Energy, environment and growth nexus in South Asia Data from Pakistan, India, Bangladesh, Nepal, Sri Lanka Muhammed Zeshan, Vaqar Ahmed

1. Energy, environment and growth nexus in
South Asia
Muhammad Zeshan & Vaqar Ahmed
Environment, Development and
Sustainability
A Multidisciplinary Approach to the
Theory and Practice of Sustainable
Development
ISSN 1387-585X
Environ Dev Sustain
DOI 10.1007/s10668-013-9459-8
1 23

2. Your article is protected by copyright and all
rights are held exclusively by Springer Science
+Business Media Dordrecht. This e-offprint
is for personal use only and shall not be selfarchived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.
1 23

3. Author's personal copy
Environ Dev Sustain
DOI 10.1007/s10668-013-9459-8
REVIEW
Energy, environment and growth nexus in South Asia
Muhammad Zeshan • Vaqar Ahmed
Received: 27 January 2013 / Accepted: 17 April 2013
Ó Springer Science+Business Media Dordrecht 2013
Abstract The present study investigates the energy, environment and growth nexus for a
panel of South Asian countries including Bangladesh, India, Pakistan, Sri Lanka and
Nepal. The simultaneous analysis of real GDP, energy consumption and CO2 emissions is
conducted for the period 1980–2010. Levin panel unit root test and Im test panel unit root
both indicate that all the variables are I (1). In addition, Kao’s panel Cointegration test
specifies a stable long-term relationship between all these variables. Empirical findings
show that a 1 % increase in energy consumption increases output by 0.81 % in long run
whereas for the same increase in CO2 emission output falls by 0.17 % in long run. Panel
Granger causality tests report short-run causality running from energy consumption to CO2
emissions and from CO2 emissions to GDP.
Keywords
Energy Á Environment Á Economic Growth Á South Asia
1 Introduction
Rising economic growth in South Asia is escalating the energy demand, and more energy
inputs are required to cater this demand. This region witnesses a positive growth trend over
the last three decades, from 1981 to 2010. It is interesting to note that most of the countries
are following the same growth pattern indicating a strong impact of regional policies on
growth. During the period of analysis, the highest average growth rate was observed in
India that was 6.2 %, whereas Pakistan observed the second highest average growth rate
that was 5 %. On the other hand; Sri Lanka, Bangladesh and Nepal witnessed 4.9, 4.8 and
4.6 % growth rates, respectively (please see Fig. 1 for details).
There are serious concerns about rising demand for energy inputs and the volume of
greenhouse gas (GHG) emissions (Zeshan 2013, Shahbaz and Dube 2012; Shahbaz et al. 2012).
M. Zeshan (&) Á V. Ahmed
Sustainable Development Policy Institute, Islamabad, Pakistan
e-mail: zeshan@sdpi.org; muh.zeshan@gmail.com
V. Ahmed
e-mail: vaqar@sdpi.org
123

4. Author's personal copy
M. Zeshan, V. Ahmed
Correspondingly, the countries with higher usage of energy consumption are adding more
CO2 emissions in the environment. The rising energy demand has created energy crisis and
environmental degradation. On the one hand, energy resources are depleting quickly,
whereas on the other hand, it is causing environmental degradation. Around the globe,
these problems have forced the governments to closely monitor and supervise energy
markets (ECSSR 2004).
At this stage, universal environment friendly energy policies are essential because the
rising CO2 emissions might bleak the future prospects of the sustainable development.
South Asia requires such energy efficient measures that could ensure the minimum CO2
emissions. Ozturk (2010) argues that higher energy consumption increases CO2 emissions
in the environment, however, the use of efficient production technology might reduce these
emissions over time (Shahbaz et al. 2010; Chang 2010). Almost all the South Asian
countries follow the same pattern for energy consumption but with the exception of
Bangladesh. Nonetheless, sometimes it tends to follow the same pattern but with much
variation. Throughout the period of analysis, average growth rate for the energy consumption in Bangladesh was the highest, 4.5 %. India and Pakistan had the second highest,
4.3 %, whereas Nepal and Sri Lanka observed 2.8 and 2.6 %, respectively (please see
Fig. 2 for details).
The analysis of a causal relationship between the energy consumption, CO2 emissions
and economic growth provides important findings to policy makers. In empirical literature,
the long-term and short-term causal relationships have much importance for energy
assessment policies. The direction of causality suggests the relevant energy policies that
Fig. 1 Economic growth in South Asia 1981–2010
Fig. 2 Energy consumption in South Asia 1981–2010
123

5. Author's personal copy
Energy, environment and growth
might be helpful for sustained economic growth (Payne 2010; Ozturk 2010, Squalli 2007).
However, most of the empirical studies have focused on the developed countries (Hossain
2011; Apergis and Payne 2009a), and a scanty literature is available for developing
countries (Huang et al. 2008). The prevailing gaps in literature focusing on regional
analysis especially for South Asian have motivated us in examining the causal linkages
among energy consumption, CO2 emissions and GDP in South Asia.
The empirical literature follows a country by country analysis which is not robust
because of less number of observations, on the other hand, the panel data provide a robust
analysis (Arouri et al. 2012; Wang et al. 2011). Hence, the present study plans to investigate the cointegrating relationship and the causal links among the variables in a panel
framework. The causality tests between economic growth, energy consumption and CO2
emissions would be performed in three steps. First it examines order of integration in
variables and employs panel unit root tests offered by Levin et al. (2002) and Im et al.
(2003). Second Kao’s (1999) panel cointegration test is used to find the long-term relationship between variables. Third it applies Panel Granger causality tests to determine the
direction of causality and adjustment mechanism in the system. The rest of the study is as
follows. Section II provides a literature review, and Section III presents data and an
account of the econometric methodology. Section IV discusses the results, whereas Section
V concludes the study and provides policy recommendations.
2 Literature review
Over the last few decades, extensive efforts are directed to discover the impact of energy
consumption and CO2 emissions on the economic growth. It works through two wellestablished directions of empirical literature. One of them discusses the relationship
between CO2 emissions and economic growth discussed in the context of Environmental
Kuznet curve (EKC) hypothesis. In this scenario, an initial rise in income causes environmental degradation. However, when income level reaches a specific point, people
become more conscious about their environmental responsibilities and environmental
degradation starts falling (Fodha and Zaghdoud 2010; Shahbaz et al. 2010). In contrast,
Dinda (2004) argues that these findings are not universal because the direction of causality
is not a like for each country. If CO2 emissions are causing economic growth, then CO2
emissions might be the result of the production process.
On the other hand, the seminal work of Kraft and Kraft (1978) provides important
insights into energy consumption and economic growth. It finds a unidirectional causal
relationship between the energy consumption and economic growth; the direction of
causality was from economic growth to energy consumption. This piece of work paved the
way for voluminous literature on finding causal linkages between energy consumption and
economic growth (Abosedra and Baghestani 1989; Bentzen and Engsted 1993).
The literature on energy consumption and economic growth is extended under four
different hypotheses that are based on direction of causality. First one is growth hypothesis
which argues that energy consumption is imperative for economic development. Energy
inputs facilitate production process and are complements to factors of production. An
economy would be energy dependent if higher economic growth is obvious in response to
rising energy consumption. To be more specific, it suggests unidirectional causality running from energy consumption to economic growth (Akarca and Long 1980).
Second one is known as conservation hypothesis which specifies that a country should
adopt conservation policy if higher energy consumption is unable to boost the economic
123

6. Author's personal copy
M. Zeshan, V. Ahmed
growth. In this case, direction of causality is from economic growth to energy consumption
showing rising energy consumption owing to higher income level. For Taiwan, Cheng and
Lai (1997) applied cointegration method to examine the interaction between economic
growth and energy consumption. It took data for the period 1955–1993 and confirmed the
conservation hypothesis. Using the same methodology, Wietze and Van Montfort (2007)
worked for Turkey and ended with the same conclusion.
Third one is neutrality hypothesis which asserts no causal relationship between energy
consumption and economic growth. In this case, energy conservation policies would not be
harmful for sustainable economic growth (Akarca and Long 1980). Finally, the fourth one
is feedback hypothesis. It presumes the bidirectional causal relationship between energy
consumption and economic growth, in this case both might be considered as complements.
It implies that any change in energy policies might cause significant effect on economic
growth and vice versa (Yang 2000; Paul and Bhattacharya 2004).
The following table provides the brief literature summary which considers two criterion.
First as most of the recent economic studies are working with panel data because it provide
robust results as compared to time series data (Lee and Chang 2008; Apergis and Payne
2009b) that is why present study has focused mainly on panel data studies. However, some
important time series studies are also reviewed. Second keeping in view the objectives of
the present study, it analyzes the collective relationship between energy consumption, CO2
emissions and economic growth (see Table 1).
3 Methodology and data
Levin panel unit root test operates under the null of a collective unit root in all the variables
in panel against the collective stationary. It is as follows,
Dxi;t ¼ ai þ di t þ #t þ qi xi;tÀ1 þ ei;t ; i ¼ 1; 2; 3; . . .; N; t ¼ 1; 2; 3; . . .; T
ð1Þ
It captures cross-sectional fixed affects with the help of a, whereas unit-specific time
trend is denoted with #. As the equation carries lagged dependent variable which presumes
slope homogeneity for all the units, it becomes important to capture unit-specific fixed
effects. Null of this test is specifies qi = 0 (for each i) against the alternative of qi = q0.
It develops a correction factor to produce standard distribution of pooled OLS estimates.
The assumption of collective stationarity of all variables is the major shortcoming of this
test. Having all the variables integrated or stationary is not a necessary condition in
econometrics, a fraction of variables might be integrated, while other might be stationary.
Im panel unit root test overcomes this problem and it discusses heterogeneity in qi under
different hypothesis. For Eq. (1), its works under the following null and alternative
hypotheses;
HA : qi 0;
H0 : qi ¼ 08i
i ¼ 1; 2; . . .; N1; i ¼ N1 þ 1; N1 þ 2; . . .; N:
In this scenario, null assumes that all the variables are non-stationary against the
alternative that a part of variables are stationary. If each variable in hand is non-stationary,
but there exits such a linear combination between the variables that make the system
stationary then the set of variables must be cointegrated. Using modified Dickey-Fuller
(DF) type and Augmented Dickey-Fuller (ADF) type tests, it employs Kao’s (1999) panel
cointegration test for finding a unique cointegrating vector. Given that all the variables are
I (1), this test is investigates long-run relationship between the variables. It is as follows,
123

7. Authors
Huang et al. (2008)
Apergis and Payne (2009a)
Apergis and Payne (2010)
Al-Mulali (2011)
Hossain (2011)
Niu et al. (2011)
Wang et al. (2011)
Arouri et al. (2012)
2
3
4
5
6
7
8
9
Soytas et al. (2007)
Ang (2007)
Soytas and Sari (2009)
11
10
Section II Time series studies
Coondoo and Dinda (2002)
1
Section I Panel studies
S. No.
12
1960–2000
960–2004
1960–2000
1981–2005
1995–2007
1971–2005
1971–2007
1980–2009
1992–2004
1971–2004
1972–2002
1960–1990
Time period
Granger causality test
Granger causality test
Cointegration, VECM, Granger
causality test
Bootstrap Panel Cointegration
Panel Cointegration, Panel VECM
Panel Granger causality test
Panel cointegration Granger
causality test
Cointegration test, Granger causality
Panel VECM
Panel Cointegration, VECM,
Granger causality test
Panel GMM, Panel VAR model
Panel Granger causality test
Methodology
Table 1 Literature review for energy, environment and growth nexus
Turkey
United States
France
MENA countries
28 provinces in China
Asia–Pacific countries
Newly industrialized countries
MENA countries
Panel of common wealth
independent states
Central America
Panel of 82 countries, Low income group
Middle-income group
High-income group
Panel North America, Western
Europe and Eastern Europe
Countries
Panel of Central American, South
American, Japan and Oceania
Panel of Asian and African countries
Countries
CO2 ? E
Y = CO2
E ? CO2
Y ? E, CO2
E?Y
E ? CO2
CO2 ? E
GDP ? E
E ? CO2
Y ? CO2, E
E, CO2$Y
E ? Y, CO2
E$Y
Short-run:
E ? Y, CO2
Long-run:
E$Y, CO2
E = CO2
Y?C
Y?E
CO2 ? Y
Y ? CO2
CO2$Y
Results
Author's personal copy
Energy, environment and growth
123

8. Authors
Zhang and Cheng (2009)
Chang (2010)
Fodha and Zaghdoud (2010)
Shahbaz et al. (2010)
Shahbaz et al. (2012)
S. No.
13
14
15
16
17
Table 1 continued
123
1971–2009
1971–2009
1961–2004
1981–2006
1960–2007
Time period
Cointegration, granger causality
Granger causality test
Cointegration, causality test
Granger causality tests
VECM, Granger causality Test
Methodology
Pakistan
Tunisia
China
China
Countries
Y ? CO2
E ? CO2
Y ? CO2
E ? CO2
Y ? CO2
Y ? E, CO2
Y?E
E ? CO2
Results
Author's personal copy
M. Zeshan, V. Ahmed

9. Author's personal copy
Energy, environment and growth
yit ¼ ai þ xit b þ uit
ð2Þ
where ai indicates country-specific constant term, b is slope of parameter, uit specifies
stationary error term, yit and xit both are unit root processes such that I (1). Both the DF
type test and ADF type test can be conducted in the following from:
^
^
uit ¼ q uitÀ1 þ vit
ð3Þ
and
uit ¼ q uitÀ1 þ
^
^
p
X
u D uitÀ1 þ vit
^
ð4Þ
j¼1
^
where residuals uit can be retained from Eq. (2); null and alternative hypotheses can be
specified as; H0 : q ¼ 1; HA : q1. Kao (1999) suggested different DF tests which are
based on the assumption of exogeneity of regressor. It also suggested its extended version
similar to ADF type test. All these tests work with nuisance parameters of long-run
conditional variance X. Asymptotic distribution of these tests converges to standard normal
distribution as N ? ? and T ? ?.
After specifying the long-term relationship between the variables, present study aspires
to investigate the direction of causality between the variables. If two integrated variables
are cointegrated, dynamic error correction mechanism can be utilized to discover the
direction of causality. Technically speaking, it is specified in the form of traditional vector
autoregression (VAR) framework augmented with one time period lagged error term
recovered from cointegrated vector. It is as follows,
X
X
h
DGDPi tÀp þ
h
DECi tÀp þ
DGDPit ¼ c1i þ
p 11 ip
p 12 ip
X
ð5Þ
Â
h
DCO2i tÀp þ l1i ECTi tÀ1 þ e1t
p 13 ip
X
X
DECit ¼ c2i þ
h21 ip DGDPi tÀp þ
h
DECi tÀp þ
p
p 22 ip
X
ð6Þ
Â
h
DCO2i tÀp þ l2i ECTi tÀ1 þ e2t
p 23 ip
X
X
DCO2it ¼ c3i þ
h
DGDPi tÀp þ
h
DECi tÀp þ
p 31 ip
p 32 ip
X
ð7Þ
Â
h33 ip DCO2i tÀp þ l3i ECTi tÀ1 þ e3t
p
where D is first difference operator, ECT is error correction term and p specifies lag length.
ECTit is the estimated residual derived from long run Eq. (2), lit shows the speed of
convergence parameter for each variable in the system. To measure granger causality, it
takes the help of F-test with a collective null that all the coefficients of another variable are
zero against the alternative of at least one of the coefficients in nonzero one by one for each
variable. A stable system requires at least one significant coefficient for all the error
correction terms in the system. It measures the speed of convergence if there is some
exogenous shock in the system.
The present study uses annual panel data that cover the period 1980–2010. The panel of
South Asian countries comprises Bangladesh, India, Pakistan, Sri Lanka and Nepal.
Following Al-mulali (2011) and Chang (2010), it uses three variables approach including
GDP (real GDP, constant 2005 international $), EC (energy consumption, constant 2005 kt
of oil equivalent) and CO2 (CO2 Emissions, kg per 2005 PPP $ of GDP). All the variables
123

10. Author's personal copy
M. Zeshan, V. Ahmed
Table 2 Results of panel unit root tests
Name of variable
Levin and Lin test (see foot note 1)
Im–Pesaran–Shin test
Level
Level
Unadjusted t-statistic
Adjusted t-statistic
First difference
w-t-bar statistic
GDP
-0.90
1.24
3.05
-3.28***
EC
-3.77
1.15
1.26
-1.44*
CO2
-3.23
0.88
2.27
-3.51***
*** and * indicate 1 and 10 % level of significance
are transformed in natural logarithm. Our data source is World Development Indicators
(WDI).
4 Results
Standard econometric techniques require the stationary data for empirical analysis. If a
variable is non-stationary, first difference makes it stationary, but this procedure wipes out
long-run information in the data. Kao’s (1999) panel cointegration technique preserves the
long-run information in data and provides robust results. Levin and Im both tests indicate
that all the three variables in this regression are integrated at levels (see Table 2 below).
Furthermore, first difference of all the three variables makes them stationary specifying
that all these series are I (1).1
Kao’s (1999) panel cointegration test portrays a unique cointegrating relationship
between the variables. As all the variables are in natural logarithms, so estimated coefficients represent elasticities. Long-run energy consumption elasticity of income is 0.81
indicating that a 1 % increase in energy consumption will bring 0.81 percent increase in
GDP in long run (See Table 3 below). It signifies that higher energy consumption might
contribute to economic growth significantly in emerging economies. On the other hand,
higher CO2 emissions are also affecting economic growth significantly. A 1 % increase in
CO2 emission reduces the GDP growth by 0.17 % in the long run. It indicates that CO2
emissions are much more detrimental for the South Asia because of its deteriorating
implications. There is a need for regional policy making to address this issue of rising CO2
emission.
VECM tests unearth the direction of short-term and long-term causality in the system.
Results illustrate that short-term causality runs from energy consumption to CO2 emissions
specifying that higher energy consumption results in more CO2 emissions, this fact is also
consistent with the Fig. 1. On the other hand, the short-term causality is running from CO2
emissions to GDP indicating that these emissions are detrimental for the sustained economic development in the short run.
The absence of any causal relationship between energy consumption and economic
growth assures the presence of neutrality hypothesis in South Asia. Zeshan (2013) argues
that if energy does not granger cause economic growth, it implies that energy is operating at
sub-optimal level. In such a situation, conservation policies can bring the society back to the
1
As Levin panel unit root test requires strongly balanced panel data so it is unable to operate with first
difference.
123

11. Author's personal copy
Energy, environment and growth
Table 3 Results of Kao’s panel
cointegration test
ADF
t-statistics
p value
-3.75**
0.03
Long-run coefficients: (GDP is dependent variable)
EC
-0.17***
Intercept
*** and ** indicate 1 and 5 %
significance level, respectively
0.81***
CO2
0.33
Table 4 Results of causality tests
Dependent variable
Short-run causality
DGDP
DEC
Long-run causality
DCO2
DGDP
–
1.15
2.71*
-0.55
DEC
0.32
–
1.18
2.14**
DCO2
1.96
6.37**
–
2.14**
** and * indicate 5 and 10 % significance level, respectively
optimal path. It implies that South Asia should adopt the conservation policies which would
also reduce the CO2 emissions in environment without impeding the economic growth.
Finally, the coefficients of error correction terms associated with the variables DEC and
DCO2 portray that the system is convergent in long run. The long-run causality indicates
that all the short-term disturbances in the system are corrected by adjusting the energy
consumption and CO2 emissions. However, GDP is unable to response in short run and one
possible factor might be inertia in GDP (please see last column of Table 4 for details).
5 Conclusion
Increasing energy demand is causing not only the energy crisis but is also depleting the
energy resources. Higher energy consumption escalates the proportion of CO2 emissions in
environment which causes pollution. Over the globe, governments are closely monitoring
this CO2 emission and are trying to supervise energy markets, same is the situation in
South Asia.
In the policy making process, causal linkages between the macro-social indicators are
very important. The information about long-run and short-run causality between energy
consumption, CO2 emission and economic growth is much helpful in devising energy
policies. For this purpose, the present study has focused on a panel of South Asian
countries that include Bangladesh, India, Pakistan, Sri Lanka and Nepal. It employs real
GDP, energy consumption CO2 emissions for the period of 1980–2010.
Both Levin and Im panel unit root tests specify that all the variables are I (1). In this
situation, the use of panel cointegration would be beneficial because it preserves the longrun information in data. Kao’s panel cointegration test finds a stable long-run relationship
between the variables. It illustrates that CO2 emissions are affecting South Asia significantly and a 1 % increase in carbon emission might reduce GDP growth by 0.17 % in the
long run. However, energy consumption positively affects the economic growth and a 1 %
increase in energy consumption escalated the economic growth by 0.81 % in the long run.
123

12. Author's personal copy
M. Zeshan, V. Ahmed
Panel Granger causality tests report short-run causality running from energy consumption to CO2 emissions which shows that higher energy consumption results in more
CO2 emissions in South Asia. Furthermore, it indicates that casualty is running from CO2
emissions to GDP indicating that CO2 emissions are detrimental to economic growth.
Moreover, the absence of any causal relationship between the energy consumption and
economic growth assures the evidence of neutrality hypothesis. Error correction coefficients portray that the system is convergent in long run and energy consumption and CO2
emissions would adjust them to rectify any short-run disturbance in the system.
6 Policy implications
• The CO2 emissions are adversely affecting the economic growth, and there is a need to
invest in environment friendly technologies.
• The South Asian countries should set regional environment protection targets to
overcome the increasing pollution in the region.
• The South Asian countries should meet at least once a year to discuss the devastating
impact of rising CO2 emissions in the region and also to devise strategies to cope with
these environmental challenges.
• The absence of any causal relationship between energy consumption and economic
growth assures the presence of neutrality hypothesis. The adoption of conservation
policies might reduce CO2 emission without impeding the economic growth.
• The South Asian countries use obsolete energy production technologies that are less
economic and are impeding the economic growth. It should gradually move to the
environment friendly technologies which are more efficient.
• The idea of regional energy market and open regional trade between the South Asian
countries would result in economies of scale and also more secure energy supplies.
• This regional interdependency will also reduce the hostile tendency of conflict which is
impeding the regional economic growth.
References
Abosedra, S., & Baghestani, H. (1989). New evidence on the causal relationship between United States
energy consumption and Gross National Product. Journal of Energy and Development, 14, 285–392.
Akarca, A. T., & Long, T. V. (1980). On the relationship between energy and GNP: a reexamination.
Journal of Energy and Development, 5, 326–331.
Al-Mulali, U. (2011). Oil Consumption, CO Emission and Economic Growth in MENA Countries. Energy,
36(10), 6165–6171.
Ang, J. B. (2007). CO2 emissions, energy consumption, and output in France. Energy Policy, 35(10),
4772–4778.
Apergis, N., & Payne, J. E. (2009a). CO2 emissions, energy usage and output in Central America. Energy
Policy, 37(8), 3282–3286.
Apergis, N., & Payne, J. E. (2009b). Energy consumption and economic growth in Central America:
evidence from a panel cointegration and error correction model. Energy Economics, 31, 211–216.
Apergis, N., & Payne, J. E. (2010). The emissions, energy consumption, and growth nexus: Evidence from
the commonwealth of independent states. Energy Policy, 38(1), 650–655.
Arouri, M.H., Ben Youssef, A., M’Henni, H., & Rault, C. (2012). Energy consumption, economic growth
and CO2 emissions in Middle East and North African Countries. CESifo Group Munich, Working
Paper Series, 3726.
Bentzen, J., & Engsted, T. (1993). Short- and long-run elasticities in energy demand : a Cointegration
approach. Energy Economics, 15, 9–16.
123

13. Author's personal copy
Energy, environment and growth
Chang, C. C. (2010). A multivariate causality test of carbon dioxide emissions, energy consumption and
economic growth in China. Applied Energy, 87(11), 3533–3537.
Cheng, B. S., & Lai, T. W. (1997). An investigation of co-integration and causality between energy
consumption and economic activity in Taiwan. Energy Economics, 19, 435–444.
Coondoo, D., & Dinda, S. (2002). Causality between income and emission: a country Group specific
econometric analysis. Ecological Economics, 40(3), 351–367.
Dinda, S. (2004). Environmental Kuznets curve hypothesis: a survey. Ecological Economics, 49, 431–455.
Emirates Center for Strategic Studies and Research, ECSSR. (2004). Asian energy markets: Dynamics and
trend. Emirates Center for Strategic Studies and Research.
Fodha, M., & Zaghdoud, O. (2010). Economic Growth and Pollutant Emissions in Tunisia: An empirical
analysis of the environmental Kuznets curve. Energy Policy, 38(2), 1150–1156.
Hossain, M. S. (2011). Panel estimation for CO2 emissions, energy consumption, economic growth, trade
openness and urbanization of newly industrialized countries. Energy Policy, 39(11), 6991–6999.
Huang, B. N., Hwang, M. J., & Yang, C. W. (2008). Causal relationship between energy consumption and
GDP growth revisited: A dynamic panel data approach. Ecological Economics, 67(1), 41–54.
Im, K. S., Pesaran, M. H., & Shin, Y. (2003). Testing For Unit Roots in Heterogeneous Panels. Journal of
Econometrics, 115(1), 53–74.
Kao, C. (1999). Spurious Regression and Residual-Based Tests for Cointegration in Panel Data. Journal of
Econometrics, 90, 1–44.
Kraft, J., & Kraft, A. (1978). On the relationship between energy and GNP. Journal of Energy and
Development, 3, 401–403.
Lee, C. C., & Chang, C. P. (2008). New evidence on the convergence of per capita carbon dioxide emissions
from panel seemingly unrelated regressions augmented Dickey-Fuller tests. Energy, 33, 1468–1475.
Levin, A., Lin, C. F., & Chu, C. (2002). Unit root test in panel data: Asymptotic and finite sample properties.
Journal of Econometrics, 108, 1–24.
Niu, S., Ding, Y., Niu, Y., Li, Y., & Luo, G. (2011). Economic growth, energy conservation and emissions
reduction: A comparative analysis based on panel data for 8 Asian-Pacific countries. Energy Policy,
39(4), 2121–2131.
Ozturk, I. (2010). A Literature Survey on Energy-Growth Nexus. Energy Policy, 38, 340–349.
Paul, S., & Bhattacharya, R. N. (2004). Causality between energy consumption and economic growth in
India: a note on conflicting results. Energy Economics, 26, 977–983.
Payne, J. E. (2010). Survey of the international evidence on the causal relationship between energy consumption and growth. Journal of Economic Studies, 37, 53–95.
Shahbaz, M., & Dube, S. (2012). Revisiting the relationship between coal consumption and economic
growth: cointegration and causality analysis in Pakistan. Applied Econometrics and International
Development, 12(1).
Shahbaz, M., Lean, H.H., & Shabbir, M.S. (2010). Environmental Kuznets curve and the role of energy
consumption in Pakistan. Development Research Papers.
Shahbaz, M., Zeshan, M., & Afza, T. (2012). Is energy consumption effective to spur economic growth in
Pakistan? New evidence from bounds test to level relationships and granger causality tests. Economic
Modelling, 29(6), 2310–2319.
Soytas, U., & Sari, R. (2009). Energy consumption, economic growth, and carbon emissions: Challenges
faced by an EU candidate member. Ecological Economics, 68(6), 1667–1675.
Soytas, U., Sari, R., & Ewing, B. T. (2007). Energy consumption, income, and carbon emissions in the
United States. Ecological Economics, 62, 482–489.
Squalli, J. (2007). Electricity consumption and economic growth: bounds and causality analyses of OPEC
countries. Energy Economics, 29, 1192–1205.
Wang, S. S., Zhou, D. Q., Zhou, P., & Wang, Q. W. (2011). CO2 emissions, energy consumption and
economic growth in China: A panel data analysis. Energy Policy, 39(9), 4870–4875.
Wietze, L., & Van Montfort, K. (2007). Energy consumption and GDP in Turkey: Is there a co-integration
relationship? Energy Economics, 29, 1166–1178.
Yang, H. Y. (2000). A note on the causal relationship between energy and GDP in Taiwan. Energy
Economics, 22, 309–317.
Zeshan, M. (2013). Finding the cointegration and causal linkages between the electricity production and
economic growth in Pakistan. Economic Modeling, 31, 344–350.
Zhang, X. P., & Cheng, X. M. (2009). Energy Consumption, Carbon Emissions, and Economic Growth in
China. Ecological Economics, 68(10), 2706–2712.
123