Renewable and Non-renewable Energy Consumption and Economic Growth: Empirical Evidence from Panel Error Correction Model

Abstract

This study examines the relationship between energy consumption and economic growth for BRICS countries within a multivariate panel framework for 1990–2012. The Pedroni (1999–2004) panel cointegration test shows a long-run relationship among GDP per capita renewable energy consumption, non-renewable energy consumption, and gross fixed capital formation. Finally, we apply panel error correction mechanism which reveals unidirectional causality from economic growth to renewable and non-renewable energy consumption. The results support the conservation hypothesis. In other words, no strong relationship is found between energy consumption to economic growth. These results indicate economic growth is the significant variable which boosts energy consumption in the BRICS countries. Higher the economic growth, higher will be the energy consumption.

Keywords

Economic growth VECM energy consumption BRICS renewable energy

Introduction

Energy is an essential input for production and consumption. It is a controversial topic with regard to the traditional neo-classical growth model, which treats land, labor, and capital as major input sources for production. Energy serves as a major element for economic growth in the era of liberalization, privatization, and globalization. The role of energy is important in generating income and employment, and thus economy heavily depends on it. The renewable sources of energy get attention due a number of attributes. The rising demand for oil, dependency on foreign energy sources, and carbon emission leading to environmental degradation are some factors which kindle interest in renewable sources of energy.

The link between energy consumption and economic growth has an important policy implication. The country which is dependent on energy, where the causality runs from energy consumption to growth, would be very much cautious with the policy making because any negative shock on energy supply will be detrimental to economic growth. The other proposition says that economic growth determines the energy consumption (where the causality runs from economic growth to energy consumption), the country can change the policy with regard to energy consumption because it has a very little effect on economic growth (Ouédraogo, 2010).

The relationship between energy consumption and gross domestic product (GDP) growth in the BRICS countries is synthetized into four hypotheses given by Apergis and Payne (2009a, 2011), Bowden and Payne (2010), Ozturk (2010), Payne (2010), and Soytas and Sari (2003). The growth hypothesis proposes that energy consumption plays a significant role in economic growth both directly and indirectly and/or as a compliment to capital and labor in the production process. The existence of unidirectional causality from energy consumption to economic growth supports the growth hypothesis. The conservative hypothesis entails that energy conservation approach reduces the carbon dioxide emissions, improves efficiency and wastes which would not affect economic growth. The existence of unidirectional causality from economic growth to energy consumption provides the confirmation of conservative hypothesis. The feedback hypothesis postulates that energy consumption and economic growth are dependent on each other, and this hypothesis is supported by the existence of bidirectional causality between economic growth and energy consumption. The neutrality hypothesis supports that there is a negligible role of energy consumption in determining the economic growth. The absence of causality between energy consumption and economic growth supports the neutrality hypothesis.

The aim of this article is to find out the relationship between energy consumption and economic growth in the emerging countries BRICS. The reason for choosing BRICS countries as a representative sample is that, BRICS countries are the ones which account for 37.2 percent of total world energy demand and 40 percent of the CO₂ emission with energy consumption of 38.7 percent in 2011 due to the higher pressure on the mineral coal in the energy matrix. We intend to explore the aspect of sustainable economic growth, rather than just economic growth in the emerging economies BRICS. The reduction commitments of CO₂ by many countries across the globe have led to increased focus on energy-related issues.

The article is arranged as follows: the next section will provide the literature novelty. The third section presents the data, methodology, and results. The final section will provide the final conclusion.

Literature Review

Resources are the major source of energy for economic growth. Several studies have examined the relationship between economic growth and energy consumption. The economists got fascinated with the concept of (Halicioglu, 2009) who tries to explain the relationship between energy consumption and economic growth. It was the Kraft and Kraft (1978) who lead a pioneering work to find out the relationship between energy consumption and economic growth.

Bowden and Payne (2010) studied the USA over the period 1949–2006 and checked the sectorial relationship between renewable and non-renewable energy consumption and economic growth by using Toda-Yamamota causality procedure within the multivariate framework including labor and gross capital formation. The results did not find any relationship between renewable energy consumption and real GDP by taking industrial sector into consideration and supports the neutrality hypothesis, the results shows unidirectional positive relationship between energy consumption and GDP, which indicates the presence of growth hypothesis, whereas causality testing shows bi-directional results from energy consumption to real GDP in commercial and residential sectors which supports feedback hypothesis, and negative relationship between industrial non-renewable energy consumption to GDP, which gives evidence of growth hypothesis.

Sadorsky (2009) uses panel correction model to test the relationship between renewable energy consumption and economic growth in 18 emerging countries from the time period of (1994–2003). The result shows the increase in income has a statistically significant impact on the per capita energy consumption of renewable energy. The results support the conservation hypothesis.

Bilgili and Ozturk (2015) studied G7 countries from the period 1980–2009. The study uses panel unit root, panel cointegration, OLS, and dynamic OLS, are run throughout homogeneous and heterogeneous variance structures to examine the relationship. The results reveal a significant positive relationship from GDP, human capital, biomass consumption, and capital stock. However, the causality test supports growth hypothesis from biomass consumption to economic growth for G7 countries. For the policy purposes, biomass consumption can be an appropriate instrument for the economic growth for G7 countries.

Alper and Oguz (2016) investigate the causality among economic growth, renewable energy consumption, capital, and labor for new EU member countries for the period of 1990–2009, by using asymmetric causality test approach and autoregressive distributed lag (ARDL) approach. The empirical results support that renewable energy consumption has positive impact on economic growth for all investigated countries. But only for Bulgaria, Estonia, Poland, and Slovenia, a statistically significant impact on economic growth was found. And it also supports neutrality hypothesis for Cyprus, Estonia, Hungary, Poland, and Slovenia, while the conservation hypothesis is present for Czech Republic. The fact is that there is a causal relationship from economic growth to renewable energy consumption and the growth hypothesis is supported for Bulgaria, referring to causality from energy consumption to economic growth.

Kula (2014) examined whether a long-run relationship between per capita renewable electricity consumption and GDP exists by employing panel integration and co-integration techniques for a dynamic heterogeneous panel of 19 Organisation for Economic Co-operation and Development (OECD) countries over the period 1980–2008. Their findings support the existence of a long-run equilibrium relation between renewable electricity consumption and GDP. Further, the evidence points to unidirectional causality from a GDP to renewable electricity consumption.

We summarize the literature under the various segments in Table 1. Fuinhas and Marques (2012) and Apergis and Payne (2012) confirmed the validity of feedback hypothesis. Huang, Hwang, and Yang (2008), Menyah and Wolde-Rufael (2010), and Ocal and Aslan (2013) confirmed the validity of conservation hypothesis. Menegaki (2011), Soytas, Sari, and Ewing (2007), and Tugcu, Ozturk, and Aslan (2012) confirmed the validity of neutrality hypothesis. Bowden and Payne (2009) validates growth hypothesis; and Apergis and Payne (2009b) find evidence for both growth and feedback hypothesis. The results of Lee (2006), Ozturk (2010), Akinlo (2008), and Pao and Fu (2013) were mixed, and they found existence of feedback, the conservation, and the neutrality hypothesis.

Table 1.

Summary of Recent Literature Review on Energy Consumption and Economic Growth

Study	Methodology/Countries and Period	Confirmed Hypothesis
Fuinhas and Marques (2012)	Autoregressive distributed lag (ARDL), Portugal, Italy, Greece, Spain, and Turkey (1965–2009)	Feedback hypothesis
Apergis and Payne (2012)	Panel error correction model, 80 countries (1990–2007)	Feedback hypothesis in both the short and long run
Huang et al. (2008)	Generalized method of moment system, 82 countries (1972–2002)	Conservation hypothesis
Menyah and Wolde-Rufael (2010)	Granger causality test, USA (1960–2007)	Conservation hypothesis
Ocal and Aslan (2013)	ARDL approach and Toda–Yamamoto causality test, Turkey (1990–2010)	Conservation hypothesis
Menegaki (2011)	Multivariate panel framework, 27 European Countries (1997–2007)	Neutrality hypothesis
Soytas et al. (2007)	Toda–Yamamoto procedure, USA (1960–2004)	Neutrality hypothesis
Tugcu et al. (2012)	Hatemi-J causality tests, G-7 countries, (1980–2009)	Neutrality hypothesis for France, Italy, Canada, and USA
Bowden and Payne (2009)	Toda–Yamamoto long-run causality test, USA (1949–2006)	Growth hypothesis
Apergis and Payne (2009b)	Panel co-integration and Granger causality, 11 Common Wealth of Independent States (1991–2005)	Growth and the feedback
Lee (2006)	Toda–Yamamoto procedure, 11 major industrialized countries (1947–1974)	Feedback, conservation, and the neutrality
Ozturk et al. (2010)	Panel cointegration and panel causality tests, 51 countries (1971–2005)	Feedback and the conversation hypothesis
Akinlo (2008)	ARDL bounds test, 11 countries in sub-Saharan Africa (1980–2003)	Feedback, the conservation, and the neutrality hypothesis
Pao and Fu (2013)	Error correction model, Brazil (1980–2010)	Conservation hypothesis (EG and EC) Feedback hypothesis (EG and REC)

Source: Authors’ compilation.

This study employs Pedroni (1999, 2004) panel cointegration and Panel VECM to estimate the causality between economic growth and energy consumption in BRICS countries from the time period 1990–2012.

Data, Methodology, and Results

The study uses panel data for BRICS countries for the period 1990–2012. The framework includes GDP in constant 2010 US dollars, real gross fixed capital (GFC) in constant 2010 US dollars and we have calculated non-renewable electricity consumption (total electricity consumption minus renewable electricity consumption) in billion kilowatt hours for each country. Data cover has been obtained from the World Bank and energy data from US Energy Information Administration (EIA). The general framework of the model is

Y = f (R E, N R E, G F C)

(1)

where Y is the GDP per capita in constant 2010 US dollars, RE is the renewable electricity consumption billion kilowatt hours, NRE is the non-renewable electricity consumption defined in billion kilowatt hours, and GFC represents real gross fixed capital formation in constant 2010 US dollars.

Table 2.

Results of Unit Roots Tests

Variable	LLC		IPS		ADF-Fisher
	Level	Ist diff.	Level	Ist diff.	Level	Ist diff.
GDP PC	1.3952	4.3546*	4.7137	3.6938*	0.3932	31.301*
RE	1.6122	7.1640*	0.7052	7.4745*	18.398	63.143*
NRE	1.0053	6.0708*	1.8524	5.3134*	8.1063	48.074*
GFC	–0.3938	5.2795*	0.2704	4.8426*	7.1697	41.772*

Source: Authors’ compilation.

Notes: Critical values at the 1 percent significance level denoted by *. Panel unit root test includes intercept only.

In order to assess the integration and unit root among the variables for panel cointegration numerous panel unit root tests have been performed. The panel unit root test based on ADF proposed by Levin, Lin, and Chu t-test (2002, hereafter LLC) which assumes homogeneity in the dynamic panel of auto regression in all panel units. The Im, Pesaran, and Shin (2003), and the Fisher-ADF and PP test all allow for individual unit root processes, the tests are all characterized by the combining of individual unit root tests to derive a panel-specific result and has the advantage of allowing much heterogeneity across all panel units. Levin, Lin, and Chu (LLC), Breitung, and Hadri (2000) tests suppose there is a common unit root process for both the tests. Both tests are different on the basis of null-hypothesis of unit root while Hadri is based on homogeneous and heterogeneous long variance and null hypothesis of no unit root. Table 2 reveals the result of panel unit root test which shows all variables are integrated of order one.

The Pedroni (1999, 2004) cointegration is better than others because it allows heterogeneous intercept and trend coefficients across cross-sections. We construct the bellow regression equation to determine the long-run relationship among the variables. Equation 2 results are given in Table 3.

y_{i, t} = σ_{i} + δ_{i} t + β_{1 i} x R E_{i t} + β_{2 i} N R E_{i t} + β_{3 i} G F C_{i t} + \in_{i t}

(2)

Pedroni constructs various methods for testing null hypothesis of no cointegration (ρ_i = 1). There is also homogeneous hypothesis which is calculated within the dimension and heterogeneous is referred between the dimension or group statistic. The null hypothesis of no cointegration H₀: δ_i – 1 for all i. The alternative hypothesis H₀: δ_i – δ < 1 or all i. Pedroni (1999, 2004) provides two cointegration tests. One test is based on within the dimension approach, including four statistics panel V, panel ρ, panel PP, and panel ADF statistics. These inferential pool all auto regression coefficients to estimate residuals taking a common factor into account across all countries. The group test is based between the dimension approach, including three statistic group ρ, group PP, and group ADF statistics. These are based on average of individual auto regression coefficients with the residuals for each country. Table 3 reports all the seven test statistics, most of the tests reject the null hypothesis of no cointegration at the 1 percent level of significance.

Table 3.

The Results of Panel Cointegration Tests

Within Dimension Test Statistics		Between Dimension Test Statistics
Panel v-statistic	–1.5025*	Group ρ-statistic	1.0381
Panel ρ-statistic	0.7003	Group pp-statistic	–1.6971*
Panel PP-statistic	–2.1941*	Group ADF-statistic	–1.4650*
Panel ADF-statistic	–2.1305*

Source: Authors’ compilation.

Notes: Of the seven tests, the panel v-statistic is a one-sided test where large positive values reject the null hypothesis of no cointegration, whereas large negative values for the remaining test statistics reject the null hypothesis of no cointegration. Critical values denoted by * are statistically significant at the 1 percent level.

Panel vector error correction model is estimated to infer the causal dynamics following Pesaran, Shin, and Smith (1999). We first estimate long-run model employing the two-step Engle and Granger (1987) procedure defined in Equation (2) whose lagged value serve as the error correction term in the estimation of error correction model which are generated from the estimated residuals as follows:

\begin{matrix} Δ G D P_{i t} = θ_{1 j} + \sum_{k = 1}^{q} θ_{11 i k} Δ G D P_{i t - k} + \sum_{k = 1}^{q} θ_{12 i k} Δ R E_{i t - k} + \sum_{k = 1}^{q} θ_{13 i k} Δ N R E_{i t - k} + \sum_{k = 1}^{q} θ_{14 i k} Δ G F C_{i t - k} \\ + λ_{1 i} ε_{i t - 1} + u_{1 i t} \end{matrix}

(3)

\begin{matrix} Δ R E_{i t} = θ_{2 ​ j} + \sum_{k = 1}^{q} θ_{21 i k} Δ G D P_{i t - k} + \sum_{k = 1}^{q} θ_{22 i k} Δ R E_{i t - k} + \sum_{k = 1}^{q} θ_{23 i k} Δ N R E_{i t - k} + \sum_{k = 1}^{q} θ_{24 i k} Δ G F C_{i t - k} \\ + λ_{2 i} ε_{i t - 1} + u \end{matrix}

(4)

\begin{matrix} Δ N R E_{i t} = θ_{3 j} + \sum_{k = 1}^{q} θ_{31 i k} Δ G D P_{i t - k} + \sum_{k = 1}^{q} θ_{32 i k} Δ R E_{i t - k} + \sum_{k = 1}^{q} θ_{33 i k} Δ N R E_{i t - k} + \sum_{k = 1}^{q} θ_{34 i k} Δ G F C_{i t - k} \\ + λ_{3 i} ε_{i t - 1} + u_{3 i t} \end{matrix}

(5)

\begin{matrix} Δ G F C_{i t} = θ_{4 j} + \sum_{k = 1}^{q} θ_{41 i k} Δ G D P_{i t - k} + \sum_{k = 1}^{q} θ_{42 i k} Δ R E_{i t - k} + \sum_{k = 1}^{q} θ_{43 i k} Δ N R E_{i t - k} + \sum_{k = 1}^{q} θ_{44 i k} Δ G F C_{i t - k} \\ + λ_{4 i} ε_{i t - 1} + u_{4 t} \end{matrix}

(6)

The statistical significance of F-statistic determines the short-run causality which is associated with the right hand side of Equations (3) to (6). While statistical significance of respective error-correction term using t-test signifies long-run causality.

Table 4 exhibits results of panel error correction model. In Equation (3), renewable energy consumption, non-renewable energy consumption, and gross fixed capital each have a negative and statistically insignificant effect on economic growth in short run. In Equation (4), non-renewable energy consumption, economic growth, and gross fixed capital each have a negative and statistically insignificant effect on renewable energy consumption in short run. With regard to Equation (5), non-renewable energy consumption has a negative insignificant impact on renewable energy consumption, economic growth, and gross fixed capital in short run. Lastly, in Equation (6) for gross fixed capital formation, economic growth, non-renewable energy consumption, and renewable energy consumption each have positive and statistically significant impact in short run.

Table 4:

Panel Causality Tests

Dependent Variable Independent Variables
	Short Run								Long Run
	ΔGDP_it_–1	ΔGDP_it_–1	ΔRE_it_–1	ΔRE_it_–2	ΔNRE_it_–1	ΔNRE_it_–2	ΔGFC_it_–1	ΔGFC_it_–2	ECT
ΔGDP_it	_	_	–0.000804	0.005423	–0.024707	–0.020353	–0.11597	–0.05460	–0.000953
(Prob)	-_	_	(0.1441)	(0.1441)	(0.1441)	(0.1441)	(0.1441)	(0.1441)	(0.3569)
ΔRE_it	0.634871	0.288491	_	_	–0.049225	0.066769	–0.44465	0.019917	–0.00455
(Prob)	(0.8775)	(0.8775)	_	_	(0.8775)	(0.8775)	(0.8775)	(0.8775)	(0.0198) ***
ΔNRE_it	1.131095	–0.658683	0.006172	0.006263	_	_	–0.20521	0.141071	–0.078354
(Prob)	(0.8860)	(0.8860)	(0.8860)	(0.8860)	_	_	(0.8860)	(0.8860)	(0.0001) ***
ΔGFC_it	0.482140	0.016874	0.016985	0.014277	–0.031145	0.003938	_	_	–0.030810
(Prob)	(0.0576) *	(0.0576) *	(0.0576) *	(0.0576) *	(0.0576) *	(0.0576) *	_	_	(0.0374) ***

Source: Authors’ compilation.

Notes: Wald test short-run result obtained with Wald test, respect to short-run changes in the independent variables. Johansen Fisher panel cointegration test the lag length is 1, VAR causality test lag 1 2. ECT represents the coefficient of the error-correction term. Probability values are also given with the corresponding F-statistic. * indicates significant at 10 percent level and *** indicates significant at 1 percent level.

The results of short run signify the absence of causality from non-renewable energy consumption to economic growth and renewable energy consumption to economic growth. On the other hand, we find unidirectional relationship from gross fixed capital formation to economic growth and renewable energy consumption, which points out that economic growth catalysis investment in fixed assets, as proxied by gross fixed capital formation and prevailing capital intensive infrastructure is associated with renewable energy sources. There is an absence of bidirectional causality in the short run.

With respect to the long-run dynamic panels captured by error correction terms in Equations (3) to (6), there is a unidirectional causality between renewable energy consumption and economic growth also between non-renewable energy consumption to economic growth in the long run. The speed of adjustment is slow and each variable is deviating from long-run equilibrium. The unidirectional causality supports conservation hypothesis which reveals that renewable energy consumption and non-renewable energy consumption will not affect or may have a little effect on economic growth.

There is no study in the literature for BRICS countries that investigates energy consumption (renewable and non-renewable energy consumption) and economic growth linkage. However, we can compare our study with those who support conservation hypothesis for other countries. Pao and Fu (2013) use error correction model for the period 1980–2010 to test the relationship between economic growth and energy consumption. The results support conservation hypothesis for Brazil. Ozturk et al. (2010) employed panel cointegration and panel causality test for 51 countries to testify the relationship between energy consumption and economic growth from the period 1971 to 2005. The conclusion supports conservation hypothesis. Ocal and Aslan (2013) investigate the relationship between economic growth and energy consumption over the period 1990–2010 by using Toda-Yamamoto causality for Turkey. The estimation supports conservation hypothesis. Huang et al. (2008) studied 82 countries from the period 1972–2002 and supports the conservation hypothesis by using generalized method of moment system.

The results of study revel that there is a unidirectional causality between energy consumption and economic growth for BRICS countries. When energy consumption has a positive impact on economic growth, it means use of energy has greater benefits than the externalities caused by energy use. Despite renewable source of energy gaining importance as a source of energy consumption, non-renewable source of energy cannot be set aside. With regard to substitutability, the development of renewable energy sector may provide relief from carbon emissions. The interdependence between energy consumption and economic growth suggests policies that influence economic growth. BRICS countries have potential for renewable energy sources, which can be a stimulus to economic growth because it meets energy needs of the time. İt is the dire need for these countries to take lead with respect to design and implementation of policies to move toward greater use of renewable source of energy with less dependence on non-renewable sources of energy.

Conclusion

Renewable energy consumption has emerged as a clean source of energy, and can ease the growing concerns of negative environmental impacts. This study aims to investigate the relationship between renewable energy consumption, non-renewable energy consumption, economic growth, and capital within a multivariate panel error correction framework for BRICS countries over the period 1990–2012.

In the literature of energy economics, there is no study which has explored the relationship between energy consumption and economic growth for BRICS countries. The results of Pedroni panel cointegration test (1999–2004) show there is a long-run relationship between renewable energy consumption, non-renewable energy consumption, GDP, and gross capital formation. Moreover, the results of panel error correction mechanism show there is a unidirectional causality running from economic growth to renewable and non-renewable energy consumption. The results may be different from the literature of energy economics because of different methodologies and the developing stages of these countries. This study suggests conservation hypothesis for BRICS countries. The causal relationship between economic growth and energy consumption has important policy implication. When economic growth brings energy consumption, the use of energy will set back economic growth under a conservation hypothesis. Policy makers should encourage promoting renewable energy consumption across countries for the sustainable economic growth (Hirschl, 2009).

References

Akinlo

A. E.

(2008). Energy consumption and economic growth: Evidence from 11 Sub-Sahara African countries. Energy Economics, 30(5), 2391–2400.

Alper

& Oguz

(2016). The role of renewable energy consumption in economic growth: Evidence from asymmetric causality. Renewable and Sustainable Energy Reviews, 60, 953–959.

Apergis

& Payne

J. E.

(2009a). Energy consumption and economic growth in Central America: Evidence from a panel cointegration and error correction model. Energy Economics, 31(2), 211–216.

Apergis

& Payne

J.E.

(2009b). Energy consumption and economic growth: Evidence from the Commonwealth of Independent States. Energy Economics, 31(5), 641–647.

Apergis

& Payne

J.E.

(2011). Renewable and non-renewable electricity consumption–growth nexus: Evidence from emerging market economies. Applied Energy, 88(12), 5226–5230.

Apergis

& Payne

J.E.

(2012). Renewable and non-renewable energy consumption-growth nexus: Evidence from a panel error correction model. Energy Economics, 34(3), 733–738.

Bilgili

& Ozturk

(2015). Biomass energy and economic growth nexus in G7 countries: Evidence from dynamic panel data. Renewable and Sustainable Energy Reviews, 49, 132–138.

Bowden

& Payne

J. E.

(2009). The causal relationship between US energy consumption and real output: A disaggregated analysis. Journal of Policy Modeling, 31(2), 180–188.

Bowden

& Payne

J.E.

(2010). Sectoral analysis of the causal relationship between renewable and non-renewable energy consumption and real output in the US. Energy Sources, Part B: Economics, Planning, and Policy, 5(4), 400–408.

10.

Engle

R. F.

& Granger

C. W.

(1987). Co-integration and error correction: Representation, estimation, and testing. Econometrica: Journal of the Econometric Society, 55(2), 251–276.

11.

Fuinhas

J. A.

& Marques

A. C.

(2012). Energy consumption and economic growth nexus in Portugal, Italy, Greece, Spain and Turkey: An ARDL bounds test approach (1965–2009). Energy Economics, 34(2), 511–517.

12.

Halicioglu

(2009). An econometric study of CO₂ emissions, energy consumption, income and foreign trade in Turkey. Energy Policy, 37(3), 1156–1164.

13.

Hirschl

(2009). International renewable energy policy—Between marginalization and initial approaches. Energy Policy, 37(11), 4407–4416.

14.

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.

15.

K. S.

Pesaran

M. H.

& Shin

(2003). Testing for unit roots in heterogeneous panels. Journal of Econometrics, 115(1), 53–74.

16.

Kraft

& Kraft

(1978). Relationship between energy and GNP. Journal of Energy and Development (United States), 3(2), 401–403.

17.

Kula

(2014). The long-run relationship between renewable electricity consumption and GDP: Evidence from panel data. Energy Sources, Part B: Economics, Planning, and Policy, 9(2), 156–160.

18.

Lee

C.C.

(2006). The causality relationship between energy consumption and GDP in G-11 countries revisited. Energy Policy, 34(9), 1086–1093.

19.

Levin

Lin

C. F.

& Chu

C. S. J.

(2002). Unit root tests in panel data: Asymptotic and finite-sample properties. Journal of Econometrics, 108(1), 1–24.

20.

Menegaki

A. N.

(2011). Growth and renewable energy in Europe: A random effect model with evidence for neutrality hypothesis. Energy Economics, 33(2), 257–263.

21.

Ocal

& Aslan

(2013). Renewable energy consumption–economic growth nexus in Turkey. Renewable and Sustainable Energy Reviews, 28, 494–499.

22.

Ouédraogo

I. M.

(2010). Electricity consumption and economic growth in Burkina Faso: A cointegration analysis. Energy Economics, 32(3), 524–531.

23.

Ozturk

(2010). A literature survey on energy–growth nexus. Energy Policy, 38(1), 340–349.

24.

Pao

H. T.

& Fu

H. C.

(2013). Renewable energy, non-renewable energy and economic growth in Brazil. Renewable and Sustainable Energy Reviews, 25, 381–392.

25.

Payne

J. E.

(2010). Survey of the international evidence on the causal relationship between energy consumption and growth. Journal of Economic Studies, 37(1), 53–95.

26.

Pedroni

(1999). Critical values for cointegration tests in heterogeneous panels with multiple regressors. Oxford Bulletin of Economics and Statistics, 61(s1), 653–670.

27.

Pesaran

M. H.

(2004). Panel cointegration: Asymptotic and finite sample properties of pooled time series tests with an application to the PPP hypothesis. Econometric Theory, 20(3), 597–625.

28.

Pesaran

M. H.

Shin

& Smith

R. P.

(1999). Pooled mean group estimation of dynamic heterogeneous panels. Journal of the American Statistical Association, 94(446), 621–634.

29.

Sadorsky

(2009). Renewable energy consumption and income in emerging economies. Energy Policy, 37(10), 4021–4028.

30.

Soytas

& Sari

(2003). Energy consumption and GDP: Causality relationship in G-7 countries and emerging markets. Energy Economics, 25(1), 33–37.

31.

Soytas

Sari

& Ewing

B. T.

(2007). Energy consumption, income, and carbon emissions in the United States. Ecological Economics, 62(3), 482–489.

32.

Tugcu

C. T.

Ozturk

& Aslan

(2012). Renewable and non-renewable energy consumption and economic growth relationship revisited: Evidence from G7 countries. Energy Economics, 34(6), 1942–1950.