Anda di halaman 1dari 36

THE JOURNAL OF ENERGY

AND DEVELOPMENT

Amine Lahiani, Avik Sinha,


and Muhammad Shahbaz,
“Renewable Energy Consumption, Income,
CO2 Emissions, and Oil Prices in
G7 Countries: The Importance of Asymmetries,”
Volume 43, Number 2

Copyright 2018
RENEWABLE ENERGY CONSUMPTION, INCOME,
CO2 EMISSIONS, AND OIL PRICES IN G7
COUNTRIES: THE IMPORTANCE OF
ASYMMETRIES

Amine Lahiani, Avik Sinha, and Muhammad Shahbaz*

Introduction

A rising level of anthropogenic greenhouse gas (GHG) emissions is aggravating


the global warming problem.1 Contemporary research suggests that human
interaction with nature in the form of unsustainable economic growth is a root
cause of global warming.2 The environmental pressure in the form of ambient air

*Amine Lahiani, Associate Professor at the University of Orléans, received his Ph.D. from the
University of Paris Ouest la Defense in France and the University of Geneva (Switzerland).
He has been a visiting professor/researcher at numerous institutions with his main focus on
econometrics applied to economics and finance. The author has published more than 30
papers that have appeared in journals such as Energy Economics, Energy Policy, Applied
Economy, Economic Modelling, The Quarterly Review of Economics and Finance, and Economics
Bulletin.
Avik Sinha, Assistant Professor in the M.B.A. Department of Atria Institute of Technology
(Bangalore, India), received his F.P.M. in economics from the Indian Institute of Management
Indore. His research interests include energy economics, environmental economics, and applied
econometrics. He has published more than 30 papers appearing in journals such as Journal of
Cleaner Production, Ecological Indicators, Renewable and Sustainable Energy Reviews, Renewable
Energy, and Journal of Developing Areas.
Muhammad Shahbaz is Full Professor at Energy and Sustainable Development (ESD),
Montpellier Business School. He was Principal Research Officer at COMSATS Institute of
Information Technology, Lahore, Pakistan. The author received his Ph.D. in economics from the
(continued)

The Journal of Energy and Development, Vol. 43, Nos. 1 and 2


Copyright Ó 2018 by the International Research Center for Energy and Economic Development
(ICEED). All rights reserved.
157
158 THE JOURNAL OF ENERGY AND DEVELOPMENT

pollution is primarily created by continuous consumption of fossil fuels. The rising


level of GHG emissions has manifested itself in several forms, including the rise
in surface temperature, melting of glaciers, the rise in sea levels, changes in the
rainfall patterns, changes in the patterns of El Niňo and La Niňa, and numerous
others ways.3
Economic growth cannot take place without the consumption of energy. As
a nation moves along its growth path, demand for energy rises along with it.
According to the U.S. Energy Information Administration (EIA)’s International
Energy Outlook 2016, energy demand will increase by 48 percent from 2012 to
2040.4 The EIA also has forecasted that global crude oil consumption will rise
from 45.5 million barrels per day in 2012 to 46.1 million barrels per day in 2040.5
Similarly, crude oil is expected to be priced at U.S. $76 per barrel in 2040, which is
a rough estimate based upon the required expected expenditures in the oil sector in
order to meet growing future energy demand. This situation provides nations with
the opportunity to shift from non-renewable to renewable energy sources (i.e.,
solar, wind, tidal, waste, and others). Compared to non-renewables, renewable
energy sources are eco-friendly and non-depletable. Nations are gradually re-
alizing the potential and significance of renewable energy. By the end of 2015,
nearly 66 countries have issued biofuel mandates at central or provincial levels.6
Along with the problem of staggering energy demand, the rising level of
emissions is catalyzing countries to explore the possibilities of generating more
renewable energy. The continuous consumption of fossil fuel has raised the level
of carbon dioxide emissions (CO2) in the atmosphere. According to the EIA, by
2040 electricity generation will increase by 69 percent, which is far below the
projected level of energy demand.7 Additionally, the projected generation of coal-
based energy is expected to increase by 25 percent. As such, policy makers also
are concerned with the resulting CO2 emissions and a need to move toward cleaner
and renewable energy sources. The proposal of the Intended Nationally De-
termined Contributions (INDCs) for the reduction of CO2 emissions put forward at
the 21st Conference of Parties (COP21) meetings held in Paris demonstrated that
countries are moving toward greater utilization of renewable energy sources.8
Moreover, the EIA also has projected a global expansion of nearly 29 percent in
renewable energy generation by 2040.9

National College of Business Administration and Economics, Lahore, Pakistan. His research focuses
on financial, energy, and tourism economics, energy finance, and development. Dr. Shahbaz has
published more than 200 research papers in refereed publications, having a cumulative impact factor
of 320, including Journal of Banking and Finance, Energy Economics, Energy Policy, Renewable
and Sustainable Energy Reviews, Energy, Economics Bulletin, and Tourism Management, as
a sampling. He is among the world’s top 25 authors in the “Economics” category as compiled by
IDEAS and among the top five authors in the “Economics” category in developing countries as
compiled by the World Bank’s Chief Economist David Mckenzie.
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 159

The rising demand for renewable energy is instrumental in addressing energy


security issues as well as CO2 emissions-related issues. Owing to these growing
problems, policy makers across the globe gradually are paying more attention to
renewable energy sources. Renewable energy generation in 2012 was only 23
percent of the entire energy generation across the globe.10 As for individual G7
countries, their share of renewables (as a percentage of primary energy) were 17
percent for Canada, 13 percent for France, 12 percent for Germany, 12 percent for
Italy, 6.9 percent for Japan, 3.8 percent for the United Kingdom, and 13 percent for
the United States.11 Canada’s rising share of renewables deserves special attention.
Apart from its massive hydropower generation infrastructure— 376 terawatts
(TW)—the nation is gradually expanding the scope of renewable energy sources
into tidal (20 megawatts—MW), wind (6.2 gigawatts—GW), and biofuel (1.9
billion liters).12
Global investment in the renewable energy sector in 2015 reached U.S. $285.9
billion, an increase of 4.9 percent relative to 2014.13 Interestingly, developing
economies have taken the leadership position in investment in renewable energy
totaling U.S. $156 billion relative to U.S. $130 billion for developed economies.
This may partly be explained by the decrease in subsidies for wind and solar
photovoltaic (PV) power projects in Europe and North America, falling prices of
wind and solar PV technologies, and rising demand for renewable energy in de-
veloping and emerging economies.14 This may suggest that renewable energy
markets in these G7 countries are gradually losing their significance among
investors.
This is the motivation of researchers for investigating the drivers of renewable
energy production in G7 countries. Previous studies evaluating the main de-
terminates of renewable energy production have provided mixed results. Fur-
thermore, drivers of renewable energy production might change under different
scenarios in G7 countries. For instance, some of these drivers are (1) the rising
demand and growing attractiveness of the renewable energy market in developing
and emerging economies; (2) a fall in oil prices due to an economic slowdown in
China, which is the market leader in solar power generation; (3) increased policy
level concerns regarding CO2 emissions; and (4) a higher level of pursuit of re-
newable energy sources in the Asian and African nations following the COP21
meeting. Therefore, keeping the 2020 objectives of G7 nations requires an un-
derstanding of the relationship among renewable energy generation, economic
growth, CO2 emissions, and oil prices.
The primary purpose of this paper is to investigate the long-run relationship and
short-run dynamics among renewable energy consumption, CO2 emissions, and
oil prices in G7 countries. This work contributes to the existing literature in three
essential ways. First, we use the multivariate nonlinear autoregressive distributed
lags (NARDL) method, proposed by Y. Shin et al.,15 to assess the asymmetric
association both in the long run and short run between renewable energy
160 THE JOURNAL OF ENERGY AND DEVELOPMENT

consumption, real income, CO2 emissions, and oil prices. The intricacy of the
economic structure and mechanisms may result in possible non-linearity and
asymmetry among these variables. If the changes in structural and policy regimes,
national and global events, and economic disturbances are taken into consider-
ation, then its multifarious impacts on time-series data can be observed. Second,
we carry out the empirical work using quarterly data over 1955 through 2014
following M. Shahbaz et al.16 for a robust empirical analysis by avoiding the
sample size problem.17 In the literature of energy economics, the majority of
studies have used annual data and a number of studies have pointed out the dif-
ferential consequences of choice of frequency (for more details, see M. Shahbaz
et al.).18 The quarterly frequency also is useful in capturing potential non-linearity
and asymmetry in the relationships. Third, we test for unit roots in individual
variables using the Zivot and Andrews method, which allows for endogenous
structural breaks in variables.19
Our empirical evidence indicates that income has a significant effect on re-
newable energy consumption in a symmetric manner in the United States, the
United Kingdom, France, and Germany and in an asymmetric manner in Japan,
but renewable energy consumption is insensitive to income for Italy in the long
run. Carbon emissions positively affect renewable energy consumption in a sym-
metric manner for the Untied States, France, Germany, Japan, and Italy in the long
run. Renewable energy consumption is asymmetrically affected by carbon emis-
sions in Canada but insignificant for the United Kingdom in the long run. Re-
newable energy consumption is asymmetrically influenced by oil prices in the
United States and symmetrically in the United Kingdom and France. This suggests
a dire need to establish a global green energy environment as our findings have
important implications for energy policy makers around the world.

Literature Review

There has been voluminous research on the association between energy con-
sumption, economic growth, and emissions (some of the recent studies are
N. Apergis, S. Asongu, S. Bedir and V. Yilmaz, P. Chen et al., M. Robaina-Alves
et al., K. Saidi and S. Hammami, M. Shahbaz et al., A. Sinha and S. Sen, among
many others).20 Most of the studies carried out were based on non-renewable
fossil-fuel consumption and the present context calls for evaluating and assessing
the impact of renewable energy consumption on economic growth and CO2
emissions. Therefore, we will focus on the recent studies investigating the impacts
of renewable energy consumption.
A. Marques et al. analyzed the determinants of renewable energy supply for
24 European countries during the time span of 1990-2006.21 By applying fixed
effects vector decomposition (FEVD), the authors have found that deployment of
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 161

renewable energy is held back by traditional non-renewable energy sources and


CO2 emissions. S. Rafiq and K. Alam22 replicated P. Sadorsky’s23 renewable
energy consumption function for energy investors in emerging economies. They
noted that income and CO2 emissions have positive and significant effects on
renewable energy consumption; however, oil prices affect renewable energy
consumption negatively but insignificantly.
A. Marques and J. Fuinhas replicated the previous study for 21 European
countries for the period of 1990-2006 and found that the lobbying effect of tra-
ditional non-renewable energy sources restrained renewable energy utilization in
the presence of structural breaks.24 The subsequent study by A. Marques and
J. Fuinhas analyzed 24 European countries over the 1990-2006 time span and they
applied a generalized method of moments (GMM) approach in this case.25 Apart
from the lobbying effect of traditional non-renewable energy sources on renew-
able energy sources, their study uncovered that the awareness level, climate
change and CO2 emissions targets, and the rising price level of non-renewable
energy sources impeded further development of the renewable energy market in
Europe.
A. Tiwari analyzed the association between renewable energy consumption,
economic growth, and CO2 emissions for India.26 By applying innovations anal-
ysis, the author found that renewable energy consumption leads to an increase in
gross domestic product (GDP) but a decline in CO2 emissions. M. Aguirre and
G. Ibikunle examined the determinants of renewable energy growth in BRICS
(Brazil, Russia, India, China, and South Africa) nations for 1990-2010.27 By ap-
plying FEVD, panel corrected standard errors (PCSE), and generalized least
squares (GLS) estimation techniques, they found that, although renewable energy
consumption can address environmental degradation issues, stringent and rising
energy demand conditions hinder policy makers’ ability to increase the share of
renewable energy in the energy mix.
N. Apergis and J. Payne undertook the same analysis for 25 of the Organization
for Economic Cooperation and Development (OECD) countries over the time span
of 1980-2011.28 They considered per-capita renewable energy consumption, per-
capita real GDP, per-capita CO2 emissions, and real oil prices in their model. By
applying panel cointegration and an error-correction model, they discovered ev-
idence of the feedback hypothesis between all the variables. In a nearly similar
study by N. Apergis and J. Payne that was carried out for the analysis of seven
Central American nations, the authors looked at evaluating the impact of the es-
tablishment of the Energy and Environment Partnership with Central America
initiative for developing the utilization of renewable energy sources in 2002.29
They noted that the influence of renewable energy consumption on coal and oil
prices strengthened during the post-2002 period. During the same period the re-
searchers found a greater sensitivity of real GDP per capita to carbon emissions
per capita. A subsequent study by N. Apergis and J. Payne was carried out for 11
162 THE JOURNAL OF ENERGY AND DEVELOPMENT

South American nations with similar results obtained in that assessment.30 Indeed,
renewable energy consumption was found to be positively sensitive to variations
of real GDP per capita, CO2 emissions per capita, and real oil prices in the long
run. Their empirical results also indicated a feedback effect indicating the im-
portance of renewable energy consumption in improving the growth of output and
in mitigating CO2 emissions.
A. Omri and D. Nguyen analyzed the association between renewable energy
consumption, CO2 emissions, real oil prices, per-capita GDP, and trade openness
for 64 countries over the period 1990 to 2011.31 By applying the system GMM
method on the dataset, they reported that CO2 emissions and trade openness are the
major determinants of renewable energy consumption. Furthermore, A. Omri et al.
disclosed that the results obtained in the previous study did not change with the
application of different methodologies, thereby reaffirming the robustness of the
findings.32 A. Omri et al. investigated the determinants of renewable energy
consumption for developed and developing economies.33 By using the simulta-
neous equation GMM approach, they noted that income, CO2 emissions, and oil
prices affect renewable energy consumption positively but oil consumption is
negatively and significantly affected.
S. Rafiq et al. analyzed the relationships among output, CO2 emissions, and
renewable energy generation for India and China during the 1972 to 2011 period.34
Following a multivariate error-correction approach, they found evidence of the
conservation hypothesis between CO2 emissions and renewable energy generation
and between renewable energy generation and output for the short run. Addi-
tionally, their research supported the feedback hypothesis for the long run in the
case of India. In terms of the Chinese context, they found evidence of the con-
servation hypothesis between CO2 emissions and renewable energy generation
and the growth hypothesis between renewable energy generation and output for
the short run, whereas the feedback hypothesis was discovered for the long run.
A. Tiwari et al. analyzed the association between economic growth and re-
newable and nonrenewable energy production for 12 sub-Saharan African coun-
tries during the period of 1971 to 2011.35 Following both linear and hidden
cointegration techniques, they found mixed results in terms of the effects of
conservation policies on economic growth, thereby impacting the future of energy
security.
One of the recent studies by A. Vaona demonstrated an innovative approach of
finding the association between renewable energy generation and import de-
mand.36 The author analyzed 26 different countries over a range of periods and
utilizing different estimation methods. The result of this study was that renewable
energy generation can ensure the sustainability of imports by reducing the de-
pendence on fossil-fuel-based energy.
B. Lin et al. explored the macroeconomic determinants of renewable electricity
consumption for the Chinese economy.37 These researchers determined that
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 163

economic growth and financial development have a positive impact on renewable


electricity consumption. Foreign direct investment, trade openness, and fossil
fuels consumption have negative effects on renewable electricity consumption.
Similarly, E. Dogan and F. Serker applied the Dumitrescu-Hurlin non-causality
approach to examine the causal relationship between CO2 emissions and their
determinants.38 They found the feedback effect between renewable energy con-
sumption and CO2 emissions but non-renewable energy consumption is caused by
CO2 emissions.39
It can be observed from this brief overview that renewable energy is part of
sustaining economic growth in various ways and is gradually turning out to be
a more salient issue. Thus, renewable energy impacts economic issues as well as
playing a role in addressing the challenges of climate change. Following the
COP21 meeting, this issue has gained more prominence at the global scale and,
therefore, given this background, the present study might find its own relevance in
the existing body of literature. A nonlinear and asymmetric association between
renewable energy consumption and its determinants may be owing to the uncertain
consequences of macroeconomic and environmental reforms and reforms imple-
mented in energy markets. In order to analyze these uncertainties, these two
factors need to be considered within the model framework. Over and above, we
have not come across any study to date that has analyzed the nonlinear and
asymmetric causal association between economic growth and energy consumption
for the case of G7 countries. As described by S. Chiou-Wei et al.,40 the intricacy of
this association can be elucidated by shifts in regimes and economic affairs, i.e.,
shifts in economic scenarios, movements of energy prices, and developments in
energy policies.

Overview of Energy Policies in G7 Countries

The economic growth of G7 economies entails a higher level of energy con-


sumption to sustain higher levels of industrialization. Across every sector in these
nations, consumption of commercial electricity is facilitating economic growth.
As these economies are developed and industrially privileged, historically they
have been able to exert a pull on potential investment and this scenario is illus-
trated in their GDP growth (figure 1). As of 2014, these countries accounted for
nearly a 45.52-percent share of global income, a 35.23-percent share of global
fossil fuel consumption, and a 27.22-percent share of worldwide CO2 emissions.41
During 1960 to 2014, the per-capita income of G7 economies has grown nearly
32.59 times, whereas the global per-capita income has grown only 2.32 times.42 In
order to catalyze and sustain this growth, the G7 demand for energy also has
risen and an increasing portion of this demand is being met by renewable energy.
Figure 1 depicts the G7 versus world GDP growth. The graph shows that before
164 THE JOURNAL OF ENERGY AND DEVELOPMENT

Figure 1
COMPARISON OF GROSS DOMESTIC PRODUCT (GDP) GROWTH: G7 ECONOMIES
a
VIZ-A-VIZ THE WORLD, 1970-2015
(in annual %)

a
G7 countries are Canada, France, Germany, Italy, Japan, United Kingdom, and United States.

year 2000, G7 economies GDP growth was greater than that of the world while
thereafter the world GDP growth turned higher. We will now take a closer look at
the individual members of the G7 and their energy policies.
Canada: The reforms in the energy policies of Canada started after the 1973 oil
crisis and these policies were targeted at energy conservation and reduction in CO2
emissions.43 In order to achieve this, Canada has been harnessing its natural re-
sources and geographical diversities in a way that it can generate a substantial
amount of clean and renewable energy. As of 2014, renewable energy constituted
nearly 19 percent of the primary energy supply of Canada. Owing to its geo-
graphical richness and large number of streams, hydropower generation in Canada
amounts to nearly 59.2 percent of the total electricity production.44 Canadian wind
and biomass energy generation amounts to nearly 3.5 percent and 1.4 percent of the
country’s electricity generation, respectively.45 Based on the New Brunswick’s 2007
Climate Change Action Plan (CCAP), the fifth assessment of CCAP for 2014–2020
has projected a reduction of CO2 emissions by 10 percent below their 1990 level by
2020 and a 75-percent to 80-percent reduction by 2050.46 In order to achieve this, the
government of New Brunswick also has embraced a sustainable economic growth
path by promoting energy efficiency and renewable energy sources.
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 165

Japan: As Japan is a net importer of crude oil (the second largest in the world),
any fluctuations in oil price can have a direct impact on its trade deficit. A rise in
liquefied natural gas (LNG) prices in the Asian market through 2014 caused se-
rious pressure on the Japanese economy.47 Following the Fukushima Daiichi nu-
clear accident in 2011, the Japanese government actively focused on ways to bring
more energy diversity and employ alternative and innovate energy sources to the
energy sector in order to sustain the economy and address rising energy demand.
As a result, the government initiated the fourth Strategic Energy Plan (SEP), which
focused on energy security, economic efficiency, and environmental protection.
This plan was targeted at reducing the share of fossil-fuel-based energy sources
and increasing the share of nuclear and renewable energy sources by 2030.48
Following the COP21 meeting in 2015, Japan also set a target of reducing CO2
emissions by 26 percent by 2030 and 80 percent by 2050.49 However, the in-
troduction of feed-in tariff mechanisms in 2012 with the desired goal of catalyzing
renewable energy generation resulted in the opposite and unintended conse-
quences. On one hand, the capacities of solar photovoltaics (PV), wind, and
geothermal energy were increased substantially while, on the other hand, the
continuous rise in the subsidies for renewable energy generation, in turn, resulted
in severe cost implications, which affected their economic growth.50
France: Though France has been a sizeable exporter of crude oil, falling oil
prices along with environmental concerns are compelling the nation to shift its
energy base to more renewable energy sources. One sign of this is the country
looking at ways to better store renewable energy and its acquisition of the battery
manufacturing firm Saft is indicative of this.51 France’s commitment to increasing
renewables can be seen in its National Renewable Energy Action Plan (NREAP),
which calls for the share of renewable energy in final energy consumption to reach
23 percent by 2020. In 2015, the parliament of France passed a comprehensive
energy and climate law, according to which the share of renewable energy gen-
eration should reach 40 percent of total energy generation by 2030.52 As of 2014,
the share of renewable energy in France’s total electricity production was nearly
19.5 percent with hydropower accounting for 13.8 percent (nearly 28 TW), wind
energy at 3.5 percent of generation (close to 9,100 MW), and PV at 1.2 percent of
production (approximately 5,300 MW).53 Moreover, the decommissioning of coal-
fired power plants further augmented the growth in renewable energy generation.
Germany: This country has set an ambitious GHG emission reduction target of
40 percent below 1990 levels and is on track to achieve a 32-percent emission
reduction. Traditionally, Germany has used a large amount of coal, which it wants
to replace with renewables. Germany’s electricity generation is fueled by about
24.7 percent brown coal, 18.3 percent hard coal, and 8.3 percent natural gas.54 In
2013, the coalition government in Germany revised the renewable energy targets
with the goal of renewable energy consumption reaching 18 percent of total
166 THE JOURNAL OF ENERGY AND DEVELOPMENT

energy consumption by 2020, 30 percent by 2030, 45 percent by 2040, and 60


percent by 2050.55 Presently, the share of renewable energy generation accounts
for 33.9 percent of total energy generation; primarily consumed in the electricity,
heating, and transportation sectors. In order to hedge against possible fluctuations in
crude oil prices and to shift the energy base toward a more sustainable energy mix,
the government is looking into different forms of biofuel, hydropower, PV, and
wind energy. Recently, the government has spurred on growth in geothermal energy
generation by bringing in a feed-in tariff. However, the increased uncertainty in the
European Union and policy-level voids are creating problems for renewable energy
solution providers due to increased costs and lack of supportive infrastructure.56
Italy: The country is having challenges in meeting electricity demand due to
inadequate existing energy infrastructure that results in Italy relying upon elec-
tricity imports. As of 2010, 12.9 percent of Italy’s electricity supply was imported.
Thus, the trade balance of Italy is highly dependent on the crude oil price
movements via the electricity import route.57 Moreover, Italy is also the fourth
largest importer of natural gas in the world. As of 2010, 67.2 percent of Italy’s
energy demand was met by electricity generated from fossil fuel based sources and
only 20.6 percent from renewable energy.58 A consequence of the Fukushima
Daiichi nuclear accident in 2011 was that Italy abandoned its nuclear energy
program. Italy’s energy policy must meet multiple goals of providing enough
energy to sustain its economic growth, ensuring energy security due to its reliance
on imports, and abiding by the EU renewable energy directive targets, according
to which it needs to generate at least 17 percent of final energy consumption from
renewable energy sources by 2020.59 The Italian government has initiated the
National Renewable Energy Action Plan and has set its own renewable energy
generation target of 26.4 percent by 2020.60 In order to achieve this target, the
government has introduced feed-in tariffs, tradable renewable energy, market
premiums, and reverse auctions. In 2012, Italy was ranked as the second largest
solar market in the world with an installed capacity of 17,000 MW.61 However,
subsidization of PV energy at this level might affect Italy’s competitiveness at
a global scale as it can impact economic growth patterns by increasing cost bur-
dens. Therefore, the government of Italy is trying to develop other renewable
energy sources like hydropower and wind power.
The United States: During 1980 to 2010, U.S. crude oil imports increased by
200 percent,62 thereby exposing the economy to high crude oil price volatility.63 In
order to combat this situation, the United States facilitated the development of
renewable energy sources and, as a result, by 2011 the energy generated from
renewables exceeded that of nuclear. The renewable energy sources in the United
States include hydropower, wind energy, PV, and geothermal energy. In the
pursuit of expanding sources of renewable energy, agricultural waste is being
transformed into bioethanol and biodiesel. These renewable energy sources are
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 167

produced domestically and it is expected that these forms of clean fuels can not
only reduce the cost of oil imports, but also help to ensure energy security at the
grassroots level. Though there has been some protest against these types of bio-
fuels owing to their contribution toward environmental degradation, the in-
troduction of second-generation biofuels has been replacing them rapidly. That
being said, it should be noted that the American Recovery and Reinvestment
Act of 2009 injected huge amounts of direct investment and tax credits into the
U.S. renewable energy sector.64 Additionally, the U.S. government introduced the
SunShot initiative in 2011 to lower the production costs of PV systems by 75
percent by 2020 in order to increase the cost competitiveness of solar energy.65
The United Kingdom: Last, like the rest of the G7 countries, the United
Kingdom is exposed to the crude oil price volatility due to the rising volume of
crude oil imports. The January 2008 Energy Bill was enacted to ensure energy
supply security with a special focus on offshore natural gas and oil development.66
This framework also focused on addressing the problem of increasing CO2
emissions and climate change. As of 2012, the United Kingdom reached the
turning point of becoming a net energy importer as the nation had become more
reliant on natural gas and coal, representing 41 percent and 29 percent of the
energy mix, respectively, while renewable energy was only 12 percent and nuclear
accounted for 18 percent.67 Among the U.K.’s renewable energy sources, biomass
energy has the maximum share of 36.8 percent, followed by onshore wind energy
at 29.4 percent.68 However, growth in the population and transportation sector has
highlighted the country’s energy security issues and the existing size of the re-
newable energy market has not been sufficient to meet rising energy demand. As
of 2014, the United Kingdom introduced new incentives for hybrid vehicles along
with energy efficiency programs such as the Energy Performance Certification for
buildings built after 2007 to help ameliorate the situation.69 In order to address
energy security, economic growth, and climate-related goals, the U.K. government
needs to look into other emerging forms of renewable energy, such as hydropower
and PV systems. The government might look to transfer the cost of imports toward
feed-in tariffs of these energy systems in order to boost these sectors. However, for
energy security to be ensured, the government should consider more nuclear
power generation, which is presently running below its full capacity; however, due
to political issues, the future of nuclear power plants in the United Kingdom is still
quite uncertain.70

Methodology and Data

Methodology: Recent studies reveal that tests of cointegration using linear


econometric methods may lead to erroneous conclusions. Y. Shin et al.71 propose
the nonlinear autoregressive distributed lag (NARDL) model that allows for
168 THE JOURNAL OF ENERGY AND DEVELOPMENT

nonlinearity and asymmetry while testing for long-run and short-run relations. The
linear ARDL model has the following form:

DRECt = r0 + rREC RECt1 + rY Yt1 + rCO2 CO2;t1 + rOP OPt1


Xp1 Xq1 Xq1
+ i=1
a i DRECt1 + i=0
ui DYti + b DCO2;ti
i=0 i
Xq1
+ i=0
gi DOPti + et ð1Þ
where RECt, Yt, CO2t, and OPt, refer to renewable energy consumption, real gross
domestic product (measure of income), CO2 emissions, and real oil prices, re-
spectively; Δ denotes change, and p represents the lag orders for the variables,
respectively. The ARDL model (1) is suitable when the relationship among the
variables is linear and symmetric. Absent linear and symmetric relations, the
model may generate erroneous results. The NARDL model proposed by Y. Shin
et al.72 allows for nonlinearity and asymmetry. Decomposing exogenous variables
into positive Xt+ and negative Xt– partial sums such that:
X
t X
t   X
t X
t  
Xtþ ¼ DXjþ ¼ 
max DXj ; 0 and Xt ¼ 
DXj ¼ min DXj ; 0
j¼1 j¼1 j¼1 j¼1

And introducing long-run and short-run asymmetries in equation (1) leads to the
following general form of the NARDL model (Y. Shin et al.):73

DRECt ¼r0 þ rREC RECt1 þ rþ þ   þ


Y þ Y t1 þ r Y  Y t1 þ r COþ COþ
2;t1
Xp
2
  þ þ  
þ rCO2 CO2;t1 þ rOPþ OPt1 þ rOP OPt1 þ a DREC ti
i¼1 i
Xp  
þ uþ þ 
i DY ti þ ui DY ti


Xp
i¼1
 
þ þ  
þ i¼1
b i DCO 2;ti þ b i DCO 2;ti
Xp  
þ i¼1
gþ þ 
i DOPti þ g i DOPti þ et


ð2Þ

Long-run asymmetries are tested using a Wald test of the respective null hy-
potheses, where



¼ r þ 
Y  ; rCOþ ¼ r CO
2
; and rþ
OPþ
¼ r
OP:
2

The long-run effects of positive and negative shocks of exogenous variables on


renewable energy consumption are computed as
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 169

LX þ ¼ rþ 
X =rREC and LX ¼ r X =r REC


where Xt denotes, respectively, Yt, CO2, and OPt. Similarly, short-run asymmetry
of the reaction of renewable energy consumption to changes in explanatory var-
iables is tested using a Wald test of the null hypotheses where
ai + = a 
i ; bi + = bi ; and g i + = g i for i = 1; . . . ; p:

Failure to reject long-run and short-run asymmetries leads to the traditional


ARDL model while non-rejection of either long-run or short-run asymmetry leads
to the NARDL model with short-run or long-run asymmetry.
Once the appropriate NARDL model is selected based on an information cri-
teria and the results of long-run and short-run asymmetry tests are completed, we
proceed to compute the asymmetric responses of renewable energy consumption
to unit positive and negative changes in income, CO2 emissions, and oil prices
using numerical derivatives of RECt relative to the desired factor as follows:
Xh @REC tþj Xh
@REC tþj
mh þ ¼ þ and m 
¼
j¼0 @X t h
j¼0
@X t
 
with Xt 2 Yt ; CO2;t ; OPt and h ¼ 0; 1; . . . . . . ð3Þ
where h ! ‘; mþ þ
þ
h ! LX ; and mh ! LX .
þ
74

Data: The data on renewable energy consumption in kilograms (kgs) of oil


equivalent are from BP’s Global Statistical Review.75 The data on real gross do-
mestic product (in constant 2010 U.S. dollars) and CO2 emissions (metric tons) are
obtained from the World Bank’s World Development Indicators.76 The data on
crude oil prices (West Texas intermediate) were collected from the Federal Re-
serve Bank of St. Louis.77 The study covers the period of 1955 to 2014 for G7
countries. Total population is used to transform all the variables into per-capita
units with the exception of real oil prices.

Empirical Results and Their Interpretations

Table 1 presents the descriptive statistics and stochastic properties of the data
series. The average renewable energy consumption ranges from 1,382.809 kt in
the United Kingdom to 6,562.210 kt in the United States. It is worth noting that
energy consumption is more important in North America (United States and
Canada) in comparison to that in Europe and Japan. The previous empirical results
can be explained by at least two reasons. First, the United States produces most of
the world’s utilized energy. Indeed, the United States is the second largest energy
170 THE JOURNAL OF ENERGY AND DEVELOPMENT

Table 1
a
LONG-RUN (WLR) AND SHORT-RUN (WSR) ASYMMETRY TESTS

WLR WSR

GDP 0.661 0.798


[0.422] [0.378]
CO2 0.188 0.276
United States
[0.668] [0.603]
OP 3.748* 9.728***
[0.061] [0.004]
GDP 1.209 0.689
[0.289] [0.420]
CO2 0.687 2.800
United Kingdom
[0.420] [0.115]
OP 0.167 1.790
[0.689] [0.201]
GDP 3.694* 2.195
[0.062] [0.147]
CO2 5.289** 0.030
Canada
[0.07] [0.863]
OP 0.109 2.423
[0.743] [0.128]
GDP 0.045 0.078
[0.832] [0.781]
CO2 1.707 0.019
France
[0.199] [0.892]
OP 1.445 9.478***
[0.237] [0.004]
GDP 1.865 0.04
[0.181] [0.835]
CO2 0.001 0.377
Germany
[0.974] [0.543]
OP 0.177 0.117
[0.677] [0.734]
GDP 4.374** 3.349*
[0.045] [0.077]
CO2 0.781 0.013
Japan
[0.383] [0.910]
OP 2.673 0.050
[0.112] [0.824]

(continued)
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 171

Table 1 (continued)
a
LONG-RUN (WLR) AND SHORT-RUN (WSR) ASYMMETRY TESTS

WLR WSR

GDP 0.605 0.106


[0.445] [0.748]
CO2 0.043 0.461
Italy
[0.838] [0.504]
OP 10.110*** 0.181
[0.004] [0.675]

a
WSR and WLR refer to the Wald statistics for the short- and long-run symmetry null hypotheses.
GDP = income as measured by gross domestic product; CO2 = carbon dioxide emissions; and OP =
oil prices. The numbers in the brackets are the p-values.
***, **, and * indicates rejection of the null of symmetry at 1 percent, 5 percent, and 10 percent,
respectively.

producer in the world but it only has 4.5 percent of the world’s population. Second,
the U.S. industry has an older capital-intensive industrial base, which is very
costly to upgrade toward more energy efficient processes. The maximum re-
newable energy consumption was recorded in the 1970s in the United States and
Germany, by the late 1990s in the United Kingdom, and in the 2000s in Japan,
France, Italy, and Canada. Renewable energy consumption is negatively skewed
and displays excess kurtosis. The Jarque-Bera statistic indicates the non-normality
of renewable energy consumption for all countries with the exception of the
United Kingdom.
The average income is the highest for the United Kingdom and, not surpris-
ingly, the lowest for Italy, which is caused by that country’s recent economic
crisis. Yt shows negative skewness and positive excess kurtosis but not for the
United States. The Jarque-Bera test fails to reject normality of income for all
countries but Japan and Italy where income weakly deviates from normality.
CO2 emissions and oil prices are, on average, the highest in the United States
and the lowest, respectively, in Italy and Japan. CO2 emissions are negatively
skewed while oil prices are positively skewed with the exceptions of the United
Kingdom and Canada. CO2 emissions and oil prices show positive excess kurtosis.
The Jarque-Bera test rejects the null of normality for CO2 emissions and oil prices
in most of the countries. Even distribution on both sides of the mean of a bell
curve symbolizes the symmetric distribution of the data. However, this distribu-
tion is an idealistic one and can be hardly found in any practical scenario. This, in
turn, leads toward the reliance on asymmetric empirical analysis. The presence of
non-normality, verified by the Jarque-Bera statistics, is handled by the asymmetric
172 THE JOURNAL OF ENERGY AND DEVELOPMENT

ARDL model as it has the capability of taking the nonlinearities of the data into
account.78
Before proceeding to the nonlinear ARDL bounds testing approach, we need to
test the order of integration of renewable energy consumption, income, CO2
emissions, and oil prices. In doing so, we have applied the Zivot-Andrews (ZA)
unit root test that accommodates a single unknown structural break stemming in
the variables. The choice of performing the ZA test rather than the augmented
Dickey-Fuller (ADF) or Phillips-Perron (PP) tests is justified by the possible
existence of a structural change in the time series due to the occurrence of crises
and geopolitical events during the 1955 to 2014 period. The results are reported in
table 2. The empirical results indicate that all the variables are I(1) for all countries
with the exception of Germany where RECt, Yt, and CO2 are found to be stationary
at I(0).
We now estimate the models specified in equations (1) and (2) to explain the
impact of income, CO2 emissions, and oil prices on renewable energy consump-
tion in G7 countries. The results are reported in table 2. We find that loading on the
lagged renewable energy consumption is negative and significant for all G7
countries indicating that the estimated NARDL model is stable for these countries.
The pass-through of income on renewable energy consumption reveals that
income has a symmetric and positive long-run effect on renewable energy con-
sumption in the United States, France, and Germany. Income has a weak sym-
metric and negative effect on renewable energy consumption in the United
Kingdom. Income is found to have no long-run effect on renewable energy con-
sumption in Italy. Additionally, income impacts renewable energy consumption in
an asymmetric manner in Canada and Japan. Indeed, an increase in income will
increase renewable energy consumption in Japan but decreases renewable energy
consumption in Canada in the long run. However, a decrease of income raises
renewable energy consumption in both countries. It is worth noting that renewable
energy consumption in the United Kingdom is the most sensitive to variations of
income in the long run as it shows the highest long-run coefficient relative to
income in absolute value. This suggests that in the United Kingdom, policy makers
may want to target lower carbon prices for poorer energy users. Otherwise, in
practice this can lead to problems in identifying people eligible for support, ad-
ministrative complexity, and take-up as well as problems of reducing the
incentives for those affected to work if support is conditional on income or em-
ployment status.79 In the short run, contemporaneous changes in income positively
impact renewable energy consumption in the United States, Canada, and Italy but
do not impact renewable energy consumption in the United Kingdom, France, and
Germany. In Japan, contemporaneous increases of income and one-period lagged
changes of income cause renewable energy consumption to go up. The empirical
results reveal that income plays an important role in raising demand for renewable
energy. These results are consistent with the findings of P. Sadorsky for G7
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 173

Table 2
a
ESTIMATION RESULTS FOR G7 COUNTRIES

United States United Kingdom LR & SR Canada GDP & France OP


OP LR & SR Symmetries CO2 LR SR Asymmetry
Asymmetries Asymmetry
RECt1 -0.207** RECt1 -0.182** RECt1 -0.112 RECt1 -0.453***
(0.092) (0.071) (0.103) (0.109)
GDPt1 0.076** GDPt1 -0.496* GDPþt1 -0.081 GDPt1 0.400***
(0.038) (0.295) (0.073) (0.098)
CO2t1 0.174** CO2t1 -0.584 GDPt1 0.113 CO2t1 0.177***
(0.080) (0.665) (0.263) (0.044)
OPþ
t1 -0.139** OPt1 1.441** CO2þt1 0.185 OPt1 0.048***
(0.056) (0.690) (0.161) (0.011)
OP
t1 0.037* DRECt1 0.283* CO2
t1 0.013 DRECt1 0.051
(0.022) (0.169) (0.077) (0.148)
DRECt1 -0.091 DRECt2 -0.050 OPt1 -0.002 DCO2t 0.584***
(0.137) (0.110) (0.023) (0.071)
DRECt2 -0.155*** Const. 1.451** DRECt1 -0.077 DOPþt 0.180**
(0.052) (0.552) (0.168) (0.068)
DGDPt 0.240*** DGDPt 0.280** Const. 3.161***
(0.077) (0.128) (0.762)
DCO2t 0.687*** DCO2t 0.459**
(0.060) (0.075)
DOPt1 0.298*** Const. 0.908
(0.104) (0.824)
Const. 1.736**
(0.776)

LGDP 0.367** LGDP -2.722* LGDPþ -0.723 LGDP 0.884***


[0.021] [0.072] [0.349] [0.000]
LCO2 0.844*** LCO2 -3.201 LGDP 1.009 LCO2 0.391***
[0.000] [0.365] [0.644] [0.000]
LOPþ -0.674 LOP 7.900** LCO2þ 1.654** LOP 0.106***
[0.113] [0.022] [0.020] [0.000]
LOP 0.180* LCO2 0.117
[0.088] [0.856]
LOP -0.020
[0.919]

AIC -389.237 AIC -78.009 AIC -320.726 AIC -287.705


SIC -356.548 SIC -57.067 SIC -291.879 SIC -258.859
J-B 0.882 J-B 8.051 J-B 0.697 J-B 293.700
[0.643] [0.018] [0.706] [0.000]
174 THE JOURNAL OF ENERGY AND DEVELOPMENT

Table 2 (continued)
a
ESTIMATION RESULTS FOR G7 COUNTRIES

Germany Japan Italy


LR & SR Symmetries GDP LR & SR Asymmetries OP LR Asymmetry
RECt1 -0.151* RECt1 -0.329*** RECt1 -0.349*
(0.086) (0.103) (0.190)
GDPt1 0.307 GDPþ
t1 0.151*** GDPt1 -0.034
(0.238) (0.043) (0.102)
CO2t1 0.507 GDP
t1 1.008*** CO2t1 0.300
(0.364) (0.205) (0.214)
OPt1 0.003 CO2t1 0.230*** OPþ
t1 0.036
(0.080) (0.080) (0.022)
DRECt1 0.189 OPt1 0.003 OP
t1 -0.004
(0.156) (0.014) (0.041)
DRECt2 0.087 DRECt1 -0.065 DRECt1 0.172
(0.147) (0.167) (0.224)
Const. 1.119* DRECt2 -0.134 DRECt2 0.154
(0.607) (0.081) (0.206)
DRECt3 0.016 DGDPt 0.521**
(0.073) (0.225)
DRECt4 0.031 DCO2t 0.623***
(0.073) (0.165)
DGDPþ
t 0.454*** DCO2t1 -0.354*
(0.149) (0.190)
DGDP
t1 -0.694* DOPt2 -0.053**
(0.357) (0.027)
DCO2t 0.609*** Const. 1.995*
(0.074) (1.012)
Const. 1.945***
(0.608)

LGDP 2.028* LGDPþ 0.459*** LGDP -0.097


[0.064] [0.000] [0.719]
LCO2 3.354** LGDP 3.067*** LCO2 0.859***
[0.009] [0.001] [0.000]
LOP 0.023 LCO2 0.701*** LOPþ 0.104*
[0.967] [0.000] [0.099]
LOP 0.011 LOP -0.012
[0.792] [0.915]

(continued)
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 175

Table 2 (continued)
a
ESTIMATION RESULTS FOR G7 COUNTRIES

Germany Japan Italy


LR & SR Symmetries GDP LR & SR Asymmetries OP LR Asymmetry
AIC -16738 AIC -272.235 AIC -272.768
SIC -141.078 SIC -236.103 SIC -232.261
J-B 1273 J-B 4.266 J-B 1.444
[0.000] [0.118] [0.456]

LR = long run; SR = short run; OP = oil price; REC = renewable energy consumption; GDP =
gross domestic product; Const = constant; AIC = Akaike information criteria; SIC = Schwarz
information criteria; and J-B = Jarque-Bera statistic. Only significant short-run coefficients are
reported in this table 2. Standard errors of the estimated coefficients are in parenthesis. The p-values
+ –
of statistical tests are in brackets. LX and LX indicate the positive and negative long-run
coefficients, respectively. Lag orders of the three NARDL models are selected according to the
Akaike and Schwarz Information criteria. p = 3, q = 2 for USA, UK, and Germany; p = 2, q = 2 for
Canada and France; p = 5, q = 2 for Japan; p = 3, q = 4 for Italy.
***, **, and * indicates rejection of the null of symmetry at 1 percent, 5 percent, and 10 percent,
respectively.

countries, S. Rafiq and K. Alam for emerging economies, A. Omri et al. for de-
veloped and developing countries, and B. Lin et al. for China, who reported that
income is positively linked with renewable energy consumption.80
The pass-through of CO2 emissions on renewable energy consumption in-
dicates that emissions show a symmetric and positive long-run effect on renewable
energy consumption in the United States, France, Germany, Japan, and Italy but
was found to be insignificant in the case of the United Kingdom. In actuality,
countries have adopted different policies to promote the use of renewable energy.
Nonetheless, the different experiences show that some policies failed in achieving
their goals. Indeed, the United States adopted a production tax credit for wind,
closed-loop biomass, and poultry waste in order to make them more cost effec-
tive to produce electricity from renewable energy rather than from fossil fuels.
However these tax advantages are uncertain owing to the fact that no one knows
when the laws will be repealed. In fact, if there is a lack of credit the companies do
not receive compensation and it may no longer be profitable to produce electricity
from these renewable energy sources. A similar reluctance and uncertainty was
observed in Germany and France. In Germany the amount of money set aside to
support renewable energy use has been already reached. In France, the bidders
receiving financial aid do not always translate into renewable energy projects
on the ground. In the case of the United Kingdom the government adopted
a “feed-in-tariffs with contract for difference”81 to encourage low-carbon elec-
tricity projects.82 Renewable energy consumption is the most sensitive to CO2
176 THE JOURNAL OF ENERGY AND DEVELOPMENT

emissions in Germany with a long-run coefficient equal to 3.354 and the least
sensitive in France with a long-run coefficient equal to 0.391. The previous
results could be due to the differences of policies adopted by the various
countries and the extent to which those policies were successful. This can be
attributed to the share of renewable energy in the energy mix of Germany and
France, i.e., the percentage of renewable energy consumption in Germany is
more than 30 percent, whereas in France it is near 11 percent.83 Moreover, in
Canada an increase of CO2 emissions causes renewable energy consumption to
increase while a decrease of emissions does not impact renewable energy con-
sumption significantly.
Turning to the analysis of the short-run effect of CO2 emissions on renewable
energy consumption, empirical results reveal that contemporaneous changes of
emissions have a symmetric and positive impact on renewable energy consump-
tion in the United States, Canada, France, Japan, and Italy. Moreover, in Italy,
carbon emissions have a positive cumulative effect on renewable energy con-
sumption. Additionally, changes in CO2 emissions do not show any significant
short-run impact on renewable energy consumption in the United Kingdom and
Germany. In France, CO2 emissions impact renewable energy consumption
asymmetrically; a contemporaneous increase of CO2 emissions causes renewable
energy consumption to increase while a contemporaneous decrease of the former
does not significantly impact the latter. Overall, our empirical analysis indicates
the positive impact of CO2 emissions on renewable energy consumption. This
empirical finding is similar to the research results of P. Sadorsky, S. Rafiq and
K. Alam, and A. Omri and D. Nguyen, who noted the significant role of CO2
emissions in stimulating renewable energy consumption.84
The pass-through of oil prices on renewable energy consumption reported in
table 2 shows that in the long run, increases in oil prices cause renewable energy
consumption to increase in the United Kingdom and France with a higher mag-
nitude witnessed in the United Kingdom. Nevertheless, oil price changes impact
renewable energy consumption insignificantly in the long run in Canada, Ger-
many, and Japan. Oil prices show an asymmetric impact on renewable energy
consumption in the long run for the United States and Italy. Indeed, an increase of
oil prices tends to raise renewable energy consumption in Italy but a decrease of
oil prices causes renewable energy consumption to decrease in the United States
and to increase in Italy. Thus, oil prices have a significant short-run effect on
renewable energy consumption in three out of the seven countries. A one-lagged
period change in oil prices positively impacts renewable energy consumption in
the United States, whereas a two-lagged period change in oil prices negatively
impacts renewable energy consumption in Italy. However, oil price changes drive
renewable energy consumption in an asymmetric manner in France, where it is
noted that a contemporaneous increase of oil prices causes renewable energy
consumption to rise. The positive impact of oil prices on renewable energy
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 177

Figure 2
a
DYNAMIC MULTIPLIERS FOR THE UNITED STATES

a
GDP = gross domestic product; REC = renewable energy consumption; OP = oil prices.
178 THE JOURNAL OF ENERGY AND DEVELOPMENT

Figure 3
a
DYNAMIC MULTIPLIERS FOR THE UNITED KINGDOM

a
GDP = gross domestic product; REC = renewable energy consumption; OP = oil prices.
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 179

Figure 4
a
DYNAMIC MULTIPLIERS FOR CANADA

a
GDP = gross domestic product; REC = renewable energy consumption; OP = oil prices.
180 THE JOURNAL OF ENERGY AND DEVELOPMENT

Figure 5
a
DYNAMIC MULTIPLIERS FOR FRANCE

a
GDP = gross domestic product; REC = renewable energy consumption; OP = oil prices.
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 181

Figure 6
a
DYNAMIC MULTIPLIERS FOR GERMANY

a
GDP = gross domestic product; REC = renewable energy consumption; OP = oil prices.
182 THE JOURNAL OF ENERGY AND DEVELOPMENT

Figure 7
a
DYNAMIC MULTIPLIERS FOR JAPAN

a
GDP = gross domestic product; REC = renewable energy consumption; OP = oil prices.
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 183

Figure 8
a
DYNAMIC MULTIPLIERS FOR ITALY

a
GDP = gross domestic product; REC = renewable energy consumption; OP = oil prices.
184 THE JOURNAL OF ENERGY AND DEVELOPMENT

consumption in G7 countries is consistent with existing studies in energy


literature such as P. Sadorsky, S. Rafiq and K. Alam, and A. Omri and D.
Nguyen.85
Figures 2 through 8 depict the predicted asymmetric adjustment paths from
an initial long-run equilibrium to a new long-run equilibrium following
a unitary positive or negative shock hitting each variable of income, CO2
emissions, and oil prices. The asymmetry path is calculated as a linear com-
bination of the dynamic multipliers associated with positive and negative shocks.
We also report the predicted adjustment paths following a positive (positive
change curve) and negative (negative change curve) together with lower and
upper 95-percent confidence bands for asymmetry for a given prediction ho-
rizon. The asymmetry path for income shows a greater reaction of renewable
energy consumption to a decrease in income. When the symmetric effect of
income on renewable energy consumption is selected as the best specification,
it is clear from the graphs of the dynamic multipliers that the effect of income
on renewable energy consumption stabilizes after almost eight quarters. Re-
garding the adjustment paths of renewable energy consumption following
positive and negative shocks of CO2 emissions, the graphs show that for Canada
the impact of a positive unitary shock to CO2 emissions is greater than that
of a negative unitary shock. The cumulative renewable energy consumption
response to oil prices shocks is significantly negative in the United States. The
new equilibrium between the two variables is reached after 20 quarters, in-
dicating that the effects of oil price shocks are persistent. For the other coun-
tries, the new equilibrium after an oil price shock is reached faster after
10 quarters.

Conclusion and Policy Implications

Problems surrounding deteriorating environment quality have led policy


makers worldwide to push forward on a greener economic path by promoting the
use of renewable energy in lieu of fossil-fuel-based energy. Indeed, the rise in
greenhouse gas emissions is resulting in environmental pressures consisting of
increasing surface temperatures, melting of glaciers, rising sea levels, and
changing rainfall patterns. Promoting the use of renewable energy requires the
adoption of appropriate energy policies that encourage the development of in-
expensive alternative energy sources such that industry embraces the transition
process toward increasing the use of renewable energy and abating pollutant
emissions.
In the formulation of appropriate energy policies that will translate into the
successful adoption of more renewable energy, a parametric analysis of the sce-
nario was required. Keeping our focus on G7 countries, we employed the
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 185

nonlinear autoregressive distributed lags model to investigate the asymmetric


transmissions of economic growth, CO2 emissions, and oil prices to renewable
energy consumption in the long and short run using quarterly data over the
period from first quarter 1955 through fourth quarter 2014. This empirical
setting helps in assessing the effects of positive and negative changes of real
economic growth, CO2 emissions, and oil prices on the levels of renewable
energy consumption in G7 countries, while accounting for the potential long-
run cointegration relationship between renewable energy consumption and its
determinants. The results show that renewable energy consumption is sig-
nificantly sensitive to economic growth in all countries in the long run, with
the exception of the United Kingdom, where economic growth is found to
have no influence on renewable energy consumption. Moreover, an asymmetric
long-run relationship between renewable energy consumption and economic
growth is observed in Japan. Similarly, CO2 emissions asymmetrically influence
renewable energy consumption in Canada and linearly in the United States,
France, Germany, Japan, and Italy in the long run. In the United Kingdom, no
long-run relationship is found between renewable energy consumption and CO2
emissions. Oil prices demonstrate a nonlinear influence on renewable energy
consumption in the United States and this effect is linear in the United Kingdom
and France. However, no evidence of such an impact is found in Canada, Ger-
many, Japan, and Italy. In the short run, results reveal a nonlinear pass-through of
income to renewable energy consumption in Japan and an asymmetric trans-
mission of oil prices to renewable energy consumption in the United States and
France. Additionally, CO2 emissions have a symmetric impact on renewable en-
ergy consumption in the short run for all G7 countries.
The empirical findings are crucial for policy makers to draw up a new energy
policy roadmap that aims at promoting the use of renewable energy consumption.
Sustainable development is possible via variables like income, CO2 emissions, and
oil prices, and the results of this study demonstrate the nature of the long-run and
short-run influences of these variables on renewable energy consumption, par-
ticularly in the case of G7 countries. In addition, while making any energy policy
concerning the promotion of renewable energy, the decision makers in G7 nations
should consider oil prices and global CO2 emissions. However, policy makers
should recognize that there are differences in the effects of economic growth and
national CO2 emissions that are observable across countries. Consequently, com-
mon energy policies should be coupled with country-specific strategies via domestic
levels of plans.
186 THE JOURNAL OF ENERGY AND DEVELOPMENT

NOTES
1
The Intergovernmental Panel on Climate Change (IPCC), Climate Change 2014: Synthesis
Report (Geneva: IPCC, 2014).
2
M. Shahbaz, F. A. Jam, S. Bibi, and N. Loganathan, “Multivariate Granger Causality between
CO2 Emissions, Energy Intensity and Economic Growth in Portugal: Evidence from Cointegration
and Causality Analysis,” Technological and Economic Development of Economy, vol. 22, no. 1
(2016), pp. 47–74; M. Shahbaz, M. K. Mahalik, S. H. Shah, and J. R. Sato, “Time-Varying Analysis
of CO2 Emissions, Energy Consumption, and Economic Growth Nexus: Statistical Experience in
Next 11 Countries,” Energy Policy, vol. 98, issue C (2016), pp. 33–48; and A. Sinha, “Trilateral
Association between SO2/NO2 Emission, Inequality in Energy Intensity, and Economic Growth: A
Case of Indian Cities,” Atmospheric Pollution Research, vol. 7, no. 4 (2016), pp. 647–58.
3
T. M. L. Wigley, “Could Reducing Fossil-Fuel Emissions Cause Global Warming?,” Nature,
vol. 349 (February 7, 1991), pp. 503–06, and A. Sinha and J. Bhattacharya, “Is Energy-Led Eco-
nomic Growth Causing Climatic Shift? A Causal Analysis for India,” International Journal of
Green Economics, vol. 8, no. 2 (2014), pp. 120–33.
4
U.S. Energy Information Administration (EIA), International Energy Outlook 2016 (Wash-
ington, D.C.: EIA, 2016).
5
Ibid.
6
A. Nabavi-Pelesaraei, H. Hosseinzadeh-Bandbafha, P. Qasemi-Kordkheili, H. Kouchaki-Penchah,
and F. Riahi-Dorcheh, “Applying Optimization Techniques to Improve Energy Efficiency and GHG
(Greenhouse Gas) Emissions of Wheat Production,” Energy, vol. 103, issue C (2016), pp. 672–78, and
Renewable Energy Policy Network for the 21st Century (REN21), Renewables Global Status Report
(Paris: REN21, 2016).
7
U.S. Energy Information Administration, International Energy Outlook 2016.
8
International Institute for Sustainable Development (IISD), Climate Change Policy & Practice
(Winnipeg: IISD, 2015), available at http://climate-l.iisd.org/events/unfccc-cop-21.
9
U.S. Energy Information Administration, International Energy Outlook 2016.
10
Ibid.
11
Renewable Energy Policy Network for the 21st Century, Renewables Global Status Report.
12
Based on the state-level policies, Canada has introduced the “Greenest City 2020” initiative,
which has the objective of attaining a carbon-neutral and waste-free ecosystem by 2020.
13
United Nations Environmental Programme (UNEP) Center, UNEP Frontiers 2016 Report
(Nairobi: UNEP, 2016).
14
Renewable Energy Policy Network for the 21st Century, Renewables Global Status Report.
15
Y. Shin, B. Yu, and M. Greenwood-Nimmo, “Modelling Asymmetric Cointegration and Dy-
namic Multipliers in an ARDL Framework,” in Festschrift in Honor of Peter Schmidt–Econometric
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 187

Methods and Applications, eds. W. C. Horrace and R. C. Sickles (New York: Springer Science &
Business Media, 2014).
16
M. Shahbaz, T. H. V. Hoang, M. K. Mahalik, and D. Roubaud, “Energy Consumption, Fi-
nancial Development and Economic Growth in India: New Evidence from a Nonlinear and
Asymmetric Analysis,” Energy Economics, vol. 63, no. 3 (2017), pp. 199–212.
17
We applied the quadratic match-sum method to transform data from annual to quarter fre-
quency following R. Sbia, M. Shahbaz, and H. Hamdi, “A Contribution of Foreign Direct In-
vestment, Clean Energy, Trade Openness, Carbon Emissions and Economic Growth to Energy
Demand in UAE,” Economic Modelling, vol. 36, issue C (2014), pp. 191–97.
18
M. Shahbaz et al., “Energy Consumption, Financial Development and Economic Growth in
India: New Evidence from a Nonlinear and Asymmetric Analysis.”
19
E. Zivot and D. Andrews, “Further Evidence of the Great Crash, the Oil Price Shock and Unit
Root Hypothesis,” Journal of Business and Economic Statistics, vol. 10, no. 3 (1992), pp. 251–70.
20
N. Apergis, “Environmental Kuznets Curves: New Evidence on Both Panel and Country-Level
CO2 Emissions,” Energy Economics, vol. 54, issue C (2016), pp. 263–71; S. Asongu, G. El
Montasser, and H. Toumi, “Testing the Relationships between Energy Consumption, CO2 Emis-
sions, and Economic Growth in 24 African Countries: A Panel ARDL Approach,” Environmental
Science and Pollution Research, vol. 23, no. 7 (2016), pp. 6563–573; S. Bedir and V. M. Yilmaz,
“CO2 Emissions and Human Development in OECD Countries: Granger Causality Analysis with
a Panel Data Approach,” Eurasian Economic Review, vol. 6, no. 1 (2016), pp. 97–110; P. Y. Chen,
S. T. Chen, C. S. Hsu, and C. C. Chen, “Modeling the Global Relationships among Economic
Growth, Energy Consumption and CO2 Emissions,” Renewable and Sustainable Energy Reviews,
vol. 65 (2016), pp. 420–31; M. Robaina-Alves, V. Moutinho, and R. Costa, “Change in Energy-
Related CO2 (Carbon Dioxide) Emissions in Portuguese Tourism: A Decomposition Analysis from
2000 to 2008,” Journal of Cleaner Production, vol. 111, part B (2016), pp. 520–28; K. Saidi and S.
Hammami, “Economic Growth, Energy Consumption and Carbon Dioxide Emissions: Recent
Evidence from Panel Data Analysis for 58 Countries,” Quality & Quantity, vol. 50, no. 1 (2016),
pp. 361–83; M. Shahbaz et al., “Multivariate Granger Causality between CO2 Emissions, Energy
Intensity and Economic Growth in Portugal: Evidence from Cointegration and Causality Analysis;”
M. Shahbaz et al., “Time-Varying Analysis of CO2 Emissions, Energy Consumption, and Economic
Growth Nexus: Statistical Experience in Next 11 Countries;” and A. Sinha and S. Sen, “Atmo-
spheric Consequences of Trade and Human Development: A Case of BRIC Countries,” Atmo-
spheric Pollution Research, vol. 7, no. 6 (2016), pp. 980–89.
21
A. C. Marques, J. A. Fuinhas, and J. P. Manso, “Motivations Driving Renewable Energy in
European Countries: A Panel Data Approach,” Energy Policy, vol. 38, no. 11 (2010), pp. 6877–885.
22
S. Rafiq and K. Alam, “Identifying the Determinants of Renewable Energy Consumption in
Leading Renewable Energy Investor Emerging Countries,” presentation at the 39th Australian
Conference of Economists (ACE 2010), September, 27-29, 2010, Sydney, Australia.
23
P. Sadorsky, “Renewable Energy Consumption, CO2 Emissions and Oil Prices in the G7
Countries,” Energy Economics, vol. 31, no. 3 (2009), pp. 456–62.
24
A. C. Marques and J. A. Fuinhas, “Do Energy Efficiency Measures Promote the Use of Re-
newable Sources?,” Environmental Science and Policy, vol. 14, no. 4 (2011), pp. 471–81.
188 THE JOURNAL OF ENERGY AND DEVELOPMENT
25
A. C. Marques and J. A. Fuinhas, “Drivers Promoting Renewable Energy: A Dynamic Panel
Approach,” Renewable and Sustainable Energy Reviews, vol. 15, no. 3 (2011), pp. 1601–608.
26
A. K. Tiwari, “A Structural VAR Analysis of Renewable Energy Consumption, Real GDP and
CO2 Emissions: Evidence from India,” Economics Bulletin, vol. 31, no. 2 (2011), pp. 1793–806.
27
M. Aguirre and G. Ibikunle, “Determinants of Renewable Energy Growth: A Global Sample
Analysis,” Energy Policy, vol. 69, issue C (2014), pp. 374–84.
28
N. Apergis and J. E. Payne, “The Causal Dynamics between Renewable Energy, Real GDP,
Emissions and Oil Prices: Evidence from OECD Countries,” Applied Economics, vol. 46, no. 36
(2014), pp. 4519–525.
29
N. Apergis and J. E. Payne, “Renewable Energy, Output, CO2 Emissions, and Fossil Fuel
Prices in Central America: Evidence from a Nonlinear Panel Smooth Transition Vector Error
Correction Model,” Energy Economics, vol. 42, issue C (2014), pp. 226–32.
30
N. Apergis and J. E. Payne, “Renewable Energy, Output, Carbon Dioxide Emissions, and Oil
Prices: Evidence from South America,” Energy Sources, Part B: Economics, Planning, and Policy,
vol. 10, no. 3 (2015), pp. 281–87.
31
A. Omri and D. K. Nguyen, “On the Determinants of Renewable Energy Consumption: In-
ternational Evidence,” Energy, vol. 72, no. 2 (2014), pp. 554–60.
32
A. Omri, S. Daly, and D. K. Nguyen, “A Robust Analysis of the Relationship between Re-
newable Energy Consumption and its Main Drivers,” Applied Economics, vol. 47, no. 28 (2015),
pp. 2913–923.
33
A. Omri, N. B. Mabrouk, and A. Sassi-Tmar, “Modeling the Causal Linkages between Nuclear
Energy, Renewable Energy and Economic Growth in Developed and Developing Countries,” Re-
newable and Sustainable Energy Reviews, vol. 42 (2015), pp. 1012–022.
34
S. Rafiq, H. Bloch, and R. Salim, “Determinants of Renewable Energy Adoption in China and
India: A Comparative Analysis,” Applied Economics, vol. 46, no. 22 (2014), pp. 2700–710.
35
A. K. Tiwari, N. Apergis, and O. R. Olayeni, “Renewable and Nonrenewable Energy Pro-
duction and Economic Growth in Sub-Saharan Africa: A Hidden Cointegration Analysis,” Applied
Economics, vol. 47, no. 9 (2015), pp. 861–82.
36
A. Vaona, “The Effect of Renewable Energy Generation on Import Demand,” Renewable
Energy, vol. 86, issue C (2016), pp. 354–59.
37
B. Lin, O. E. Omoju, and J. U. Okonkwo, “Factors Influencing Renewable Electricity Con-
sumption in China,” Renewable and Sustainable Energy Reviews, vol. 55, no. 2 (2016), pp. 687–96.
38
E. Dogan and F. Seker, “Determinants of CO2 Emissions in the European Union: The Role of
Renewable and Non-Renewable Energy,” Renewable Energy, vol. 94, issue C (2016), pp. 429–39.
39
D. Ince, H. Vredenburg, and X. Liu, “Drivers and Inhibitors of Renewable Energy: A
Qualitative and Quantitative Study of the Caribbean,” Energy Policy, vol. 98, issue C (2016), pp.
700–12, reported that informal institutions play an important role in renewable energy development
for the Caribbean region.
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 189
40
S. Z. Chiou-Wei, C. F. Chen, and Z. Zhu, “Economic Growth and Energy Consumption
Revisited – Evidence from Linear and Nonlinear Granger Causality,” Energy Economics, vol. 30,
no. 2 (2008), pp. 3063–076.
41
World Bank, World Bank Indicators (Washington, D.C.: World Bank, 2015).
42
Ibid.
43
D. Yergin, “Ensuring Energy Security,” Foreign Affairs, vol. 85, no. 2 (2006), pp. 69–82.
44
Natural Resources Canada (NRC), “About Renewable Energy,” 2014, available at https://
www.nrcan.gc.ca/energy/renewable-electricity/7295.
45
Ibid.
46
International Energy Agency (IEA), Transitioning to a Low-Carbon Economy: New Bruns-
wick Climate Change Action Plan (2014–2020) (New Brunswick: IEA, 2014), available at https://
www.iea.org/policiesandmeasures/pams/canada/name-152556-en.php.
47
J. Stern, “International Gas Pricing in Europe and Asia: A Crisis of Fundamentals,” Energy
Policy, vol. 64, issue C (2014), pp. 43–8.
48
S. Ohnishi, M. Fujii, M. Ohata, I. Rokuta, and T. Fujita, “Efficient Energy Recovery through
a Combination of Waste-to-Energy Systems for a Low-Carbon City,” Resources, Conservation and
Recycling, vol. 128 (2016), pp. 394–405.
49
R. Boyd, J. Turner, and B. Ward, “Intended Nationally Determined Contributions: What Are
the Implications for Greenhouse Gas Emissions in 2030?,” Policy Paper, Centre for Climate
Change Economics and Policy and Grantham Research Institute on Climate Change and the En-
vironment, London, 2015.
50
N. Ayoub and N. Yuji, “Governmental Intervention Approaches to Promote Renewable
Energies—Special Emphasis on Japanese Feed-In Tariff,” Energy Policy, vol. 43, issue C (2012),
pp. 191–201.
51
T. Macalister, “Green Really is the New Black as Big Oil Gets a Taste for Renewables,” The
Guardian, May 21, 2016, available at https://www.theguardian.com/business/2016/may/21/oil-
majors-investments-renewable-energy-solar-wind.
52
T. Patel, “France to Adopt Law to Cut Nuclear Dependency, Carbon Emissions,” Bloomberg,
July 22, 2015.
53
Réseau de transport d´électricité (RTE), RET Electricity Report for 2014 (Paris: RTE,
2015).
54
B. Burger, Power Generation in Germany – Assessment of 2016 (Freiburg, Germany:
Fraunhofer Institute for Solar Energy Systems ISE, 2016).
55
L. Hong, H. Lund, B. V. Mathiesen, and B. Möller, “2050 Pathway to an Active Renewable
Energy Scenario for Jiangsu Province,” Energy Policy, vol. 53, issue C (2013), pp. 267–78.
190 THE JOURNAL OF ENERGY AND DEVELOPMENT
56
C. Chilcoat, “Germany about to Make Big Changes to Its Renewables Policy,” Oil Price,
2016, available at http://oilprice.com/Alternative-Energy/Solar-Energy/Germany-About-To-Make-
Big-Changes-To-Its-Renewables-Policy.html.
57
Renewable Energy Policy Network for the 21st Century (REN21), “Italy (2013) – Policy and
Regulatory Overviews,” available at http://www.reegle.info/policy-and-regulatory-overviews/IT.
58
Ibid.
59
P. Spugnoli, R. Dainelli, L. D’Avino, M. Mazzoncini, and L. Lazzeri, “Sustainability of
Sunflower Cultivation for Biodiesel Production in Tuscany within the EU Renewable Energy
Directive,” Biosystems Engineering, vol. 112, no. 1 (2012), pp. 49–55.
60
A. Jäger-Waldau, M. Szabó, N. Scarlat, and F. Monforti-Ferrario, “Renewable Electricity in
Europe,” Renewable and Sustainable Energy Reviews, vol. 15, no. 8 (2011), pp. 3703–716.
61
G. Barbose, N. R. Darghouth, S. Weaver, D. Feldman, R. Margolis, and R. Wiser, “Tracking
US Photovoltaic System Prices 1998–2012: A Rapidly Changing Market,” Progress in Photovol-
taics: Research and Applications, vol. 23, no. 6 (2015), pp. 692–704.
62
U.S. Energy Information Administration (EIA), US Imports of Crude Oil and Petroleum
Products (Washington, D.C.: EIA, 2011).
63
C. Baumeister and G. Peersman, “The Role of Time‐Varying Price Elasticities in Accounting
for Volatility Changes in the Crude Oil Market,” Journal of Applied Econometrics, vol. 28, no. 7
(2013), pp. 1087–109.
64
C. D. Strobel, “American Recovery and Reinvestment Act of 2009,” Journal of Corporate
Accounting & Finance, vol. 20, no. 5 (2009), pp. 83–5.
65
A. Schwartz, “DOE’s SunShot Program Aims to Reach Competitive Solar By 2020,” Fast
Company, February 7, 2011.
66
E. Cuce, “An Overview of Domestic Energy Consumption in the UK: Past, Present and Fu-
ture,” International Journal of Ambient Energy, vol. 37, no. 4 (2016), 428–35.
67
I. MacLeay, K. Harris, and A. Annut, Digest of United Kingdom Energy Statistics 2013
(London: Department of Energy and Climate Change, Government of the United Kingdom, 2013).
68
Ibid.
69
G. J. Offer, M. Contestabile, D. A. Howey, R. Clague, and N. P. Brandon, “Techno-economic
and Behavioural Analysis of Battery Electric, Hydrogen Fuel Cell and Hybrid Vehicles in a Future
Sustainable Road Transport System in the UK,” Energy Policy, vol. 39, no. 4 (2011), pp. 1939–950,
and S. Wang, C. Yan, and F. Xiao, “Quantitative Energy Performance Assessment Methods for
Existing Buildings,” Energy and Buildings, vol. 55 (December 2012), pp. 873–88.
70
D. Maddox, “Nuclear Disaster Casts Shadow over Future of UK’s Energy Plans,” The
Scotsman, March 30, 2012, available at http://www.scotsman.com/news/uk/nuclear-disaster-casts-
shadow-over-future-of-uk-s-energy-plans-1-2205614.
71
Y. Shin et al., op. cit.
G7 COUNTRIES: THE IMPORTANCE OF ASYMMETRIES 191
72
Ibid.
73
Ibid.
74
Ibid.
75
BP, Global Statistical Review (London: BP, 2015), available at https://www.bp.com/en/global/
corporate/energy-economics/statistical-review-of-world-energy/renewable-energy.html.
76
World Bank, op. cit.
77
Federal Reserve Bank of St. Louis, “Economic Research, Crude Oil Prices: West Texas
Intermediate (WTI)—Cushing, Oklahoma,” available at https://fred.stlouisfed.org/series/
DCOILWTICO/downloaddata.
78
Y. Shin et al., op. cit.
79
A. Advani, S. Bassi, A. Bowen, S. Fankhauser, P. Johnson, A. Leicester, and G. Stoye, Energy
Use Policies and Carbon Pricing in the UK (London: Institute for Fiscal Studies, 2013), available
at: https://www.ifs.org.uk/comms/r84.pdf.
80
P. Sadorsky, op. cit.; S. Rafiq and K. Alam, op. cit.; A. Omri et al., “Modeling the Causal
Linkages between Nuclear Energy, Renewable Energy and Economic Growth in Developed and
Developing Countries;” and B. Lin et al., op. cit.
81
The mechanism consists of a long-term contract between an electricity generator and a con-
tract counter party, which will enable the generator to have a constant level of revenues at a pre-
agreed price, called the strike price.
82
Z. Abdelmouleh, R. A. M. Alammari, and A. Gastli, “Review of Policies Encouraging Re-
newable Energy Integration & Best Practices,” Renewable and Sustainable Energy Reviews, vol.
45, issue C (2015), pp. 249–62.
83
U.S. Energy Information Administration, International Energy Outlook 2016.
84
P. Sadorsky, op. cit.; S. Rafiq and K. Alam, op. cit.; and A. Omri and D. K. Nguyen, op. cit.
85
Ibid.