Qual Quant DOI 10.1007/s11135-011-9468-3

Electricity consumption and economic growth empirical evidence from Pakistan Muhammad Shahbaz · Mete Feridun

© Springer Science+Business Media B.V. 2011

Abstract The present article uses the Autoregressive Distributed Lag (ARDL) bounds testing procedure to identify the long run equilibrium relationship between electricity consumption and economic growth. Toda Yamamoto and Wald-test causality tests have identified the direction of the causal relationship between these two variables in the case of Pakistan in the period between 1971 and 2008. Ng-Perron and Clement-Montanes-Reyes unit root tests are used to handle the problem of integrating orders for variables. The results suggest that the two variables are in a long run equilibrium relationship and economic growth leads to electricity consumption and not vice versa. Keywords

Electricity consumption · Economic growth · Pakistan · Causality

1 Introduction Pakistan is an economy where productive activities are restrained by its underdeveloped energy infrastructure. It is a frequent phenomenon that intentionally-engineered electrical power outages, i.e. load-sheddings are frequently used to meet increasing demand for electricity in the economy. Khan and Ahmed (2009) report that the current electricity production of Pakistan is around 11,500 MW per day, whereas the authors estimate the economy’s electricity need to jump from around 15,000 MW electricity per day to around 20,000 MW per day by 2010. This gap between the supply and demand in the energy sector in Pakistan suggests the magnitude of the energy crisis the economy faces.

M. Shahbaz (B) COMSATS Institute of Information Technology, M.A. Jinah Campus, Lahore, Pakistan e-mail: [email protected]; [email protected] M. Feridun Department of Banking and Finance, Faculty of Business and Economics, Eastern Mediterranean University, Gazi Magosa, Mersin 10, Turkey e-mail: [email protected]

123

M. Shahbaz, M. Feridun 36000 600

32000

500

28000

400

24000

300

20000

200

16000

100

12000

0 1975

1980

1985

1990

1995

2000

2005

Electricity consumption per capita (in KWH) Real GDP per capita

Fig. 1 Graphical representation of the data

It is against this backdrop that the present article aims at identifying the long run equilibrium relationship and the causal link between electricity consumption and economic growth in the case of Pakistan. Despite the plethora of studies in the related literature (see, for example, Kraft and Kraft 1978; Akarca and Long 1980; Erol and Yu 1987; Stern 1993; Asafu-Adjaye 2000; Oh and Lee 2004; Narayan and Singh 2007; Reynolds and Kolodziej 2008; Narayan and Prasad 2008; Wolde-Rufael 2004, 2009; Apergis and Payne 2009; Chandran et al. 2009; Bowden and Payne 2009; Soytas and Sari 2009), to date, there exists no empirical study in the literature on the nexus between electricity consumption and economic growth in the case of Pakistan. The present article aims at filling this gap in the literature. From a theoretical standpoint, the direction of the causal relationship between electricity consumption and economic growth is not very straightforward: as countries develop they will start relying more on manufacturing sectors hence need to consume more energy. On the other hand, increased use of energy may lead to more efficient production and, hence, to faster economic growth. In the face of the energy crisis in Pakistan, an investigation of the nature of the relationship between electricity consumption and economic growth in this country may be of interest to both policy makers and practitioners. An interesting aspect of the existing literature is that the relationship between energy consumption and growth may differ between short run and long run even within the same country. For this reason, the present article uses the Autoregressive Distributed Lag (ARDL) bounds testing procedure to identify the long run equilibrium relationship between electricity consumption and economic growth as well as causality tests to identify the direction of the causal relationship between these two variables using annual data from 1971 to 2008. The problem of unit root is handled by Ng-Perron and Clement-Montanes-Reyes unit root tests. The rest of the article is structured as follows. Section 2 reviews the literature. Section 3 introduces the data and methodology. Section 4 presents the empirical results, and Sect. 5 points out the conclusions that emerge from the study (Fig. 1).

2 Literature review The relationship between energy consumption and growth has been studied extensively in the literature. The seminal empirical study is by Kraft and Kraft (1978) who have found unidirectional causality from GNP to energy consumption using annual US data. In another early study, Erol and Yu (1987) have examined the GDP-Energy Consumption relationship

123

Electricity consumption and economic growth from Pakistan

for England, France, Italy, Germany, Canada and Japan and found diverse empirical evidence on the relation between energy consumption and economic growth. Stern (1993) used multivariate framework including capital and labor force into the model of energy consumption and GDP and Asafu-Adjaye (2000) estimated the causal relationship between energy consumption and income for India, Indonesia, Philippines, and Thailand. The recent existing literature has divided into two parts i.e. panel studies and country case studies (selected work on such issue is explained in Table 1).1 The causality is running from GDP per capita to electricity consumption has been found in some panel studies, such as, Yoo (2006) have investigated relationship between electricity consumption and economic growth in ASEAN member economies including Indonesia, Malaysia, Singapore, and Thailand.2 The time period of study is used from 1971 upto 2002. Their empirical results suggest oneway causal relation from economic growth to energy consumption (electricity consumption) only in Indonesia and Thailand. In Malaysia and Singapore, bi-variate causal link is found between electricity consumption and economic growth. Similarly, Chen et al. (2007) have discussed issue of causality between electricity consumption and economic growth using panel cointegration approach in 10 industrialized and low income countries of Asian region. The panel causality shows bi-variate causality for said variables but causality is running from GDP per capita to electricity consumption per capita in heterogeneous causality approach. On contrary, for Fiji Islands, Narayan and Singh (2007) document uni-variate causality from electricity consumption to economic growth and cointegration for both variables. Narayan and Smyth (2009) have checked the issue of causality for said variables in the case of Middle Eastern countries. Their empirical evidence shows that 0.04 percent GDP can be increase due to the increase in electricity consumption by 1 percent significantly. Ozturk (2010) and Payne (2010) have conducted review based study on issue of energy consumption (electricity consumption) and economic growth. The review based evidence shows that electricity consumption does have positive impact on economic growth through stylized facts. On contrary, Narayan and Prasad (2008) employ bootstrap testing to examine causality between electricity consumption and economic growth for 30 OECD countries. They note that electricity consumption seems to cause real GDP per capita in Australia, Iceland, Italy, the Slovak Republic, the Czech Republic, Korea, Portugal, and the UK. Yoo and Kwak (2010) further explore link between electricity consumption and GDP per capita in seven South American countries including Argentina, Brazil, Chile, Columbia, Ecuador, Peru, and Venezuela employing time series data starting from 1975 up to 2006. The results show uni-variate causal link from electricity consumption to economic growth proxies by real GDP for Argentina, Brazil, Chile, Columbia, and Ecuador but two-way causality for said variables for the case of Venezuela. Additionally, there is no causal relationship between electricity consumption and economic growth in the case of Peru. Studies in literature have also presented bivariate and diverse empirical evidence for said variables. For instance, Squalli and Wilson (2006) examined said issue for GCC countries. They have employed ARDL bounds testing and Toda and Yamamoto (1995) causality approach to check the direction of causality between electricity consumption and economic growth. Cointegration is found for all countries included in sample but evidence on causality is mixed. Squalli (2007) seems to inspect the causality issue in the case of OPEC economies. ARDL bounds testing approach has been used to investigate cointegration between variables which has confirmed long run relation between electricity consumption and economic growth. Their empirical evidence about causality between electricity consumption and economic 1 Both multi and countries case studies. 2 Brunei, Cambodia, Laos, Myanmar, and Vietnam have been dropped due to unavailability of reliable data.

123

123 U.S

Sri Lanka Shanghai

From 1955 upto 1990

From 1955 upto 1993

From 1963 upto 1993

From 1954 upto 1997

From 1950 upto 1997

From 1960 upto 1996

From 1947 upto 1974

From 1947 upto 1974

From 1947 upto 1974

From 1947 upto 1974

From 1960 upto 1998

From 1952 upto 1999

From 1954 upto 2003

Masih and Masih (1996)

Cheng and Lai (1997)

Cheng (1999)

Yang (2000)

Ghosh (2002)

Hondroyiannis et al. (2002)

Soytas and Sari (2003)

Soytas and Sari (2003)

Soytas and Sari (2003)

Soytas and Sari (2003)

Morimoto and Hope (2004)

Wolde-Rufael (2004)

Lee and Chang (2005)

Philippines

Japan Cyprus Turkey

From 1970 upto 2002

From 1971 upto 2002

From 1960 upto 2001

From 1960 upto 2004

From 1968 upto 2004

From 1982 upto 2006

Yoo (2005)

Yoo and Kim (2006)

Lee (2006)

Zachariadis and Pashourtidou (2007) Halicioglu (2007)

Hu and Lin (2008)

Taiwan

Indonesia

Korea

Turkey

Altinay and Karagol (2005)

Sweden

From 1965 upto 2000

From 1950 upto 2000

Hatemi and Irandoust (2005)

Taiwan

Italy

Korea

South Africa

Argentina

Greece

India

Taiwan

Brazil

Taiwan

India

Jamica

N. A

From 1970 upto 1986

U.S

Yu and Choi (1985)

Yu and Hwang (1984)

Country

Ramcharran (1990)

N. A

From 1947 upto 1979

Akarca and Long (1980)

Time span

Authors

ARDL Bounds Testing for Cointegration and Engle-Granger Causality Approach Hansen and Seo (2002) Threshold Approach for Cointegration

VECM Cointegration and Causality Tests

Toda and Yamamoto (1995) Causality Test

VAR Granger Approach

ECM Ganger Causality Test

Dolado and Lütkepohl (1996) Causality Test

A leveraged bootstrap approach

Johnson (1990) Cointegration and Granger Causality Tests

Toda and Yamamoto (1995) Causality Test

OLS Regression and ECM Model

ECM Ganger Causality Test

ECM Ganger Causality Test

ECM Ganger Causality Test

ECM Ganger Causality Test

ECM Ganger Causality Test

Engle-Granger Causality Test

Hsiao’s Granger Causality Test

Engle-Granger Causality Test

Engle-Granger Causality Test

ECM Granger Causality

Engle-Granger Causality Test

Engle-Granger Causality Test

Engle-Granger Causality Test

Sims Causality Test

Methodology used

Table 1 An Overview of Previous and Selected Studies on Electricity Consumption and Economic Growth

EC←GDP

EC←GDP

EC↔GDP

EC→GDP

EC←GDP

EC↔GDP

EC→GDP

EC→GDP

EC→GDP

EC→GDP

ES→GDP

EC←GDP

EC←GDP

Neutral

EC←GDP

GDP↔EC

EC←GDP

EC↔GDP

EC→GDP

EC←GDP

EC→GDP

EC→GDP

EC→GDP

Neutral

EC← GDP

Direction of Causality

M. Shahbaz, M. Feridun

From 1949 upto 2006

Bowden and Payne

From 1971 upto 2002

From 1980 upto 2003

From 1971 upto 2001

From 1971 upto 2001

From 1980 upto 2003

From 1991 upto 2005

Yoo (2006)

Squalli (2007)

Chen et al. (2007)

Ciarreta and Zarraga (2008)

Akinlo (2008)

Apergis and Payne (2009)

From 1971 upto 2004

From 1971 upto 2001

Wolde-Rufael (2006)

Wolde-Rufael (2009)

From 1971 upto 2001

Wolde-Rufael (2005)

From 1974 upto 2002

From 1970 upto 1990

Murray and Nan (1996)

Narayan and Smyth (2009)

From 1960 upto 1984

Ebohon (1996)

Multi-country Case Studies

Time span

Authors

Table 1 continued

Iran, Israel, Kuwait, Oman, Syria, Saudi Arabia Algeria, Benin, South Africa

China, Indonesia, Hong Kong, India, Malaysia, Korea, Taiwan, Philippines, Singapore, Thailand Austria, Belgium, Denmark, Finland, France, Sweden, Norway, Germany, Italy, Luxembourg, Netherlands, Switzerland Gambia, Ghana, Senegal, Cameroon, Kenya Commonwealth of Independent States

Iran, Qatar, Saudi Arabia, Iraq, Kuwait, UAE

Cameroon, Ghana, Nigeria, Senegal, Zambia, Zimbabwe, Algeria, Congo Republic, Kenya, South Africa, Sudan Indonesia, Malaysia, Singapore, Thailand

Algeria, Congo DR, Egypt, Ghana, Ivory Coast,

15 countries

Nigeria, Tanzania

U.S

Country

Toda and Yamamoto (1995) Causality Test

(Pedroni, 1999) and (Pedroni, 2004) for Panel Cointegration and ECM causality for short run Bootstrapped Causality Testing Approach

ARDL Bounds Testing /VAR

Panel Cointegration, GMM, Panel Causality Test

Engle-Granger Causality Test and Hsiao’s Version of Granger Causality Method ARDL Bounds Testing,Toda and Yamamoto (1995) Causality Approach Pedroni panel Cointegration, ECM, Panel Causality Test

ARDL Bounds Testing, Toda and Yamamoto (1995) Causality Approach ARDL Bounds Testing,Toda and Yamamoto (1995) Causality Approach

Engle-Granger Causality Test

Engle-Granger Causality Approach

Toda and Yamamoto causality for long run

Methodology used

EC↔GDP

MENA Panel, EC ↔GDP

EC→GDP

EC↔GDP

EC→GDP for long span of time

EC↔GDP

EC←GDP (Indonesia, Thailand) EC↔GDP (Malaysia, Singapore) EC↔GDP

EC↔GDP

EC↔GDP

Mixed results

EC↔GDP

EC→GDP

Direction of Causality

Electricity consumption and economic growth from Pakistan

123

M. Shahbaz, M. Feridun

growth is diverse. Similarly, Sinha (2009) has explored said issue for 88 developing and developed economies. The empirical results indicate two-way casual relationship between said variables not only for long run but also for short span of time. Ozturk and Acaravci (2010) have conducted a study to explore the nature of relation between energy consumption and economic growth in South African economies including Albania, Bulgaria, Hungary and Romania. The empirical results show long run association between two variables i.e. electricity consumption per capita and GDP per capita or economic growth in Hungary and two-way causality between said variables. Furthermore, there is no cointegration between electricity consumption and GDP per capita in Albania, Bulgaria and Romania which does allow in estimating short run error correction model.3 Furthermore, Acaravci and Ozturk (2009) have explored causality issue between electricity consumption per capita and GDP per capital in 15 transition economies namely Albania, Belarus, Bulgaria, Czech Republic, Estonia, Latvia, Lithuania, Macedonia, Moldova, Poland, Romania, Russian Federation, Serbia, Slovak Republic and Ukraine. The Pedroni panel and error correction method do not provide support for cointegration for the variables and economic growth is not stimulated by an increase in electricity consumption in such economies. Yoo and Lee (2010) have examined relationship between electricity consumption and economic growth. Their empirical evidence indicates inverted -U-shaped association between electricity consumption and per capita income4 in cross-country analysis. The literature also reveals the one way causality running from GDP per capita to electricity consumption per capita in country case studies. For instance, Jumbe (2004) has used ADF-cointegration to examine long run relationship between said variables in the case of Malawi. The results posit uni-variate causal relationship from economic growth to electricity consumption. This leads to conclude that “a permanent rise in GDP may cause a permanent growth in electricity consumption”. Similarly, Narayan and Smyth (2005) have examined association between electricity consumption, real income and employment for Australian economy. In long span of time, increase in real income tends to cause electricity consumption and vice versa in shorter period of time. Moreover, Mozumder and Marathe (2007) note down one-way causality is running from economic growth to electricity consumption but electricity consumption does not seem to cause economic growth for the case Bangladesh. Furthermore, Pao (2009) has examined relationship between electricity consumption and economic growth over the period of 1980 up to 2007 using cointegration and error correction model for the economy of Taiwan. Both variables are cointegrated and economic growth leads to cause the electricity consumption both in short and long span of times. Ouédraogo (2010) has conducted a study on the relation between electricity consumption and economic growth for Burkina Faso over the period of 1963–2008. The results indicate the uni-variate causality is running from economic growth to electricity consumption with significant feedback. Finally, Ciarreta and Zarraga (2010) seem to use time series data to examine causality direction for the case of Spain. Toda and Yamamoto (1995) and Dolado and Lütkepohl (1996) have been employed while empirical evidence indicates linear causality running from economic growth to electricity consumption. On contrary, literature on causal relationship from electricity consumption to economic growth is also exited. For example, Shiu and Lam (2004); Yuan et al. (2007, 2008) seem to employ Johansen and Juselius (1990) cointegration and short run granger causality approaches for cointegration and direction of causality between electricity and GDP per capita for China. The results intimate one-way causal relation running from electricity consumption 3 They have used ARDL bounds testing for cointegration analysis. 4 Their sample is comprised of OECD and non-OECD countries.

123

Electricity consumption and economic growth from Pakistan

to economic growth over 1978–2004. Moreover, Ho and Sui (2006) have found long run relationship between said variables and concluded that electricity consumption leads the causation to economic growth in the case of Hong Kong. Furthermore, Chandran et al. (2009) also reinvestigate link between electricity consumption and real GDP for the case of Malaysia. They have found cointegration among electricity consumption, consumer prices and real GDP by employing ARDL bounds testing. In short span of time, results show that electricity consumption leads to cause economic growth. The relationship between electricity consumption and economic growth is investigated by Abosedra et al. (2009) using time series but monthly data for the case of Lebanon.5 The results seem to support for cointegration between both variables and electricity consumption causes economic growth or GDP. Some studies present bivariate casual relation for said variables such as Yoo (2005) have noted bi-variate causal association between energy or electricity consumption and economic growth in the case of Korea. The findings suggest that electricity consumption has direct impact on economic growth and economic growth further seems to boost the energy or electricity consumption. Similarly, Zamani (2006) documents unidirectional relationship from economic growth to total energy and bidirectional link between economic growth and gas as well as economic growth and petroleum products consumption over the period of 1963 upto 2003 in the case of Iran. In addition, Tang (2009) seems to examine the link between electricity consumption and economic growth for the case of Malaysia. The empirical evidence does not indicate any cointegrating vector between electricity consumption and economic growth. Further, he has applied MWALD test for Granger causality and documented bivariate causality between both variables using quarterly data over the period of 1972:1 to 2003:4. The issue of causality between electricity consumption and economic growth has been investigated by Tsani (2010) in the case of Greece over the period of 1960 up to 2006. Tansi has used both energy consumption in industrial, residential and transport for aggregated. The empirical exercise reveals that two-way causal relation exists between industrial and residential energy consumption and economic growth but no link is found between transport energy consumption and GDP per capita. In the case of South Africa, Odhiambo (2009) seems to investigate relationship between electricity consumption, employment and economic growth. The results of tri-variate causality indicate two-way causal association between electricity and economic growth and one-way from employment to GDP per capita for Tanzania. Furthermore, Akinlo (2009) has toil to investigate the association between energy consumption proxies by electricity consumption and real GDP for the case of Nigeria. The empirical shows cointegration for both variables and electricity consumption seems to cause real GDP. The results of Hodrick–Prescott (HP) filter also seems to decompose fluctuations from series of electricity consumption and economic growth. In the case of India, Ghosh (2009) have conducted a study on the nexus between electricity supply, employment and real GDP. ARDL multivariate approach is used to investigate long run relation among variables. The results indicate that cointegration among variables has been established. Furthermore, electricity consumption and GDP per capita are responsible for change in employment levels in the country. In the case of Pakistan only a few studies investigated the causality between energy consumption and economic growth. Anjum and Butt (2001) have examined relationship between energy consumption proxies by petroleum consumption and economic growth employing Hsiao’s version of causality. Their findings suggest one-way causality is running from economic growth to petroleum consumption or energy consumption. Although, there is no casual relation between gas sector and economic growth but power sector leads to cause economic 5 They have data of monthly tourism imports as a proxy for GDP.

123

M. Shahbaz, M. Feridun

growth. Finally, energy consumption causes employment in that sector. On the other hand, Alam and Butt (2002) have used cointegration and granger causality approaches for energy consumption and economic growth. Their empirical evidence provides supports for cointegration and found bidirectional causal relation for said variables in the case of Pakistan. But, Siddiqui (2004) seems to document that energy consumption does have positive impact to stimulate economic activity but causality is reversed for petroleum products. Recently, Khan and Qayyum (2007) examine causal relationship between energy consumption and economic growth in South Asian Countries, namely, Pakistan, Bangladesh, India and Sri Lanka. Their results suggest causality running from energy consumption to economic growth. The authors also document evidence that each is energy dependent and energy crisis may adversely affect the economic growth and declining trend both in income and employment. Zahid (2008) provides a study on energy consumption and economic growth in South Asian Countries using different indicators of energy consumption and provides mixed results regarding energy consumption-economic growth causality. Qazi and Riaz (2008) have also conducted a study on relationship between energy consumption and economic growth. Their empirical evidence intimates bivariate causal relationship between energy consumption and economic growth for short span of time but in long span of time direction of causality is running from economic growth to energy consumption. As can be seen, the existing studies on Pakistan yielded contradicting results on the link between energy consumption and economic growth. This shows that relationship between electricity consumption and economic growth is widely ignored as a case study for Pakistan to date.6 The objective of the present study, therefore, is to fill this gap in the literature and main motivation for researchers. 3 Data and methodology The time series data used in the present analysis is in annual frequency and spans the period from 1971 to 2008. Data series have been obtained from the World Bank’s world development indicators (WDI) CD-ROM. Economic growth (denoted by GDP) is proxied by real GDP per capita and electricity consumption (denoted by EC) is proxied by electricity consumption per capita in KWH. This paper follows the ARDL bounds testing approach to cointegration developed by Pesaran and Pesaran (1997); Pesaran et al. (2000), and latter on by Pesaran et al. (2001) to examine long run relationship between economic growth and electricity consumption. The autoregressive distributive lag model can be applicable with out investigating the order of integration (Pesaran and Pesaran 1997). Haug (2002) has argued that ARDL approach to cointegration provides better results for small sample data set such as in our case as compared to traditional approaches to cointegration i.e. Engle and Granger (1987); Johansen and Juselius (1990) and Phillips and Hansen (1990). Another advantage of ARDL bounds testing is that unrestricted model of ECM seems to take satisfactory lags that captures the data generating process in a general-to-specific framework of specification (Laurenceson and Chai, 2003). However, Pesaran and Shin (1999) contented that, “appropriate modification of the orders of ARDL model is sufficient to simultaneously correct for residual serial correlation and problem of endogenous variables”. The equations of unrestricted error correction methods are being modeled as:

6 Even no study seems to use electricity consumption as a proxy for energy consumption.

123

Electricity consumption and economic growth from Pakistan

L EC = α◦ + α1 t +

m  i=1

α2 L ECt−i +

m 

α3 LG D Pt−i + α4 L ECt−1

i=0

+α5 LG D Pt−1 + ηi m m   LG D P = β◦ + β1 t + β2 LG D Pt−i + β3 L ECt−i i=1

(1)

i=0

+β4 LG D Pt−1 + β5 L ECt−1 + μi

(2)

The ARDL bounds testing approach to cointegration depends upon the tabulated critical values by Pesaran et al. (2001) to take decision about cointegration among variables. The null hypothesis of no cointegration is α2 = β2 = α3 = β3 = 0 in both models. The alternative hypothesis of cointegration among variables isα2  = β2  = α3  = β3  = 0. Next step is to compare the calculated F-statistics with LCB (lower critical bound) and UCB (upper critical bound) by Pesaran and Pesaran (1997) or Pesaran et al. (2001). There is cointegration among variables if calculated value of F-statistics is more than UCB. If LCB is more than computed F-statistics then hypothesis of no cointegration may be accepted. Finally, if calculated F-statistics is between lower and upper critical bounds then decision about cointegration is inconclusive. The third stage includes conducting standard Granger causality tests augmented with a lagged error-correction term. The Granger representation theorem suggests that there will be Granger causality in at least one direction if there exists cointegration between the variables provided the variables are integrated at I(1). Engle and Granger (1987) cautioned that if the Granger causality test is conducted at first difference through vector auto regression (VAR) method then it will be misleading in the presence of cointegration. Therefore, an inclusion of an additional variable to the VAR method such as the error-correction term would help us to capture the long run relationship. To this end, an augmented form of Granger causality test is involved to the error-correction term and it is formulated in a bi-variate pth order vector error-correction model (VECM) which is as follows:          p  L ECt k d11 (L) d12 (L) γ EC Tt−1 L ECt−i = 1 + + 1 d21 (L) d22 (L) LG D Pt k2 LG D Pt−i λ1 EC Tt−1 i=1     C1 η + + 1 (3) C2 η2 where  is a difference operator, ECT representing the error-correction term derived from long run cointegrating friendship via ARDL model, C (i = 1, 2) is constant and η (i = 1, 2) are serially uncorrelated random disturbance term with zero mean. Through the ECT, the VECM provides new directions for Granger causality to appear. Long-run causality can be revealed through the significance of the lagged ECTs by t test, while F-statistic or Wald test investigate short run causality through the significance of joint test with an application of sum of lags of explanatory variables. Additionally, Toda-Yamamoto approach is also used as this approach is known to overcome the problem of invalid asymptotic critical values when causality tests are performed in the presence of non-stationary series and can be used regardless of the existence of cointegration between the underlying variables in the model (Zapata and Rambaldi 1997). The augmented test of non-causality test developed by Toda and Yamamoto (1995) is applicable in level vector auto regressions (VARs) irrespective of whether variables are integrated at same order of integration or not. VAR can be estimated with out true lag order k but it is applicable with (k +d) lag order where d indicates possible order of integration for variables.

123

M. Shahbaz, M. Feridun

The Toda and Yamamoto (1995) causality test is examined by performing hypothesis exercise disregarding the additional lags k + 1, . . . , k + d in vector auto regression (VAR). The Toda-Yamamoto causality technique involves the estimation of the following models:

LG D P = α◦ +

k+d max 

α2 LGDPt−1 +

i=1

LEC = β◦ +

k+d max 

k+d max 

α3 L ECt−i + η1

(4)

β3 LG D Pt−i + η2

(5)

i=1

β2 L ECt−i +

k+d max 

i=1

i=1

where, GDP and EC are as defined earlier. In the models, each variable is regressed on each other with lag order starting from 1 towards k + d max, η1 and η2 are the error terms, k is the optimal lag order and d is the maximum order of integration of the variables in the concerned system. Since the procedure requires a VAR only in levels, it does not lead to a loss of information as it would happen in the case of differencing. For this reason, the procedure can be used only as a long-run test.

4 Empirical results There is need to find out order of integration for said macroeconomic variables. In such, situation, unit root analysis ensures us that no variable is integrated at I(2) to keep away from spurious results. It is inferred by Ouattara (2004) that if any variable is integrated at I(2) then computation of F-statistics for cointegration becomes senseless. Pesaran et al. (2001) critical bonds are based on assumption such as variables should be stationary at I(0) or I(1). Therefore, application of unit root tests is still necessary to ensure that no variable is integrated at I(2) or beyond. Ng-Perron and Clement-Montanes-Reyes have been used to inspect the order of integration. The results of both tests are reported in Tables 2 and 3. The results disclose that both variables are integrated at I(1). The likeness of order of integration supports to use ARDL bounds testing approach for cointegration. The initial step for ARDL approach to apply is the selection of appropriate lag length. Considering the small sample data set we can not take lag more than 2 on basis of minimum value of FPE and AIC. Literature reveals that the calculation of ARDL F-statistics is quite sensitive to the selection of lag order in the model (see, for instance, Bahmani-Oskooee and Brooks 1999; Bahmani-Oskooee et al. 2006 and Bahmani-Oskooee and Harvey (2006)). As

Table 2 Unit root estimation Variables

MZa

MZt

MSB

MPT

Ng-Perron Test at Levels LEC

−2.9066

−1.1422

0.3929

29.5889

LGDP

−3.9775

−1.3994

0.3518

22.7751

Ng-Perron Test at 1st Difference LEC

−14.7854c

−2.7185

0.1838

6.1658

LGDP

−27.2043a

−3.6881

0.1355

3.3496

a ,c The significance at 1, 10% levels

123

Electricity consumption and economic growth from Pakistan Table 3 Table Clemente-Montanes-Reyes unit root tests with two structural breaks Innovative outliers

Additive outliers

t-stat

TB1

TB2

Decision

t-stat

TB1

TB2

Decision

EC

−2.287

1983

2002

I (1)

−2.534

1989

2002

I (1)

GDPC

−4.275

1978

2002*

I (1)

−8.642∗

1984*

1993*

I (0)

∗ The significance at 5% level

Table 4 VAR Lag Length Selection Lag

LogL

LR

FPE

AIC

0

−686.6287

NA

2.61e+13

39.40736

1

−531.5021

274.7957

6.19e+09

31.05726

2

−520.7361

17.22562

5.67e+09*

30.95635∗

3

−513.5612

10.24980

6.51e+09

31.06064

LR sequential modified LR test statistic (each test at 5% level), FPE Final prediction error, AIC Akaike information criterion ∗ Lag order selected by the criterion Table 5 Cointegration test: bounds test Model for estimation

F-statistics

Lag

FEC (EC/G D P)

6.321*

2

FGDPC (GDP/EC)

4.221***

2

Critical bounds

Lower bound

Upper bound

1%

4.428

5.898a

%5

3.368

4.590

10%

2.893

4.008

a Critical values obtained from Narayan (2005) from page-1990. The lag selection is based on AIC and SBC.

* and *** denotes the probability and the significant level at 0.01 and 0.10, respectively

can be seen in Table 4, several selection criteria have been considered but appropriate lag length is selected as 2 years is based on FPE and AIC. The results of the ARDL bounds tests shown in Table 5, suggest the rejection of the null hypothesis of no long run relationship at the 10% level of significance when GDP is treated as the dependent variable and EC is treated as its long run forcing variable. As can be seen from the table, the estimated F-statistic is greater than the upper bound critical values suggested by Narayan (2005) at the 10% level in the case where GDP is the dependent variable and EC is the independent variable. On the other hand, when EC is treated as the dependent variable and GDP is treated as its long run forcing variable, the estimated F-statistic is above the upper bound critical value provided by Narayan (2005) at 1% level of significance, As a result, it can be concluded that there exists a strong long run equilibrium relationship between EC and GDP. Next, parameter stability tests are used to check if the relationship between the two variables is stable. Following Pesaran and Pesaran (1997), Brown et al. (1975) stability tests

123

M. Shahbaz, M. Feridun 15 10 5 0 -5 -10 -15 86

88

90

92

94

96

CUSUM

98

00

02

04

06

08

02

04

06

08

5% Significance

Fig. 2 CUSUM plot 1.6

1.2

0.8

0.4

0.0

-0.4 86

88

90

92

94

CUSUM of Squares

96

98

00

5% Significance

Fig. 3 CUSUMSQ plot

known as cumulative (CUSUM) and cumulative sum of squares (CUSUMSQ) are used. The CUSUM and CUSUMSQ statistics are updated recursively and plotted against the breaks points. If the plots of CUSUM and CUSUMSQ statistics stay with in the critical bonds of 5% level of significance, the null hypothesis of all coefficients is the given regression are stable can not be rejected. As evident from Figs. 2 and 3, all plots of CUSUM and CUSUMSQ statistics are well within the critical bounds, implying that the coefficients in the error-correction model are stable. Although the evidence obtained so far has identified the relationship between EC and GDP, the results are not sufficient to identify whether the direction of the relationship is from EC to GDP or vice versa. The existence of a long run relationship between the variables in question is only a necessary but not a sufficient condition for rejecting the non-causality hypothesis (Morley 2006). In other words, the existence of cointegrating relationships among the underlying variables suggests that there must be Granger-causality in at least one direction.

123

Electricity consumption and economic growth from Pakistan Table 6 Short run Wald-test Granger causality

Dependent variable

R2

DEC

DGDP

Ecmt−1

_

3.78099

−0.00151

(0.0214)

(0.0035)

Lag 1 Wald-statistics DEC

Probability values are given in parentheses Table 7 Toda Yamamoto Causality analysis

DGDP

2.04971

_

(0.1297)

Dependent Variable

−0.00832

0.53351 0.30698

(0.2169)

F-statistics * EC

GDP

EC

_

9.0395

GDP

0.88303

_

Direction of Causality at Lag-1 (0.0049) (0.3540) Dependent Variable

EC

GDP

EC

_

5.1114

GDP

0.22457

_

Direction of Causality Lag-2 (0.0121) (0.8001) Dependent Variable

EC

GDP

EC

_

3.8313

GDP

0.0782

_

Direction of Causality Lag-3 (0.0203) * Probability values are given in parentheses

(0.9712)

As Groenewold and Tang (2007) suggest, Granger-causality tests are applicable regardless of the orders of integration of the underlying variables if it has been established that there exists long run equilibrium between the underlying series. Since the results obtained from the ARDL bounds tests indicate the existence of a long run equilibrium relationship among the pairs, the use of Granger causality tests are justified. Table 6 reports the results of the short run Wald-test Granger causality tests. The evidence obtained suggests the existence of strong causality running from GDP to EC with a positive coefficient at the 5% level both in long and short runs. On the other hand Table 7 presents the results of the Toda Yamamoto causality analysis, which has yielded even stronger evidence of a uni-directional causal link from GDP to EC. Therefore, the empirical results obtained from the two causality tests suggests that economic growth leads to electricity consumption in Pakistan and not vice versa. These findings are consistent with line of literature such as Jumbe (2004) for Malawi, Narayan and Smyth (2005) for Australia, Mozumder and Marathe (2007) for Bangladesh, Pao (2009) for Taiwan, Ouédraogo (2010) for Burkina Faso and Ciarreta and Zarraga (2010) for Spain and contradict with Aqeel and Butt (2001).

123

M. Shahbaz, M. Feridun

5 Conclusions The present article used the Autoregressive Distributed Lag (ARDL) bounds testing procedure to identify the long run equilibrium relationship between electricity consumption and economic growth as well Toda Yamamoto and Wald-test causality tests to identify the direction of the causal relationship between these two variables in case of Pakistan. The results suggest that the two variables are in a long run equilibrium relationship and that economic growth leads to electricity consumption in Pakistan and not vice versa. Although these findings lend support to some earlier studies that report evidence of a unidirectional causality running from GDP to electricity or energy consumption such as Ghosh (2002); Fatai et al. (2004) and Hatemi and Irandoust (2005), they contradict with the results of some of the existing studies. More specifically the results have failed to lend support to Erol and Yu (1987) who identified unidirectional causality running from energy consumption to GDP for Canada, Narayan and Singh (2007) who found that energy consumption affects real GDP positively in the long run in G7 countries; Wolde-Rufael (2009) who found that energy consumption causes economic growth in seventeen African countries; and to more recent studies by Apergis and Payne (2009) who document evidence of short run and long run causality from energy consumption to economic growth for six central American countries. The policy implication that emerges from the study is quite straightforward: The findings suggest that that economic growth causes electricity consumption in Pakistan but not vice versa. It implies that government should initiate environmental friendly policies initiatives such as electricity consumption conservation. This leads to conclude that electricity consumption conservation polices do not seem to impede the speed of economic growth for small economy like Pakistan. The balance between environment and economic growth can be filed through the exploration and utilization of other environmental friendly sources of energy such as solar, hydro, and wind power instead of fossil fuel.

References Abosedra, S., Dah, A., Ghosh, S.: Electricity consumption and economic growth: the case of Lebanon. Appl. Energy 86, 429–432 (2009) Acaravci. A., Ozturk, I.: Electricity consumption-growth nexus: Evidence from panel data for transition countries. Energy Econ. 32(2), 604–608 (2009) Akarca, A.T., Long, T.V.: On the relationship between energy and GDP: a reexamination. J. Energy Dev. 5, 326– 331 (1980) Akinlo, A.E.: Energy consumption and economic growth: evidence from 11 Sub-Sahara African countries. Energy Econ. 30, 2391–2400 (2008) Akinlo, A.E.: Electricity consumption and economic growth in Nigeria: evidence from cointegration and co-feature analysis. J. Policy Model. 31, 681–693 (2009) Alam, S., Butt, M.S.: Causality between energy consumption and economic growth in Pakistan: an application of cointegration and error correction modeling techniques. Pac. Asian J. Energy 12, 151–165 (2002) Altinay, G., Karagol, E.: Electricity consumption and economic growth: evidence from Turkey. Energy Econ. 27, 849–856 (2005) Apergis, N., Payne, J.E.: Energy consumption and economic growth: Evidence from the Commonwealth of Independent States. Energy Econ. 31, 641–647 (2009) Aqeel, A., Butt, M.S.: The Relationship between energy consumption and economic growth in Pakistan. AsiaPac. Dev. J. 8, 101–110 (2001) Asafu-Adjaye, J.: The Relationship between energy consumption, energy prices and economic growth: time series evidence from Asian developing countries. Energy Econ. 22, 615–625 (2000) Bahmani-Oskooee, M., Brooks, T.J.: Bilateral J–curve between US and her trading partners. Weltwirtschaftliches Archiv 135, 156–165 (1999)

123

Electricity consumption and economic growth from Pakistan Bahmani-Oskooee, M., Harvey, H.: How sensitive are Malaysia’s bilateral trade flows to depreciation?. Appl. Econ. 38, 1279–1286 (2006) Bahmani-Oskooee, M., Economidou, C., Goswami, G.: Bilateral J-curve between the UK vis-à-vis her major trading partners. Appl. Econ. 38, 879–888 (2006) Bowden, N., Payne, J.E.: The causal relationship between US energy consumption and real output: a disaggregated analysis. J. Policy Model. 31, 180–188 (2009) Brown, R.L., Durbin, J., Ewans, J.M.: Techniques for testing the constance of regression relations overtime. J. Royal Stat. Soc. 37, 149–172 (1975) Chandran, V.G.R., Sharma, S., Madhavan, K.: Electricity consumption–growth nexus: the case of Malaysia. Energy Policy 38, 606–612 (2009) Chen, S.T., Kuo, H.I., Chen, C.C.: The relationship between GDP and electricity consumption in 10 Asian countries. Energy Policy 35, 2611–2621 (2007) Cheng, B.S., Lai, T.W.: An investigation of co-integration and causality between energy consumption and economic activity in Taiwan. Energy Econ. 19, 435–444 (1997) Cheng, B.S.: Causality between energy consumption and economic growth in India: an application of cointegration and error-correction modeling. Indian Econ. Rev. 34, 39–49 (1999) Ciarreta A., Zarraga A.: Economic growth and electricity consumption in 12 European Countries: a causality analysis using panel data. Working Paper, Department of Applied Economics III (Econometrics and Statistics), University of the Basque Country. Available at: http://www.et.bs.ehu.es/biltoki/EPS/dt200804. pdf. (2008) Ciarreta, A., Zarraga, A.: Electricity consumption and economic growth in Spain. Appl. Econ. Lett. 17, 1–5 (2010) Dolado, J.J., Lütkepohl, H.: Making Wald test work for cointegrated VAR systems. Econom. Theory 15, 369– 386 (1996) Ebohon, O.J.: Energy, economic growth and causality in developing countries: A case study of Tanzania and Nigeria. Energy Policy 24, 447–453 (1996) Engle, R.F., Granger, C.W.J.: Cointegration and error correction: representation, estimation and testing. Econometrica 55, 251–276 (1987) Erol, U., Yu, E.S.H.: On the causal relationship between eenrgy and income for industrialized countries. J. Energy Dev. 13, 113–122 (1987) Fatai, K., Oxley, L., Scrimgeour, F.G.: Modeling the causal relationship between energy consumption and GDP in New Zealand, Australia, India, Indonesia, The Philippines and Thailand. Math. Comput. Simul. 64, 431–445 (2004) Ghosh, S.: Electricity consumption and economic growth in Taiwan. Energy Policy 30, 125–129 (2002) Ghosh, S.: Electricity supply, employment, and real GDP in India: evidence from cointegration and Grangercausality tests. Energy Policy 37, 2926–2929 (2009) Groenewold, N., Tang, S.H.K.: Killing the Goose that Lays the Golden Egg: institutional change and economic growth in Hong Kong. Econ. Inquiry 45, 787–799 (2007) Halicioglu, F.: Residential electricity demand dynamics in Turkey. Energy Economics 29, 199–210 (2007) Hansen, B.E., Seo, B.: Testing for two regime threshold cointegration in vector error-correction models. J. Econom. 110, 293–318 (2002) Hatemi, A., Irandoust, M.: Energy consumption and economic growth in Sweden: a leveraged bootstrap approach, 1965–2000. Int. J. Appl. Econometrics Quant. Stud. 4, 1–20 (2005) Haug, A.: Temporal aggregation and the power of cointegration tests: a Monte Carlo study. Oxford Bull. Econ. Stat. 64, 399–412 (2002) Ho, C., Sui, K.: A dynamic equilibrium of electricity consumption and GDP in Hong Kong: an empirical investigation. Energy Policy 35, 2507–2513 (2006) Hondroyiannis, G., Lolos, S., Papapetrou, E.: Energy consumption and economic growth: assessing the evidence from Greece. Energy Economics 24, 319–336 (2002) Hu, J.L., Lin, C.H.: Disaggregated energy consumption and GDP in Taiwan: a threshold co-integration analysis. Energy Economics 30, 2342–2358 (2008) Johansen, S., Juselius, K.: Maximum likelihood estimation and inference on cointegration with applications to the demand for money. Oxford Bull. Econ. Stat. 52(2), 169–210 (1990) Jumbe, C.B.L.: Cointegration and causality between electricity consumption and GDP: empirical evidence from Malawi. Energy Economics 26, 61–68 (2004) Khan, A.M., Ahmed, U.: Energy Demand in Pakistan: A Disaggregate Analysis, Pakistan Institute of Development Economics, Islamabad. Available at http://mpra.ub.uni-muenchen.de/15056/ MPRA Paper No. 15056 (2009) Khan, A.M., Qayyum, A.: Dynamic modeling of energy and growth in South Asia. Pak. Dev. Rev. 46, 481– 498 (2007)

123

M. Shahbaz, M. Feridun Kraft, J., Kraft, A.: On the relationship between energy and GDP. J. Energy Dev. 3, 401–403 (1978) Laurenceson, J., Chai, J.C.H.: Financial reform and economic development in China. Edward Elgar, Cheltenham, UK (2003) Lee, C.C.: The causality relationship between energy consumption and GDP in G-11 countries revisited. Energy Policy 34, 1086–1093 (2006) Lee, C.C., Chang, C.P.: Structural breaks, energy consumption, and economic growth revisited: evidence from Taiwan. Energy Econ. 27, 857–872 (2005) Masih, A.M.M., Masih, R.: Energy consumption, real income and temporal causality: results from a multycountry study based on cointegration and error-correlation modeling techniques. Energy Econ. 18, 165– 183 (1996) Morimoto, R., Hope, C.: The impact of electricity supply on economic growth in Sri Lanka. Energy Econ. 26, 77–85 (2004) Morley, B.: Causality between economic growth and migration: an ARDL bounds testing approach. Econ. Lett. 90, 72–76 (2006) Mozumder, P., Marathe, A.: Causality relationship between electricity consumption and GDP in Bangladesh. Energy Policy 35, 395–402 (2007) Murray, D.A., Nan, G.D.: A definition of the gross domestic product–electrification interrelationship. J. Energy Dev. 19, 275–283 (1996) Narayan, P.N.: The saving and investment Nexus for China: evidence from co-integration tests. Appl. Econ. 37, 1979–1990 (2005) Narayan, P.K., Prasad, A.: Electricity consumption-real GDP causality nexus: evidence from a bootstrapped causality test for 30 OECD countries. Energy Policy 36, 910–918 (2008) Narayan, P.K., Singh, B.: The electricity consumption and GDP nexus for the Fiji Islands. Energy Econ. 29, 1141–1150 (2007) Narayan, P.K., Smyth, R.: Electricity consumption, employment and real income in Australia: evidence from multivariate Granger causality tests. Energy Policy 33, 1109–1116 (2005) Narayan, P.K., Smyth, R.: Multivariate granger-causality between electricity consumption, exports, and GDP: evidence from a panel of Middle Eastern countries. Energy Policy 37, 229–236 (2009) Odhiambo, N.M.: Energy consumption and economic growth nexus in Tanzania: An ARDL bounds testing approach. Energy Policy 37, 617–622 (2009) Oh, W., Lee, K.: Causal relationship between energy consumption and GDP revisited: the case of Korea 1979–1999. Energy Econ. 26, 51–59 (2004) Ouattara, B.: Foreign Aid and Fiscal Policy in Senegal. Mimeo University of Manchester, Manchester (2004) Ouédraogo, M.: Electricity consumption and economic growth in Burkina Faso: a cointegration analysis. Energy Econ. 3, 524–531 (2010) Ozturk, A.: A literature survey on energy-growth nexus. Energy Policy 38, 340–349 (2010) Ozturk, I., Acaravci, A.: The causal relationship between energy consumption and GDP in Albania, Bulgaria, Hungary and Romania: evidence from ARDL bound testing approach. Appl. Energy. 87, 1938–1943 (2010) Pao, H.: Forecast of electricity consumption and economic growth in Taiwan by state space modeling. Energy 34, 1779–1791 (2009) Payne, J.: A survey of the electricity consumption-growth literature. Appl. Energy 87, 3723–3731 (2010) Pedroni, P.: Critical values for cointegration tests in heterogeneous panels with multiple regressors. Oxford Bull. Econ. Stat. 61, 653–670 (1999) Pedroni, P.: Panel cointegration: asymptotic and finite sample properties of pooled time series tests with an application to the PPP hypothesis. Econom. Theory 20, 597–625 (2004) Pesaran, M.H., Pesaran, B.: Working with Microfit 4.0: interactive econometric analysis. Oxford University Press, Oxford (1997) Pesaran, M.H., Shin, Y.: An autoregressive distributed lag modelling approach to cointegration analysis. In: Strom, S. (eds.) Econometrics and economic theory in 20th Century: the Ragnar Frisch Centennial Symposium, Chapter 11, Cambridge University Press, Cambridge (1999) Pesaran, M.H., Shin, Y., Smith, R.J.: Structural analysis of vector error correction models with exogenous I(1) variables. J. Econom. 97, 293–343 (2000) Pesaran, M.H., Shin, Y., Smith, R.J.: Bounds Testing Approaches to the Analysis of Level Relationships. J. Appl. Econom. 16, 289–326 (2001) Phillips, P.C.B., Hansen, B.E.: Statistical inference in instrumental variables regression with I(1) processes. Rev. Econ. Stud. 57(1), 99–125 (1990) Qazi, M.A.H., Riaz, S.: Causality between energy consumption and economic growth: the case of Pakistan. Lahore J. Econ. 13, 45–58 (2008) Ramcharran, H.: Electricity consumption and economic growth in Jamaica. Energy Econ. 12, 65–70 (1990)

123

Electricity consumption and economic growth from Pakistan Reynolds, D.B., Kolodziej, M.: Former Soviet Union oil production and GDP decline: Granger causality and the multi-cycle Hubbert curve. Energy Econ. 30, 271–289 (2008) Shiu, A., Lam, P.L.: Electricity consumption and economic growth in China. Energy Policy 32, 47–54 (2004) Siddiqui, R.: Energy and Economic Growth in Pakistan. Pak. Dev. Rev. 43, 175–200 (2004) Sinha, D.: The energy consumption-GDP Nexus: panel data evidence from 88 Countries. MPRA Paper-18446 (2009) Soytas, U., Sari, R.: Energy consumption and GDP: causality relationship in G-7 countries and emerging markets. Energy Econ. 25, 33–37 (2003) Soytas, U., Sari, R.: Energy consumption, economic growth, and carbon emissions: challenges faced by an EU candidate member. Ecol. Econ. 68, 1667–1675 (2009) Squalli, J., Wilson, K.: A bounds analysis of electricity consumption and economic growth in the GCC. Working Paper -06-09, EPRU. Zayed University, Abu Dhabi (2006) Squalli, J.: Electricity consumption and economic growth: bounds and causality analyses for OPEC members. Energy Econ. 29, 1192–1205 (2007) Stern, D.I.: Energy use and economic growth in the USA: a multivariate approach. Energy Econ. 15, 137– 150 (1993) Tang, C.F.: A re-examination of the relationship between electricity consumption and economic growth in Malaysia. Energy Policy 36, 3077–3085 (2009) Toda, H.Y., Yamamoto, T.: Statistical inferences in vector autoregressions with possibly integrated processes. J. Econom. 66, 225–250 (1995) Tsani, S.Z.: Energy consumption and economic growth: a causality analysis for Greece. Energy Econ. 32, 582–590 (2010) Wolde-Rufael, Y.: Disaggregated energy consumption and GDP, the case of Shanghai, 1952–1999. Energy Econ. 26, 69–75 (2004) Wolde-Rufael, Y.: Energy demand and economic growth: the African experience. J. Policy Model. 27, 891– 903 (2005) Wolde-Rufael, Y.: Electricity consumption and economic growth: a time series experience for 17 African countries. Energy Policy 34, 1106–1114 (2006) Wolde-Rufael, Y.: Energy consumption and economic growth: the experience of African countries revisited. Energy Econ. 31, 217–224 (2009) Yang, H.Y.: A note on the causal relationship between energy and GDP in Taiwan. Energy Econ. 22, 309– 317 (2000) Yoo, S.: The causal relationship between electricity consumption and economic growth in ASEAN countries. Energy Policy 34, 3573–3582 (2006) Yoo, S.H.: Electricity consumption and economic growth: evidence from Korea. Energy Policy 33, 1627– 1632 (2005) Yoo, S., Kwak, S.: Electricity consumption and economic growth in seven South American countries. Energy Policy 38, 180–188 (2010) Yoo, S., Lee, J.: Electricity consumption and economic growth: A cross-country analysis. Energy Policy 38, 622–625 (2010) Yoo, S.H., Kim, Y.: Electricity generation and economic growth in Indonesia. Energy 31, 2890–2899 (2006) Yu, E.S.H., Hwang, B.K.: The relationship between energy and GNP: further results. Energy Econ. 6, 186– 190 (1984) Yu, E.S.H., Choi, J.Y.: The causal relationship between energy and GNP: an international comparison. J. Energy Dev. 10, 249–272 (1985) Yuan, J., Zhao, C., Yu, S., Hu, Z.: Electricity consumption and economic growth in China: cointegration and co-feature analysis. Energy Econ. 29, 1179–1191 (2007) Yuan, J., Kang, J., Zhao, C., Hu, Z.: Energy consumption and economic growth: evidence from China at both aggregated and disaggregated levels. Energy Econ. 30, 3077–3094 (2008) Zachariadis, T.: Exploring the relationship between energy use and economic growth with bivariate models: new evidence from G-7 countries. Energy Econ. 29, 1233–1253 (2007) Zachariadis, T., Pashourtidou, N.: An empirical analysis of electricity consumption in Cyprus. Energy Econ. 29, 183–198 (2007) Zahid, A.: Energy–GDP relationship: A causal analysis for the five countries of South Asia. Appl. Econom. Int. Dev. 1, 167–180 (2008) Zamani, M.: Energy consumption and economic activities in Iran. Energy Econ. 29, 1135–1140 (2006) Zapata, H.O., Rambaldi, A.N.: Monte Carlo evidence on cointegration and causation. Oxford Bull Econ. Stat. 52, 285–298 (1997)

123