Professional Documents
Culture Documents
Impact of fossil fuels energy consumption, energy policies, and urban sprawl T
on carbon emissions in East Asia and the Pacific: A panel investigation
Imran Hanif
Department of Economics, NUR International University, Model Town, Lahore, Pakistan
A R T I C LE I N FO A B S T R A C T
Keywords: This paper employs the Environmental Kuznets Curve (EKC) hypothesis in studying the impact of energy con-
Carbon emissions sumption, economic growth, and urbanization on carbon emissions in developing economies. Data from twelve
Energy consumption developing East Asian and Pacific countries for the period 1990 to 2014 has been used for analysis. The empirical
Energy management results, estimated through application of the System Generalized Method of Moment (GMM), show that energy
Economic growth
consumption, economic growth, and urbanization all significantly increase carbon emissions, thereby adding to
Urbanization
serious environmental challenges in the region. Moreover, public policies designed to shift economies from non-
renewable to renewable energy resources have been taken into account, and their impact was found to have been
insignificant in the sampled countries. However, the results depict an inverted U-shaped relationship between
per capita economic growth and carbon emissions, confirming the existence of an EKC in the region. The em-
pirical findings suggest that energy conservation and policies fostering use of renewables helped to improve
economic growth, control carbon emissions, and support the rising energy demand resulting from increased
urban sprawl. In order to improve the influence of public policies in controlling carbon emissions, our analysis
finds a need for stronger initiative to be taken in regional dialogues and policy frameworks, and for greater
collaboration among diverse stakeholders often characterized by differing values, interests, and objectives.
https://doi.org/10.1016/j.esr.2018.04.006
Received 7 November 2017; Received in revised form 7 April 2018; Accepted 15 April 2018
Available online 24 April 2018
2211-467X/ © 2018 Elsevier Ltd. All rights reserved.
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
Table 1
Averages of Key Indicators in East Asia ans Pacific (1990-2014).
Source: Based on World Bank Data (1990−2015)
Countries CO2 Emissions (Kilotons) Environment Rating (Rating points) Energy Use (Kg of oil) Fossil Fuels Use (% of total) GDP Per Capita (in US$) Urbanization
(in millions)
coal alone are causing about 50,000 premature deaths and 0.4 million estimation model may lead to biased results. Moreover, carbon emis-
new cases of bronchitis every year. In the past few decades, due to these sions are a product of energy consumption as well as other factors,
environmental changes, noxious hazes are also on the rise in the region, known and unknown; the omission of any important indicator in model
causing serious safety and health concerns – particularly in Indonesia, construction may lead to mis-specification bias. Quantification of the
Thailand, Malaysia, Brunei, and Singapore. These rapidly increasing impact of energy consumption, energy policies, urban sprawl, and
hazes are perilous for local inhabitants, adding further economic costs economic growth on carbon emissions is tricky and demands effective
to the list of negative consequences. Consequently, populations from control over simultaneity and mis-specification bias in the model. Thus,
developing East Asian and Pacific nations have grown more sensitive to a methodology based on a two-step System Generalized Method of
environmental damages and are demanding well-defined policies to Moment (GMM) has been adopted to tackle problems of simultaneity,
preserve the natural environment. Considering the consequences of this measurement error, and omitted variable bias, in order to report the
burning issue, the developing countries of East Asia and the Pacific are most concentrated results and to suggest effective and efficient policy
now addressing various environmental concerns as cross-border issues. frameworks for controlling carbon emissions. Our empirical findings
In the context of such efforts, there is a dire need at the regional level to confirm an inverted U-shape relationship between per capita real GDP
observe the Environmental Kuznets Curve (EKC) test results and to and per capita carbon emissions in the East Asia and Pacific region. The
inspect the impact of non-renewable energy consumption, energy results further depict an insignificant influence of environmental
management policies, and urban sprawl on carbon emissions – which management policies toward achieving sustainable development and
the present study undertakes to do. toward controlling carbon emissions in East Asia and the Pacific, and
A sizeable literature has already determined the impact of economic elsewhere.
growth and fossil fuels consumption on carbon emissions, including: The following section explains our data and methodology. Section 3
[1,3,6,18,23,31,43,46,59,60,67,38,13,34]. Additionally, individual estimates results, section 4 interprets those results, and section 5 pre-
studies have focused on different pollutants [26,27] used sulphur di- sents the conclusions.
oxide emissions [54]; examined sulphur dioxide and nitrogen oxide
[31,60,6] all examined CO2 emissions to determine the impact of eco-
nomic development on environmental degradation. Moreover, many 2. Data and methodology
studies have employed a Granger causality test and reported unidirec-
tional as well as bidirectional causalities between economic growth and Energy is essential for economic progress as well as for the attain-
carbon dioxide emissions for various regions ment of sustainable development. However, it is prerequisite of sus-
[40,14,41,49,25,50,51,55,4]. Empirical findings are inconsistent due to tainable development to foster economic growth without harming the
variations in time-span, indicators examined, regions or countries se- environment. Commonly, developing countries are facing multi-fold
lected, and models followed; thus the empirical evidence is still seen as challenges to improve economic growth. Owing to antiquated tech-
inadequate for recommending effective policy to improve energy effi- nologies, developing economies are bound to make excessive use of
ciency and environmental quality, particularly in the East Asian and fossil fuels in order to meet the growing demand for energy and to
Pacific economies. produce more goods. Such economies are therefore experiencing stag-
The present study is an attempt to draw together analyses of the nant economic growth while contributing ever more towards the gen-
impacts of fossil fuels consumption, per capita growth, and energy eration of toxic gases, excessively consuming fossil fuels and adding to
management policies on carbon emissions in these developing coun- environmental decline in the form of air and land degradation. As a
tries. In addition, we consider the rising pace of industrialization and general summary of statistics concerning the economies in East Asia
rural-to-urban migration, which exert increasing strains on natural re- and the Pacific, the average values of carbon emissions, fossil fuels
sources via increased consumption of energy, foods, and other goods, consumption, economic growth, and urbanization are reported in
and which are equally imperative in determining the impact of growing Table 1.
urbanization on carbon emissions in these developing countries. In addition, the East Asian and Pacific regions are highly vulnerable
Furthermore, very few studies have so far addressed energy manage- to climate change owing to their geography, climate, socioeconomic
ment, energy consumption, and carbon emission issues in terms of East conditions, and demographic factors, as well as to the great sensitivity
Asia and the Pacific; the present study hopes to fill that gap in the lit- of their natural assets (such as forests and biodiversity) to climate
erature by using the most recent panel data for the period 1990 to 2014. change. Rising emissions of carbon dioxide are endangering species,
Generally speaking, variables of energy consumption, growth, and increasing air pollution, causing damage to coral reefs, and increasing
urbanization are seen to ‘cause’ one another, but simultaneity becomes deforestation and water insecurity. Our summary of key environmental,
an unavoidable problem in the case of bidirectional causality, and an economic, and demographic factors in Table 1 allows assessment of the
heterogeneity of the countries included in the panel.
17
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
The panel consists of the twelve developing countries of East Asia the logarithms are used to examine the impact of independent variables
and the Pacific, namely: Cambodia, China, Indonesia, the Korean (which are measured in different units) on carbon emissions to examine
Republic, Lao PDR, Malaysia, Mongolia, Myanmar, Philippines, changes in percentage, rather than unit changes.
Thailand, Timor-Leste, and Vietnam. Other developing countries such
as American Samoa, Fiji, Kiribati, Marshall Islands, Micronesia, Palau, 3. Empirical testing
Papua New Guinea, Samoa, Solomon Islands, Tonga, Tuvalu, and
Vanuatu have not been considered due to data deficiency. Therefore, Summary statistics of dependent and independent variables without
this study has used unbalanced panel data from twelve developing logarithm are given in Table 2. The average value of carbon dioxide
countries of East Asia and the Pacific for the period 1990 to 2014. Data emissions was found to be 540,682.9 kilotons (kt), with the highest
were collected from the World Bank's World Development Indicators figure reported in China in 2013 (about 10,249,463 kt) and the lowest
(WDIs) and from the annual reports of Environmental Protection in Timor-Leste in 2002 (about 161.34 kt). In the developing economies
Agencies (EPAs). of East Asia and the Pacific, the average value of per capita economic
Considering the objectives of the study, we have used the log of growth for the period reached about US$ 2315.55. The highest per
yearly per capita carbon dioxide emissions (ED) in kilotons as a de- capita gross domestic product was around US$ 10,878.39, reported in
pendent variable. The log of real annual per capita economic growth Malaysia, while the lowest was in Myanmar, at about US$ 186.9 for the
(EG) in 2010 US$ and the log of the square of annual per capita eco- year 1991. The average value of per capita energy use reported about
nomic growth (EG2) were used to examine the impact of economic 911.17 kg of oil equivalent; the highest such value was reported in
growth on carbon emissions, and to test the Environmental Kuznets Malaysia, at 3019.81 kg of oil equivalent in 2013, while the lowest was
Curve hypothesis for these East Asia and Pacific economies. Moreover, in Timor-Leste, at 59.23 kg of oil equivalent in the year 2005. Similarly,
to examine the impact of non-renewable energy consumption, the log of the average value of urban agglomeration (of more than 1 million) as a
annual per capita fossil fuels energy consumption in kilograms of oil percentage of total population was approximately 14.86%; the highest
equivalent (EC) was used as an independent variable. Similarly, to ex- value was reported in Cambodia (around 51.04% in 1993) while the
amine the influence of urban expansion (UE) on carbon emissions, the lowest was reported in Mongolia (about 1.91% in 2015). As far as en-
log of population growth in urban agglomerations of more than 1 ergy management policies are concerned, the lowest effect on en-
million was used. Finally, to examine the impact of energy management vironmental degradation was reported in Timor-Leste during the period
policies (EP), the logs of yearly country ratings based on the quality of 2011-15, while the highest value was reported in Lao PDR for 2008-12,
energy management policies and quality of institutions for sustain- and also in Vietnam for 2014-15. For further details, the country-wise
ability were employed in the model. The effectiveness of energy policies summary statistics are reported in Appendix 1.
and the quality of energy management policies for environmental sus- The results of a pairwise correlation matrix are reported in Table 3,
tainability assess the extent to which energy policies have fostered the presenting the intensity of multicollinearity among the regressors. The
protection and sustainable management of pollution [65]. Assessments results show that carbon emission (ED) is weakly correlated with per
of the effectiveness and quality of energy management policies were capita economic growth (EG), the square of per capita economic growth
made on the basis of multidimensional criteria based on energy pro- (EG2), fossil fuel energy consumption (EC), urbanization (UR), and
duction by water, air, and the consumption of fuels, including bio fuels, energy management policies (EP).
and their influence on environment. The country-wise summary of To avoid a spurious regression result, it is necessary to determine
dependent and independent variables based on average data is given in the stationarity level of the selected variables. The study used a Fisher-
Fig. 1. type unit root test to determine the stationarity in the data series, and
For the purpose of empirical investigation, and to construct the the results are reported in Table 4.
model, the study used a transformed functional form of the EKC based The results of the Fisher-type unit root test reject the null hypothesis
on the work of [29–31,58]. The simplest function form of our estimated and show that all the data series are stationary at level. Therefore, we
model can be written as follows: can estimate the model using the appropriate regression technique,
though it is also necessary to deal with endogeneity in the model. To
ED = f (EG, EG 2, θ)
test the endogeneity, we used the Durbin-Wu-Hausman test, and results
The generalized form of econometric model based on the EKC can are reported in Table 5.
be described as: The lower p-values are an indication of acceptance of an alternative
hypothesis, thus confirming endogeneity in the variables. Because our
EDit = β0 + β1 EGit + β2 EGit2 + Ψnπit + μit (1) data is cross-country and data series may suffer from heteroscedasticity,
In equation (1), πit represents the additional set of independent the Breush-Pagan test for homoscedasticity was applied to control the
variables and Ψn is the slope of n number of independent variables, outliers, and its results are reported in Table 6. The lower p-values
while μit shows the idiosyncratic error term. Finally, our examination of reject the null hypothesis, which shows that the data series do suffer
the impact of energy management policies, non-renewable energy from heteroscedasticity.
consumption, economic growth, and urbanization on carbon emissions The study used a robust version of System GMM, generally re-
in the East Asia and Pacific region can be written as: commended to control heterogeneity in the presence of endogenous
explanatory variables. Because the result of the Durbin-Wu-Hausman
log EDit = β0 + β1 logEGit + β2 logEGit2 + β3 logECit + β4 logURit + β5 logEPit test shows that the explanatory variables in the estimated model are
+ μit endogenous, and thus correlated with the error term, the Fixed Effects
(2)
(FE) are therefore inconsistent [66]. Moreover, the variables used in the
In equation (2) the log transformation is used because it is expected model are stationary at level, but the use of OLS will not consistently
that per capita real GDP (EG) and carbon emissions (ED) do not have a estimate the coefficient unless there is no heterogeneity, and there may
linear relationship, and may have an exponential relationship. Here, to be an upward bias. FE is also biased, usually downward, and the bias
test the EKC, the impact of per capita real GDP on per capita carbon can be substantial unless the T is large. However, GMM using the
emissions is determined, and it is expected that a proportional increase Arellano-Bond conditions is considered the most robust: it employs only
in per capita GDP would increase the carbon emissions to a certain the moment conditions implied by the AR (1) model, and it properly
level, beyond which the increase in per capita GDP would either flatten removes heterogeneity. System GMM adds extra moment conditions
or decrease the level of carbon emissions. Moreover, the log transfor- that may be false, so it is important to mention that if the System GMM
mation helps to reduce the problem of multicollinearity [8]. Therefore, estimate seems reasonably precise, and there is no evidence of weak
18
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
Table 2 Table 3
Summary statistics. Correlation matrix.
Variables Obs. Mean Median Maximum Minimum Std. Dev. log ED log EG log EG2 log EC log UR log EP
19
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
Table 4
Results based on fisher-type unit root tests.
Ho: All panels contain unit roots
Table 5 Table 7
Tests of endogeneity. Results of two-step GMM (robust) estimation.
Dependent Variable: log ED Independent Variables: log EG, log EG2, log EC, log UR, log EP Dependent variable = log ED
20
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
21
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
programs to deal with environmental and energy conservational issues. and geographical contexts. Likewise, as many prevailing environmental
Moreover, there is an urgent need to switch from non-renewable to issues extend across national boundaries, it is imperative that regional
renewable energy sources, and to introduce unilateral moves in the dialogues be held to develop policy frameworks for collaboration
form of tax incentives and subsidies for the installation of equipment to among stakeholders (often with diverse values, interests, and objec-
generate renewable energy. Such moves in the region will prove helpful tives) in order to tackle environmental problems at regional level. In
in reducing the consumption of fossil fuels, diminishing energy short- short, swift economic development and effective control of trans-na-
falls, and fostering environmentally friendly economic growth at the tional environmental issues in East Asia and the Pacific require regional
regional level. So far, the East Asia and Pacific region has experienced as well as sub-regional collaboration, along with equally effective en-
great difficulty in this regard due to varying economic, political, social, ergy management strategies.
Appendix 1
Table 8
Country-wise Summary Statistics
22
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
Table 8 (continued)
References [18] X. Diao, S. Zeng, C. Tam, V. Tam, EKC analysis for studying economic growth and
environmental quality: a case study in China, J. Clean. Prod. 17 (2009) 541–548.
[19] J. dos Santos Gaspar, A.C. Marques, J.A. Fuinhas, The traditional energy-growth
[1] M.M. Alam, M.W. Murad, A.H.M. Noman, I. Ozturk, Relationships among carbon nexus: a comparison between sustainable development and economic growth ap-
emissions, economic growth, energy consumption and population growth: testing proaches, Ecol. Indic 75 (2017) 286–296.
Environmental Kuznets Curve hypothesis for Brazil, China, India and Indonesia, [20] Environmental Protection Agency, Causes of Climate Change, (2016) Retrieved
Ecol. Indic 70 (2016) 466–479. from http://www3.epa.gov/climatechange/science/causes.html#ref2.
[2] M. Allen, V. Barros, J. Broome, W. Cramer, R. Christ, IPCC Fifth Assessment [21] L. Esso, Y. Keho, Energy consumption, economic growth and carbon emissions:
Synthesis Report-climate Change 2014 Synthesis Report, (2014). cointegration and causality evidence from selected African countries, Energy 114
[3] U. Al-mulali, C. Sab, The impact of energy consumption and CO 2 emission on the (2016) 492–497.
economic on the economic growth and financial development in Sub-Saharan [22] B. Ewing, R. Sari, U. Soytas, Disaggregate energy consumption and industrial output
African countries, Energy 39 (2012) 180–186. in the United States, Energy Pol. 35 (2007) 1274–1281.
[4] U. Al-Mulali, C. Tang, I. Ozturk, Estimating the environment Kuznets curve hy- [23] M. Fodha, O. Zaghdoud, Economic growth and pollutant emissions in Tunisia: an
pothesis: evidence from Latin America and the Caribbean countries, Renew. empirical analysis of the environmental Kuznets curve, Energy Pol. 38 (2010)
Sustain. Energy 50 (2015) 918–924. 1150–1156.
[5] J. Ang, Economic development, pollutant emissions and energy consumption in [24] R. Franklin, M. Ruth, Growing up and cleaning up: the environmental Kuznets curve
Malaysia, J. Policy Model 30 (2008) 271–278. redux, Appl. Geogr. 32 (2012) 29–39.
[6] N. Apergis, I. Ozturk, Testing environmental Kuznets curve hypothesis in Asian [25] V. Govindaraju, C. Tang, The dynamic links between CO 2 emissions, economic
countries, Ecol. Indic 52 (2015) 16–22. growth and coal consumption in China and India, Appl. Energy 104 (2013)
[7] M. Arellano, S. Bond, Dynamic Panel Data Estimation Using PPD: a Guide for Users, 310–318.
(1988). [26] G.E. Halkos, Environmental Kuznets Curve for sulfur: evidence using GMM esti-
[8] D. Asteriou, S.G. Hall, Applied Econometrics, Palgrave Macmillan, New York, NY, mation and random coefficient panel data models, Environ. Deve. Eco. 8 (4) (2003)
2015. 581–601.
[9] V. Azevedo, S. Sartori, L. Campos, CO2 emissions: a quantitative analysis among the [27] G.E. Halkos, Exploring the economy–environment relationship in the case of sul-
BRICS nations, Renew. Sustain. Energy 81 (2018) 107–155. phur emissions, J. Environ. Plan. Mang. 56 (2) (2013) 159–177.
[10] H. Bekhet, N. Harun, Energy essential in the industrial manufacturing in Malaysia, [28] F. Halicioglu, A dynamic econometric study of income, energy and exports in
Int. J. Econ. 4 (2012) 129. Turkey, Energy 36 (2011) 3348–3354.
[11] R. Blundell, S. Bond, Initial conditions and moment restrictions in dynamic panel [29] I. Hanif, Impact of economic growth, nonrenewable and renewable energy con-
data models, J. Econom. 87 (1998) 115–143. sumption, and urbanization on carbon emissions in Sub-Saharan Africa, Environ.
[12] J. Breitung, The local power of some unit root tests for panel data, in: B. Baltagi Sci. Pollut. Res. (2018) 1–11 https://doi.org/10.1007/s11356-018-1753-4.
(Ed.), Nonstationary Panels, Panel Cointegration, and Dynamic Panels, Advances in [30] I. Hanif, Economics-energy-environment nexus in Latin America and the caribbean,
Econometrics, vol. 15, JAI, Amsterdam, 2000, pp. 161–178. Energy 141 (2017) 170–178.
[13] A. Canas, P. Ferrao, P. Conceicao, A new environmental Kuznets curve? [31] I. Hanif, P. Gago-de-Santos, The importance of population control and macro-
Relationship between direct material input and income per capita: evidence from economic stability to reducing environmental degradation: an empirical test of the
industrialised countries, Ecol. Econ. 46 (2003) 217–229. environmental Kuznets curve for developing countries, Environ. Dev. 23
[14] S. Chaabouni, N. Zghidi, M. Mbarek, On the causal dynamics between CO 2 emis- (2016) 1–9.
sions, health expenditures and economic growth, Sustain. Cities Soc. 22 (2016) [32] J. He, S. Wang, Y. Liu, H. Ma, Q. Liu, Examining the relationship between urba-
184–191. nization and the eco-environment using a coupling analysis: case study of Shanghai,
[15] V. Chandran, S. Sharma, K. Madhavan, Electricity consumption–growth nexus: the China, Ecol. Indic 77 (2017) 185–193.
case of Malaysia, Energy Pol. 38 (2010) 606–612. [33] H. Heidari, S. Katircioğlu, L. Saeidpour, Economic growth, CO 2 emissions, and
[16] C. Chang, C. Carballo, Energy conservation and sustainable economic growth: the energy consumption in the five ASEAN countries, Int. J. Electr 64 (2015) 785–791.
case of Latin America and the Caribbean, Energy Pol. 39 (2011) 4215–4221. [34] H. Hettige, M. Mani, D. Wheeler, Industrial pollution in economic development: the
[17] T.A. das Neves Almeida, L. Cruz, E. Barata, I.M. García-Sánchez, Economic growth environmental Kuznets curve revisited, J. Dev. Econ. 62 (2000) 445–476.
and environmental impacts: an analysis based on a composite index of environ- [35] C. Hsiao, M.H. Pesaran, A.K. Tahmiscioglu, Maximum likelihood estimation of fixed
mental damage, Ecol. Indic 76 (2017) 119–130. effects dynamic panel data models covering short time periods, J. Econm 109
23
I. Hanif Energy Strategy Reviews 21 (2018) 16–24
(2002) 107–150. [55] M. Shahbaz, H. Lean, M. Shabbir, Environmental Kuznets curve hypothesis in
[36] R. Huaman, T. Jun, Energy related CO 2 emissions and the progress on CCS projects: Pakistan: cointegration and Granger causality, Renew. Sustain. Energy 16 (2012)
a review, Renew. Sustain. Energy Rev. 31 (2014) 368–385. 2947–2953.
[37] K.S. Im, S.C. Ahn, P. Schmidt, J. Wooldridge, Efficient estimation of PanelData [56] R. Shrestha, G. Anandarajah, M. Liyanage, Factors affecting CO 2 emission from the
models with strictly exogenous explanatory variables—a Monte Carlo study, J. power sector of selected countries in Asia and the Pacific, Energy Pol. 37 (2009)
Econm. 93 (1) (1999) 177–201. 2375–2384.
[38] A. Kahuthu, Economic growth and environmental degradation in a global context, [57] U. Soytas, R. Sari, Energy consumption and GDP: causality relationship in G-7
Environ. Dev. Sustain. 8 (2006) 55–68. countries and emerging markets, Energy Econ. 25 (2003) 33–37.
[39] S. Kais, H. Sami, An econometric study of the impact of economic growth and en- [58] A. Tamazian, J. Chousa, K. Vadlamannati, Does higher economic and financial
ergy use on carbon emissions: panel data evidence from fifty eight countries, development lead to environmental degradation: evidence from BRIC countries,
Renew. Sustain. Energy Rev. 59 (2016) 1101–1110. Energy Pol. 37 (2009) 246–253.
[40] Y. Kang, T. Zhao, Y. Yang, Environmental Kuznets curve for CO 2 emissions in [59] S. Tao, T. Zheng, T. Lianjun, An empirical test of the environmental Kuznets curve
China: a spatial panel data approach, Ecol. Indic 63 (2016) 231–239. in China: a panel cointegration approach, China Econ. Rev. 19 (2008) 381–392.
[41] P. Kivyiro, H. Arminen, Carbon dioxide emissions, energy consumption, economic [60] S. Wang, Q. Li, C. Fang, C. Zhou, The relationship between economic growth, en-
growth, and foreign direct investment: causality analysis for Sub-Saharan Africa, ergy consumption, and CO 2 emissions: empirical evidence from China, Sci. Total
Energy 74 (2014) 595–606. Environ. 542 (2016) 360–371.
[42] Y. Li, Y. Li, Y. Zhou, Y. Shi, X. Zhu, Investigation of a coupling model of co- [61] Q. Wang, X. Yuan, J. Zhang, R. Mu, H. Yang, C. Ma, Key evaluation framework for
ordination between urbanization and the environment, J. Environ. Manage 98 the impacts of urbanization on air environment–A case study, Ecol. Indic 24 (2013)
(2012) 127–133. 266–272.
[43] J. Liu, T. Dietz, S. Carpenter, M. Alberti, C. Folke, Complexity of coupled human [62] F. Windmeijer, A finite sample correction for the variance of linear efficient two-
and natural systems, Science 317 (2007) 1513–1516. step GMM estimators, J. Econom. 126 (2005) 25–51.
[44] K. Menyah, Y. Wolde-Rufael, Energy consumption, pollutant emissions and eco- [63] Y. Wolde-Rufael, Energy demand and economic growth: the African experience, J.
nomic growth in South Africa, Energy Econ. 32 (2010) 1374–1382. Policy Model 27 (2005) 891–903.
[45] P. Narayan, R. Smyth, Electricity consumption, employment and real income in [64] World Bank, Environment Strategy for the World Bank in the East Asia and Pacific
Australia evidence from multivariate Granger causality tests, Energy Pol. 33 (2005) Region, World Bank, Washington, DC, 2005http://documents.worldbank.org/
1109–1116. curated/en/603841468751751164/Environment-strategy-for-the-World-Bank-in-
[46] S. Nasreen, S. Anwar, I. Ozturk, Financial stability, energy consumption and en- the-East-Asia-and-Pacific-Region.
vironmental quality: evidence from South Asian economies, Renew. Sustain. Energy [65] World Bank, World Development Indicators 2008, World Bank Publications, cop,
67 (2017) 1105–1122. Washington, DC, 2008 2008.
[47] H. Pao, C. Tsai, Modeling and forecasting the CO 2 emissions, energy consumption, [66] J.M. Wooldridge, Simple solutions to the initial conditions problem in dynamic,
and economic growth in Brazil, Energy 36 (2011) 2450–2458. nonlinear panel data models with unobserved heterogeneity, J. Appl. Econm 20 (1)
[48] S. Rafiq, R. Salim, I. Nielsen, Urbanization, openness, emissions, and energy in- (2005) 39–54.
tensity: a study of increasingly urbanized emerging economies, Energy Econ. 56 [67] J. Wu, Landscape sustainability science: ecosystem services and human well-being
(2016) 20–28. in changing landscapes, Landsc. Ecol. 28 (2013) 999–1023.
[49] A. Robalino-López, J. García-Ramos, A. Golpe, 2014. System dynamics modelling [68] G. Yang, H. Wang, J. Zhou, X. Liu, Analyzing and predicting the economic growth,
and the environmental Kuznets curve in Ecuador, Energy Pol. 67 (1980–2025) energy consumption and CO2 emissions in Shanghai, Energy Environ. Res. 2
923–931. (2012) 83.
[50] B. Saboori, J. Sulaiman, Environmental degradation, economic growth and energy [69] R. Zeb, L. Salar, U. Awan, K. Zaman, M. Shahbaz, Causal links between renewable
consumption: evidence of the environmental Kuznets curve in Malaysia, Energy Pol. energy, environmental degradation and economic growth in selected SAARC
60 (2013) 892–905. countries: progress towards green economy, Renew. Energy 71 (2014) 123–132.
[51] B. Saboori, J. Sulaiman, S. Mohd, Economic growth and CO 2 emissions in Malaysia: [70] Y. Zhao, S. Wang, Y. Ge, Q. Liu, X. Liu, The spatial differentiation of the coupling
a cointegration analysis of the environmental Kuznets curve, Energy Pol. 51 (2012) relationship between urbanization and the eco-environment in countries globally: a
184–191. comprehensive assessment, Ecol. Modell 360 (2017) 313–327.
[52] P. Sadorsky, Financial development and energy consumption in Central and Eastern [71] S. Ziaei, Effects of financial development indicators on energy consumption and CO
European frontier economies, Energy Pol. 39 (2011) 999–1006. 2 emission of European, East Asian and Oceania countries, Renew. Sustain. Energy
[53] K. Saidi, S. Hammami, The impact of CO 2 emissions and economic growth on Rev. 42 (2015) 752–759.
energy consumption in 58 countries, Energy Reports 1 (2015) 62–70. [72] M. Zilio, M. Recalde, GDP and environment pressure: the role of energy in Latin
[54] T. Selden, D. Song, Environmental quality and development: is there a Kuznets America and the Caribbean, Energy Pol. 39 (2011) 7941–7949.
curve for air pollution emissions? J. Environ. Econ 27 (1994) 147–162.
24