CO2 emissions, energy consumption, economic growth, and financial development in GCC countries: Dynamic simultaneous equation models

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

crossmark

CO2 emissions, energy consumption, economic growth, and financial development in GCC countries: Dynamic simultaneous equation models 1. Introduction

The relationship between energy consumption and economic growth, as well as economic growth and environmental pollution, has been the subject of intense research in the last three decades. However, the empirical evidence remains controversial due to many influencing factors. The existing literature reveals that empirical studies differ substantially; for example, some investigate single countries and others examine multiple countries, see [1–10],. The results of these studies carry various policy implications in an economic context. An assessment of the existing literature suggests that most studies have focused either on the nexus of economic growth and energy consumption or of economic growth and environmental pollutants, but only a limited number of studies has tested these two relationships in the same framework. Rapid global economic growth resulted in a 1.4% increase in overall emissions over 2011, reaching a total of 34.5 billion tonnes in 2012. The CO2 emissions trend reflects energy-related human activities which were determined by economic growth, particularly in emerging countries. In 2012, a decoupling of the increase in CO2 emissions from global economic growth (in gross domesticc product) took place, which points to a shift toward less fossil fuel-intensive activities. Furthermore, it reflects enhanced use of renewable energy and increased energy saving. Actually, 90% of the CO2 emissions originate from fossil-fuel combustion and therefore are determined by the energy demand or the level of energy-intensive activity. High energy demand predicts high levels of use in power generation, industries, and road transport. However, changes in energy efficiency and shifts in the fuel mix, especially from carbon-intensive coal to low-carbon gas or from fossil fuels to nuclear or renewable energy, can cut the overall global emissions level [11]. The relationship between energy consumption and economic growth, as well as economic growth and environmental pollution, has been the subject of intense research in oil-rich countries (Oman, United Arab Emirates (UAE), Kindom of Saudi Arabia (KSA), Bahrain, Qatar, and Kuwait). These countries are blessed with abundant renewable energy resources; however, their economies are still highly dependent on the export of fossil fuel products. These countries hold around 23.5% of the world's natural gas and 40% of proven oil reserves. However, only 0.61% of the world's population resides in Gulf Cooperation Council (GCC) countries, but they contribute about 2.4% of the total global greenhouse gas (GHG) emissions [12]. In the present scenario, the economic growth of GCC states is better represented by their gross domestic product (GDP) and per capita energy consumption, which are much higher than in other developing countries [13]. Most of these countries’ domestic energy needs are served by fossil fuels, which eventually are responsible for their high per capita GHG emissions. Furthermore, it is significant that GCC states stand in the top 25 countries for CO2 emissions per capita in global rankings [14]. Growing energy consumption in conjunction with environmental threats poses a practical challenge for GCC countries. There is urgent need to plan renewable energy (RE) technologies which can combat the future challenges. In this context, GCC has a unique opportunity to channel more investment projects in placement of RE technologies to enhance future energy security and cutting off CO2 emissions. So, the aim of this study is to offers a discussion to policymakers to develop comprehensive and conversation energy policies to achieve long term sustainability in GCC energy systems. Thus, it is appropriate to investigate the relationship and causality among economic growth, financial development, CO2 emissions, and energy consumption in one framework. The structure of this paper is as follows. Section 2 contains an overview of GCC countries. Section 3 summarizes the previous studies and Section 4 discusses the estimation methodology. Section 5 provides the empirical analysis, which includes data and the estimation results. Lastly, conclusions and policy implications are presented in Section 6. 2. GCC context A combination of brisk economic expansion and population growth is fueling a rapid increase in energy demand in the GCC countries [16]. Their energy consumption has grown 74% since 2000 and is projected to nearly double its current level by 2020 [17]. The GCC countries are facing a dual challenge; one aspect involves having to maintain the domestic energy demand and the other involves controlling emissions. All the GCC nations are projected to experience a substantial rise in energy demand, with Qatar leading the others in its energy-demand growth rate (its share of GCC energy demand is projected to increase from around 10% to 15% between 2010 and 2020). Qatar is estimated to have earned $55 billion from net oil exports and this sector accounted for 57.8% of the country's GDP in 2012 [18]. The GCC countries present an increasingly attractive opportunity for industrial and energy companies looking to take advantage of low-cost hydrocarbon inputs and favorable tax regimes for the production of higher value products such as steel, aluminum, refined fuels, petrochemicals, and plastics. The recent growth in GCC energy demand and electricity demand, in particular, has resulted in rising global concern over CO2 emissions and climate change. In the absence of carbon-reduction technologies, alternative sources of energy, and significant energy efficiency measures, CO2 emissions will continue to increase. http://dx.doi.org/10.1016/j.rser.2016.11.089

1364-0321/ © 2016 Elsevier Ltd. All rights reserved.

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Fig. 1. Energy use and CO2 emissions by GCC Countries (1990–2011). Source: World Bank Indicators (2014): Retreived from: http://data.worldbank.org/country.

Overall, only 0.6% of the global population is living in GCC countries, but the region contributes 2.4% of the global greenhouse gas emissions. CO2 emissions per capita, energy intensities, and CO2 emissions per GDP in the GCC countries are higher than the average of 25 European Union (EU) countries and Organization for Economic Cooperation and Development (OECD) countries. Considering all the above, it is clear that energy efficiency can be improved in the region [19]. This resource-rich group (the GCC countries) is characterized by extremely high levels of energy use and carbon emissions per capita. The reason behind such high levels is the scarcity of water. These countries lack sufficient renewable water resources within their borders and must depend on other non-renewable water sources to satisfy the needs of their growing populations. In most cases, especially in Saudi Arabia and UAE, they have adopted desalination of sea water as a solution to the water challenges. However, given the energy-intensive nature of the desalination process, they have to rely on fossil fuels to operate the desalination plants, which add to their energy demand and subsequently high emissions. Abu Dhabi is estimated to devote more than half its domestic energy use to desalination [20]. The trends in energy consumption and emissions in all the GCC countries are close to each other (see Fig. 1). The GCC economies have benefitted from historically high oil prices and expanded oil production, with expansionary fiscal policies and low interest rates providing additional stimulus. However, nominal GDP increases resulting from higher oil prices – assuming that they are not shortlived – represent “real” income and are not just an expression of inflationary price increases. In 2009, for example, the GCC countries’ total real GDP increased marginally, while their combined nominal GDP measured in U.S. dollars fell by almost 19%, primarily as a result of lower oil prices. The GCC countries are leading the regional recovery as oil prices have rebounded and the GCC financial sector is stabilizing. Also, they posted strong economic growth at 7.2% and 6.0% for 2011 and 2012, respectively, but in 2013 the pace slowed a bit to 4%. While the 2009 collapse in oil prices led to a sharp decline in both nominal oil GDP and fiscal revenues in that year, continued increases in government spending kept real non-oil GDP growth positive in all the GCC countries except Kuwait and Oman. Clearly, in these countries, positive growth in real non-oil GDP does not preclude a major decline in real income, or vice versa. Fig. 2 depicts the increase in GDP per capita from 1995 through 2013. In addition, GCC countries make up the most financially developed region among the group of oil exporters, with the largest energy reserves and 118

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Fig. 2. GDP per capita by GCC Countries (1995–2013). Source: World Bank Indicators (2014): Retreived from: http://data.worldbank.org/country.

a positive correlation between oil revenues and financial development [21]. This is justified by theory such that greater oil production facilitates greater finance in two ways. First, oil trades are largely conducted in dollars, the capital needs in oil production are also large, and drilling projects are usually long term. As a result, oil-intensive economies will typically demand more financial transactions than other countries. Second, oil production contributes to GDP, and richer countries in general are involved in greater amounts of financial activity. 3. Literature review Exploration of the energy-environment-economy nexus has attracted attention from researchers in different countries over time. Generally, past studies in this field can be divided into two lines of research [22]. The first focuses on the relationship between economic growth and energy consumption, which dates back to the pioneering work of [23]. The second line of research focuses on the relationship between economic growth and the environment and is closely related to examining the inverted U-shaped relationship between environmental pollutants and economic growth that is, testing the validity of the environmental Kuznets curve (EKC) hypothesis [24–26]. Based on the previous literature, we divided past studies into subsections in the literature review, as follows: Section 3.1 summarizes the findings on energy and growth nexus; Section 3.2 highlights the relationship among CO2 emissions, GDP growth, and energy consumption; Section 3.3 elaborates the findings for CO2 emissions, energy consumption, and financial development; and Section 3.4 contains the findings for GCC countries. 3.1. Studies on energy consumption and GDP growth Payne [27,28] presented a detailed survey of the energy-growth and electricity-growth literature. Recently [29], explained the complex relationship between energy use and the economic process; issues such as the scarcity of energy resources, energy theory of value, and growth approaches are closely related to the relationship between energy and development. They traced the implications of the energy-GDP causality dialogue in the context of the growth debate, where the energy-development link plays a decisive role. They also investigated the possible existence of a fundamental “macro” direction of causality between energy use and economic growth that is not influenced by study-specific characteristics or events; they used a meta-analysis of 158 studies on causality between energy and GDP, covering 1978 through 2011. Furthermore, Menegaki [30] also discussed a meta-analysis of 51 studies published in the last two decades, with worldwide data since 1949, on the relationship between energy consumption and GDP growth. His results yielded evidence that the long-run elasticity of GDP growth with respect to energy consumption is not independent of the method and data employed for co-integration. Shahbaz et al. [31] investigated the relationship between energy (renewable and nonrenewable) consumption and economic growth using the Cobb–Douglas production function in Pakistan over the period of 1972 through 2011. They found co-integration among renewable energy consumption, nonrenewable energy consumption, economic growth, capital, and labor. However [32], found strong evidence of a bidirectional relationship between energy consumption and GDP in Lebanon, in both the short run and the long run. This indicated that energy is a limiting factor in economic growth in Lebanon. From a policy perspective, confirmation of the feedback hypothesis warns against the use of policy instruments geared toward restricting energy consumption, as they may lead to adverse effects on economic growth. Ozturk [33] presented a detailed energy-growth survey. Other studies have also highlighted this relationship. For example, Tang [34] discussed this nexus in Malaysia; Eggoh et al. [35] elaborated this relationship for African countries; Dergiades et al. [36] studied Greece; Araç and Hasanov [37] studied Turkey; Al-mulali et al.[38] investigated high income, upper middle income, lower middle income, and high income countries; Fuinhas and Marques [39] examined the interaction between primary energy consumption and growth in Portugal, Italy, Greece, Spain, and Turkey; Ouedraogo [40] examined the community of West African States (ECOWAS); Ocal and Aslan [41] looked at Turkey; Wang et al. [42] studied China; Gross [43] studied the US; Ohler and Fetters [44] studied 20 OECD countries; Apergis and Payne [45] studied Central America. 3.2. Studies on CO2 emissions, GDP growth, and energy consumption The second strand of research explained the growth-energy nexus while including emissions and Acaravci and Oztruk [46] examined the causal relationship between these variables by using the F-bound testing approach for 19 European countries. In addition, their Granger causality test indicated a bidirectional relationship between energy consumption and CO2. Recently, Saboori et al. [47] examined the causality relationship among CO2, energy consumption, and economic growth in OECD countries. Their result suggested a bidirectional relationship between CO2 and economic growth. Menyah and Wolde-Rufael [48,49] applied a modified version of the Granger causality test and found unidirectional causality running from pollutant emissions to economic growth, from energy consumption to economic growth, and from energy consumption to CO2 emissions, all without feedback. The study explored the causal relationship among CO2 emissions, renewable and nuclear energy consumption, and real GDP for the US over the 1960–2007 period. Using a modified version of the Granger causality test, they found unidirectional causality running from nuclear energy consumption to CO2 emissions without feedback but no causality running from renewable energy to CO2 emissions. Also, other researchers have highlighted this energy-growth and emissions relationship, see [50–53]. 119

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Furthermore, Almulali and Sab [54] investigated the impact of total primary energy consumption and CO2 emissions on economic development in 16 emerging countries. The main recommendation of that study was to increase investment and government spending on green energy projects to increase the share of green energy in these countries’ total energy consumption. This can be considered a good solution for their energy woes. Moreover, Almulali et al. [55] explored the bi-directional long-run relationship among energy consumption, CO2 emissions, and economic growth in Latin American and Caribbean countries, while Onafowora and Owoye [56] examined the long-run and dynamic temporal relationships among economic growth, energy consumption, population density, trade openness, and CO2 emissions in Brazil, China, Egypt, Japan, Mexico, Nigeria, South Korea, and South Africa based on the EKC hypothesis. The estimated results showed that the inverted U-shaped EKC hypothesis holds in Japan and South Korea. In the other six countries, the long-run relationship between economic growth and CO2 emissions follows an N-shaped trajectory and the estimated turning points are much higher than the sample mean. In addition, the results indicated that energy consumption Granger-causes both CO2 emissions and economic growth in all the countries. In Malaysia, Lau et al. [57] explored the co-integration relationship among CO2, exports, institutional quality, and economic growth and found a positive relationship between CO2 and institutional quality. Recently, the Malaysian economy has been the subject of other relevant studies on the relationship among emissions, energy consumption, and GDP [58–62]. 3.3. Studies on energy consumption, CO2, and financial development The third strand of literature included financial development regarding emissions and energy consumption. Among these studies, Tamazian et al. [63] established a link between financial development and environmental performance. The authors documented evidence that a wellfunctioning financial sector may provide higher financing at lower costs. This is also true for investment in environmental projects. Sadorsky [64] investigated the relationship between financial development and energy consumption for Central and Eastern European frontier economies. The Table 1 Summary of key findings in GCC Countries. Authors

Country

Variables

Methodology

Causality

Al-Iriani[72]

GCC

EC, GDP

LGDP→ LEC

Mehrara[73]

Oil exporting countries MENA

EC, GDP

Panel Cointegration Panel Cointegration Panel Cointegration

LOC↔ LCO2

Panel data analysis Panel cointegration, Panel causality.

LGDP→ LCO2

Al-Mulali[74]

CO2, GDP, oil consumption (OC) GDP, CO2, EC

LGDP→ LEC

Arouri et al. [75] Farhani and Rajeb[76]

12 MENA Countries MENA

Al-mulali and Lee[78]

GCC

FD, EC, GDP, URB, TR.

Panel data, Granger causality.

LEC ↔LGDP LFD ↔ LGDP LTD ↔LGDP LTD ↔LFD LTD ↔LUR

Hamdi et al. [77]

Bahrain

Electricity consumption, FDI, GDP

ARDL, VECM Granger causality

LEC ↔LGDP LEC ↔LFDI

Ozcan[78]

12 Middle East Countries GCC

GDP, CO2, EC

Panel data analysis

LGDP→ LCO2 LEC →LCO2

GDP, EC

Panel data analysis

LGDP→ LEC

Oil exporting countries MENA countries

GDP, EC

Panel cointegration

LEC →LGDP

GDP, CO2, EC

LGDP↔ LCO2

Al-mulalli and TANG[79] Damette andSeghir[80]) Omri[81].

Alkhathlan and Javid[82]. Tang and Abosedra[83],. Farhani et al. [84]. Sbia et al. [85].

EC, GDP, CO2

LGDP→ LCO2

Saudi Arabia

GDP, CO2

Simultaneous equations models ARDL

MENA countries

EC, Tourism, GDP

GMM estimator

LEC →LGDP

MENA countries UAE

CO2, Trade openness, GDP CO2, GDP, Trade openness, FDI, EC.

Panel data analysis ARDL

LCO2→LGDP

120

LEC →LGDP

LGDP→ LEC

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

result showed a positive and statistically significant relationship between financial development and energy consumption when financial development is measured using banking variables such as deposit money bank assets to GDP and financial system deposits to GDP. Jalil and Feridun [65] investigated the impact of financial development, economic growth, and energy consumption on environmental pollution in China from 1953 through 2006 using the autoregressive distributed lag (ARDL) model. The results revealed a negative sign for the coefficient of financial development, suggesting that financial development in China has not taken place at the expense of environmental health. Moreover, Oztruk and Acaravci [66] conducted an extended research study by including openness with financial development in the Turkish economy. Al-mulali and Sab [67] highlighted that energy consumption played an important role in increasing both economic growth and financial development but resulted in high pollution. The findings indicated that countries should increase energy productivity by increasing energy efficiency. This can be accomplished through implementation of energy-savings projects, energy conservation, and energy infrastructure outsourcing to achieve financial development and GDP growth in sub-Saharan African countries. Moreover, Islam et al. [68] suggested that financial development can reduce energy use by increasing energy efficiency in Malaysia. In further studies, Shahbaz and Lean [69] and Shahbaz et al. [70] demonstrated that financial development resulted in increased energy consumption and confirmed that energy use, financial development, and international trade have a positive impact on economic growth in China. In addition, Boutabba [71] argued that financial development has a long-run positive impact on carbon emissions, implying that financial development slows environmental degradation.

3.4. Studies in GCC countries Table 1 summarizes studies related to CO2 emissions, energy consumption, and GDP growth in GCC countries. From the above mentioned past studies review, we notice that various findings arise out of the use of different datasets, econometric methodologies, sample periods and countries. Therefore, we come to an understanding that there are limited numbers of studies for the GCC countries on this issue. So, the main contribution of this paper is to fill the gap in recent studies by providing fresh evidence to the cross-linkage between economic development, energy consumption and CO2 emissions by incorporating the role of financial element in effective long term renewable energy policies. The financial perspective can have brought various insights in this region due to rapid economic growth and sustainability targets.

4. Data and methodology Various approaches have been employed to examine the relationship between CO2 emissions and economic growth by modeling the relationship in an equilibrium framework with an aggregate growth model, see [86,87]. However, a different strand of studies has developed the relationship in a single equation model like the Cobb-Douglas production function [88,57 and 77]. Recently, Boutabba [71] and Shahbaz et al. [69] incorporated financial development as a prospective determinant of CO2 emissions to augment the single equation model. The main objective of the present paper is to investigate the long-run and short-run relationships between CO2 emissions and their determinants (Yt, ECt, and FDt) from 1980 through 2011 in GCC countries. Following the empirical literature (Section 3), it is plausible to construct the long-run relationship between the variables above in a linear form, with a view to testing the long-run, short-run, and causality relationships among these variables in GCC countries.

lnCO2t = β0 + α1 ln Yt + α2 lnECt + α3 lnFDt + εt

(1)

where CO2 represents the carbon dioxide emissions metric ton per capita; Y is the real GDP per capita (at constant price, 2005=$100 US) as a proxy for economic growth; EC is energy consumption; and FD is financial development as a proxy for domestic credit to the private sector as a percentage of GDP, obtained from the banking sector, including the gross credit to various sectors with the exception of credit to the central government [68]. β0 is constant term; α1,…, α3 are the coefficients of the model; and εt is the error term. The study used annual data over the period of 1980–2011 for the GCC countries. The world Development Indicators prepared by World Bank [89] are the source of data to this study. All variables have been transformed into natural logarithms (ln) to help mobilize stationarity in the variance-covariance matrix [90]. Furthermore, as Lau et al. [57] suggested, the log-linear model specification can generate more efficient and more symmetric results and avoid heteroscedasticity problems. To investigate the long-run and short-run relationships among CO2 emissions and their determinants, several steps of the methodology were used. In economic and financial analysis, stationarity tests are conducted to determine the unit root test of time series data. However, the current study employed the augmented Dickey Fuller (ADF), Phillips-Perron (P-P), and Kwiatkowski, Phillips, Schmidt, and Shin (KPSS) tests to detect the level of stationarity at I(0), I(1), or I(d) [91–94]. In general, if the time series data are not stationary, the regression analysis would not be true; it would be a spurious regression [95]. Before processing a bounds testing approach, one needs to ensure that the variables are not at the I(2) stationary level to avoid spurious results [96]. Also, if the variables are stationary at I(2), the computed F-statistics are not valid because bounds testing is based on the assumption that the variables are stationary at I(0), I(1), or both. Therefore, implementation of a bounds test procedure might still be necessary to ensure that none of the variables is stationary at the I(2) level [97]. However, in an economic view, the variables will be co-integrated if they have a long-run or equilibrium relationship with each other [98]. Also, the co-integration test was applied to determine which model would be suitable for the current study. The Engle and Granger [99] and JohansenJuselius (J-J) [100] techniques require that all variables (regressors) in the system be stationary and with an equal order of integration. Pesaran et al. [96] developed a model to introduce a surrogate co-integration technique known as the ARDL bounds testing approach. This approach has many advantages over the previous co-integration techniques. First, it uses more appropriate considerations than the J-J and Engle-Granger techniques for testing the co-integration among variables in a small sample [101]. Comparatively, the Johansen co-integration techniques need a large data sample for validity. Second, the approach can be applied whether the underlying variables are purely I(0) or purely I(1), or mixed, while other co-integration techniques require that all the variables be integrated on the same order [96]. Third, the ARDL application allows the variables to have different optimal lags, while this is impossible with conventional co-integration procedures [102]. Finally, the approach has become increasingly popular in recent years [103,104]. Based on these advantages of the ARDL model, this study employed the bounds test to examine the equilibrium relationships among the variables. To investigate this relationship, the unrestricted error correction model (UECM), ARDL approach, was formulated for each variable for each country, as in Eq. (2). 121

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

⎡ β01 ⎤ ⎡ δ11 ⎡ lnCO2 ⎤ ⎢ ⎥ k ⎢ ⎢ ⎥ β02 ⎥ δ ⎢ lnY ⎥ = (1−B ) ⎢ + ∑ (1−B ) ⎢ 21 ⎢ ⎥ β ⎢ δ31 03 ⎢ lnEC ⎥ i =1 ⎢⎣ δ41 ⎣ lnFD ⎦t ⎢⎣ β ⎥⎦ 04 ⎡ η11 η12 η13 η14 ⎤ ⎢η η η η ⎥ + ⎢ η21 η22 η23 η24 ⎥ × ⎢ 31 32 33 34 ⎥ ⎣ η41 η42 η43 η44 ⎦

δ12 δ22 δ32 δ42

δ13 δ23 δ33 δ43

δ14 ⎤ ⎡ ln CO2 ⎤ ⎥ δ24 ⎥ ⎢ lnY ⎥ ⎥ ×⎢ δ34 ⎥ ⎢ ln EC ⎥ ⎥ δ44 ⎦ ⎣ ln FD ⎦t − i

⎡ ln CO2 ⎤ ⎡ ε1 ⎤ ⎢ ⎥ ⎢ε ⎥ ⎢ lnY ⎥ + ⎢ ε2 ⎥ 3 ⎢ ln EC ⎥ ⎢ ⎥ ⎣ ln FD ⎦t −1 ⎣ ε4 ⎦t

(2)

where (1-B) is the first difference operator; β01,…., β04 are the constant terms, δ11,…., δ44 represent the short-run coefficients; and η11,…., η44 represent the long-run coefficients. For testing the existence of a short-run relationship among the above variables (Eq. (2)), the H0 and H1 hypotheses were formulized as: H0: No short-run relationship; δij =0, and H1: There is a short-run relationship; δij ≠0). However, for testing the existence of a long-run relationship, the H0 and H1 hypotheses were formulated (H0: No long-run relationship; ηij =0, and H1: There is a long-run relationship; ηij ≠0), where i, j=1,.., 4. For accepting and rejecting decisions (H0 or H1), Pesaran et al. [96] confirmed the following procedures: If Fs > upper bound then reject H0 and the variables are co-integrated. If Fs < lower bound then accept H0 and the variables are not co-integrated. If Fs ≥ lower bound and ≤ upper bound then the decision is inconclusive. where: Fs is F-statistic. However, Eq. (3) is used to show the presence of a bidirectional or unidirectional Granger causality relationship, whether one variable causes the other variable or not, and examines the short- and long-term causality among the variables.

⎡ β01 ⎤ ⎡ ϕ11,1 ϕ12,1 ϕ13,1 ϕ14,1 ⎤ ⎡ ⎡ ΔlnCO2 ⎤ ⎥ Δ ln CO2 ⎤ ⎢ ⎥ ⎢ ⎢ ⎥ ⎥ β02 ⎥ ⎢ ϕ21,1 ϕ22,1 ϕ23,1 ϕ24,1⎥ ⎢ ΔlnY ⎢ lnY Δ ⎢ ⎥ = +⎢ ⎥ ⎢ Δ ln EC ⎥ + ...... ⎢ ⎥ β ϕ ϕ ϕ ϕ lnEC Δ 03 ⎢ ⎥ ⎥ ⎢ 31,1 32,1 33,1 34,1⎥ ⎢ ⎣ ΔlnFD ⎦t ⎢⎣ β ⎥⎦ ⎢ ϕ ⎣ Δ ln FD ⎦t −1 04 ⎣ 41,1 ϕ42,1 ϕ43,1 ϕ44,1 ⎥⎦ ⎡ ϕ11, m ϕ12, m ϕ13, m ϕ14, m ⎤ ⎡ δ1 ⎤ ⎡ ε1 ⎤ ⎢ ⎥ ⎡ Δ ln CO2 ⎤ ⎢ ⎥ ⎥ ⎢ε ⎥ ⎢ ϕ21, m ϕ22, m ϕ23, m ϕ24, m ⎥ ⎢ ΔlnY δ2 ⎥ ⎢ + ECTt −1 + ⎢ ε2 ⎥ ⎢ϕ ⎥ ⎢ Δ ln EC ⎥ 3 δ ϕ ϕ ϕ ⎢ ⎥ 3 ⎥ ⎢⎣ ε4 ⎥⎦ ⎢ 31, m 32, m 33, m 34, m ⎥ ⎢ ⎢ ⎥ t ⎢⎣ ϕ41, m ϕ42, m ϕ43, m ϕ44, m ⎥⎦ ⎣ Δ ln FD ⎦t − m ⎣ δ4 ⎦ (3) Table 2 Unit root tests. Source: Output of E.Views package, version 7.2. Models

lnCo2 lnY lnEC lnFD Δ lnCo2 Δ lnY Δ lnEC Δ lnFD Oman lnCo2 lnY lnEC lnFD Δ lnCo2 Δ lnY Δ lnEC Δ lnFD Qatar lnCo2 lnY lnEC lnFD Δ lnCo2 Δ lnY Δ lnEC Δ lnFD

KSA

UAE

ADF

P-P

KPSS

ADF

P-P

KPSS

−2.424 −4.888*** 0.220 −2.892* −5.714*** −3.097** −6.675*** −4.796***

−2.424 −4.831*** −1.523 −2.799* −5.713*** −3.101** −10.567*** −5.338***

0.173 0.319 0.728** 0.734** 0.134 0.456* 0.004 0.247

−2.846* −0.973 −1.710 −1.133 −11.830*** −3.504** −5.183*** −4.326***

0.477** 0.604** 0.188 0.682** 0.0318 0.149 0.601** 0.076

1.566 −3.629** −0.538 −1.666 −8.281*** −3.719*** −3.865*** −4.752***

2.001 −3.629** −0.079 −1.671 −7.624*** −3.584** −7.931*** −4.741***

0.718** 0.752*** 0.735** 0.673** 0.209 0.035 0.185 0.109

0.284 1.374 −2.225 −1.823 −3.148** −3.467** −6.780*** −5.064

0.460** 0.521** 0.350* 0.110 0.119 0.199 0.096 0.156

−1.555 0.628 −1.540 −3.203** −4.156*** −4.036*** −5.311*** −5.434***

−1.883 0.046 −1.639 −3.218** −4.105*** −4.105*** −5.314*** −5.433***

0.248 0.492** 0.405* 0.251 0.128 0.021 0.131 0.251

−2.877* −1.359 −1.474 −1.094 −5.765*** −3.512** −5.169*** −4.049*** Kuwait −0.624 1.434 −0.408 −1.798 −3.940*** −3.467** −3.458** −5.010*** Bahrain −2.382 −1.021 −2.545 −0.785 −7.080*** −4.624*** −8.583*** −4.826***

−2.319 −1.146 −2.622 −0.889 −7.733 −4.624**8 −9.124*** −4.787***

0.281 0.528** 0.341 0.583** 0.182 0.197 0.392* 0.112

Notes: (1) ***, **,* denotes the significant level of 1%, 5%, 10%, respectively; (2) For ADF and P-P tests, H0 = series has a unit root, while for KPSS test H0 = series is stationary; (3) Critical values for ADF test are: - 3.646 (1%), −2.954 (5%), −2.615 (10%); (4) Critical values for P-P test are: - 3.639 (1%), −2.951 (5%), −2.614 (10%); (5) Critical values for KPSS test are: 0.739 (1%), 0.463 (5%), 0.347 (10%).

122

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

where: β01,…, β04 are the constant terms; ϕ11,…, ϕ44 are the short-run coefficients for the variables; ECTt-1 are the error correction terms (long-run coefficient of the variables); δ1,…., δ4 are the coefficients of the error-correction terms; and ε1 t,…., ε4 t are the white noise error terms. The causal relationships can be investigated in two ways. The first is the short-run Granger causalities that are examined by the Wald test or F-statistic for the significance of the relevant ϕ coefficients of the first difference series. The second is the long-run causalities detected via the t-test for the significance of the relevant δ coefficients on the lagged error correction term. The presence of a significant relationship in I (1) variables provides evidence of the direction of the short-run causality, while the long-run causality is provided by the t-test of lagged ECT. Theoretically, it is possible for one Granger variable to cause the other; while in actual evidence, no causal relationship can be detected between the two variables. Eventually, the word (causality) according to Granger causality does not mean that movements of one variable cause movements of another; it means only a correlation between the current value of one variable and the past values of others [105].

5. Results and analaysis Table 2 presents the results of the ADF, P-P, and KPSS tests of the unit root test for GCC countries. The table shows that all the variables are stationary after the first difference. This can be seen by comparing the observed values (in absolute terms) of the ADF, PP, and KPSS test statistics with the critical values of the test statistics at the 1%, 5%, and 10% levels of significance. The results provide strong evidence of stationarity in both level and first difference. Therefore, the null hypothesis is accepted and it is sufficient to conclude that there is a unit root in the variables at the vrious levels except for some variables like Y in Saudi Arabia and Oman, FD in Saudi Araia and Qatar, and CO2 in UAE. Thus, the null hypothesis of non-stationarity is rejected and it is safe to conclude that the variables are stationary at I(0). On the other hand, the rest of the variables are stationary at I(1). This implies that the variables are mutually integrated in order zero and one, that is, I(0) and I(1), which represents a motivation to apply the ARDL approach. A long-run equilibrium relationship among carbon emissions per capita, energy consumption per capita, real GDP per capita, and financial development at the 1% significance level in Saudi Arabia, Oman, Kuwait, and Qatar have been found, but there is no long-run relationship among these variables in the case of UAE and the result is inconclusive for Bahrain (see Table 3). Table 3 co-integration results among the variables (F- bound testing with intercept). Models

FlnCo2 (lnCO2 / lnY, lnEC, lnFD) FlnY (lnY / ln CO2, lnEC, lnFD) FlnEC (lnEC / lnCO2, lnY, lnFD) FlnFD (lnFD / lnCO2, lnY, lnEC,)

KSA

UAE

F-statistic

Decision

F-statistic

Decision

4.933***

Cointegration

1.314

No cointegration

4.245**

Cointegration

12.209***

Co-integration

5.436***

Cointegration

2.077

No cointegration

0.745

No cointegration

1.122

No cointegration

Oman FlnCo2 (lnCO2 / lnY, lnEC, lnFD) FlnY (lnY / ln CO2, lnEC, lnFD) FlnEC (lnEC / lnCO2, lnY, lnFD) FlnFD (lnFD / lnCO2, lnY, lnEC,)

Kuwait

8.584***

Cointegration

5.289***

Co-integration

2.5804

No cointegration

3.853**

Co-integration

0.1886

No cointegration

1.124

No cointegration

5.000***

Cointegration

11.910***

Co-integration

Qatar FlnCo2 (lnCO2 / lnY, lnEC, lnFD) FlnY (lnY / ln CO2, lnEC, lnFD) FlnEC (lnEC / lnCO2, lnY, lnFD) FlnFD (lnFD / lnCO2, lnY, lnEC,)

Bahrain

12.566***

Cointegration

2.912

Inconclusive

1.131

No cointegration

0.108

No cointegration

9.944***

Cointegration

1.104

No cointegration

2.326

No cointegration

11.014***

Co-integration

Notes: (1) ***, ** denotes significant level of 1%, 5%, respectively; (2) Critical values for F-bound testing are: 4.223–5.928 (1%), 3.002–4.252 (5%), and 2.493–3.566 (10%) [106]. used with an intercept and no trend.

123

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Table 4 ARDL Results. Regressors

KSA

UAE

Oman

Kuwait

Qatar

Bahrain

Short-run coefficients Δln Y ΔlnEC ΔlnFD Constant ECTt−1

−0.013 0.483* 0.0619 0.868 −0.419***

0.060 −0.045 −0.155 2.366 −0.612***

−0.316 0.046 0.016 7.468 −0.422**

−0.440*** 0.843*** −0.233*** 0.099 −0.476***

0.217 0.430 0.162* −3.218 −0.188

0.096 0.582 −0.005 −2.867 −0.464***

Long-run coefficients lnY lnEC lnFD Constant DW-test R2-adjusted Serial correlation χ2(1) Hetroscedasticity χ2(1) Normality χ2(2) RSS

−0.024 0.040 0.147 2.069 1.701 0.558 1.826 [0.177] 0.879 [0.348] 2.28 [0.320] 31.631

0.098 −0.074 −0.253 3.862 1.763 0.412 2.390 [0.122] 0.502 [0.478] 29.14[0.000] 13.496

−0.106 0.641 0.491 −1.890 1.969 0.550 0.669 [0.413] 2.063 [0.151] 1.570[0.456] 25.456

0.926 −0.722 −0.075 0.935 2.017 0.979 0.004 [0.944] 0.597 [0.440] 0.49 [0.781] 20.540

−0.444 2.283 0.865 −17.079 1.650 0.785 1.148 [.284] 3.891 [.490] 0.345 [.841] 21.309

−0.207 0.781 0.250 6.767 1.967 0.516 0.175[.989] 0.26[0.608] 4.84[0.089] 0.255

Notes: (1) ***, **,* denotes significant level of 1%, 5%, 10%, respectively; (2) RSS = residual sum of square.

These results confirm that a long-run relationship can be detected in the CO2 models in KSA, Oman, Qatar, and Kuwait. In contrast, the result is inconclusive for Bahrain, with a co-integration in the FD model. For UAE, the co-integration is detected in the Y model, only without any evidence of co-integration in the CO2 model. The short- and long-term coefficients of all the variables in GCC countries are presented in Table 4. Table 5 VECM Granger causality test results. Short-run Causality lnY

ΔlnCO2 t ΔlnYt ΔlnECt ΔlnFDt

– 1.027 0.078 2.218

1.191 – 4.400** 4.934**

3.089* 0.056 – 1.006

3.118* 0.073 0.428 –

−0.419*** −0.297*** −0.404*** −0.258**

lnCO2→lnEC LnCO2→lnFD lnEC→lnY lnFD→lnY

UAE ΔlnCO2 t ΔlnYt ΔlnECt ΔlnFDt

– 2.198 0.090 3.073*

0.084 – 0.308*** 31.343

5.695** 5.131** – 12.193***

0.023 0.163** 0.092 –

−0.612*** −0.365*** −0.135** −0.206

lnCO2→lnEC lnY↔lnEC lnY→lnFD lnFD→lnCO2 lnFD→lnEC

Oman ΔlnCO2 t ΔlnYt ΔlnECt ΔlnFDt

– 4.161** 3.348* 3.576*

7.030*** – 2.214*** 1.273

2.080 0.776*** – 0.670

0.133 0.480** 3.183* –

−0.422** −0.296*** −0.675*** −0.400***

lnCO2↔lnY lnY↔lnEC lnY→lnFD lnEC→lnCO2 lnEC→lnFD lnFD→lnCO2

Kuwait ΔlnCO2 t ΔlnYt ΔlnECt ΔlnFDt

– 4.035** 2.407 5.425**

0.589*** – 1.273 0.953

4.628** 1.575 – 1.598

2.303 1.844 1.374 –

−0.476*** −0.079 −0.040 −0.481

lnCO2↔lnY lnCO2→lnEC lnFD→lnCO2

Qatar ΔlnCO2 t ΔlnYt ΔlnECt ΔlnFDt

– 1.027 0.078 2.218

1.191 – 4.400** 4.93473**

3.089* 0.056 – 1.00611

3.118* 0.073 0.428 –

−0.188 −0.027 −0.303*** −0.225**

lnCO2→lnEC lnCO2→lnFD lnEC→lnY lnFD→lnY

Bahrain ΔlnCO2 t ΔlnYt ΔlnECt ΔlnFDt

– 3.0295** 0.3848 1.2166

0.000 – 2.345 0.293

1.044 1.169 – 3.054**

0.698 2.456 1.266 –

−0.464*** −0.038 −0.695*** −0.204*

t-1

lnFD

Causality direction

lnCO2 t-1

Variables

lnEC

Long-run Causality ECTt-1 t-1

t-1

KSA

Notes: ***, **,* denotes significant level of 1%, 5%, 10%, respectively.

124

lnY→lnCO2 lnFD→lnEC

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Fig. 3. CUSUM and CUSUMQ tests for LSPI model with intercept.

Table 4 shows that the CO2 model points to significant relationships between CO2 emissions and energy consumption in KSA. This reflects the importance and impact of energy consumption on CO2. Moreover, the financial development, energy consumption, and economic growth variables play a significant role and strongly influence the CO2 emissions in Kuwait. For Qatar, the result indicates that the most important variable with a significant impact is financial development. These findings are consistent with [85,71,53 and 107]. In addition, Table 4 shows that all error correction term coefficients (ECTs) are negative and statistically significant at the 1% confidence level except for Qatar. These values indicate that any deviation from the short-run disequilibrium among the variables is corrected in each period to return to the long-run equilibrium level. Furthermore, they reveal that the rate of adjustment in returning to equilibrium for UAE is much faster than for the other GCC countries in absolute value. The diagnostic tests in our analysis suggest that error terms of short run models are normally distributed (with the exemption of UAE model); free from serial correlation, heteroscedasticity, and ARCH problems across all the six models. Ramsey reset further provides that the functional forms are well specified. Table 5 summarizes the multivariate causal relationships among the variables (VECM Granger causality). Unidirectional causality relationships run from CO2 to energy consumption in Saudi Arabia, UAE, and Qatar. In Oman and Kuwait, the results display bidirectional causality between the two variables, which is consistent with [71]. In addition, the results reveal unidirectional causality relationships running from CO2 to financial development in Saudi Arabia and Qatar, and for Bahrain the relationship runs from ecnomic growth to CO2. Another bidirectional relationship is found between CO2 and economic growth for Oman and Kuwait, which is consistent with [108,109, 81 and 84]. 125

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Plot of Cumulative Sum of Recursive Residuals

Plot of Cumulative Sum of Squares of Recursive Residuals

Plot of Cumulative Sum of Recursive Residuals

Plot of Cumulative Sum of Squares of Recursive Residuals

Plot of Cumulative Sum of Recursive Residuals

Plot of Cumulative Sum of Squares of Recursive Residuals

Fig. 3. (continued)

In addition, due to the structural changes in these economies, it is likely that macroeconomic series may be subject to one or multiple structural breaks. Thus, the stability of the short-run and long-run coefficients is checked through the CUSUM and the CUSUMSQ tests [110,111]. Fig. 3 presents the plots of CUSUM and CUSUMSQ tests for all GCC countries that fall inside the critical bounds of the 5% significance level. This implies that the estimated parameters are stable for all countries except Oman (CUSUMQ) over the 1996–2001 period. The results agree with pertinent literature, see [46,33,112,77 and 106]. To confirm the results of short- and long-run relationships, the credibility test of the impulse response function (IRF) was performed. Fig. 4 reveals the reaction of CO2 to one standard deviation shock in each variable of GDP, EC, and FD in GCC countries. Fig. 4 also shows the negative response in CO2 due to standard shocks stemming from EC in KSA until the fourth year, which then converts to a positive response after that; it responded negatively in UAE, Kuwait, and Qatar after the sixth, fourth, and eighth years, respectively. The response of CO2 to economic growth is positive until the tenth year in Oman and positive in KSA and Kuwait after the fifth year. In contrast, the CO2 responded negatively to economic growth in UAE and Qatar. The CO2 emission responds positively due to a standard shock in financial development in KSA, Kuwait, Oman, and Qatar, while it responds negatively in UAE until the tenth year (see Fig. 4).

6. Conclusions and policy implications This study explores the relationships among CO2 emissions, financial development, energy consumption, and economic growth in GCC 126

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Fig. 4. Impulse response function (IRF).

countries. To test the long-term relationship among the above variables, the study employed the ARDL bounds testing approach for the period of 1980–2011. Despite some variables being stable at I(0), the stationarity results indicated strong evidence that all variables are stationary after the first difference, I(1). The empirical results proved the existence of a long-term equilibrium relationship among CO2 and real GDP per capita, energy consumption, and financial development in all GCC countries except UAE. From the short-run disequilibrium among the variables is corrected in each period to return to the long-run equilibrium level. Furthermore, they reveal that the rate of adjustment in returning to equilibrium for UAE is much faster than for the other GCC countries in absolute value. While in the long run 1% significance level in Saudi Arabia, Oman, Kuwait, and Qatar have been found, but there is no long-run relationship among these variables in the case of UAE and the result is inconclusive for Bahrain. Furthermore, the Granger causality test suggests unidirectional causality running from CO2 to energy consumption in Saudi Arabia, UAE, and Qatar this means that in these countries the energy conservation policies will not affect environment. While Oman and Kuwait display bidirectional causality between these variables this displays feedback implication. Lastly, in Oman, a unidirectional relationship runs from financial development to CO2, which implies that FD plays a substantial role in CO2 growth. In addition, the results of the CUSUM and CUSUMQ tests confirm the stability of the CO2 models in all GCC countries. This is because the 127

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Fig. 4. (continued)

coefficients stay within the 5% critical lines. Thus, the results are consistent with previous studies [47,84,88,103,112–117]. Impulse response functions analysis revealed that there are certain CO2 shocks linking with EC in KSA specifically in fourth year and later become positive. The strong dynamic unidirectional causality results suggest that energy consumption leads to economic growth in Saudi Arabia and Qatar. This indicates that an increase in energy consumption could result in deterioration of environmental quality by increasing CO2 emissions in the country. These findings match our impulse response test result, which explains the negative response in CO2 due to shocks in energy use. Whereas, in the coming few years it also displays negative trend for UAE, Kuwait, and Qatar respectively. The findings of this study have important policy implications for GCC not only in terms of environmental perspective but also offering the allocation of financial resources for future planning. The GCC countries have recently adopted a more anticipatory approach to addressing environmental issues on the international, national, and regional levels. However, the remodel strategies have not yet resulted in the development of consistent policies on ecological modernization [14]. This paper can be seen as an evaluation of GCC policy to support economic growth by encouraging financial development and efficient use of energy. The strong dynamic co-integration results indicate that CO2 emissions are increasing in KSA, Oman, Qatar, and Kuwait. However, UAE and Bahrain are predicting different result in this context. In the case of UAE, the global carbon agenda was planned to reduce CO2 emissions by 30% by 2030, with the government offering attractive advantages for investors in clean energy.

128

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

Fig. 5. National financial strategies of the GCC countries. Source [118]:

Consequently, investments in renewable energy and clean energy have increased significantly in recent periods. Furthermore, UAE has ongoing Shams and Noor projects while the pioneering Masdar city relies on renewable energy for about 7% of its energy needs. Recently, Qatar was named to host the 2022 World Cup; this is expected to boost the penetration of renewable energy and the government has declared plans to cool the stadiums using solar power (6%). Considering that financial development as measured by domestic credit to the private sector as a share of GDP increases the demand for energy, the findings show that financial development has also resulted in environmental degradation in Oman, Kuwait, and Bahrain. Fig. 5 shows the GCC countries future vision looking lease oil dependence, financial, economic and diversification planning perspective. All GCC visionary policies show that the governments of these countries are already on the right path to improve the economies by implementing extensive strategies by encouraging financial development. This can also help them to achieve long run economic growth and sustainable energy. Considering Fig. 5, it can elaborate individual country policy measures in GCC region. Firstly, Saudi Arabia has the longest and most elaborate tradition of planning among the GCC countries. In 2004, a Long-Term Strategy, which spans the time period of 2005–24, was published to raise the national economy to the level of advanced economies, which implies doubling the per capita income. It was established in response to notably the emerging challenges of providing productive employment to Saudi national manpower and improving the quality of life. There is also upmost desire to increase the role of non-oil production in the economy and increase diversification and expansion. Moreover, the Ninth Development Plan tackled these problems by enhancing private sector participation in the country but opening new investment plans and openness. Secondly, in case of UAE the government launched vision 2021 having series of aims for the development of economy in 2010. It aims to become one of the best places in the world to do business and have string path to sustainable development in a future that will less reliant on oil. This means expanding new 129

Renewable and Sustainable Energy Reviews 70 (2017) 117–132

strategic sectors to channel their energies into industries and services where can build a long-term competitive advantage. A step ahead, government encourages innovation, research and development in energy sector to strengthen the regulatory framework of producing high-value added sectors in order to enhance the looking ahead perspective. Thirdly, Bahrain Vision 2030 was prepared by the Economic Development Board (EDB), outlines a future path for the development of the Bahraini economy to build on a much-increased role for the private sector. The aim is to increase productivity in the private sector and create an environment which is conducive to entrepreneurship and innovation, and which will create knowledge-based and high-value-adding companies and economic activities. This can help to bring new technical innovation in energy pathways by locating investment projects. Fourthly, in Kuwait the aim of the State Vision 2035 plan is to turn Kuwait into a regional trade and financial hub for the northern Gulf through economic development, diversification and GDP growth. It also fosters the involvement of private sector investment which can further enhance the state-led development. Fifthly, looking Oman Vision 2020 pursues the aim of ‘providing suitable conditions for economic take off’, which implies diversification by increasing the non-oil production in the country. It also aims to achieve substantial changes in the structure of the national economy by diversifying the production base, enhancing the role of the private sector in the economy and developing human resources. Concerning improvement in the investment environment is the diversification strategy to liberalize of foreign ownership policies. Furthermore, there is emphasized to be a free market economy and an actives private sector. This, in combination with rapidly depleting oil resources, makes it a matter of urgency for Oman to implement a diversified and production-oriented policy. Lastly, Qatar Vision 2030 with long term development goals encompasses not only detailed plans for specific projects and initiatives, but also economic and institutional changes. Development within the economic realm is to be guided by three overriding policies: sound economic management, responsible exploitation of oil and gas, and suitable economic diversification. There is ongoing planning to enable private sector for increased involvement in the future economic, social and environmental investment projects. Looking all the above policies regarding GCC it can also help them to attain commitment with the United Nations Framework Convention on Climate Change (UNFCCC) and the Kyoto Protocol, Bhatto et al. [119] mentioned that this liability will reduce greenhouse gas emissions in the GCC countries in near future. One important suggestion is that because energy prices are heavily subsidized by the Saudi government and these energy price distortions are primarily responsible for the implausibly high energy intensity in the country, the Saudi government can effectively implement an energy conservation policy through energy price reforms and fuel substitution. Financial development performance shows causality running from FD to emissions in UAE, Oman, and Kuwait. Future research on energy consumption and the real GDP nexus in GCC countries can investigate this nexus in different sectors of the economy, such as agriculture, transport, commerce, industry, and households. Such studies can contribute to energy policy design as they would form a micro foundation for the aggregate macro economy. Researchers can also apply unit root testing with single and two unknown structural breaks in the series developed using the newly developed co-integration approach by Bayer and Hanck [120] to reexamine the relationships among energy use, financial development, carbon emissions, and economic growth in GCC countries using sectoral data. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37]

Huang BN, Hwang MJ, Yang CW. Causal relationship between energy consumption and GDP growth revisited: a dynamic panel data approach. Ecol Econ 2008;67:41–54. Akbostanci E, Türüt-Aşik S, Tunç GI. The relationship between income and environment in Turkey: is there an environmental Kuznets curve?. Energy Policy 2009;37:861–7. Zhang XP, Cheng XM. Energy consumption, carbon emissions, and economic growth in China. Ecol Econ 2009;68:2706–12. Fodha M, Zaghdoud O. Economic growth and pollutant emissions in Tunisia: an empirical analysis of the environmental Kuznets curve. Energy Policy 2010;38:1150–6. Belke A, Dobnik F, Dreger C. Energy consumption and economic growth: new insights into the cointegration relationship. Energy Econ 2011;33:782–9. Hamit-Haggar M. Greenhouse gas emissions, energy consumption and economic growth: a panel cointegration analysis from Canadian industrial sector perspective. Energy Econ 2012;34:358–64. Wesseh PK, Jr, Zoumara B. Causal independence between energy consumption and economic growth in Liberia: evidence from a non-parametric bootstrapped causality test. Energy Policy 2012;50:518–27. Ghosh S, Kanjilal K. Long-term equilibrium relationship between urbanization, energy consumption and economic activity: empirical evidence from India. Energy 2014;66:324–31. Robalino-López A, García-Ramos J, Golpe AA, Mena-Nieto A. System dynamics modeling and the environmental Kuznets curve in Ecuador (1980–2025). Energy Policy 2014;67:923–31. Mensah JT. Carbon emissions, energy consumption and output: a threshold analysis on the causal dynamics in emerging African economies. Energy Policy 2014;70:172–82. PBL Netherlands Environmental Assessment Agency. Trends in Global CO2 Emissions: 2013 Report, ISBN: 978-94-91506-51-2; 2013. BP. BP Statistical Review of World Energy, June 2008, London; 2008. Qader MR. Electricity consumption and GHG emissions in GCC countries. Energies 2009;2:1201–13. Reiche D. Renewable Energy Policies in the Gulf countries: a case study of the carbon-neutral ‘‘Masdar City’’ in Abu Dhabi. Energy Policy 2010;38:378–82. World Bank. World Development Indicators for Population Growth. Accessed at: 〈http://data.worldbank.org/indicator/SP.POP.GROW〉; 2011. Kinninmont J. The GCC in 2020: Resources for the Future. EconomistIntelligence Unit. 2010; 2010. EIA. US Energy Information Administration, Retrieved from: 〈www.eia.gov/countries/analysisbriefs/Qatar/qatar.pdf〉; 2014. Doukas H, Patlitzianas KD, Kagiannas A, Psarras J. Renewable energy sources and rational use of energy development in the countries of the GCC: myth or reality?. Renew Energy 2006;31:755–70. Elgendy K. Two Trends of Energy and Carbon Emissions in the Arab World. Carbon Middle East Sustainable Cities, Available at: 〈http://www.carboun.com/energy〉; 2011. Knoop T. Global finance in emerging market economies. Routledge: eBook Publication; 2013. p. 183–6. Soytas U, Sari R. Energy consumption, economic growth, and carbon emissions: challenges faced by an EU candidate member. Ecol Econ 2009;68:1667–75. Kraft J, Kraft A. On the relationship between energy and GNP. J Energy Dev 1978;3:401–3. Selden T, Song D. Environmental quality and development: is there a Kuznets curve for air pollution emissions?. J Environ Econ Manag 1994;27:147–62. Friedl B, Getzner M. Determinants of CO2 emissions in a small open economy. Ecol Econ 2003;45:133–48. Galeotti M, Lanza A. Desperately seeking environmental Kuznets. Environ Model Softw 2005;20:1379–88. Payne JE. Survey of the international evidence on the causal relationship between energy consumption and growth. J Econ Stud 2010;37:53–95. Payne JE. A survey of the electricity consumption–growth literature. Appl Energy 2010;87:723–31. Kalimeris P, Richardson C, Bithas K. A meta-analysis investigation of the direction of the energy-GDP causal relationship: implications for the growth-degrowth dialogue. J Clean Prod 2014;67:1–13. Menegaki AN. On energy consumption and GDP studies; a meta-analysis of the last two decade. Renew Sustain Energy Rev 2014;29:31–6. Shahbaz M, Zeshan M, Afza T. Is energy consumption effective to spur economic growth in Pakistan? New evidence from bounds test to level relationships and Granger causality tests. Econ Model 2012;29:2310–9. Dagher L, Yacoubian T. The causal relationship between energy consumption and economic growth in Lebanon. Energy Policy 2012;50:795–801. Ozturk I, Acaravci A. CO2 emissions, energy consumption and economic growth in Turkey. Renew Sustain Energy Rev 2010;14:3220–5. Tang CF. A re-examination of the relationship between electricity consumption and economic growth in Malaysia. Energy Policy 2008;36:3077–85. Eggoh JC, Bangake C, Rault C. Energy consumption and economic growth revisited in African countries. Energy Policy 2011;39:7408–21. Dergiades T, Martinopoulos G, Tsoulfidis L. Energy consumption and economic growth: parametric and non-parametric causality testing for the case of Greece. Energy Econ 2013;36:686–97. Araç A, Hasanov M. Asymmetries in the dynamic interrelationship between energy consumption and economic growth: evidence from Turkey. Energy Econ 2014;44:259–69.

130

Renewable and Sustainable Energy Reviews 70 (2017) 117–132 [38] Al-mulali U, Lee JYM. Estimating the impact of the financial development on energy consumption: evidence from the GCC (Gulf Cooperation Council) countries. Energy 2013;60:215–21. [39] Fuinhas JA, Marques AC. Energy consumption and economic growth nexus in Portugal, Italy, Greece, Spain and Turkey: an ARDL bounds test approach (1965–2009). Energy Econ 2012;34:511–7. [40] Ouedraogo NS. Energy consumption and economic growth: evidence from the economic community of West African States (ECOWAS). Energy Econ 2013;36:637–47. [41] Ocal O, Aslan A. Renewable energy consumption–economic growth nexus in Turkey. Renew Sustain Energy Rev 2013;28:494–9. [42] Wang Y, Wang Y, Zhou J, Lu G. Energy consumption and economic growth in China: a multivariate causality test. Energy Policy 2011;39:4399–406. [43] Gross C. Explaining the (non-) causality between energy and economic growth in the U.S.—a multivariate sectoral analysis. Energy Econ 2012;34:489–99. [44] Ohler A, Fetters I. The causal relationship between renewable electricity generation and GDP growth: a study of energy sources. Energy Econ 2014;43:125–39. [45] Apergis N, Payne JE. The renewable energy consumption–growth nexus in Central America. Appl Energy 2011;88:343–7. [46] Acaravci A, Ozturk I. On the relationship between energy consumption, CO2 emissions and economic growth in Europe. Energy 2010;35(12):5412–20. [47] Saboori B, Sapri M, bin Baba M. Economic growth, energy consumption and CO2 emissions in OECD (Organization for Economic Co-operation and Development)'s transport sector: a fully modified bi-directional relationship approach. Energy 2014;66:150–61. [48] Menyah K, Wolde-Rufael Y. Energy consumption, pollutant emissions and economic growth in South Africa. Energy Econ 2010;32:1374–82. [49] Menyah K, Wolde-Rufael Y. CO2 emissions, nuclear energy, renewable energy and economic growth in the US. Energy Policy 2010;38:2911–5. [50] Iwata H, Okada K, Samreth S. Empirical study on the environmental Kuznets curve for CO2 in France: the role of nuclear energy. Energy Policy 2010;38:4057–63. [51] Alam MJ, Begum IA, Buysse J, Huylenbroeck GV. Energy consumption, carbon emissions and economic growth nexus in Bangladesh: cointegration and dynamic causality analysis. Energy Policy 2012;45:217–25. [52] Shahbaz M, Lean HH. Does financial development increase energy consumption? The role of industrialization and urbanization in Tunisia. Energy Policy 2012;40:473–9. [53] Shahbaz M, Lean HH, Shabbir MS. Environmental Kuznets curve hypothesis in Pakistan: cointegration and Granger causality. Renew Sustain Energy Rev 2012;16:2947–53. [55] Al-mulali U, Fereidouni HG, Lee JY, Che Sab CNB. Examining the bi-directional long run relationship between renewable energy consumption and GDP growth. Renew Sustain Energy Rev 2013;22:209–22. [56] Onafowora OA, Owoye O. Bounds testing approach to analysis of the environment Kuznets curve hypothesis. Energy Econ 2014;44:47–62. [56] Al-mulali U, Sab CNBC. Energy consumption, pollution and economic development in 16 emerging countries. J Econ Stud 2013;40:686–98. [57] Lau LS, Choong CK, Eng YK. Investigation of the environmental Kuznets curve for carbon emissions in Malaysia: do foreign direct investment and trade matter?. Energy Policy 2014;68:490–7. [58] Bekhet HA, Yasmin T. Economic Growth and Pollutant Emission in Malaysia: An Empirical Analysis of Environmental Kuznets Curve. 16th AFBE Conference, Putrajaya, Malaysia, 370-387; 2012. [59] Bekhet HA, Yasmin T. Exploring EKC, trends of growth patterns and air pollutants concentration level in Malaysia: A Nemerow Index Approach. IOP Conference Series: Earth and Environmental Science 16, 1-4; 2013a. [60] Bekhet HA, Yasmin T. Disclosing the relationship among CO2 Emissions, Energy Consumption, Economic Growth and Bilateral Trade between Singapore and Malaysia: An Econometric Analysis, World Academy of Science, Engineering and Technology, International Conference on Sustainable Energy and Environmental Engineering Singapore, 1213th September; 2013b. [61] Bekhet HA, Yasmin T. Disclosing the relationship among CO2 emissions, energy consumption, economic growth and bilateral trade between Singapore and Malaysia: an econometric analysis. World Acad Sci Eng Technol, Int J Soc, Hum Sci Eng 2013;7:321–6. [62] Bekhet HA, Yasmin T. Analyzing energy consumption, carbon emissions, gross fixed capital formation and GDP in Malaysia: cointegration and causality Tests. In: Proceedings of the 6th international econimics & business manegement conference (IEBMC) Kuantan, 2-4 October, Malaysia; 2013d. [63] Tamazian A, Chousaa JP, Vadlamannatia KC. Does higher economic and financial development lead to environmental degradation: evidence from BRIC countries?. Energy Policy 2009;37:246–53. [64] Sadorsky P. Financial development and energy consumption in Central and Eastern European frontier economies. Energy Policy 2011;39:999–1006. [65] Jalil A, Feridun M. The impact of growth, energy and financial development on the environment in China: a cointegration analysis. Energy Econ 2011;33:284–91. [66] Ozturk I, Acaravci A. The long-run and causal analysis of energy, growth, openness and financial development on carbon emissions in Turkey. Energy Econ 2013;36:262–7. [67] Al-mulali U, Sab CNBC. The impact of energy consumption and CO2 emission on the economic growth and financial development in the Sub Saharan African countries. Energy 2012;39:180–6. [68] Islam F, Shahbaz M, Ahmed AU, Alam MM. Financial development and energy consumption nexus in Malaysia: a multivariate time series analysis. Econ Model 2013;30:435–41. [69] Shahbaz M, Khan S, Tahir MI. The dynamic links between energy consumption, economic growth, financial development and trade in China: fresh evidence from multivariate framework analysis. Energy Econ 2013;40:8–21. [70] Shahbaz M, Hye QMA, Tiwari AK, Leitão NC. Economic growth, energy consumption, financial development, international trade and CO2 emissions in Indonesia. Renew Sustain Energy Rev 2013;25:109–21. [71] Boutabba MA. The impact of financial development, income, energy and trade on carbon emissions: evidence from the Indian economy. Econ Model 2014;40:33–41. [72] Al-Iriani MA. Energy–GDP relationship revisited: anan example from GCC countries using panel causality. Energy Policy 2006;34:3342–50. [73] Mehrara M. Energy consumption and economic growth: the case of oil exporting countries. Energy Policy 2007;35:2939–45. [74] Al-Mulali U. Oil consumption, CO2 emission and economic growth in MENA countries. Energy Policy 2011;36:6165–71. [75] Arouri MEH, Youssef AB, M’henni H, Rault C. Energy consumption, economic growth and CO2 emissions in Middle East and North African countries. Energy Policy 2012;45:342–9. [76] Farhani S, Rajeb JB. Energy consumption, economic growth and CO2 emissions: evidence from panel data for MENA region. Int J Energy Econ Policy 2012;2:71–81. [77] Hamdi H, Sbia R, Shahbaz M. The nexus between electricity consumption and economic growth in Bahrain. Econ Model 2014;38:227–37. [78] Ozcan B. The nexus between carbon emissions, energy consumption and economic growth in Middle East countries: a panel data analysis. Energy Policy 2013;62:1138–47. [79] Al-mulali U, Tang CF. Investigating the validity of pollution haven hypothesis in the gulf cooperation council (GCC) countries. Energy Policy 2013;60:813–9. [80] Damette O, Seghir M. Energy as a driver of growth in oil exporting countries?. Energy Econ 2013;37:193–9. [81] Omri A. CO2 emissions, energy consumption and economic growth nexus in MENA countries: evidence from simultaneous equations models. Energy Econ 2013;40:657–64. [82] Alkhathlan K, Javid M. Energy consumption, carbon emissions and economic growth in Saudi Arabia: an aggregate and disaggregate analysis. Energy Policy 2013;62:1525–32. [83] Tang CF, Abosedra S. The impacts of tourism, energy consumption and political instability on economic growth in the MENA countries. Energy Policy 2014;68:458–64. [84] Farhani S, Shahbaz M, Sbia R, Chaibi A. What does MENA region initially need: grow output or mitigate CO2 emissions?. Econ Model 2014;38:270–81. [85] Sbia R, Shahbaz M, Hamdi H. A contribution of foreign direct investment, clean energy, trade openness, carbon emissions and economic growth to energy demand in UAE. Econ Model 2014;36:191–7. [86] Jorgenson DW, Wilcoxen PJ. Kneese AV, Sweeney JL, editors. Energy, the environment, and economic growth, 3. Handbook of Natural Resource and Energy Economics; 1993. p. 1267–349. [87] Kumbaroğlu G, Karali N, Arıkan Y. CO2, GDP and RET: an aggregate economic equilibrium analysis for Turkey. Energy Policy 2008;36:2694–708. [88] Shahbaz M, Mutascu M, Azim P. Environmental Kuznets curve in Romania and the role of energy consumption. Renew Sustain Energy Rev 2013;18:165–73. [89] World Bank. GDP Annual Growth Rate. Retreived from: data.worldbank.org/indicator; 2014. [90] Chang T, Fang W, Wen LF. Energy consumption, employment, output and temporal causality: evidence from Taiwan based on co-integration and error-correction modeling techniques. Appl Econ 2001;33:1045–56. [91] Dickey DA, Fuller WA. Distribution of the estimators for autoregressive time series with a unit root. J Am Stat Assoc 1979;74(366):427–31. [92] Dickey DA, Fuller WA. The likelihood ratio statistics for autoregressive time series with a unit root. Econom J Econom Soc 1981;49:1057–72. [93] Phillips PCB, Perron P. Testing for a unit root in time series regression. Biometrika 1988;75:335–46. [94] Kwon CS, Shin TS. Co-integration and causality between macroeconomic variables and stock market returns. Glob Financ J 1999;10(1):71–81. [95] Gujarati DN, Porter DC. , 5th ed.Basic econometrics. Higher education. New York: McGraw-Hill; 2009. [96] Pesaran MH, Shin Y, Smith RJ. Bound testing approaches to the analysis of level relationship. J Appl Econ 2001;16:289–326. [97] Ahmad I, Qayyum A. Effect of Government Spending and Macro-Economic Uncertainty on Private Investment in Services Sector: Evidence from Pakistan. Published in: European Journal of Economics, Finance and Administrative Sciences 11, 84-96; 2008. [98] Bekhet HA, Mugableh M. ‘Investigating equilibrium relationship between macroeconomic variables and Malaysian stock market index through bounds tests approach’. Int J Econ Financ 2012;4(10):69–81. [99] Engle RF, Granger CWJ. Co-integration and error-correction: representation, estimation and testing. Econometrica 1987;55:251–76. [100] Johansen S, Juselius K. Maximum likelihood estimation and inference on co-integration with application to money demand. Oxf Bull Econ Stat 1990;52(2):169–210. [101] Ghatak S, Siddiki J. The use of ARDL approach in estimating virtual exchange rate in India. J Appl Stat 2001;28:573–83.

131

Renewable and Sustainable Energy Reviews 70 (2017) 117–132 [102] Ozturk I, Acaravci A. Electricity consumption and real GDP causality nexus: evidence from ARDL bounds testing approach for 11 MENA countries. Appl Energy 2011;88:2885–92. [103] Jayaraman TK, Choong C. Growth and oil price: a study of causal relationships in small Pacific Island countries. Energy Policy 2009;37:2182–9. [104] Bekhet HA, Matar A. Risk-adjusted performance: a two-model approach application in Amman Stock Exchange. Int J Bus Soc Sci 2012;3:34–45. [105] Brooks C. Introductory econometrics for finance, 8th ed.. Cambridge: Cambridge University Press; 2008. [106] Narayan PK. The saving and investment nexus for China: evidence from co-integration tests. Appl Econ 2005;37:1979–90. [107] Shahbaz M, Nasreen S, Ling CH, Sbia R. Causality between trade openness and energy consumption: what causes what in high, middle and low income countries?. Energy Policy 2014;70:126–43. [108] Tiwari AK. Comparative performance of renewable and nonrenewable energy source on economic growth and CO2 emissions of Europe and Eurasian countries: a PVAR approach. Econ Bull 2011;31:2356–72. [109] Tiwari AK. A structural VAR analysis of renewable energy consumption, real GDP and CO2 emissions: evidence from India. Econ Bull 2011;31:1793–806. [110] Pesaran M, Pesaran B. Microfit 4.0 (Windows Version). New York: Oxford University Press Inc; 1997. [111] Bahmani-Oskooee M, Bohl MT. German monetary unification and the stability of the German M3 money demand function. Econ Lett 2000;66:203–8. [112] Shahbaz M, Ozturk I, Afza T, Ali A. Revisiting the environmental Kuznets curve in a global economy. Renew Sustain Energy Rev 2013;25:494–502. [113] Ghosh S. Examining carbon emissions economic growth nexus for India: a multivariate cointegration approach. Energy Policy 2010;38(6):3008–14. [114] Bekhet HA, Matar A. Causality of macroeconomic variables impacting the stock market index: time series approach in Amman Stock Exchange. Conference on Asian Forum on Business Education (AFBE). UNITEN, Selangor, Malaysia (July, 9–10/2012); 2012b. [115] Bekhet HA, Matar A. The influence of global financial crisis on Jordanian equity market: VECM approach. Int J Monet Econ Financ 2013;6:285–301. [116] Bekhet HA, Matar A. Co-integration and causality analysis between stock market prices and their determinates in Jordan. Econ Model 2013;35:508–14. [117] Al-mulali U, Lee JYM, Mohammed AK, Sheau-Ting L. Examining the link between energy consumption, carbon dioxide emission, and economic growth in Latin America and the Caribbean. Renew Sustain Energy Rev 2013;26:42–8. [118] Hvidt M. Economic Diversification in GCC Countries: Past Record and Future Trends. Retrieved from: 〈http://www.lse.ac.uk/middleEastCentre/kuwait/documents/economicdiversification-in-the-gcc-countries.pdf〉; 2013. [119] Bhutto AW, Bazmi AA, Zahedi G, Klemes JJ. A review of progress in renewable energy implementation in the Gulf Cooperation Council countries. J Clean Prod 2014;71:168–80. [120] Bayer C, Hanck C. Combining non‐cointegration tests. J Time Ser Anal 2013;34:83–95.

Hussain Ali Bekhet,, Ali Matar, Tahira Yasmin Graduate Business School, Universiti Tenaga Nasional (UNITEN), Malaysia Jadara university, Irbid 21110, Jordan College of Graduate Studies, Universiti Tenaga Nasional (UNITEN), Malaysia E-mail address: [email protected]

⁎

Corresponding author.

132