Natural gas consumption and economic growth nexus: Panel data analysis for GCC countries

Renewable and Sustainable Energy Reviews 51 (2015) 998–1003

Contents lists available at ScienceDirect

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

Natural gas consumption and economic growth nexus: Panel data analysis for GCC countries Ilhan Ozturk b,n, Usama Al-Mulali a a b

Faculty of Business, Multimedia University, 75450 Melaka, Malaysia Faculty of Economics and Administrative Sciences, Cag University, 33800 Mersin, Turkey

art ic l e i nf o

a b s t r a c t

Article history: Received 25 January 2015 Accepted 4 July 2015

This study investigates the relationship between natural gas energy consumption and economic growth by including trade openness, total labor force and gross fixed capital formation as a major determinants of GDP growth within the multivariate framework model in Gulf Cooperation Council (GCC) countries. A panel GDP model is constructed taking the period of 1980–2012. The result revealed that natural gas energy consumption is cointegrated with GDP growth in the investigated countries. In addition, based on the panel dynamic ordinary least square (DOLS) and the fully modified ordinary least square (FMOLS), this study concluded that the natural gas energy consumption affects the GCC’s countries GDP growth positively in the long run. Furthermore, the results from the Granger causality test revealed bidirectional causality between natural gas energy consumption and GDP growth which confirms the feedback hypothesis. From the outcome of this research, a number of policy implications were provided for the GCC countries. & 2015 Elsevier Ltd. All rights reserved.

Keywords: Natural gas energy consumption GDP growth GCC countries Panel data analysis

Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2. Literature review . . . . . . . . . . . . . . . . . . . . 3. Data, methodology and empirical results . 4. Conclusion and policy implications . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

1. Introduction The Gulf Cooperation Council (GCC) countries have witnessed a substantial increase in the economic growth which represented an average increase of 50% [80] in the last three decades. The economies of the GCC countries are fossil fuel (oil and natural gas) based economies. These natural resources represent more than 75% of total exports and government revenues. Moreover, the GCC countries produce more than 21% of the world oil production and have a share of 36% of the world oil reserves. Furthermore, the GCC countries are also one of the major natural gas producers in the world. The natural

n

Corresponding author. Tel./fax: þ 90 324 6514800. E-mail addresses: [email protected] (I. Ozturk), [email protected] (U. Al-Mulali). http://dx.doi.org/10.1016/j.rser.2015.07.005 1364-0321/& 2015 Elsevier Ltd. All rights reserved.

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. 998 . 999 . 999 1002 1002

gas production of these countries represents more than 11% of the world natural gas production. In addition, these countries hold more than 10% of the world natural gas reserves [31]. This abundance of fossil fuels made these resources a major source of energy in the GCC countries. Natural gas consumption plays a vital role in electricity generation in the GCC countries because more than 80% of its electricity comes from natural gas [80]. This source of energy is considered to be the cleanest fossil fuel energy because of its low production of CO2 emissions. Thus, it can be a good solution for reducing the levels of environmental pollution. In spite of the substantial literature that explored the energy–GDP relationship, none of the previous studies had investigated the natural gas energy consumption–GDP relationship in the GCC countries, even though they have witnessed a considerable increase in natural gas consumption in the last three decades. Therefore, the aim of this study is to fill this gap and

I. Ozturk, U. Al-Mulali / Renewable and Sustainable Energy Reviews 51 (2015) 998–1003

contribute literature by investigating the natural gas consumption and GDP growth relationship in the 7 GCC countries (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates (UAE)) for the 1980–2012 period.

2. Literature review Many studies explored the relationship between energy consumption (from renewable and non-renewable sources) and GDP growth. The conclusions of these studies varied. Some studies found that a bidirectional causality exists between energy consumption and GDP growth. This relationship is called the feedback hypothesis which denotes that energy consumption and GDP growth are mutually determined. Another group of studies indicated that there is a unidirectional causality from energy consumption to GDP growth. This relationship is termed the growth hypothesis which implies that conducting energy conservation might adversely affect GDP growth. A unidirectional causality from GDP growth to energy consumption was found in a number of studies. This relationship is called the conservation hypothesis. This hypothesis implies that energy conservation policies might adversely affect economic growth. Lastly, a number of studies found no causal relationship between the two variables. This phenomenon is called the neutrality hypothesis. This indicates that energy consumption and GDP growth are not correlated. Thus, the utilization conservation polices on energy consumption will have no affect economic growth [60]. The feedback hypothesis was found in a number of countries and regions such as Southeast Asian countries: Malawi [40], Malaysia [79], China [88,42,77,90] Taiwan [85], Bangladesh [3], Pakistan [72], in ten South east Asian countries [26], and South Korea [54,35]; in African countries such as Algeria [25], Burkina Faso [58], Tunisia [23], West African states [59], South Africa [56] and Nigeria [30]; in European countries such as Poland [36], Portugal, Italy, Greece, Spain and Turkey [33], Portugal [71], Eurasian countries [7,9] and Russia [89]; in Middle Eastern countries such as Lebanon [29] and the entire Middle East [53]. In addition, this relationship was also found in the Americas [15,69], Barbados [47], Pacific Island countries [48], in different developed and developing countries [11–13,17], emerging countries [16], and the OECD countries [10,27]. The growth hypothesis confirmed in a number of European countries [27], Turkey [5], Soviet Republics [24], and Greece [28]; in a number of Middle Eastern countries such as Israel [19], and Lebanon [1]; in South East Asian countries and ASEAN [44], Hong Kong [37], China [82,76,81], and Shanghai [83]; in a number of Pacific countries: Australia [75], and Fiji Islands [50]; in South American countries ([8], 2010); in African countries such as Tanzania [57]. Lastly, the growth hypothesis was found in the G7 [51] and OECD countries [52]. The conservation hypothesis was discovered by a number of studies in different countries such as Switzerland [21], Pakistan [39], in the GCC countries [6], New Zealand [22], in oil exporting countries, Bangladesh [49], India [34], South Korea [55] and Turkey [46]. Furthermore, the neutrality hypothesis was detected in Turkey [4,61], transition countries [2], in the US. [64,65,86,63], in 11 MENA countries [62], and Taiwan [84]. However, a limited number of researchers studied the link between natural gas energy consumption and GDP growth. For instance, Kum et al. [41] found a unidirectional causality from natural gas consumption to economic growth in the G-7 countries. Similar results were discovered in Pakistan by Shahbaz et al. [73]. However, Aqeel and Butt [18] and Siddiqui [78] did not find such relationship in Pakistan. In addition, Hossein et al. [38] also did not find any Granger causality relationship for all the OPEC countries in the long run. On the other hand, Sari et al. [70] found a unidirectional causality from economic growth to natural gas consumption for USA. Morever,

999

Apergis and Payne [14] found a bi-directional relationship between natural gas consumption and GDP growth in a panel of 67 countries. The same results were discovered by Lim and Yoo [45] in South Korea, Pakistan [74], Iran [87], and Taiwan [43]. The examined review of literature indicates the diverse conclusions that previous studies reached. The results obtained from the 83 studies are as follows: 52% of them indicated that feedback relationship between energy consumption and GDP growth exists, 28% of the studies showed the presence of the growth hypothesis, 10% of the studies indicated the existence of the conservation hypothesis, and 10% of the studies showed the presence of the neutrality hypothesis. It is crucial to note that the studies which investigated the energy–GDP growth relationship in the GCC countries are relatively limited. Moreover, there have been no studies that investigated the natural gas-GDP growth relationship in the GCC countries despite its prominent significance in electricity production. Thus, this study will explore the relationship between natural gas energy consumption and GDP growth in GCC countries and fill the gap in the energy economics literature.

3. Data, methodology and empirical results The sample period is 1980–2012 and annual data is utilized for 7 GCC countries which are Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates. Four variables are used in this study to build a GDP growth model for the GCC countries. Following Sadorsky’s [69] GDP growth model, this study used trade openness, total labor force and gross fixed capital formation as major determinants of GDP growth. Furthermore, natural gas energy consumption is used to examine the effect of natural gas energy consumption on the GCC countries’ GDP growth. The gross domestic product (GDP) is utilized as the dependent variable measured in millions of constant 2000 US dollars. In addition, four independent variables are utilized, namely, gross fixed capital formation (GFC) measured in millions of constant 2000 US dollars, total trade of goods and services (TD) as an indicator of openness measured in constant 2000 US dollars, total labor force (L) measured in thousands of workers and natural gas energy consumption (NG) measured in billion cubic feet. The data source for the GDP, TD, GFC and L are retrieved from the [32]. In addition, the data for NG is sourced from the [31]. Panel model is utilized because of a number of advantages such as, its capability to control the serial correlation and the individual heterogeneity which increases the degree of freedom and it is more reliable and efficient compared to the individual time series analysis [20]. The panel GDP growth model: GDP it ¼ f ðGFC it þ Lit þ TDit þNGit Þ

ð1Þ

Each variable is presented in its natural log. Moreover, the error term is added to the GDP growth model. The model can be written as follows: GDP it ¼ β 1i GFC it þ β2i L þ β3i TDit þ β 4i NGit þ εit

ð2Þ

The β1i, β2i, β3i and β4i represent the slop coefficients, i represent cross section (1…6 GCC countries), t is the time period (1980–2012), and ε is the error term. The initial step in the econometric analysis is to analyze the stationarity of the variables by using the two unit root tests, namely, the Fisher-ADF and Fisher-PP. These tests are considered by a combination of individual unit root tests to attain panel results. The PP and the ADF unit root tests work under the null and the alternative hypothesis. The former indicates that the variables have a panel unit root which indicates that they are not stationary. The latter explains that the variables are stationary, i.e. do not have

1000

I. Ozturk, U. Al-Mulali / Renewable and Sustainable Energy Reviews 51 (2015) 998–1003

a panel unit root. The panel unit root test, in general, became popular among researchers due to its high power compared to the normal times series unit root tests [20]. Table 1 reviews the unit root test results. All the variables are not significant at levels, which signify that the null hypothesis is not rejected. Therefore, the variables have a panel unit root. However, the variables were significant at the first difference, hence; the null hypothesis of panel unit root can be rejected. Consequently, the variables are considered stationary at the first difference. The cointegration test is utilized in this study as all the variables were found to be stationary at the first difference. The Pedroni [66,67] cointegration test is utilized to ascertain the long run relationship between the variables. Pedroni proposes several cointegration tests that allow heterogenous intercepts and trend coefficients across cross-sections. Based on the following regression: yit ¼ αi þ δit þ β1 x1;it þ β2 x2it þ ⋯ þ β ki xkit þ ϵit

ð3Þ

t¼1,…,T; i¼1,…,N; j¼1,…,k; and y and x are assumed to be integrated of order 1 i.e. I(1). The parameters αi and δi are individual entity and time effects respectively, which can be set to zero if desired. Under the null hypothesis of no cointegration, the residuals εit is I(1). Pedroni constructs various statistics for testing the null hypothesis of no cointegration: Panel v-statistics, panel p-statistics, panel t-statistic (non-parametric), panel t-statistic (parametric), Group ρ-statistics, Group t-statistics (nonparametric), and the Group statistics (parametric). The first four statistics are referred to as within-dimension or panel statistics test, and the remainder test are referred to as between dimension test. Table 2 reviews the results for the Pedroni cointegration test which indicate that four of the statistics were significant; hence, the null hypothesis of no cointegration is rejected. Moreover, the results also indicate that gross GDP growth, fixed capital formation, total trade of goods and services, total labor force, and natural gas energy consumption are cointegrated. Since the variables are cointegrated the study can proceed with the dynamic ordinary least square (DOLS) and fully modified OLS (FMOLS). The DOLS is a single cointegration equation which is Table 1 Panel unit root tests results. Panel I: ADF-Fisher Chi-square Variables

GDP L TD GFC NG

Level

First difference

Intercept

Intercept and trend

Intercept

Intercept and trend

0.13507 0.06192 0.20582 14.0523 14.8311

10.5037 0.76870 9.73244 11.4384 0.95232

45.1839nnn 25.5215nnn 43.3437nnn 54.5904nnn 139.042nnn

50.7564nnn 25.0326nnn 45.0501nnn 48.7668nnn 124.078nnn

Panel II: PP-Fisher Chi-square Variables Level Intercept GDP L TD GFC NG

0.14632 0.11674 0.17063 13.2317 14.3420

Intercept and trend 4.21914 0.78367 5.24661 14.2930 0.56278

First difference Intercept nnn

92.4861 126.675nnn 96.9886nnn 166.287nnn 157.033nnn

Intercept and trend 121.938nnn 140.597nnn 155.573nnn 676.157nnn 655.447nnn

The unit root tests were conducted with individual trends and intercept for each variable lag length were selected automatically using the Schwarz Information Criteria (SIC). nnn

Indicates statistical significance at the 1% levels.

Table 2 The results of Pedroni’s cointegration tests. Tests

Statistics

p-Values

Panel v-statistic Panel ρ-statistic Panel PP-statistic Panel ADF-statistic Group ρ-statistic Group PP-statistic Group ADF-statistic

1.228320  0.796385  2.800653nnn  2.686659nnn 0.708918  1.354382nnn  3.263770nnn

0.1097 0.2129 0.0025nnn 0.0036nnn 0.7608 0.0878nnn 0.0024nnn

nnn

Denotes significance at the 1% levels. Lag length and bandwidth are selected by Schwarz Information Criterion (SIC) and the Bartlett kernel Newey–West estimator.

designed to indicate the long run relationship between the independent and the dependent variables. The panel DOLS was proposed by Pedroni [67]. The DOLS method involves the expansion of the cointegration regression by creating lags and leads of Δ Xit resulting cointegration equation error term is orthogonal to the full history of the stochastic regressor innovations: r X

yit ¼ X 0it β þ D01it γ 1 þ

ΔX it þ j δ þ v1it

ð4Þ

j ¼ q

Adding q lags and r leads of the different regressors eliminate all the long run correlation between the residuals. The panel pooled FMOLS will be implemented to analysis the long run cointegration relationship between the dependent and the independent variables. The pooled FMOLS was proposed by Phillips and Moon [68]. This cointegrating regression is more able to prevent spurious regression generated from the involvement of the I(1) variables which cause misleading results. The pooled FMOLS estimator is presented below: !1 N X T N X T   X X þ ^β ¼ ~ ~ 0 X X 0 X~ y~  λ^ ð5Þ it

FP

i¼1t ¼1

it

it

it

12

i¼1t ¼1

where X~ it y~ it are the corresponding data removed from the individual deterministic trends and λ represents the cointegration regressors. It is important to know that the pooled FMOLS estimator simply sums across cross-sections separately in the numerator and denominator. Table 3 shows the DOLS test results. The results reveal that the gross fixed capital formation increases GDP growth significantly in two countries only, Kuwait and Qatar, while in the rest of the countries the variable was not significant. In addition, total trade has a positive significance for all the countries. The results show that the increase in total trade of goods and services by 1% will increase GDP in Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the UAE by 0.809479%, 1.017340%, 0.738366%, 0.968287%, 0.694709%, and 0.757457% respectively. The prominent significance of total trade in increasing GDP growth in these countries is due to the fact that trade is the largest contributor to the GDP growth because it constitutes an average of 95% of their total GDP. In addition, total labor force has a positive effect on GDP growth in three countries, namely, Oman, Saudi Arabia, and the UAE. The increase in total labor by 1% will increase GDP growth in Oman by 0.549313%, Saudi Arabia by 0.438392% and the UAE by 0.485438%. Furthermore, the natural gas energy consumption affects GDP growth positively in the all countries. 1% increase in natural gas energy consumption will increase GDP growth in Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the UAE by 0.509217%, 0.130749%, 0.143401%, 0.085060%, 0.139875%, and 0.664224%, respectively. Since this study uses the panel data methodology in the analysis, the policy implications will be based on the panel results. The panel results reveal that the gross fixed capital formation, total

I. Ozturk, U. Al-Mulali / Renewable and Sustainable Energy Reviews 51 (2015) 998–1003

1001

Table 3 The results of Panel group mean DOLS. Country

Dependent variable: GDP GFC

Bahrain Kuwait Oman Qatar Saudi Arabia UAE Group panel

 0.003033 0.017789nn 0.003532nn  0.004181 0.001300  0.006267 0.001523nn

TD (  1.060048) (2.854058) (2.520390) (  1.117959) (0.144171) (  1.038731) (2.301881)

L

0.809479nnn 1.017340nnn 0.738366nnn 0.968287nnn 0.694709nnn 0.757457nnn 0.830939 nnn

(10.99033) (26.97520) (26.17907) (38.44702) (14.56789) (28.70135) (24.31014)

NG

0.065846 0.193845 0.549313n 0.039284 0.438392nn 0.485438nn 0.295353

0.509217nnn 0.130749nn 0.143401nnn 0.085060nn 0.139875nn 0.664224nnn 0.945421nnn

(0.584732) (0.192813) (1.896352) (1.093284) (2.214930) (2.5482747) (1.4217309)

(4.826902) (2.491137) (6.332647) (2.558772) (2.622578) (10.15888) (4.831819)

Figures in the parenthesis ( ) are the t-statistics. nnn

Denotes significance at the 1% levels. Denotes significance at the 5% levels. Denotes significance at the 10% levels.

nn n

Table 4 The results of Panel group mean FMOLS. Country

Dependent variable: GDP GFC

Bahrain Kuwait Oman Qatar Saudi Arabia UAE Group panel

0.008228 (1.628701) 0.018729n(1.880035) 0.012642 nnn (7.203871) 0.886214 (1.633568) 0.008933 (0.190579) 0.023202 (1.389126) 0.159658nn (2.32098)

TD

L nnn

1.019925 1.079677nnn 0.697563nnn 0.941164nnn 0.681076nnn 0.759217nnn 0.863104nnn

(4.113593) (3.261828) (6.526267) (3.120125) (6.442399) (4.618733) (4.680491)

NG

0.0593928 0.2983059 0.4859324 0.005947 0.594930nn 0.285747nnn 0.288376

0.172723n 0.263533nnn 0.151975nnn 0.254395nn 0.230333nn 0.559897nn 0.272143nn

(1.048385) (0.584932) (1.209305) (0.578723) (2.385943) (3.28574) (1.515504)

(1.9272) (3.806739) (2.382831) (2.658702) (2.291580) (1.985938) (2.508832)

Figures in the parenthesis ( ) are the t-statistics. nnn

Denotes significance at the 1% levels. Denotes significance at the 5% levels. Denotes significance at the 10% levels.

nn n

trade, and natural gas energy consumption have a long run positive effect on the GCC countries’ GDP growth. 1% increase in gross fixed capital formation will increase GDP growth by 0.001523%, also the increase in total trade of goods and services by 1% will increase GDP growth by 0.830930%. The primary finding in the DOLS results is that natural gas energy consumption is significant in increasing GDP growth to the point that 1% increase in natural gas energy consumption will increase GDP growth in the investigated countries by 0.945421%. The results also showed that total labor force has no significant effect on the GDP growth in these countries. The results for the FMOLS results revealed in Table 4 were consistent with the DOLS outcomes. At the county level, the gross fixed capital formation has a significant positive effect only on Oman and Kuwait GDP growth, the increase in the gross fixed capital formation by 1% will increase Oman and Kuwait GDP growth by 0.018729% and 0.012642% respectively. Moreover, trade openness has a significant positive effect on all GGC’s GDP growth. The increase in trade openness by 1% will increase the GDP growth of Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the UAE by 1.019925%, 1.079677%, 0.697563%, 0.941164%, 0.681076%, 0.759217% respectively. However, labor force has a long run significant positive long run effect on GDP growth of Saudi Arabia and the UAE, the increase in labor force by 1% will increase GDP growth by 0.594930% and 0.285747% correspondingly. The FMOLS results also show that natural gas energy consumption increases GDP growth for all the investigated countries, the increase of natural gas energy consumption by 1% will increase the GDP growth of Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the UAE by 0.172723%, 0.263533%, 0.151975%, 0.254395%, 0.230333%, and 0.559897% correspondingly. In addition, based on the panel level the FMOLS results reveal that the gross fixed capital formation, trade openness, and natural gas energy consumption increases GDP growth

in the GCC countries. The increase of these variables by 1% will increase GDP growth by 0.159658%, 0.863104% and 0.272143%, respectively. However, labor has not long run effect on GDP in these countries. Granger causality based on the vector error-correction model (VECM) is utilized since the variables are cointegrated to show the causality between the variables. The VECM Granger causality can capture short run and long run relationship based on the Fstatistics and the lagged error correction term ect(  1). The VECM Granger causality model is represented in the following equation: 3 ΔGDP it 6 ΔGFC 7 6 it 7 7 6 6 ΔTDit 7 ¼ 7 6 6 ΔL 7 4 it 5 ΔNGit 2

2 3 β11p α1 6β 6α 7 6 21p 6 27 X r 6 7 6 6β 6 α3 7 þ 6 31p 7 6 6 α 7 p1 6 6 β41p 4 45 4 2

α5

2

β12p β22p β32p β42p β52p

β51p

φ1

3

2

ε1it

β14p β24p β34p β44p β54p

β15p 32 ΔGDP it  p 3 β25p 7 76 ΔGFC it  p 7 7 76 7 6 β35p 7 ΔTDit  p 7 76 7 76 7 6 7 Δ L β45p 54 it  p 5 ΔNGit  p β55p

3

6φ 7 6ε 7 6 27 6 2it 7 7 7 6 6 7 7 6 φ þ 6 3 7ect it  1 þ 6 6 ε3it 7 6φ 7 6ε 7 4 45 4 4it 5

φ5

β13p β23p β33p β43p β53p

ð6Þ

ε5it

The i represents the cross section (1…6 GCC countries), t donates the time (1980–2009), εit is the error term, and the ect is the lagged error correction term. Table 5 represents the Granger causality test results. In the short run, there is a feedback causality between GDP growth and gross fixed capital formation, total trade and GDP growth, gross fixed capital formation and total labor. In addition, a unidirectional causal relationship was found from gross fixed capital formation to total trade, total trade to total labor force, natural gas consumption to total trade. The main finding in the short run causality results show feedback causality between natural gas consumption and

1002

I. Ozturk, U. Al-Mulali / Renewable and Sustainable Energy Reviews 51 (2015) 998–1003

Table 5 The panel Granger causality tests results. The independent variables The short run causality ΔGDP ΔGDP ΔGFC ΔTD ΔL ΔNG

– 2.185230nn 158.1360nnn 0.438297 37.41216nnn

The long run causality

ΔGFC

ΔTD nnn

6.313612 – 6.240717nnn 2.584739nn 5.323556nnn

ΔL nnn

167.4415 1.632863 – 2.4539681nn 36.46632nnn

1.084574 2.198473n 0.785632 – 0.48322

ΔNG

ect (  1) nnn

41.94244 1.480746 43.23629nnn 0.567387 –

 3.426946nnn  3.525034nnn  2.338774nn  1.9859832n –3.679758nnn

nnn

Denotes statistical significance at the 1% levels. Denotes statistical significance at the 5% levels. Denotes statistical significance at the 10% levels.

nn n

GDP growth exists. This, in turn, represents the feedback hypothesis. Furthermore, there is a bi-directional long run causal relationship between all the variables based on the lagged error correction term ect(  1),

4. Conclusion and policy implications The aim of this study was to explore the natural gas energy consumption–GDP growth relationship in the 7 GCC countries (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates). The main motive behind investigating the relationship between natural gas energy consumption and GDP growth is due to lack of studies that explored this relationship in the GCC countries. In addition, it is important to investigate this relationship because natural gas energy consumption is significant due to its importance as a major source of energy in the GCC’s electricity power generation. To realize the goal of this study, a panel GDP model is constructed taking the period of 1980–2012. The results from the Pedroni cointegration test revealed that total trade of goods and services, total labor force, gross fixed capital formation and natural gas energy consumption are cointegrated with GDP growth. Moreover, the DOLS and the FMOLS results showed that these variables have a positive long run effect on the GDP growth in the GCC countries. Most importantly, the natural gas energy consumption has a significant positive long run effect on the GCC countries’s GDP growth. Furthermore, the results retrieved from the Granger causality revealed a bi-directional causal relationship between GDP growth and gross fixed capital formation, total trade and GDP growth, gross fixed capital formation and total labor. In addition, a unidirectional causal relationship was found from gross fixed capital formation to total trade, total trade to total labor force, natural gas consumption to total trade. The Granger causality test results also reveal bidirectional causality is found between natural gas consumption and GDP growth supporting the feedback hypothesis. This signifies that the natural gas energy consumption and GDP growth in the GCC countries effect and jointly determine each other in the same time. Since these countries have an abundant of natural gas reserves, this study recommends that the GCC countries should increase their investment on natural gas energy projects which will result in a larger share of natural gas energy consumption of the total energy consumption. This source of energy is important because it’s a clean source of energy compared to other fossil fuels. Thus, natural gas can be a good solution in reducing pollution and maintaining a clean and healthy environment in this region. Moreover, it is important that these countries should increase their investment on energy saving and renewable energy investments and reduce the amount of fossil fuels energy consumption that is used to achieve economic growth.

References [1] Abosedra S, Dah A, Ghosh S. Electricity consumption and economic growth, the case of Lebanon. Appl Energy 2009;86:429–32. [2] Acaravci A, Ozturk I. Electricity consumption-growth nexus: evidence from panel data for transition countries. Energy Econ 2010;32:604–8. [3] Ahamad MG, Islam AKMN. Electricity consumption and economic growth nexus in Bangladesh: revisited evidences. Energy Policy 2011;39:6145–50. [4] Altinay G, Karagol E. Structural break, unit root, and the causality between energy consumption and GDP in Turkey. Energy Econ 2004;26:985–94. [5] Altinay G, Karagol E. Electricity consumption and economic growth: evidence from Turkey. Energy Econ 2005;27:849–56. [6] Al-Iriani MA. Energy–GDP relationship revisited: an example from GCC countries using panel causality. Energy Policy 2006;34:3342–50. [7] Apergis N, Payne JE. Energy consumption and economic growth: evidence from the Commonwealth of Independent States. Energy Econ 2009;31:641–7. [8] Apergis N, Payne JE. Energy consumption and economic growth in Central America: evidence from a panel cointegration and error correction model. Energy Econ 2009;31:211–6. [9] Apergis N, Payne JE. Renewable energy consumption and growth in Eurasia. Energy Econ 2010;32:1392–7. [10] Apergis N, Payne JE. Renewable energy consumption and economic growth: evidence from a panel of OECD countries. Energy Policy 2010;38:656–60. [11] Apergis N, Payne JE. A panel study of nuclear energy consumption and economic growth. Energy Econ 2010;32:545–9. [12] Apergis N, Payne JE. Coal consumption and economic growth: evidence from a panel of OECD countries. Energy Policy 2010;38:1353–9. [13] Apergis N, Payne JE. Energy consumption and growth in South America: evidence from a panel error correction model. Energy Econ 2010;32:1421–6. [14] Apergis N, Payne JE. Natural gas consumption and economic growth: a panel investigation of 67 countries. Appl Energy 2010;87:2759–63. [15] Apergis N, Payne JE. The causal dynamics between coal consumption and growth: evidence from emerging market economies. Appl Energy 2011;88:5226–30. [16] Apergis N, Payne JE. The renewable energy consumption–growth nexus in Central America. Appl Energy 2011;88:343–7. [17] Apergis N, Payne JE. Renewable and non-renewable energy consumptiongrowth nexus: evidence from a panel error correction model. Energy Econ 2012;34:733–8. [18] Aqeel A, Butt MS. The relationship between energy consumption and economic growth in Pakistan. Asia-Pac Dev J 2001;8:101–10. [19] Aslan A, Çam S. Alternative and nuclear energy consumption economic growth nexus for Israel: evidence based on bootstrap-corrected causality tests. Prog Nucl Energy 2013;62:50–3. [20] Baltagi B. Econometric Analysis of Panel Data. John Wiley and Sons, Ltd; 2009. [21] Baranzini A, Weber S, Bareit M, Mathys NA. The causal relationship between energy use and economic growth in Switzerland. Energy Econ 2012;36:46–470. [22] Bartleet M, Gounder R. Energy consumption and economic growth in New Zealand: results of trivariate and multivariate models. Energy Policy 2010;38:3508–17. [23] Belloumi M. Energy consumption and GDP in Tunisia: cointegration and causality analysis. Energy Policy 2009;37:2745–53. [24] Bildirici ME, Kayıkçı F. Economic growth and electricity consumption in former Soviet Republics. Energy Econ 2012;34:747–53. [25] BélaÏd F, Abderrahmani F. Electricity consumption and economic growth in Algeria: a multivariate causality analysis in the presence of structural change. Energy Policy 2013;55:286–95. [26] Chen ST, Kuo HI, Chen CC. The relationship between GDP and electricity consumption in10 Asian countries. Energy Policy 2007;35:2611–21. [27] Coers R, Sanders M. The energy–GDP nexus; addressing an old question with new methods. Energy Econ 2013;36:708–15. [28] 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.

I. Ozturk, U. Al-Mulali / Renewable and Sustainable Energy Reviews 51 (2015) 998–1003

[29] Dagher L, Yacoubian T. The causal relationship between energy consumption and economic growth in Lebanon. Energy Policy 2012;50:795–801. [30] Ebohon OJ. Energy, economic growth and causality in developing countries. Energy Policy 1996;24:447–53. [31] Energy Information Administration (EIA). International energy statistics. Washington, DC. Available from: 〈http://www.eia.gov/countries/data.cfm〉; 2014 [accessed on 17.03.14]. [32] Euromonitor International from National Statistics. Global Market Information (GIMD). Available from 〈http://www.portal.euromonitor.com/portal/server.pt? space=Login&control=Redirect〉; 2014 [accessed on 20.07.14]. [33] 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. [34] Ghosh S. Electricity consumption and economic growth in India. Energy Policy 2002;30:125–9. [35] Glasure YU, Lee AR. Cointegration, error-correction, and the relationship between GDP and energy: the case of South Korea and Singapore. Resour Energy Econ 1997;20:17–25. [36] Gurgul H, Lach L. The electricity consumption versus economic growth of the Polish economy. Energy Econ 2012;34:500–10. [37] Ho CY, Siu KW. A dynamic equilibrium of electricity consumption and GDP in Hong Kong: an empirical investigation. Energy Policy 2007;35:2507–13. [38] Hossein A, Yazdan GF, Ehsan AG. The relationship between energy consumption, energy prices and economic growth: case study (OPEC countries). OPEC Energy Rev 2011;36(3):272–86. [39] Jamil F, Ahmad E. The relationship between electricity consumption, electricity prices and GDP in Pakistan. Energy Policy 2010;38:6016–25. [40] Jumbe CBL. Cointegration and causality between electricity consumption and GDP: empirical evidence from Malawi. Energy Econ 2004;26:61–8. [41] Kum H, Ocal O, Aslan A. The relationship among natural gas energy consumption, capital and economic growth: bootstrap-corrected causality tests from G-7 countries. Renewable Sustainable Energy Rev 2012;16:2361–5. [42] Lai TM, Lo WM, Choy WC, Lam YS, K.H.. The causal relationship between electricity consumption and economic growth in a Gaming and Tourism Center: the case of Macao SAR, the People’s Republic of China. Energy 2011;36:1134–42. [43] Lee CC, Chang CP. Structural breaks, energy consumption and economic growth revisited: evidence from Taiwan. Energy Econ 2005;27:857–72. [44] Lee CC, Chang CP. Energy consumption and economic growth in Asian economies: a more comprehensive analysis using panel data. Resour Energy Econ 2008;30:50–65. [45] Lim HJ, Yoo SH. Natural gas consumption and economic growth in Korea: a causality analysis. Energy Sources, Part B 2011;7:169–76. [46] Lise W, Montfort KV. Energy consumption and GDP in Turkey: is there a cointegration relationship? Energy Econ 2007;29:1166–78. [47] Lorde T, Waithe K, Francis B. The importance of electrical energy for economic growth in Barbados. Energy Econ 2010;32:1411–20. [48] Mishra V, Smyth R, Sharma S. The energy-GDP nexus: evidence from a panel of Pacific Island countries. Resour Energy Econ 2009;31:210–20. [49] Mozumder P, Marathe A. Causality relationship between electricity consumption and GDP in Bangladesh. Energy Policy 2007;35:395–402. [50] Narayan PK, Singh B. The electricity consumption and GDP nexus for the Fiji Islands. Energy Econ 2007;29:1141–50. [51] Narayan PK, Smyth R. Energy consumption and real GDP in G7 countries: new evidence from panel cointegration with structural breaks. Energy Econ 2008;30:2331–41. [52] Narayan PK, Prasad A. Electricity consumption–real GDP causality nexus: evidence from a boots trapped causality test for 30 OECD countries. Energy Policy 2008;36:910–8. [53] Narayan PK, Smyth R. Multivariate granger causality between electricity consumption, exports and GDP: evidence from a panel of Middle Eastern countries. Energy Policy 2009;37:229–36. [54] Oh W, Lee K. Energy consumption and economic growth in Korea: testing the causality relation. J Policy Model 2004;26:973–81. [55] Oh W, Lee K. Causal relationship between energy consumption and GDP revisited: the case of Korea 1970–1999. Energy Econ 2004;26:51–9. [56] Odhiambo NM. Electricity consumption and economic growth in South Africa: a trivariate causality test. Energy Econ 2009;31:635–40. [57] Odhiambo NM. Energy consumption and economic growth nexus in Tanzania: an ARDL bounds testing approach. Energy Policy 2009;37:617–22. [58] Ouedraogo IM. Electricity consumption and economic growth in Burkina Faso: a cointegration analysis. Energy Econ 2010;32:524–31. [59] Ouedraogo N. Energy consumption and economic growth: evidence from the economic community of West African States (ECOWAS). Energy Econ 2013;36: 637–47.

1003

[60] Ozturk I. A literature survey on energy-growth nexus. Energy Policy 2010;38 (1):340–9. [61] Ozturk I, Acaravci A. CO2 emissions, energy consumption and economic growth in Turkey. Renewable Sustainable Energy Rev 2010;14(9):3220–5. [62] 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. [63] Payne JE. On the dynamics of energy consumption and output in the US. Appl Energy 2009;86:575–7. [64] Payne JE, Taylor JP. Nuclear energy consumption and economic growth in the U.S.: an empirical note. Energy Sources, Part B 2010;5:301–7. [65] Payne JE. US disaggregate fossil fuel consumption and real GDP: an empirical note. Energy Sources, Part B 2011;6:63–8. [66] Pedroni P. Critical values for cointegration tests in heterogeneous panels with multiple regressors. Oxford Bull Econ Stat 1999;61:653–70. [67] Pedroni P. Panel cointegration; asymptotic and finite sample properties of pooled time seriestests with an application to the PPP hypothesis. Econometric Theory 2004;20:597–625. [68] Phillips PCB, Moon HR. Linear regression limit theory for nonstationary panel data. Econometrica 1999;67 1057–11. [69] Sadorsky P. Energy consumption, output and trade in South America. Energy Econ 2012;34:476–88. [70] Sari R, Ewing BT, Soytas U. The relationship between disaggregate energy consumption and industrial production in the United States: an ARDL approach. Energy Econ 2008;30:2302–13. [71] Shahbaz M, Tang CF, Shabbir MS. Electricity consumption and economic growth nexus in Portugal using cointegration and causality approaches. Energy Policy 2011;39:3529–36. [72] Shahbaz M, Lean HH. The dynamics of electricity consumption and economic growth: a revisit study of their causality in Pakistan. Energy 2012;39:146–53. [73] Shahbaz M, Lean HH, Farooq A. Natural gas consumption and economic growth in Pakistan. Renewable Sustainable Energy Rev 2013;18:87–94. [74] Shahbaz M, Arouiri M, Teulon F. Short- and long-run relationships between natural gas consumption and economic growth: evidence from Pakistan. Econ Modell 2014;41:219–26. [75] Shahiduzzaman Md, Alam K. Cointegration and causal relationships between energy consumption and output: assessing the evidence from Australia. Energy Econ 2012;34:2182–8. [76] Shengfeng X, sheng XM, tianxing Z, Xuelli Z. The relationship between electricity consumption and economic growth in China. Phys Procedia 2012;24:56–62. [77] Shuyun Y, Donghua Y. The causality between energy consumption and economic growth in china: using panel method in a multivariate frame work. Energy Procedia 2011;5:808–12. [78] Siddiqui R. Energy and economic growth in Pakistan. Pak Dev Rev 2004;43: 175–200. [79] Tang CF, Tan EC. Exploring the nexus of electricity consumption, economic growth, energy prices and technology innovation in Malaysia. Appl Energy 2013;104:297–305. [80] The World Bank. World Development Indicators. Washington, DC. Available from: 〈http://data.worldbank.org/data-catalog/world-development-indica tors〉; 2014 [accessed on 15.02.14]. [81] Wang Y, Wang Y, Zhou J, Zhu X, Lu G. Energy consumption and economic growth in China: a multivariate causality test. Energy Policy 2011;39: 4399–406. [82] Wei LZ, Xun-gang Z. Study on relationship of energy consumption and economic growth in China. Phys Procedia 2012;24:313–9. [83] Wolde-Rufael Y. Disaggregated industrial energy consumption and GDP: the case of Shanghai, 1952–1999. Energy Econ 2004;26:69–75. [84] Wolde-Rufael Y. Nuclear energy consumption and economic growth in Taiwan. Energy Sources, Part B 2012;7:21–7. [85] Yang HY. A note on the causal relationship between energy and GDP in Taiwan. Energy Econ 2000;22:309–17. [86] Yildirim Y, Sarac S, Aslan A. Energy consumption and economic growth in the USA: evidence from renewable energy. Renewable Sustainable Energy Rev 2012;16:6770–4. [87] Zamani M. Energy consumption and economic activities in Iran. Energy Econ 2007;29:1135–40. [88] Zhang C, Xu J. Retesting the causality between energy consumption and GDP in China: evidence from sectoral and regional analyses using dynamic panel data. Energy Econ 2012;34:1782–9. [89] Zhang YL. Interpreting the dynamic nexus between energy consumption and economic growth: empirical evidence from Russia. Energy Policy 2011;39: 2265–72. [90] Zhixin Z, Xin R. Causal relationships between energy consumption and economic growth. Energy Procedia 2011;5:2065–71.