Post on 09-Jun-2020
DEPARTMENT OF ECONOMICS
ISSN 1441-5429
DISCUSSION PAPER 26/16
Oil Curse and Finance-Growth Nexus in Malaysia: The Role of
Investment
Ramez Abubakr Badeeb, Hooi Hooi Lean* and Russell Smyth
Abstract: We empirically examine the existence of an oil curse in the finance-growth nexus in Malaysia.
We provide new insights into the oil curse phenomenon in Malaysia that challenges the
conventional argument that Malaysia is a counter-example of an oil curse country. We do not
find any significant evidence of direct effects of financial development on economic growth
and TFP. However, there are direct and positive effects on the level of investment due to
financial development and oil dependence. While we do not find statistical evidence of a direct
negative impact of oil rent on economic growth, our results reveal that the symptoms of an oil
curse exist. Specifically, we find that oil rent has a weak, indirect, impact on the finance-growth
nexus through the quantitative channel or investment quantity. The policy implications of our
findings are that the financial sector should be more involved in productive investment
activities that can strengthen its role in economic growth and that policymakers should reduce
dependence on oil and promote economic diversification.
Keywords: Oil Curse; Financial Development; Economic Growth; Investment; Malaysia.
JEL Classification: O13; O16; C22
* This version is forthcoming in Energy Economics. We thanks two referees and Jakob Madsen for
comments/advice on earlier versions of this paper.
Hooi Hooi Lean, Economics Program, School of Social Sciences Universiti Sains Malaysia
Ramez Abubakr Badeeb Economics Program, School of Social Sciences Universiti Sains Malaysia
Russell Smyth Department of Economics, Monash University, Australia
© 2016 Ramez Abubakr Badeeb, Hooi Hooi Lean and Russell Smyth
All rights reserved. No part of this paper may be reproduced in any form, or stored in a retrieval system, without the prior
written permission of the author.
monash.edu/ business-economics ABN 12 377 614 012 CRICOS Provider No. 00008C
3
1. Introduction
The relationship between natural resource (oil) dependence and economic growth has
received considerable attention among economists (Gelb, 1988; Auty, 1993; Sachs and
Warner, 1995, 1999, 2001; Gylfason, 2001; Mehlum et al., 2006; Apergis et al., 2014;
Apergis and Payne, 2014; Betz et al., 2015). In principle, oil and other non-renewable natural
resources should offer huge benefits to economies. However, empirical evidence shows that
these resources do not always lead to increased economic growth. This puzzling phenomenon
is referred to as the “natural resource curse” (Auty, 1993). The term “natural resource curse”
refers to the phenomenon of encumbering growth caused via a series of negative effects, due
to excessive dependence on natural resources, such as oil and natural gas, in a country or a
region (Shao and Yang, 2014). It occurs when natural resources interact with, and alter,
various social, political and economic factors, resulting in slower economic growth and
impeding development. “Oil curse” has often been used as a synonym for resource curse
because it is primarily observed in oil-dependent economies.
The literature on the natural resource curse consists of two strands. The first strand has
examined the direct effect of natural resource dependence on economic growth (e.g. Sachs
and Warner, 1995, 1999, 2001; Sachs, 2007; Gylfason, 2001; Mevrotas et al., 2011 and Kim
and Lin, 2015). The second strand of literature has examined the effect of natural resource
dependence on macroeconomic and political economic factors, such as human capital, good
governance, investment and savings that are associated with sustainable economic growth
(e.g., Atkinson and Hamilton, 2003; Gylfason and Zoega, 2006; Mehlum et al., 2006; Dietz et
al., 2007; Papyrakis and Gerlagh, 2007; Daniele, 2011; Blanco and Grier, 2012; Libman,
2013; Bhattacharyya and Hodler, 2014; Cockx and Francken, 2016).
4
One factor that is associated with sustainable economic development is financial
development. According to Levine (1997), there are two channels through which financial
development affects economic growth. These channels are depicted in Figure 1. The first
channel provides liquidity services and mobilizes savings, which affects the scale of
investment (quantitative channel). The second channel is the efficient allocation of resources
to productive investments, which induces economic growth (qualitative channel). Oil
dependence can influence the relationship between financial development and economic
growth in various ways. On the one hand, oil revenue might be considered an extra resource
for financial institutions, in which case the link between financial development and economic
growth would be stronger. On the other hand, high natural resource dependence may hinder
the ability of financial institutions to accumulate capital (quantitative channel) or to allocate
credit efficiently to productive investments (qualitative channel), which are negative effects.
[Insert Figure 1]
Empirical evidence concerning how natural resource dependence affects financial
development, and thus influences economic growth, is scant (Nili and Rastad, 2007; Beck,
2011; Yuxiang and Chen, 2011; Barajas et. al., 2013). Related research examines financial
sector characteristics in resource-dependent economies (Kurronen, 2015), the effect of oil
price fluctuations on the current account in oil exporting countries (Allegret and Mignon,
2014), and the effect of oil prices on financial markets (Narayan and Narayan, 2010; Kang et
al., 2014; Demirir et al., 2015; Phan et al., 2015). The latter, however, does not consider if
there is an oil curse, nor examine the effect of financial development on economic growth.
We examine the role of natural resource (oil) dependence on the finance-growth nexus in one
of the largest oil-rich economies in Southeast Asia, Malaysia. We first examine whether oil
5
dependence has a direct effect on economic growth. We then examine whether oil
dependence has an indirect effect on the finance-growth nexus via the investment quantity
(quantitative) and investment efficiency (qualitative) channels. While we do not find
evidence of a direct negative impact of oil dependence on economic growth, we do find that
oil dependence has a weak, indirect, impact on the finance-growth nexus through the
investment quantity channel. Our results are generally robust to several alternative ways of
measuring financial development as well as oil dependence.
We situate the study in Malaysia because there is considerable debate over the role of oil in
facilitating economic development in that country. Prior to the Asian financial crisis,
Malaysia was one of the fastest growing economies in the world and was widely considered
to be one of those countries that had escaped the resource curse (Gylfason, 2001; Rosser,
2006). However, recently, Doraisami (2015) has argued that over dependency on natural
resources has led the Malaysian government to engage in inefficient activities and provide
funding for unproductive investments, casting doubt on whether Malaysia really has escaped
the resource curse. This finding, along with the current destabilized economic situation in
Malaysia that is associated with the prolonged global fall in oil prices, motivates us to dig
deeper to explore empirically whether the conventional argument about Malaysia is still valid.
We contribute to the literature in the following important ways. First, as noted above, there
are very few papers that specifically examine how natural resource dependence affects the
relationship between financial development and economic growth and the associated
channels through which this occurs. We contribute to the scant evidence on the effects of oil
dependence on the relationship between financial development and economic growth and the
channels through which oil dependence affects the finance-growth nexus.
6
Second, of the few studies that explicitly examine how natural resource dependence affects
the relationship between financial development and economic growth, almost all provide
cross-sectional and/or panel evidence for large numbers of countries (Nili and Rastad, 2007;
Beck, 2011; Barajas et. al., 2013). Beck (2011) has emphasized that understanding how
natural resource dependence influences the finance-growth nexus is critical for policymakers.
He states (at p.2):
“Policymakers who care about the development of their countries need to
understand the relative importance of different policy areas and the effectiveness of
specific policies. Understanding channels through which resource abundance can
stimulate or dampen economic development can be important to develop policies to
maximize the benefits of natural capital”.
Yet, findings from large multi-country studies are of limited value for policy-making in
specific countries. The use of cross-sectional or panel data has led previous studies to ignore
the existence of large variations among countries in terms of the degree of dependence on
natural resources, type of resources or stage of financial development. One exception is
Yuxiang and Chen (2011), who discussed the direct effect of natural resource dependence on
the financial development in Chinese provinces. However, these researchers ignored the
indirect effects of natural resources on the relationship between financial development and
economic growth. We provide the first single country evidence of the effect of oil
dependence on the financial development-economic growth relationship coupled with
examination of the channels through which oil dependence affects the finance-growth nexus.
7
Third, existing studies focus on the direct relationship between natural resource dependence
and financial development in terms of its depth or size, but largely ignore the impact on
efficiency, or the ability of the financial sector to translate savings into optimum investments
in one step, and translate this investment into economic growth in the subsequent step. Unlike
the existing literature, we pay particular attention to the role played by the efficiency of the
financial sector in understanding how the natural resource dependence might depress
economic growth through the financial development channel.
Fourth, Beck (2011) calls for more detailed treatment of how natural resource dependence
affects the financial development-economic growth relationship using better measures of
natural resource dependence. In our main results, we use oil and gas rent as a proportion of
GDP to measure oil dependence. Oil and gas rent has been the preferred measure of oil
dependence in a number of recent studies on the natural resource curse (Collier and Hoeffler,
2009; Bhattacharya and Hodler, 2010, 2014). This is a better measure of oil dependence than
used in some of the previous multi-country studies that have examined how natural resource
dependence affects the financial development-economic growth relationship, such as Nili and
Rastad (2007), who use a dummy variable to denote whether a country is oil exporting to
capture oil dependence. Compared to our approach, that measure has the limitation that it
masks differences in the degree of dependence on oil among oil exporters.
Our fifth contribution is to the debate on the role of oil dependence in the specific case of
Malaysia. While we use Malaysia to illustrate arguments likely to have application to major
oil producers more generally, in so doing specifically, we add statistical evidence to
Doraisami’s (2015) observation; based on a reading of the descriptive data, on whether
Malaysia has been successful in escaping the natural resource curse.
8
The remainder of this paper is organized as follows: the literature review is presented in
section 2. In section 3, we provide an overview of the Malaysian economy. Section 4 focuses
on data and methodology, and the empirical results and discussion are presented in section 5.
Finally, section 6 concludes with policy implications.
2. Literature Review
Our study is related to four main strands of literature. The first is studies of the effect of
natural resource dependence on economic growth (Gelb, 1988; Auty, 1993; Sachs and
Warner, 1995, 2001; Sachs, 2007; Gylfason, 2001; Mehlum et al., 2006; Apergis et al., 2014;
Apergis and Payne, 2014; Betz et al., 2015; Kim and Lin, 2015). Much of the literature on the
natural resource curse is summarized in Frankel (2010). While there is wide country variation,
there is considerable empirical evidence of a natural resource curse on economic
development. Studies have shown that natural resource wealth provides the means for
autocratic andother less than democratic governments to buy off opposition and avoid
accountability and transparency. It also allows elites to hold on to power and facilitates
inefficient rent seeking that impedes economic development. This can also result in conflict,
as in some countries in Sub-Saharan Africa, that further curtails economic growth (see Collier
and Hoeffler, 2009; Beck, 2010; Bhattacharya and Hodler, 2010, 2014).
The second strand of literature to which our study is related is studies on the relationship
between financial development and economic growth. Beginning with Schumpeter (1934), a
large literature has shown that financial development can encourage economic growth by
channeling resources to highly productive investments (Patrick, 1966; McKinnon, 1973;
Shaw, 1973). These studies have shown that that financial intermediation (banks) plays an
important role in the economy by increasing saving and capital accumulation (Pagano, 1993;
9
King and Levine, 1993a, b; Levine, 1997, 2003; Hassan et al., 2011; Beck, 2011; Bittencourt,
2012).
Based on a sample of 77 countries for the period 1980–2007, Beck et al. (2014) found that, in
the long run, financial intermediation increases growth and reduces growth volatility. Both
effects have, however, become weaker over time. Law and Singh (2014) found that there is a
threshold effect in the finance-growth nexus. They found that financial development is
beneficial for economic growth up to a certain threshold; thereafter, financial development
tends to adversely affect economic growth (see also Shen and Lee, 2006; Ergungor, 2008;
Arcand et al., 2012; Cecchetti and Kharroubi, 2012; Ductor and Grechyna, 2015; Samargandi
et al., 2015)1.
The third strand of literature to which our study is related is studies of the effect of natural
resources on the financial sector and financial development. This literature includes the large
number of studies on the effect of oil prices on the financial sector, and stock market in
particular (see e.g. Narayan and Narayan, 2010; Kang et al., 2014; Demirir et al., 2015; Phan
et al., 2015). Beginning with Jones and Kaul (1996) and Sadorsky (1999), several studies
show that stock markets respond negatively to a positive oil price change; however, other
studies find no relationship (see Filis et al., 2011 for a review). Narayan and Sharma (2011)
found that the exact relationship depended on the sector in which the stock was located while
Phan et al. (2015) found that the relationship depended on whether the stock was for an oil
producer or oil consumer. Arouri and Rault (2011) found that in oil exporting countries, the
relationship between oil price shocks and stock market returns was positive. None of these
1 Refer to Ang (2009), Levine (2003) and Law and Singh (2014) for a comprehensive review on the
relationship between financial development and economic growth.
10
studies, however, consider the effect of oil, or other natural resources, on the relationship
between financial development and economic growth.
The literature to which this study directly contributes, which is how natural resource
dependence affects the finance-growth nexus, is limited to a few studies. Nili and Rastad
(2007) compared the effect of financial development on economic growth for 12 oil
economies and 132 non-oil economies over the period 1992 to 2001, finding that the quality
of investment is lower in oil economies, and that this originates from the low quality of the
financial institutions, which translates into poor economic growth. For 153 countries over the
period 1980 to 2007, Beck (2011) found that countries that depend more on natural resources
tend to have an underdeveloped financial system, in which both private credit and stock
market activities are weaker, and access to credit for businesses is more limited. Finally, for
146 countries over the period 1975 to 2005, Barajas et al. (2013) found a weaker finance-
growth nexus in oil-dependent economies than in non-oil economies.
To summarize, the existing literature on how natural resource dependence affects the finance-
growth nexus consists of large cross-sectional and panel studies of several oil and non-oil
countries. There are no single country studies of the effect of oil dependence on the financial
development-economic growth relationship that also examine the channels through which oil
dependence affects the finance-growth nexus. This is a gap in the literature that we seek to
address in the current study.
3. Malaysian Economy Overview
Malaysia is the second largest oil and natural gas producer in Southeast Asia, and the second
largest exporter of liquefied natural gas globally; in addition, it is strategically located amid
11
important routes for seaborne energy trade. According to the Oil and Gas Journal, as of
January 2013, Malaysia held proven oil reserves of 4 billion barrels, the fifth-highest reserves
in Asia-Pacific after China, India, Vietnam, and Indonesia. Malaysia also has 83 trillion cubic
feet of proven natural gas reserves, and, as of January 2013, it was the third largest natural
gas reserves holder in the Asia-Pacific region. Total oil production in 2012 was estimated at
643,000 barrels per day (bbl/d).
The oil dependence indicator has witnessed fluctuations over the last 40 years caused by
fluctuations in commodity prices in the international market. The indicator was 16% in the
late 1970s, and then declined until the mid-1990s. Since 1995, the indicator rose to 18% in
2008, before the global financial crisis, which caused oil prices to decrease, and, thus, the
natural resource dependence to decrease (WDI, 2013) (see Figure 2).
[Insert Figure 2]
Since independence in 1957, Malaysia has experienced solid growth, with an average annual
growth rate of 6% over the period 1970 to 1980. GDP growth decreased significantly during
the economic recession in the mid-1980s. Then, the country recovered from the crisis in the
mid-1980s and achieved an average GDP growth rate of at least 9% from 1990 to 1996.
However, the GDP growth rate declined to -7% during the Asian financial crisis in 1997.
Since then, the economy has shown a slow process of recovery that remained inconsistent,
with an annual growth rate of 5% from 2002 to 2008. The growth rate also declined to -1.5%
in 2009 during the global financial crisis. A critical element of success is the high savings
rate that provided capital available for investment. Figure 3 presents the time series plot of
the real GDP growth rate in Malaysia for the 1970 to 2013 period.
12
[Insert Figure 3]
The financial system in Malaysia can be broadly classified into financial institutions that
consist of a banking system, non-bank financial intermediaries and financial markets that are
composed of money and foreign exchange markets, capital markets, derivative markets and
offshore markets. However, because of limited development in the capital markets, the
Malaysian financial system is predominantly bank-based (Ang, 2009).
The banking system is the largest component of the financial system. The system’s growth
has been remarkable during the past few decades because of strong economic growth, a high
savings rate and advances in telecommunications. The deposits and the domestic credit to the
private sector have dramatically increased from approximately 30% and 20% of GDP in the
1970s to 130% and 135% of GDP, respectively at the beginning of the 2000s (refer to Figure
4). This growth was associated with a series of financial restructuring programs adopted by
the Malaysian authorities. Furthermore, after the Asian financial crisis, a series of
macroeconomic policy responses, such as capital controls and a reflationary policy, have
been adopted (Ang and Mckibbin, 2007).
[Insert Figure 4]
4. Data, Model and Methodology
4.1 Data and Variables
This study employs data for Malaysia over the period 1970-2013 2. All data are obtained from
the World Development Indicators (WDI). Real GDP per worker in constant 2005 USD
2 The reason for beginning and finidhing the study in 1970 and 2013 is the unavailability of data for a main
variable of interest (i.e., oil and gas rent) before and after those years.
13
prices is used as the economic growth measurement, whereas the measurement of oil
dependence, financial development, and investment are chosen based on the following
justifications.
Oil Dependence
Following Collier and Hoefler (2009) and Bhattacharya and Hodler (2010, 2014), among
others, we use oil and gas rent to GDP as the oil dependence indicator. According to the
World Bank, the oil and gas rent is defined as the difference between the value of production
for oil and gas at world prices and their total cost of production.
Financial Development
Due to limited development in the capital markets and the dominant role of the banking
system,3 we initially use one proxy to measure the level of financial intermediation.4 This
proxy is domestic credit to private sector as a share of GDP. It is considered to be one of the
best indicators to measure financial development, and has been widely used in the literature
(e.g., King and Levine, 1993a; Nili and Rastad, 2007; Shahbaz and Lean, 2012). This proxy
provides information regarding the commercial bank’s credit allocated to the private sector,
compared with the size of the economy as a whole. Therefore, this indicator accurately
3The Malaysian financial system is characterized by the limited development of the financial markets, where
banks are the main source of finance. The majority of the companies are not usually listed and the market
concentration ratio is rather high for Malaysia compared to other more advanced financial markets because
market capitalization is highly concentrated in the hands of the ten largest firms (see Ang and Mckibbin, 2007).
Thus, the Malaysian financial system can be described as a bank-based system rather than a market-based
system. Hence, the use of bank-based financial proxies is more appropriate to study the issue at hand. 4More proxies are reported for the robustness check. However, we could not include a variable for stock market
(e.g., market capitalization) due to the lack of this data, which only starts from 1981. Nevertheless, we perform
an unreported-test with the available data of market capitalization as percent of GDP and total value of stock
trade as percent of GDP. We do not find any significant impact of oil dependence on the finance-growth nexus
(result is available upon request). It can be inferred from this result that the oil curse is transmitted to economic
growth through the banking sector rather than the stock market. However, this result must be treated with
caution due to the data limitation.
14
measures the role of financial intermediation in channeling funds to the private sector. The
higher ratio implies more financial services, and, therefore, greater financial development.
Investment Quantity
We use gross fixed capital formation as a share of GDP as a proxy for investment quantity.
According to the World Bank, this indicator includes land improvements (fences, ditches, and
drains) and plant, machinery, and equipment purchases; it also includes the construction of
roads, railways, schools, offices, hospitals, private residential dwellings, and commercial and
industrial buildings.
Investment Efficiency/Quality
In accordance with Dasgupta et al. (2002), investment efficiency is represented by total factor
productivity (TFP). TFP is derived from a standard neoclassical Cobb-Douglas production
function:5
1LAKY a
where Y is the real GDP, K is the real physical capital stock, and L is the total labor force.
Therefore, TFP is measured as A = TFP = y/kα, where y is the output-worker ratio (Y/L) and
k is the capital-worker ratio (K/L). Ang (2009) set the capital’s income share (α) for Malaysia
to 0.3. K is constructed using the perpetual inventory method, Kt = It + (1 - ϑ) Kt-1, where I is
real investment, and ϑ is the depreciation rate, which is assumed to be 5% in accordance with
Bosworth and Collins (2003). The initial capital stock is estimated using the Solow model
steady-state value of I0/ (ϑ+g) where I0 is the initial real investment, and g is the growth rate
in real investment over the 1970-2013 period.
5See Beck et al. (2000), Ang (2009) and Farhadi et al. (2015).
15
4.2 Models
Our first objective is to identify whether oil dependence has a direct effect on economic
growth. Therefore, in the first model, we augment the neoclassical Cobb-Douglas production
function 6 by incorporating financial development and oil dependence in addition to the
capital and labor force (see Ang, 2009, Salim et al., 2014 and Lotz, 2015).
1LAKY a
where Y = aggregate GDP, L = labor, K = capital and A = TFP.
Dividing by L and taking the natural logs,
Y/L = AK(α)L(1-α-1)
Y/L = AK (α)L(-α)
Y/L = A(K/L)(α)
ln(Y/L) = ln A + αln(K/L)
Denote TFP as a function of financial development and oil dependence:
A = f (FD, OD) where FD is financial development and OD is oil dependence.
This suggests our first model:
Model (1): Y / L
t= a
0+a
1FD
t+a
2OD
t+a
3K / L
t+ e
t (1)
where Y/L is GDP per worker in constant 2005 USD prices, FD is the financial development
indicator, OD is the oil dependence indicator, K/L is capital stock per worker. 7 K is
constructed using the perpetual inventory method as described above.
For the second objective, we focus on the link between financial development and investment
quantity/quality to capture the impact of oil dependence in this relation.
6The selection of the Cobb-Douglas production function was made on the basis of its merits. The Cobb-Douglas
function is flexible and it can handle multiple inputs in its generalized form. In the presence of imperfections in
the market, it does not introduce distortions of its own; and various econometric estimation problems, like serial
correlation and heteroscedasticity, can be handled adequately and easily (Bhanumurthy, 2002). 7All variables throughout this paper are transformed into natural logarithms.
16
The rationale for this argument is to determine whether oil dependence has an indirect
weakening effect on the finance-growth nexus through the investment quantity and quality
channels. Any negative effect on one of these two channels would weaken the relation
between financial development and economic growth. This argument provides important
policy implications in terms of a futuristic oil resource management policy in the country.
Inspired by Nili and Rastad (2007), the investment quantity model can be specified as follows:
Model (2): 0 1 2 3 4( * )t t t tINV FD OD FD OD REX (2)
where INV is fixed capital formation to GDP and REX is the real exchange rate of the
Malaysian Ringgit against the US dollar. Theoretically, exchange rate fluctuation has two
opposite impacts on investment. The first is a positive effect when the exchange rate
depreciates. The marginal profit of investment increases because of higher revenue from
domestic and foreign sales. However, this positive effect is counterbalanced by the rising
variable cost and the higher price for imported capital (see Harchaoui et al., 2005). Moreover,
we include REX because of its important role in natural resource-based economies (Sachs,
1999). (FD*OD) is the interaction term between oil dependence and financial development.
This interaction term is expected to reveal the impact of oil dependence on the finance-
investment nexus. At the margin, the total effect of increasing OD can be calculated by
examining the partial derivatives of investment with respect to financial development.
t
t
t ODFD
INV31
(3)
Eq. (3) indicates how the marginal effect of the financial development on investment quantity
changes with the level of oil dependence.
If the coefficient on the interaction term was negative, this implies that a small increase in oil
dependence would then result in a weaker finance-investment nexus. This would certainly be
17
the case if δ1 is positive. If, conversely, the coefficient on the interaction term is positive, a
small increase in oil dependence would result in a stronger finance-investment nexus.
Finally, an efficient financial system performs the task of screening investment projects. With
an effective risk evaluation of different investment opportunities, the selection of the most
promising investment projects could improve the quality of investment (Ang, 2009). To
assess the impact of oil dependence on the role of financial development on the
quality/efficiency of investments, we propose the following model from Beck et al. (2000):
Model (3): tttttt EDUODFDODFDTFP 43210 )*( (4)
where TFP is the natural logarithm of total factor productivity representing investment
efficiency. TFP is calculated as described in Section 4.1. EDU is the secondary school
enrolment (% gross). EDU is a proxy for human capital, which is considered to be a main
determinant for TFP (Benhabib and Spiegel, 1994; Anwar and Sun, 2011). The interaction
term between financial development and oil dependence captures the effect of oil dependence
on the role of financial development in investment efficiency.
t
t
t ODFD
TFP31
(5)
In equation (4), the coefficient φ3 represents the effect of oil dependence in the relation
between financial development and investment efficiency. If φ3 < 0, oil dependence has a
negative effect on the role of financial development in boosting investment efficiency; this
implies a weaker finance-growth nexus. However, if φ3 > 0, oil dependence has a positive
effect on the role of financial development in boosting investment; this implies a stronger
finance-growth nexus. Eq. (5) indicates how the marginal effect of the financial development
on TFP changes with the level of oil dependence. The sign on φ3 shows whether oil
dependence has a positive or a negative effect on the role of financial development on TFP.
18
4.3 Methodology
To test the long-run relationships among the variables, we adopt the auto-regressive
distributed lag (ARDL) bounds testing approach to cointegration suggested by Pesaran et al.
(2001). Most recent studies indicate that an ARDL model is preferable for estimating the
cointegration relation because it is applicable irrespective of whether the underlying
regressors are I(0) or I(1), and it performs well for a small sample size. In addition, one of the
important statistical advantages of the ARDL approach for the cointegration test is that, as
Pesaran and Shin (1998) note, appropriate modification of the orders of the ARDL model is
adequate to correct the problem of endogeneity bias (see also Ang, 2009).
The ARDL version of the estimation models can be specified as:
0 1 2 1 3 1 4 5 6
1 01 1
7 8
0 0
t i
t i
pm
t t
i it t t t i
q r
t
i i t i
Y Y K YFD OD FD
L L L L
KOD
L
1 1 1 110 1 2 3 4 5 6
1
7 8 9 10
0 0 0 0
( * )
( * )
t t t tt
t i t it i
m
t t i
i
p q r s
tt i
i i i i
INV INV FD OD FD OD REX INV
FD OD FD OD REX
1 1 1 110 1 2 3 4 5 6
1
7 8 9 10
0 0 0 0
( * )
( * )
t t t tt
t i t it i
m
t t i
i
p q r s
tt i
i i i i
TFP TFP FD OD FD OD EDU TFP
FD OD FD OD EDU
where m,p,q,r and s are the optimal lag lengths for each variable.
The coefficients (α1, α2, α3, α4), ( 54321 ,,,, ) and (δ1, δ2, δ3, δ4, δ5) of the first portion of
the models measure the long-run relations, whereas the coefficients (α5, α6, α7, α8)
( 109876 ,,,, ) and (δ6, δ7, δ8, δ9, δ10) represent the short-run dynamics. The F-statistic is
19
used to test the existence of a long-run relation among the variables. We test the null
hypotheses, H0: α1 = α2 = α3 = α4 = 0; 054321:0 H ; H0: δ1= δ2= δ3= δ4= δ5
= 0 that there is no cointegration among the variables. The F-statistics are then compared
with the critical values provided by Narayan (2005), which are appropriate for small samples.
If the computed F-statistic is greater than the upper bound critical value, we reject the null
hypothesis of no cointegration and conclude that a steady-state equilibrium exists among the
variables. If the computed F-statistic is less than the lower bound critical value, the null
hypothesis of no cointegration cannot be rejected. However, if the computed F-statistic lies
between the critical values for the lower and upper bounds, the result is inconclusive. We
then adopt the modified version of the Granger causality test developed by Toda and
Yamamoto (1995), and Dolado and Lutkepohl (1996), hereafter (TYDL).
5. Empirical Findings and Discussion
Although the ARDL model does not require pre-testing variables, we need to ensure that that
none of the variables are integrated in order 2 or beyond. Because of its reliability for a small
sample size, the Dicky-Fuller GLS stationary test is employed to examine the time series
properties for each variable and determine its order of integration. Table 1 reveals that all the
variables are integrated in order one with the exception of OD and INV, which are I(0).
Hence, it is confirmed that the ARDL approach can be applied to analyze the long-run
relationship. To model the structural breaks, we implement the Narayan and Popp (2010) M1
and M2 two-break unit root test and the result is presented in Table 2. Both Narayan and
Popp (2010) M1 and M2 unit root tests suggest the same break dates. We incorporate the
structural breaks in the cointegration test and model estimation.
[Insert Table 1]
20
[Insert Table 2]
After investigating the time series properties for all variables, the ARDL approach is used to
examine the potential long-run equilibrium relation. This test is sensitive to the number of
lags used. Given the limited number of observations in this study, lags with a maximum of
two years have been imposed on the first difference of each variable, and the Schwarz-
Bayesian Criterion (SBC) is used to select the optimal lag length for each variable. The
results of the ARDL bound test of cointegration are tabulated in Table 3. In this table, we
divide the results into two parts. In the first part, we do the ARDL test without structural
break. In the second part, we repeat the test after adding dummies for the break points
suggested by the Narayan and Popp (2010) unit root test. In general, we do not find any
significant differences for both sets of ARDL results.
[Insert Table 3]
Table 3 shows that the calculated F statistic is higher than the upper bound critical value at 5%
in our third model. This result provides strong evidence for the existence of a long-run
relation among the variables in these two models. However, we fail to prove the existence of
a long-run relation through the F-statistic in our first and second models. According to
Kremers et al. (1992), the significant coefficient of the lagged error-correction term (ECT) is
an alternative, and more efficient, means of establishing cointegration. Our results show that
the coefficient on the lagged ECT is negative and significant in all models. This finding
confirms the existence of a long-run relation in all models.
21
Because there is cointegration among the variables, we can derive the long-run coefficient as
the estimated coefficient of the one lagged level independent variable divided by the
estimated coefficient of the one lagged level dependent variable and multiply it with a
negative sign. Conversely, the short-run coefficients are calculated as the sum of the lagged
coefficients of the first differenced variables.
[Insert Table 4]
Table 4 Panel A reveals that the level of financial development has a direct negative impact
on economic growth in the long-run. The conclusion from this result is that an increase in
credit to the private sector does not per se contribute to growth. Most important is how this
credit is used and how effectively it is allocated to finance growth-enhancing projects. Ang
and Mckibbin (2007) argued that the financial sector in Malaysia has not been playing a vital
role in allocating resources. Most of the credit provided to the private sector was issued for
the purchase of shares and real estate property, rather than for investment in productive
activities. This led to bubbles in the property sector and triggered many speculative activities
in the share market prior to the financial crisis in 1997–98. This risky behavior resulted in the
mismanagement of assets and generated many larger non-performing loans.
Nevertheless, the estimated long-run coefficient shows that oil dependence is not significant.
The absence of a significant negative impact of oil dependence on economic growth is not
surprising and supports the previous finding (Gylfason, 2001; Rosser, 2006) that Malaysia
has escaped the resource curse. Conversely, we find that capital has played a crucial role in
shaping Malaysia’s economic growth.
22
Because we could not find direct evidence of an oil curse from Model 1, we attempt to seek
an indirect symptom of the oil curse from Model 2. The empirical results pertaining to Model
2 are presented in columns 2 and 5 of Panel A in Table 4. Financial development and oil
dependence have positive and significant impacts on investment in the long run. However,
this positive impact decreases by increasing the degree of oil dependence, as indicated by the
negative sign on the interaction term between financial development and oil dependence.
Therefore, we cannot assert that Malaysia has fully escaped the oil curse because we have
found a symptom of the oil curse in the financial sector. Sachs (2007) argued that one reason
for the oil curse is that the high dependence on oil adversely affects other economic sectors,
particularly the sectors that can drive sustainable economic growth.
The result of Model 3 in columns 3 and 6 corroborates our previous argument that financial
development does not affect TFP in the long run. Moreover, the coefficients of oil
dependence and the interaction term between financial development and oil dependence are
not significant. This result suggests that the financial sector in Malaysia may not be
sufficiently developed to foster the efficiency of investment. These results expose the illusory
role of oil revenue in Malaysia’s economy. Although oil revenue does not have a significant
impact on economic growth in the long run, it would negatively affect the mechanisms of
certain growth determinants in the long-run, such as the finance-investment nexus found in
our long-run analysis
Table 4 Panel B shows the short-run results. The results from Model 1 reveal that oil
dependence has a positive, but insignificant, impact on economic growth in the short-run.
Similar to our long-run results, financial development has a negative effect and capital stock
has a positive impact on economic growth in the short run. The illusory role of oil
23
dependence is also reflected in the absence of a significant negative effect of the interaction
term on investment in columns 2 and 5. These results for the short-run can be attributed to the
fact that the oil curse symptoms most likely appear in the long-run as per the findings of
Torvik (2001), and Collier and Godris (2008).
Finally, the coefficients for the estimated lagged error correction term are negative and
significant in all models. This confirms the existence of a long-run relation among the
variables. In addition, the coefficient suggests that a deviation from the long-run equilibrium
following a shock is corrected by approximately 52%, 42% and 8% per year, respectively.
Panel C in the same table notes that all models pass all the diagnostic tests for serial
correlation, autoregressive conditional heteroskedasticity and model specification. 8 The
CUSUMSQ in Figures 5 to 10 remain within the critical boundaries for the 5% significance
level. These statistics confirm that the long-run coefficients and all short-run coefficients in
the error correction model are stable.
[Insert Figures 5-10]
Theoretically, financial intermediation affects economic growth mainly by mobilizing
savings and allocating funds to productive investment projects that will generate strong
returns. Because financial development does not have a positive effect on economic growth,
as reflected in Model 1, one may conclude that, although the financial sector channels funds
to investment, the funds are not efficiently placed into productive investments that will
expand the economy. The reason for this distortion may be the high dependence on oil rent in
Malaysia. The high dependence on oil rent increases the dominant role of government in total
8We did not find any evidence of a multicollinearity problem among our independent variables, except for
the interaction term between financial development and oil dependence. However, multicollinearity
associated with the interaction can be considered as one of the cases where we can safely ignore
multicollinearity. This is because the effect of multicollinearity is mainly on the coefficients of the
variables, and there is no effect on the p-value or the relationships in the models (see, Allison, 2012).
24
investment at the expense of private investment (Nili and Rastad, 2007). Because private
investment is more efficient than public investment, the high dependence on oil rent replaces
more efficient investment with less efficient investment, and thus, growth suffers (Banerjee,
2011). Therefore, we argue that the oil curse in Malaysia lies in crippling the mechanism of
finance-growth, i.e., investment, as shown in Model 2. The oil curse is transmitted through
the quantitative channel rather than the qualitative channel.
The ability of financial intermediaries to accumulate capital may be impaired because of the
uncertainty that stems from the volatile nature of oil prices. Subsequently, this uncertainty
weakens the incentive of banks to invest and encourages banks to pursue lower risk activities,
such as consumption or housing loans.9 The impact on the level of capital, rather than the
productivity of capital, may be the reason that oil dependence biases the finance-growth
nexus in Malaysia. These findings are in accord with the fact that, until the early 1970s, the
bulk of financial sector credits were unproductive loans that were channeled into inefficient
activities rather than into high quality investments (refer to Ghani and Suri, 1999; Hill, 2005).
Finally, the TYDL causality tests in Table 5 reveal that there is unidirectional causality
running from capital stock to economic growth. We also find that the interaction term
between financial development and oil dependence causes investment. This result supports
our argument that the interactive relationship between financial development and oil
dependence affects the level of investment in Malaysia.
(Insert Table 5)
5.1 Robustness Check
As a robustness check, we re-estimate our models using several alternative proxies for
financial development and also a new oil dependence proxy. In addition, we conduct these
9Thiel (2001) argued that, in some countries, significant numbers of bank loans are provided to finance housing
loans, instead of being channelled to finance productive investments.
25
estimations utilizing two other econometric approaches, which are FMOLS and DOLS. In
general, the outcomes of our robustness check support our previous findings with the ARDL
approach.
Our alternative financial development proxies are LIQ, which equals liquid liabilities divided
by GDP. Many researchers have used this variable as a measure of financial depth. LIQ does
not represent the effectiveness of the financial system. However, by assuming that the size of
the financial intermediary system is positively correlated with the financial system’s activities,
in our regressions, this can be used as a measure of financial development. The second proxy
used is deposits to GDP (DPS), which equals the demand, time and saving deposits in deposit
banks and other financial institutions as a share of GDP. The size of deposits is an indicator
for the potential investment quantum. The third proxy is the PCA, which is a new proxy
constructed by the previous three based on the Principal Components Analysis approach.
This proxy is able to capture most of the information from the original dataset, which consists
of the previous three financial development measures.
Additionally, we use an alternative proxy for oil dependence. This proxy is a dummy variable
for the number of years where the oil rent is greater than 10% of GDP (see Beck, 2011). The
results of our robustness check, which are tabulated in Table 6, are in line and in agreement
with our main models that confirm our stated argument and conclusion.
[Insert Table 6]
6. Conclusion and policy implications
This paper empirically examines the oil curse in Malaysia and the role of investment as a
mechanism of the oil curse. We do not find any significant evidence of the direct effect of
financial development on economic growth and TFP. However, there is a direct and positive
26
effect of financial development and oil dependence, respectively, on the level of investment.
Furthermore, a significant negative interaction term between financial development and oil
dependence exposes the oil curse in Malaysia. This finding supports Doraisami (2005, p. 107),
who stated, “Malaysia has certainly not been untouched by the [oil] curse”.
Nevertheless, given the absence of a significant impact of financial development on TFP, we
argue that the financial sector in Malaysia channels funds inefficiently into non-productive
investment. The reason for this distortion of financial development may be the high
dependence on oil rent in Malaysia. Oil dependence weakens and cripples the mechanism of
one of the most important economic growth motors, which is financial development. We
argue that the oil curse in Malaysia exists and that the curse is transmitted through the
quantitative channel rather than the qualitative channel.
Our findings contain a mix of policy implications for Malaysia. On the one hand, the
Malaysian government should be aware of the indirect risk of oil dependence on the financial
role in fostering investment activities. It is advisable to maintain the degree of oil dependence
at a low level, enhance economic diversification activities and increase the contribution of
other sectors to GDP. Additionally, the financial sector should be more involved in
productive investment activities to enhance its role in fostering economic growth. In this
regard, policymakers should pursue actions that enhance the efficiency of bank
intermediation. Introducing improvements in information on prospective borrowers,
including the establishment of credit bureaus and enhancing the legal protection of creditor
rights, are all potential areas in which quality gains can be achieved.
27
Furthermore, one of the main oil curse channels in oil-dependent economies is
mismanagement and neglect of human development. The easy access of oil rent may relieve
the government from developing its human capital, which may negatively affect the
performance of various economic sectors including the financial sector. Therefore, we
suggest that policymakers need to work to enhance human development, which has an
important positive role in investment efficiency.
28
References
Allegret, J. P., Couharde, C., Coulibaly, D. and Mignon, V. 2014. Current accounts and oil
price fluctuations in oil-exporting countries: the role of financial development. Journal of
International Money and Finance, 47, 185-201.
Allison, P.D. 2012. When Can You Safely Ignore Multicollinearity [Web log post]. Retrieved
from http://statisticalhorizons.com/blog .
Ang, J. B. 2009. Financial development and economic growth in Malaysia. London; New
York: Routledge.
Ang, J. B and McKibbin,W.J 2007. Financial liberalization, financial sector development and
growth: Evidence from Malaysia. Journal of Development Economics 84, 215-233.
Anwar, S., and Sun, S. 2011. Financial development, foreign investment and economic
growth in Malaysia. Journal of Asian Economics, 22(4), 335-342.
Apergis, N., El-Montasser, G., Sekyere, E., Ajmi, A. N., and Gupta, R. 2014. Dutch disease
effect of oil rents on agriculture value added in Middle East and North African (MENA)
countries, Energy Economics, 45 485-490.
Apergis, N., and Payne, J. E. 2014. The oil curse, institutional quality, and growth in MENA
countries: Evidence from time-varying cointegration. Energy Economics, 46, 1-9.
Arcand, J. L., Berkes, E.,Panizza, U., 2012. Too much finance? IMF Working Paper 12/161.
Arouri, M.E.H. and Rault, C., 2011. Causal relationships between oil and stock prices: Some
new evidence from gulf oil-exporting countries. Economie Internationale, 2, 41-56.
Atkinson, G., and Hamilton, K. 2003. Savings, growth and the resource curse
hypothesis. World Development, 31(11), 1793-1807.
Auty, R M., 1993. Sustaining development in mineral economies: the resource curse thesis.
Routledge, London
Baltagi, B. H., Demetriades, P. O., and Law, S. H. 2009. Financial development and openness:
Evidence from panel data. Journal of Development Economics, 89(2), 285-296.
Banerjee, P. 2011. Microeconomic Policy Environment–An Analytical Guide for Managers.
Tata McGraw-Hill.
Barajas, A., Chami, R and Yousefi, R.S 2013. The Finance and Growth Nexus Re-Examined:
Do All Countries Benefit Equally? IMF Working Paper 13/130.
Beck, T 2011. Finance and Oil: Is there a resource curse in financial development? European
Banking Center Discussion Paper, (2011-004).
Beck, T., Feyen , E., Ize, A and Moizeszowicz, F 2008. Benchmarking Financial
Development. Policy Research Working Paper No 4638, The World Bank.
Beck, T., Degryse, H., and Kneer, C. 2014. Is more finance better? Disentangling
intermediation and size effects of financial systems. Journal of Financial Stability, 10,
50-64.
29
Beck, T., and Levine, R. 2004. Stock markets, banks, and growth: Panel evidence. Journal of
Banking and Finance, 28(3), 423-442.
Beck, T., Levine, R., and Loayza, N. 2000. Finance and the Sources of Growth. Journal of
Financial Economics, 58(1), 261-300.
Benhabib, J., and Spiegel, M. M. 1994. The role of human capital in economic development
evidence from aggregate cross-country data. Journal of Monetary economics, 34 (2),
143-173.
Bertocco, G., 2008. Finance and development: is Schumpeter’s analysis stillrelevant? Journal
of Banking and Finance 32 (6), 1161–1175.
Betz, M. R., Partridge, M. D., Farren, M., and Lobao, L. 2015. Coal Mining, Economic
Development, and the Natural Resources Curse. Energy Economics, 50, 105-116
Bhanumurthy, K. V. 2002. Arguing a case for Cobb-Douglas production function. Review of
Commerce Studies, 20, 21.
Bhattacharyya, S. and Hodler, R. 2010. Natural resources, democracy and corruption.
European Economic Review, 54(4), 608-621.
Bhattacharyya, S and Hodler, R 2014. Do natural resource revenues hinder financial
development? The role of political institutions. World Development 57, 101-113.
Bittencourt, M., 2012. Financial development and economic growth in Latin America: Is
Schumpeter right?. Journal of Policy Modeling 34, 341-355.
Blanco, L., and Grier, R. 2012. Natural resource dependence and the accumulation of
physical and human capital in Latin America. Resources Policy, 37(3), 281-295.
Bosworth, B., and Collins, S. M. 2003. The empirics of growth: An update. Brookings Papers
on Economic Activity, 34(2), 113–206.
Cecchetti, S. G., and Kharroubi, E. 2012. Reassessing the impact of finance on growth (No.
381). Bank for International Settlements.
Cockx, L., and Francken, N. 2016. Natural resources: A curse on education spending? Energy
Policy, 92, 394-408.
Collier, P., and Goderis, B. 2008. Commodity prices, growth, and the natural resource curse:
reconciling a conundrum. Growth, and the Natural Resource Curse: Reconciling a
Conundrum (June 5, 2008).
Collier, P. and Hoeffler, A., 2009. Testing the neocon agenda: democracy in resource-rich
societies. European Economic Review, 53(3), 293-308.
Daniele, V. 2011. Natural resources and the ‘quality’ of economic development. The Journal
of Development Studies, 47(4), 545-573.
Dasgupta, D., Keller, J., and Srinivasan, T. G. 2002. Reform and Elusive Growth in the
Middle-East: What has Happened in the 1990s? Washington, DC: World Bank.
30
Demetriades, P., Hussein, K., 1996. Does financial development cause economic growth?
Time series evidence from 16 countries. Journal of Development Economics 51(2), 387–
411.
Demirer, R., Jategaonkar, S.P. and Khalifa, A.A., 2015. Oil price risk exposure and the cross-
section of stock returns: The case of net exporting countries. Energy Economics, 49,
132-140.
Dietz, S., Neumayer, E., and De Soysa, I. 2007. Corruption, the resource curse and genuine
saving. Environment and Development Economics, 12(01), 33-53.
Dolado, J. J., and Lütkepohl, H. 1996. Making Wald tests work for cointegrated VAR
systems. Econometric Reviews, 15(4), 369-386.
Doraisami, A. 2015. Has Malaysia really escaped the resources curse? A closer look at the
political economy of oil revenue management and expenditures. Resources Policy, 45,
98-108.
Ductor, L., and Grechyna, D. 2015. Financial development, real sector, and economic
growth. International Review of Economics and Finance, 37, 393-405.
Energy Information Administration (EIA). 2014. Malaysia Report, Available at:
http://www.eia.gov/beta/international/analysis_includes/countries_long/Malaysia/malays
ia.pdf
Ergungor, O. E., 2008. Financial system structure and economic growth: structure matters.
International Review of Economics and Finance 17(2), 292–305.
Farhadi, M., Islam, M. R., and Moslehi, S. 2015. Economic Freedom and Productivity
Growth in Resource-rich Economies. World Development, 72, 109-126.
Frankel, J.A. 2010. The natural resource curse: A survey. HKS Faculty Research Working
Paper, RWP10-005, John F. Kennedy School of Government, Harvard University.
Filis, G., Degiannakis, S., and Floros, C. 2011. Dynamic correlation between stock market
and oil prices: The case of oil-importing and oil-exporting countries. International
Review of Financial Analysis, 20(3), 152-164.
Fry, M.J 1978. Money and capital or financial deepening in economic development. Journal
of Money, Credit and Banking 10, 464–475.
Ghani, E., and Suri, V. 1999. Productivity growth, capital accumulation, and the banking
sector: Some lessons from Malaysia. World Bank Policy Research Working Paper,
(2252).
Gelb, A (ed.). 1988. Oil Windfalls: Blessing or Curse? Oxford and New York: Oxford
University Press.
Gylfason, T. 2001. Natural resources, education, and economic development. European
Economic Review, 45(4), 847-859.
Gylfason, T and Zoega, G., 2006. Natural resources and economic growth: The role of
investment. World Economy 29, 1091-1115.
31
Hassan, M. K., Sanchez, B., and Yu, J. S. 2011. Financial development and economic growth:
New evidence from panel data. The Quarterly Review of Economics and Finance, 51(1),
88-104.
Harchaoui, T. M., Tarkhani, F., and Yuen, T. 2005. The effects of the exchange rate on
investment: Evidence from Canadian manufacturing industries. Bank of Canada.
Working paper 2005-22.
Hasan, I., Wachtel, P., Zhou, M., 2009. Institutional development, financial deepening and
economic growth: evidence from China. Journal of Banking and Finance 33 (1), 157–170.
Hill, H. 2005. Malaysia Economic Development in Comparative Southeast Asian Perspective.
In Conference Paper for ‘Malaysia and Sino-Malaysian relations’, Institute of Malaysian
Studies, Centre for Southeast Asian Studies, Xiamen University, China.
Inglesi-Lotz, R. 2015. The impact of renewable energy consumption to economic growth: a
panel data application. Energy Economics.
Jalil, A., Feridun, M., Ma, Y., 2010. Finance-growth nexus in China revisited: new evidence
from principal components and ARDL bounds tests. International Review of Economics
and Finance 19 (2), 189–195.
Jones, C. and Kaul, G. 1996. Oil and the stock market. Journal of Finance, 51, 463-491.
Kang, W., Ratti, R.A. and Yoon, K.H., 2014. The impact of oil price shocks on US bond
market returns. Energy Economics, 44, 248-258.
Kim, D. H., and Lin, S. C. 2015. Natural resources and economic development: New panel
evidence. Environmental and Resource Economics, 1-29.
Kim, D. H., Lin, S. C., and Suen, Y. B., 2010. Dynamic effects of trade openness on financial
development. Economic Modelling, 27(1), 254-261
King, R.G and Levine, R 1993a. Finance and growth: Schumpeter might be right. Quarterly
Journal of Economics 108, 717–738.
King, R. G., and Levine, R. 1993b. Finance, entrepreneurship and growth. Journal of
Monetary Economics, 32(3), 513-542.
Kremers, J.J.M., Ericsson, N.R., and Dolado, J.J. 1992. The power of cointegration tests.
Oxford Bulletin of Economics and Statistics, 54, 325-348,
Kurronen, S., 2015. Financial sector in resource-dependent economies. Emerging Markets
Review, 23, 208-229.
Law, S. H., and Singh, N. 2014. Does too much finance harm economic growth? Journal of
Banking and Finance, 41, 36-44.
Law, S. H., Azman-Saini, W. N. W., and Ibrahim, M. H. 2013. Institutional quality
thresholds and the finance–growth nexus. Journal of Banking and Finance, 37(12),
5373-5381.
32
Levine, O., and Warusawitharana, M. 2014. Finance and Productivity Growth: Firm-level
Evidence. Finance and Economics Discussion Series. Federal Reserve Board,
Washington, D.C.14-17
Levine, R 1997. Financial development and economic growth: views and agenda. Journal of
Economic Literature 35, 688-726.
Levine, R. 2003. More on finance and growth: more finance, more growth?.Review-Federal
Reserve Bank of Saint Louis, 85(4), 31-46.
Liang, Q and Teng, J. Z 2006. Financial development and economic growth: Evidence from
China. China Economic Review 17, 395-411.
Libman, A. 2013. Natural resources and sub-national economic performance: Does sub-
national democracy matter?. Energy economics, 37, 82-99.
Mavrotas, G., Murshed, S. M and Torres, S. 2011. Natural resource dependence and
economic performance in the 1970–2000 period. Review of Development Economics 15,
124-138.
McKinnon, R. I 1973. Money and Capital in Economic Development. (Washington, DC:
Brookings Institution).
Mehlum, H., Moene, K., and Torvik, R. 2006. Institutions and the Resource Curse. The
Economic Journal, 116(508), 1-20.
Narayan, P.K 2005. The saving and investment nexus for China: Evidence from cointegration
tests, Applied Economics 17, 1979-1990.
Narayan, K., P., and Narayan, S. 2010 Modeling the impact of oil prices on Vietnam’s stock
prices. Applied Energy, 87, 356-361.
Narayan, P.K. and Popp, S., 2010. A new unit root test with two structural breaks in level and
slope at unknown time. Journal of Applied Statistics, 37(9), 1425-1438.
Narayan, P.K. and Sharma, S.S., 2011. New evidence on oil price and firm returns. Journal of
Banking and Finance, 35(12), 3253-3262.
Nili, M. and Rastad, M. 2007. Addressing the Growth Failure of the Oil Economies: The
Role of Financial Development. The Quarterly Review of Economics and Finance 46,
726–40.
Pagano, M. 1993. Financial markets and growth: an overview. European Economic Review,
37(2), 613-622.
Papyrakis, E., and Gerlagh, R. 2004. The resource curse hypothesis and its transmission
channels. Journal of Comparative Economics, 32(1), 181-193.
Papyrakis, E., and Gerlagh, R. 2007. Resource abundance and economic growth in the United
States. European Economic Review, 51(4), 1011-1039.
Patrick, H.T. 1966. Financial Development and Economic Growth in Underdeveloped
Countries, Economic Development and Cultural Change, 14,174-189
33
Pesaran, M.H., Shin, Y and Smith, R 2001. Bounds testing approaches to the analysis of level
relationships. Journal of Applied Econometrics 16, 289-326.
Phan, D.H.B., Sharma, S.S. and Narayan, P.K., 2015. Oil price and stock returns of
consumers and producers of crude oil. Journal of International Financial Markets,
Institutions and Money, 34, 245-262.
Rahaman, M.M., 2011. Access to financing and firm growth. Journal of Banking and Finance
35 (3), 709–723.
Rajan, R. G., and Zingales, L. 2003. The great reversals: the politics of financial development
in the twentieth century. Journal of Financial Economics, 69(1), 5-50.
Rosser A 2006 The political economy of the resource curse: a literature survey, IDS Working
Paper 268 (Institute of Development Studies, Brighton)
Rousseau, P. L., and Yilmazkuday, H.,2009. Inflation, financial development, and growth: A
trilateral analysis. Economic Systems, 33(4), 310-324.
Sachs, J. D 2007. How to handle the macroeconomics of oil wealth. Humphreys, M./Sachs,
JD/Stiglitz, JE (Ed.), 173-193.
Sachs, J and Warner, A.M 1995. Natural resources abundance and economic growth.
National bureau for Economic Research (NBER). Working Paper 5398.
Sachs, J. D., and Warner, A. M. 1999. The big push, natural resource booms and
growth. Journal of Development Economics, 59(1), 43-76.
Sadorsky, P. 1999. Oil price shocks and stock market activity. Energy Economics, 21, 449-
469.
Sachs, J. D., and Warner, A. M. 2001, “The Curse of Natural Resources”, European
Economic Review, 45, 827-838.
Salim, R. A., Hassan, K., and Shafiei, S. 2014. Renewable and non-renewable energy
consumption and economic activities: Further evidence from OECD countries. Energy
Economics, 44, 350-360.
Samargandi, N., Fidrmuc, J., and Ghosh, S. 2015. Financial development and economic
growth in an oil-rich economy: The case of Saudi Arabia. Economic Modelling, 43, 267-
278.
Samargandi, N., Fidrmuc, J., and Ghosh, S. 2015. Is the relationship between financial
development and economic growth monotonic? Evidence from a sample of middle-
income countries. World Development, 68, 66-81.
Schoar, A. 2002. Effects of corporate diversification on productivity. The Journal of
Finance, 57(6), 2379-2403.
Schumpeter, J. A 1934. The Theory of Economic Development. Translated by Redvers Opie,
Cambridge MA: Harvard University Press.
34
Shahbaz, M. and Lean, H. H., 2012. Does financial development increase energy
consumption? The role of industrialization and urbanization in Tunisia. Energy Policy,
40, 473-479.
Shao, S., and Yang, L. 2014. Natural resource dependence, human capital accumulation, and
economic growth: A combined explanation for the resource curse and the resource
blessing. Energy Policy 74, 632–642
Shaw, E. S 1973. Financial Deepening in Economic Development. Oxford University Press,
New York.
Shen, C. H., Lee, C. C., 2006. Same financial development yet different economic growth –
why? Journal of Money, Credit and Banking 38(7), 1907–1944.
Shroff, N., Verdi, R. S., and Yu, G. 2013. Information environment and the investment
decisions of multinational corporations. The Accounting Review,89(2), 759-790.
Singh, T 2008. Financial development and economic growth nexus: a time-series evidence
from India, Applied Economics 40, 1615-1627.
Stevens, P., and Dietsche, E. 2008. Resource curse: An analysis of causes, experiences and
possible ways forward. Energy Policy, 36(1), 56-65.
Thiel, M. 2001. Finance and economic growth-a review of theory and the available
evidence (No. 158). Directorate General Economic and Monetary Affairs (DG ECFIN),
European Commission.
Toda, H. Y., and Yamamoto, T. 1995. Statistical inference in vector autoregressions with
possibly integrated processes. Journal of Econometrics, 66(1), 225-250.
Torvik, R. 2001. Learning by doing and the Dutch disease. European economic review, 45(2),
285-306.
Yuxiang, K and Chen, Z 2011. Resource abundance and financial development: Evidence
from China. Resources Policy 36, 72-79.
35
Table 1 Results of Unit Root Test
Variable DF-GLS
Level 1st Difference
Y/L 0.3196 -5.6460***
K/L -2.5275 -3.4565**
FD -0.2860 -2.1820**
OD -0.9592*** -0.2789***
INV -1.8223 -4.2180***
REX -0.7042 -5.6012***
TFP 0.8333 -5.7807***
EDU -0.6735 -5.0945***
Note: *** and* denote significance at the 1% and 10% levels, respectively.
Table 2: Results of Narayan-Popp (2010) two breaks unit root test
M1
M2
Series
t-stat k
t-stat k TB1 TB2
Y/L
-0.1791 0
-3.1890 0 1984 1997
INV
-3.8367 1
-3.0818 1 1997 2000
TFP
-1.6532 0
-3.4280 1 1984 1997
Notes: M1 is Narayan and Popp’s Model 1. M2 is Narayan and Popp’s Model 2. TB is the structural break and K is the lag length.
36
Table 3: Result from ARDL Cointegration Test
Without Structural Break With Structural break
Equation Model (1) Model (2) Model (3) Model (1) Model (2) Model (3)
Optimal lag (1,0,0,1) (2,0,0,0,1) (1,0,0,1,1) (1,0,0,1) (2,0,0,0,1) (1,0,0,1,1)
F-statistic 2.7813 1.7688 5.0708** 2.6255 1.7998 4.9206**
ECt-1 -0.5177*** -0. 4152*** -0.0768*** -0.4600*** -0.3251*** -0.0687***
Critical Values K=3 K=4
1% level 5 % level 10% level 1% level 5 % level 10% level
Lower Bound 4.983 3.535 2.893 4.394 3.178 2.638
Upper Bound 6.423 4.733 3.983 5.914 4.450 3.772
Note: *** and ** denotes the significance at 1% and 5% level respectively. Critical values bounds are from Narayan (2005) with unrestricted intercept and no trend.
37
Table 4: Long Run and Short Run Analysis
Panel A. Long Run Results
Without Structural Breaks With Structural Breaks
Model (1) Model (2) Model (3) Model (1) Model (2) Model (3)
Dependent
variable Y/L INV TFP Y/L INV TFP
C 2.1177*** (9.9486)
-2.4778
(-1.3379) -0.2499
(-0.0792) 2.1328*** (9.8071)
-2.3697 (-1.4067)
-1.5035 (-0.4068)
FD -0.2654***
(-6.7480)
0.8077***
(3.0016) 0.1508
(0.4668)
-2.2693***
(-6.6144)
0.8073***
(3.2987)
0.1455
(0.4149)
OD 0.0004
(0.0294)
1.1369**
(2.1040) 0.3844
(0.5581)
0.0018
(0.1264)
1.2550**
(2.5556)
0.5137
(0.6887)
K/L 0.8271***
(23.3740) - -
0.82711***
(22.9249) - -
FD*OD - -0.3135**
(-2.2217) -0.1371
(-0.7068) -
-0.3400**
(-2.6575)
-0.1829
(-0.8502)
REX - 0.5785*
(1.9773) - -
0.5519**
(2.0995) -
EDU - - 1.6213* (1.9895)
- - 1.9783** (2.0481)
DUM1984 - - - 0.0025
(0.0475) -
0.5145
(1.0904)
DUM1997 - - - 0.0333
(0.5890)
0.3505 (1.3291)
0.1964
(0.5473)
DUM2000 - 0.5484*
(1.7630) -
Panel B. Short Run Results
ΔFD -0.1185***
(-3.1257)
0.1918
(0.9893) -0.0034
(-0.0902)
-0.1147***
(-3.2210)
0.2870
(1.5773)
-0.0036
(-0.0972)
ΔOD 0.0014
(0.1309)
0.0897
(0.2931) -0.0053
(-0.1053)
0.0025
(0.2550)
0.4320
(1.4764)
-0.0043
(-0.0920)
38
ΔK/L 0.9025***
(9.3539) - -
0.8299***
(9.3027) - -
Δ(FD*OD) - -0.0305
(-0.3835) 0.0054
(0.3880) -
-0.1267
(-1.6236)
0.0049
(0.3740)
ΔREX - 0.6799***
(3.2475) - -
0.5566***
(2.8558) -
ΔEDU - - -0.4718***
(-4.6242) - -
-0.4479***
(-3.9403)
DUM1984 - - - 0.0257
(1.6210) -
0.0631***
(4.0656)
DUM1997 - - - 0.0529***
(3.3562)
0.2585***
(4.0819)
0.0272
(1.5481)
DUM2000 - - - - 0.2677***
(3.5135) -
ECt-1
-0.5177***
[-4.2334]
-0.4152***
(-3.4667)
-0.0768***
(-5.7015)
-0.4600***
(-3.9151)
-0.3251***
(-3.3544)
-0.0687***
(-5.6848)
Panel C Diagnostic Tests
Test F-Statistic F-Statistic F-Statistic F-Statistic F-Statistic F-Statistic
LM 1.5320
[0.2303]
1.4838
[0.2420]
0.4305
[0.6542]
1.7852
[0.1836]
0.4716
[0.6286]
0.5331
[0.5926]
ARCH 0.4669
[0.6305]
0.8540
[0.4339]
0.8089
[0.4540]
0.5949
[0.5567]
1.7985
[0.1797]
1.3011
[0.2858]
RESET 0.6430
[0.4279]
1.6115
[0.2145]
0.2531
[0.6184]
1.0618
[0.3101]
1.3202
[0.2812]
0.4622
[0.5020]
Note: ***, **, * denotes the significance at 1%, 5% and 10% levels respectively. t-statistic is in the parenthesis, P-
value is in the brackets.
39
Table 5: Results of TYDL Causality Tests
χ2 Y/L→ FD 0.4277
FD→Y/L 3.7904
K/L→Y/L 8.7203**
INV→FD 1.5086
FD→INV 0.8429
(FD*OD)→INV 5.0655*
FD→TFP 0.1985
TFP→FD 3.5056
40
Table 6: Robustness Check
Model 1 Model 2 Model 3
FMOLS DOLS FMOLS DOLS FMOLS DOLS
FD proxy LIQ DPS PCA LIQ DPS PCA LIQ DPS PCA LIQ DPS PCA LIQ DPS PCA LIQ DPS PCA
Constant 3.282*** 3.134*** 1.807*** 3.412*** 3.229*** 1.642*** -0.528 -0.092 0.753 -0.198 0.193 1.459 1.468*** 1.262** 0.105 0.418 0.421 -0.977
(18.339) (16.437) (5.626) (14.193) (13.018) (4.480) (-0.277) (-0.050) (0.557) (-0.088) (0.091) (0.818) (2.777) (2.176) (0.091) (0.560) (0.516) (-0.795)
FD -0.246*** -0.209*** -0.106*** -0.307*** -0.270** -0.121*** 0.373** 0.311* 0.146** 0.378* 0.314* 0.132* -0.307* -0.283** -0.096* -0.185 -0.175 -0.067
(-3.592) (-3.300) (-5.019) (-3.513) (-3.259) (-4.967) (2.059) (1.936) (2.499) (1.828) (1.733) (1.849) (-2.027) (-2.046) (-1.705) (-0.887) (-0.832) (-1.086)
ODd -0.004 -0.0002 -0.002 0.018 0.027 0.016 3.915*** 3.399*** 0.078 5.794*** 4.869*** 0.114 0.066 0.033 -0.060 1.448 1.113 0.023
(-0.188) (-0.011) (-0.113) (0.582) (0.773) (0.677) (2.746) (2.725) (0.622) (2.926) (2.881) (0.571) (0.098) (0.055) (-0.960) (1.411) (1.144) (0.307)
K/L 0.708*** 0.703*** 0.751*** 0.723*** 0.721*** 0.768*** - - - - - - - - - - - -
(19.607) (18.661) (22.201) (17.536) (16.510) (20.374)
FD*ODd - - - - - - -0.882*** -0.787*** -0.277** -1.319*** -1.143** -0.457*** -0.028 -0.020 -0.017 -0.327 -0.258 -0.121*
(-2.852) (-2.852) (-2.601) (-3.127) (-3.130) (-2.935) (-0.193) (-0.150) (-0.312) (-1.472) (-1.202) (-1.731)
REX - - - - - - 0.455* 0.431 0.519* 0.405 0.393 0.398 - - - - - -
(1.734) (1.634) (1.912) (1.292) (1.267) (1.097)
EDU - - - - - - - - - - - - 1.529*** 1.544*** 1.538*** 1.647*** 1.629*** 1.798***
(6.169) (6.027) (5.164) (5.762) (5.119) (5.717)
Note: ***, **, * denotes the significance at 1%, 5% and 10% levels respectively. t-statistic is in the parenthesis.
41
Figures
Investment
Fig 1. Oil Dependence Transmission Mechanism into Finance-Growth Nexus
0
4
8
12
16
20
1970 1975 1980 1985 1990 1995 2000 2005 2010
Fig.2. Oil and Gas Rent to GDP (%) Source: WDI
Financial
Development
Economic
Growth
Quantitative
Qualitative
Financial
Development
Quantitative
Qualitative
Financial
Development
Quantitative
Qualitative
Oil Dependence
Economic
Growth Financial
Development
Quantitative
Qualitative
42
-8
-4
0
4
8
12
1970 1975 1980 1985 1990 1995 2000 2005 2010
Fig.3. Annual Growth Rate of GDP (%) Source: WDI
0
20
40
60
80
100
120
140
160
1970 1975 1980 1985 1990 1995 2000 2005 2010
Fig.4. Domestic Credit to Private Sector to GDP (%)
Sources: WDI
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1980 1985 1990 1995 2000 2005 2010
CUSUM of Squares 5% Significance
Fig. 5. Plot of cumulative sum of squares of recursive
residual (without structural break) (Model 1)
Fig. 6. Plot of cumulative sum of squares of recursive
residuals (with structural break) (Model 1)
-0.4
0.0
0.4
0.8
1.2
1.6
99 00 01 02 03 04 05 06 07 08 09 10 11 12 13
CUSUM of Squares 5% Significance
43
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1980 1985 1990 1995 2000 2005 2010
CUSUM of Squares 5% Significance
-0.4
0.0
0.4
0.8
1.2
1.6
01 02 03 04 05 06 07 08 09 10 11 12 13
CUSUM of Squares 5% Significance
Fig. 7. Plot of cumulative sum of squares of recursive
residual (without structural break) (Model 2)
Fig. 8. Plot of cumulative sum of squares of recursive
residuals (with structural break) (Model 2)
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12
CUSUM of Squares 5% Significance
-0.4
0.0
0.4
0.8
1.2
1.6
98 99 00 01 02 03 04 05 06 07 08 09 10 11 12
CUSUM of Squares 5% Significance
Fig. 9. Plot of cumulative sum of squares of recursive
residual (without structural break) (Model 3).
Fig. 10. Plot of cumulative sum of squares of recursive
residuals (with structural break) (Model 3).