal of Economic Behavior
& Organization 44: 31–54.
North DC. 1990. Institutions, Institutional Change, and Economic Performance. Cambridge University Press: Cambridge.
Pagano M. 1993. Financial markets and growth: an overview. European Economic Review 37: 613–622.
Pesaran MH, Shin Y, Smith RP. 1999. Pooled mean group estimation of dynamic heterogeneous panels. Journal of American
Statistical Association 94: 621–634.
Pesaran MH, Smith RP. 1995. Estimating long-run relationship from dynamic heterogeneous panels. Journal of Econometrics 68:
79–113.
Pistor K, Wellons PA, Sachs JD, Scott HS. 1998. The Role of Law and Legal Institutions in Asian Economic Development, 1960–1995.
Oxford University Press: Oxford.
Rioja F, Valev N. 2004. Does one size fit all?: a re-examination of the finance and growth relationship. Journal of Development
Economics 74: 429–447.
Rodrik D. 1997. TFPG controversies, institutions and economic performance in East Asia. National Bureau of Economic Research
Working Paper, No. 5914.
Rodrik D, Subramanian A, Trebbi F. 2002. Institutions rule: the primacy of institutions over integration and geography in economic
development. IMF Working Paper, No. 189.
Zingales L. 2003. The weak links. Federal Reserve Bank of St. Louis Review 85: 47–52.

Int J Fin Econ - 2006 - Demetriades - Finance Institutions and Economic Development

Uploaded by

Copyright:

Available Formats

Int J Fin Econ - 2006 - Demetriades - Finance Institutions and Economic Development

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Int J Fin Econ - 2006 - Demetriades - Finance Institutions and Economic Development

Uploaded by

Copyright:

Available Formats

INTERNATIONAL JOURNAL OF FINANCE AND ECONOMICS

Int. J. Fin. Econ. 11: 245–260 (2006)

FINANCE, INSTITUTIONS AND ECONOMIC DEVELOPMENT

Copyright # 2006 John Wiley & Sons, Ltd.

2. EMPIRICAL MODEL, METHODOLOGY AND DATA

2.1. Empirical model

Ait ¼ Ai0 e gi tþPit yi ð3Þ

2.2. Econometric approach

Copyright # 2006 John Wiley & Sons, Ltd.

Int. J. Fin. Econ. 11: 245–260 (2006)

Private sector credit/GDP ðPRIÞ

Domestic credit/GDP ðDOCÞ

Private sector credit/GDP ðPRIÞ

Domestic credit/GDP ðDOCÞ

Private sector credit/GDP ðPRIÞ

Domestic credit/GDP ðDOCÞ

Private sector credit/GDP ðPRIÞ

Domestic credit/GDP ðDOCÞ

Private sector credit/GDP ðPRIÞ

Domestic credit/GDP (DOC)

Private sector credit/GDP ðPRIÞ

Domestic credit/GDP ðDOCÞ

APPENDIX A: LIST OF COUNTRIES USED IN THE ESTIMATION CATEGORIZED

High-income Middle-income Low-income

You might also like