Factors influencing renewable electricity consumption in China

Renewable and Sustainable Energy Reviews 55 (2016) 687–696

Contents lists available at ScienceDirect

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

Factors influencing renewable electricity consumption in China Boqiang Lin a,n, Oluwasola E. Omoju b, Jennifer U. Okonkwo c a Collaborative Innovation Center for Energy Economics and Energy Policy, China Institute for Studies in Energy Policy, Xiamen University, Xiamen, Fujian, 361005, PR China b China Center for Energy Economics Research, School of Economics, Xiamen University, Xiamen, Fujian, 361005, PR China c Wang Yanan Institute for Studies in Economics, Xiamen University, Fujian, 361005, PR China

art ic l e i nf o

a b s t r a c t

Article history: Received 3 April 2015 Received in revised form 20 July 2015 Accepted 5 November 2015 Available online 5 December 2015

Renewable energy is an important factor in achieving a low-carbon economic development path in China. This paper investigates the factors influencing renewable electricity consumption in China. Specifically, the factors that influence the share of renewable electricity in total electricity consumption in China is investigated using data from 1980 to 2011 and employing the Johansen cointegration technique and vector error correction model. The result of the analysis shows that there is a long run relationship between renewable electricity consumption and GDP per capita, trade openness, foreign direct investment, financial development and share of fossil fuel in energy consumption. Economic development and financial development promotes renewable electricity consumption while foreign direct investment, trade openness and the lobby of conventional energy sources undermine the share of renewables in total electricity consumption in China. While the effects of shocks to the other variables appear to die out over time, the “lobby effect” is persistent and explosive. The results also show that there is a uni-directional short run causality from financial development to renewable electricity consumption and from renewable electricity consumption to trade openness. The Chinese government should pursue policies that not only increase the amount of renewable electricity, but also increase the share of renewables in total electricity consumption. & 2015 Elsevier Ltd. All rights reserved.

Keywords: Electricity consumption Renewable electricity Environmental protection China

Contents 1. 2. 3. 4.

5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 688 Literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 690 3.1. Variables and model specification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 4.1. Unit root test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 4.2. Optimal lag selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 4.3. Cointegration rank test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 4.4. Normalised cointegration coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 692 4.5. Vector error correction model: short run dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 4.6. Diagnostic tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 4.6.1. Goodness of fit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 4.6.2. Eigenvalue stability condition for stability test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 4.6.3. Test for serial correlation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 4.6.4. Test for normality of residuals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 4.7. Impulse response and variance decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 694 Conclusions and policy implications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695

n Corresponding author at: Collaborative Innovation Center for Energy Economics and Energy Policy, China Institute for Studies in Energy Policy, Xiamen University, Xiamen, Fujian, 361005, PR China. Tel.: þ86 5922186076; fax: þ865922186075. E-mail addresses: [email protected], [email protected] (B. Lin).

http://dx.doi.org/10.1016/j.rser.2015.11.003 1364-0321/& 2015 Elsevier Ltd. All rights reserved.

688

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695

1. Introduction China’s impressive economic performance in the past decades has resulted in increased energy consumption and carbon intensity. Over the past few years, China’s economy has grown at an average of over 7%, exceeding that of the United States and the European Union combined. However, this impressive economic performance has led to increase in energy consumption and carbon (CO2) emission. According to the United States Energy Information Administration (EIA), China’s primary energy consumption increased from 17.29 Quad BTU in 1980 to 103.72 Quad BTU in 2011. Similarly, her electricity net consumption increased from 261.49 billion kilowatthours in 1980 to 4207.70 billion kilowatthours in 2011 (Fig. 1), an increase of over 1500%. With the increase in total primary energy and coal-dominated electricity consumption, carbon emission also increased significantly as shown in Fig. 1. Carbon emission associated with electricity production and consumption in China is high because coal is the dominant fuel for electricity production in the country. As at 2012, China’s energy consumption-related CO2 emission stands at 8547.74 million metric tons compared to 1448.46 million metric tons in 1980, which makes it the largest CO2 emitter in the world. The current emission in China is almost that of Africa (1152.22), Europe (4305.17), Middle East (1951.80) and Central and South

Fig. 1. Electricity consumption and CO2 emission in China. Source: Data from EIA Database.

Fig. 2. Share of renewables in total electricity consumption in China, 1980–2011. Source: Data from US Energy Information Administration.

America (1339.47) combined. The current trend of high resource intensity, energy consumption and emission is not sustainable. If current trend continues, China’s emission level will undermine global effort to stem climate change and global warming. The high level of energy intensity and CO2 emission in China has attracted local and international attention, and has called for significant changes to the country’s energy strategy and policy. In response to this, measures are being put in place to address the situation. One of the key measures aimed at addresing energyrelated CO2 emission in China is increasing the share of renewable energy in the total energy mix. The share of renewable-generated electricity in total electricity consumption in China is small and has decreased over time, as shown in Fig. 2. China has the highest amount of renewable electricity net consumption in 2011 (800.96 billion Kwh) compared to the United States (527.48 Kw h), Germany (126.18 Kwh), India (160.36 Kw h) and Finland (23.39 Kw h). Yet, the share of renewable-generated electricity in total electricity net consumption in China (19.03%) is low compared to Germany (23.46%), India (21.15%) and Finland (28.86%). However, the enactment of the Renewable Energy Law of 2005 was aimed at reversing the trend and promoting renewable energy in the country. The 11th Five-Year Plan (2006–2010) targets 20% reduction in per capita GDP energy consumption and 10% reduction in two major air pollutants, while the renewable energy

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

plan of the National Development and Reform Committee (NDRC) for the same period sets a 10% of renewable in total enrgy consumption by 2010 [48]. The 12th Five-Year Plan (2011–2015) sets more ambitious goals of achieveing a share of 11.4% of non-fossil fuel (renewables and nuclear) in total energy consumption. Also, during the National People’s Congress (NPC) and Chinese People's Political Consultative Conference (CPPPC) 2014 Annual Session, the government states its commitment towards reducing environmental pollution and energy-related CO2 emissions by promoting the development, deployment and use of renewable energy. This is a significant step towards reducing energy-related CO2 emission in China and globally, given the position of China as the largest carbon emitter in the world. Reducing energy consumption and CO2 emission in China will, to a large extent, facilitate the achievement of global target for CO2 emissions and climate change mitigation. Against this background, this paper examines the influencing factors of renewable energy consumption in China. A number of studies have been conducted on renewable energy development, with substantial studies focusing on China. However, this present study differs by examining the share of renewables in total energy consumption instead of the amount of renewable energy. The main contribution of this paper to the literature is threefold. First, although there have been enormous studies on renewable energy in the field of energy and environmental economics, majority of these studies focus on developed and industrialised countries such as United States, EU and generally, OECD countries [18,31]. In contrast, this paper models and analyses the determinants of renewable electricity consumption in an emerging country. Second, most of the empirical studies analysing the drivers of and barriers to renewable energy employ panel data techniques, and do not adequately investigate country specific factors [1,27,30]. Following SSDN and IDDRI [39], deep decarbonisation of energy system requires both globally coordinated decarbonisation strategy and individual country-level decarbonisation pathways. Also, according to Vachon and Menz [42], individual country characteristics such as culture, wealth and renewable energy endowment are important drivers of renewable energy. The individual country characteristics and pathways are necessary given the significant differences in income level, resource endowment, energy consumption level and structure, technology advancement, amount of CO2 emission, energy market structure, mitigation and adaptation capabilities, and development policy goals across countries. Thus, this study takes these factors into consideration, and focuses the analysis on an individual country – China – whose decarbonisation pathway is crucial for meeting global climate change goals. Third, most of the previous studies on renewable energy use the amount of renewable energy produced or consumed as dependent variables. However, based on Aguirre and Ibikunle [1] and SSDN and IDDRI [39], it is the share of renewable energy in total energy consumption and not the amount of renewable energy consumed that is important for climate change mitigation. Therefore, this study deviates from previous studies and uses the share of renewable energy-generated electricity in total electricity consumption as the dependent variable. Therefore, the objective of this paper is to empirically investigate the drivers of and barriers to renewable electricity consumption in China.

2. Literature review There has been substantial research attention on renewable energy in recent years. Renewable energy is recognised as a viable option to enhance energy access and at the same time mitigate climate change [22]. Research on the determinants of renewable energy can be classified into panel and time series analysis,

689

developed and developing countries, investigation of individual variables, and various types of renewable energy. Marques et al. [18] analyse the drivers of renewable energy in the European Union (EU) using fixed effect vector decomposition (FEVD) technique on data spanning 1990–2006. The study focuses on political, socioeconomic and country-specific factors affecting renewable energy. The result shows that the influence of traditional energy sources and CO2 emission undermine renewables commitment while the goal of reducing energy dependency stimulates renewable energy consumption. Rafiq and Alam [32] study the determinants of renewable energy consumption in leading renewable investor emerging countries. The study uses data from six emerging economies (Brazil, China, India, Indonesia, Philippines and Turkey) and employ panel methods (FMOLS and DOLS) and autoregressive distributed lag (ARDL). The result of the study shows that income and pollutant emission are the major driver of renewables in Brazil, China, India and Indonesia while income seems to be the only driver of renewable energy in Turkey and Philippines. Omri and Nguyen [27] determine the influencing factors of renewable energy consumption in a panel of 64 countries over the period 1990–2011 using dynamic GMM panel model. They also developed subpanels of high, middle and low-income countries. They find that trade openness and increase in carbon emissions are the major influencers of renewable energy. Oil price has a negative but small impact of renewable energy development in the middle-income and global panels. According to Marques et al. [18], some studies have investigated the role of individual factors, policies and variables in promoting renewable energy adoption in different countries [42,43,46,47]. Johnstone et al. [17] provides the prospects and challenges of public policies in promoting renewable energy. Carley [5] and Menz and Vachon [20] point out the importance of state policies and financial incentives in promoting renewable energy use. Empirical evidences from Gan et al. [13] and Chien and Hu [7] show that energy security is a major promoter of renewable energy development. Chang et al. [6] investigates the link between renewable energy, GDP and energy prices, and find that countries with higher GDP have the capacity to adopt renewables regardless of their high prices. Sadorsky [34] hypothesised that high environmental concerns are significant incentives for renewable energy development and deployment. Sovacool [36] argue that the share of conventional energy sources (fossil fuels) in the total energy consumption has potential influences on the deployment of renewable energy. The impact of income, measured by the level of GDP, on renewable energy adoption has been comprehensively discussed in the literature, with most of the studies finding a strong positive impact of income on renewables [14,25,35]. From an empirical survey conducted by Peterson [30], there is little evidence that factors and financial mechanism like trade, FDI, ODA, GEF and CDM significantly enhance greenhouse gas mitigation-related technology. Popp et al. [31] investigates the impact of patenting activity on renewable energy technology in 26 OECD countries from 1991 to 2004. They find that knowledge has a small but robust effect on renewables. Similarly, Brunnschweiler [4] analyses the impact of financial sector development on renewables in non-OECD countries. Studies explaining the deployment of specific types of renewable energy are also copious in the literature. Bird et al. [3] and Menz and Vachon [20] investigate the factors promoting wind renewable energy in US states. Beckman et al. [2] investigates the determinants of on-farm (wind and solar) renewable energy adoption in the US using data from the 2009 on-farm renewable energy survey and adopting a binary-choice model. The result shows that farmers with large farm size, on-farm residence, and those adopting conservation practices are more likely to report renewable energy production while those that specialise in row crop production and use expensive machinery are likely to report

690

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

less. Adelaja and Hailu (2007) examine the projected impacts of renewable portfolio standards on wind industry development in Michigan, and find that the policy enhances wind energy development in the state. Pfeiffer and Mulder (2013) analyse the drivers of non-hydro renewable energy in 108 developing countries using two-stage estimation methods. The result of the study shows that economic and regulatory instruments, higher per capita income, stable and democratic regimes, higher schooling level improve the possibility of renewable energy adoption. On the other hand, openness, aid, increase electricity consumption, high fossil fuel production and institutional policy support programs undermine adoption of renewables. The choice of renewable energy policies has also attracted attention in the literature. Stadelmann and Castro [40] examine the domestic and international determinants of renewable energy policies in 112 developing and emerging countries using data from 1998 to 2009. The study focuses on four types of policies – renewable energy targets, feed-in-tariffs, framework policies, and other financial incentives – and employs logit-linked discrete-time events history model. The result of the study shows that domestic factors such as population and wealth are positively associated with the adoption of renewable energy policies, with endowment only driving renewable policies in some specific cases while hydro power resources undermine the adoption of targets. With respect to international factors, colonial influence and EU membership foster renewable policy adoption while climate finance mechanisms such as Global Environmental Facility (GEF) and Clean Development Mechanism (CDM) only facilitate the adoption of targets and frameworks and are ineffective on tariffs and incentives. According to Martinot [19], the design of domestic policies such as electricity sector liberalisation could affect renewable energy deployment. Mitchell et al. [21] shows that domestic factors such as employment generation, pursuit of affordable energy and possibility of developing new industries are very important drivers of renewable energy policies in developing countries. Carley [5] evaluate the effectiveness of renewable energy electricity policies in US states. There are also studies on the determinants of renewable energy development in China given the potential importance of renewable energy in reducing her carbon emission. Fengqi [12] made it a subject of his work. Other researchers have also contributed considerably to the subject [11,43,49]. Rafiq et al. [33] examine the relationship between income, CO2 emission and renewable energy generation in China and India using multivariate vector error correction model and data from 1972 to 2011. The result of the causality test shows a unidirectional relationship from output and CO2 emission to renewable energy in the short run in China. In the long run, there is a unidirectional relationship from output to renewable energy and bidirectional causality between CO2 emission and renewables. However, the previous studies on renewable energy use the amount of renewable energy produced or consumed as dependent variables. But following SSDN and IDDRI [39], increasing the share of renewable energy in total energy consumption is the key point for reducing greenhouse gas emissions and global warming. Therefore, this study differs from previous ones in the literature by using the share of renewable energy generated-electricity in total electricity consumption as the dependent variable. The results of this study is important for policies to improve the share of renewables in total electricity consumption.

3. Methods To capture the dynamic relationship between the renewable electricity and its influcencing factors in China, the Vector Error

Correction Method (VECM) and cointegration technique are employed. The VECM framework determines the direction of causality between the variables while providing estimates on both the long run and the short run. The co-integration analysis which is a property of long run equilibrium provides information about the long run relationship among the variables while the granger causality test which is a short run phenomenon provides information on the short run dynamics among the variables [36]. If renewable electricity consumption and its influencing factors are cointegrated, a VECM representation could have the following form: ΔX t ¼ A0 þ ПXt  1 þ

k X

Гj ΔXt  j þ εt

j¼1

where Δ is the difference operator, Xt is a 6x1 – dimensional vector of non stationary I(1) endogenous variables of the model, A0 is a 6x1 - dimensional vector of constant and εt is k-dimensional vector of the stochastic error term normally distributed with white noise properties N(0,σ2). П is the long run matrix that determines the number of co-integrating vectors that consist of α and β’ representing speed of adjustment towards long run equilibrium and long run parameter respectively. Г is the vector of parameters that represents the short term relationship. If the variables are integrated of the same order, then we can test for the existence of a long run cointegration relationship among the variables. There are two major methods used for cointegration analysis – Engle–Granger method [10] and Johansen and Juselius [16]. The Engle–Granger method can only be applied to a single equation model while the Johansen-Juselius can be used to test for the existence of cointegration among variables and also accurately determine the number of cointegrating vectors [24]. Given that this study deals with multiple variables, we employ the Johansen–Juselius cointegration method. The Johansen cointegration test is expressed in the following equation: ΔX t ¼

γX 1

Γ i ΔX t  1 þ ΠX t  1 þ εt

I¼1

where X t is a 6x1 vector (RE, GDPGR, OPEN, FDIG, FIN, FUEL), Δ is a symbol of difference operator, εt is a 6x1 vector of residuals. The VECM model has information about the short and long run adjustment to changes in X t via the estimated parameters Γ i and Π respectively. Here, the expression ΠX t  1 is the error correction term and Π can be factored into two separate matrices α and β, such that Π ¼ αβ0 , where β0 denotes the vector of cointegrating parameters while α is the vector of error correction coefficients measuring the speed of convergence to the long-run steady state. However, before conducting cointegration test, there is need to check the stationarity of the time series data. A stationary linear combination of economic variables implies the existence of a long run equilibrium relationship. In this paper, we use the Augmented Dickey Fuller (ADF) unit root test [9] to test for the presence of unit root in the series. The ADF test is based on the following regression: Δyt ¼ αyt  1 þ

k P

i¼1

βi Δyt  1 þ X it γ þ ut

where y is the time series being tested for a unit root at time t and T is the number of observations. Xt is the exogenous variable (a constant or a constant trend). Δ denotes the first difference operator and ut is an i.i.d. error term that is distributed independently and identically. In order to test the null hypothesis of the presence of a unit root in yt, we conduct the hypothesis testing that α1 ¼0 in the equation. If α1 is significantly less than zero, the null hypothesis of a unit root is rejected.

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

691

(In) of the variables are taken, and the equation follows the form:

3.1. Variables and model specification The dependent variable in this study is the share of renewable energy-generated electricity in total electricity consumption in China. The data is obtained by calculating the ratio of renewablegenerated electricity to total electricity consumption in China. The independent variables include real GDP per capita, trade openness, foreign direct investment, financial development and lobby of fossil fuel. GDP per capita reflects economic performance. Over the past few decades, the Chinese economy has grown at an impressive pace to emerge as the second largest economy in the world, behind the United States. This phenomenal economic performance has however been accompanied by huge energy consumption and environmental issues. A number of studies have found economic performance (represented by GDP) to have significant influence on renewable energy development [32,6]. Hence, the need to include GDP in the model. Trade openness is indicated by the share of import and export in the GDP. It is well recognised in the literature that the development of the Chinese economy has been largely enhanced by the opening up policy of 1978. Over the period from 1978 to 2014, China became the largest trading country in the world. Thus, due to the very important contribution of trade to the economy and following previous literature [27], trade openness is included in the model. Similarly, the impacts of foreign direct investment on technology transfer in host countries have been explored [38,8]. Moreover, China is one of the largest recipients of foreign direct investment in the world. Given that trade openness and FDI are important factors in the Chinese economy, they are included in the model to examine their influence on renewable energy development and test the “technology transfer” hypothesis. Some literature have examined the impact of financial development on renewable energy development, with mixed results [15,26,29,4]. Theoretically, it is believed that a developed financial sector will positively contribute to clean energy technology projects. Thus, we include financial development in the model to empirically examine whether financial development has significant impact on renewable energy technology in China. The “lobby effect” in renewable energy adoption is documented in Aguirre and Ibikunle [1]. It implies the influence of traditional energy sources in undermining the adoption of renewable energy. It is assumed that the higher the consumption of fossil fuel, the more difficult it is to adopt renewable energy. Other studies also found significant impact of the “lobby effect” on renewable energy adoption [18,30,37]. Given that China is the largest primary energy consumer and one of the top energy producers in the world, coupled with the fact that over 50% of her energy consumption is from fossil fuel, the “lobby effect” is an important factor. Thus, it is included in the model. To investigate the dynamic relationship between renewable electricity consumption and its influencing factors in China, this study specifies the following model:

InREt ¼ β0 þβ1 InRGDPPCt þ β2 InOPENt þ β3 InFDIGt þβ4 InFINt þ β5 InFUELt þ εt …………

ð2Þ

The data used in the paper are obtained from various sources. REt is obtained from the database of the United States Energy Information Administration (EIA), RGDPPCt, OPENt, FDIGt, FINt and FUELt are obtained from the World Development Indicator of the World Bank. STATA software is used to carry out the analysis.

4. Results 4.1. Unit root test Before analysing the impact of the determinants of renewable energy in China using Eq. (2) above, the properties of the time series data used in this study is tested for unit root. The Augmented Dickey Fuller (ADF) unit root test is used to examine the stationarity of the series and the result is presented in Table 1. From the table, it is observed that the variables are non-stationary at levels, but their first difference forms are stationary. Thus, the variables fulfils the condition for cointegration. 4.2. Optimal lag selection Constructing the model requires an optimal lag order. The model is used to analyse the relationship between renewable electricity consumption (InRE) and real GDP per capita (InRGDPPC), trade openness (InOPEN), foreign direct investment (InFDIG), financial development (InFIN) and conventional fuel sources (InFUEL). The FPE, HQIC, Akaike Information Criterion (AIC), Likelihood Ratio (LR) and Schwarz Bayesian Information Criterion (SBIC) are mostly used in lag selection. From Table 2, all the lag selection criteria except the FPE criterion selects lag 3. Thus, we follow the four criteria (LR, AIC, HQIC, SBIC) and select lag 3 as the optimal lag order. 4.3. Cointegration rank test The result of the Johansen test for cointegration is shown in the Table 3. Given that the trace statistics (101.0967)4the 5% critical value (94.15), we reject the null hypothesis that the cointegration rank is zero (0). However, the null hypothesis that the cointegration rank is 1 cannot be rejected because the trace statistics (61.0570) othe 5% critical value (68.52). So based on the trace statistics, it is observed that there is one cointegrating equation between renewable electricity consumption and real GDP per capita, trade openness, financial development, foreign direct investment and share of fossil fuel in total energy consumption in China. In other words, there is a long run relationship between renewable electricity consumption and these factors in China.

REt ¼ β0 þ β1 RGDPPCt þβ2 OPENt þβ3 FDIGt þ β4 FINt þ β5 FUELt þ εt ………………:…

ð1Þ

where REt is the share of renewable electricity in total electricity consumption; RGDPPCt is real GDP per capita; OPENt is trade openness and is depicted by trade as a % of GDP; FDIGt is foreign direct investment and proxied by the ratio of foreign direct investment to GDP; FINt is financial development and is indicated by domestic credit to the private sector as a % of GDP; FUELt is the share of fossil fuel in energy consumption; β0 is the constant and β1, β2,…., β5 are the coefficient of the corresponding variables; and εt is the error terms. The choice of the variables are based on theory and previous literature. To avoid heteroscedasticity, the natural log

Table 1 Summary result of ADF unit root test. Variables

DF statistics @ levels

DF statistics @ 1 st diff.

Order of integration

InRE InRGDPPC InOPEN InFIN InFUEL InFDIG

 1.357 1.274  1.628  1.689  0.231  2.010

 5.677nn  2.983n  5.310nn  5.089nn  5.595nn  3.485nn

I(1) I(1) I(1) I(1) I(1) I(1)

nn n

¼1% significance level. ¼5% significance level.

692

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

Table 2 Lag selection order criteria. Lag

LL

LR

df

P

FPE

AIC

HQIC

SBIC

0 1 2 3 4

147.449 306.46 342.212 427.005 –

318.02 71.503 169.59* –

36 36 36 36

0.000 0.000 0.000 –

1.6e  12 2.7e  16 4.1e  16 4.6e  17 -4.6e  65*

 10.1035  18.89  18.8723  22.3575* –

 10.0162  18.2791  17.7377  20.6993* –

 9.81802  16.8917  15.1611  16.9335* –

Table 3 Johansen cointegration rank test. Maximum rank Parms LL

0 1 2 3 4 5 6

42 53 62 69 74 77 78

313.59686 333.61674 346.42476 356.32437 360.65067 364.13924 364.14522

Table 4 Summary result of the vector error correction model. Eigenvalue Trace statistic 5% Critical value 101.0967 61.0570* 35.4409 15.6417 6.9891 0.0120

0.73675 0.57423 0.48314 0.25055 0.20751 0.00040

94.15 68.52 47.21 29.68 15.41 3.76

Trace test indicates 1 cointegrating eq. at 5% significance level.

4.4. Normalised cointegration coefficient The coefficients of the long run relationship between renewable electricity, real GDP growth, trade openness, foreign direct investment, financial development and conventional energy sources are presented in the estimated cointegration equation below: REt ¼ 7:27 þ 0:26RGDPPCt –0:58OPENt –0:02FDIGt þ 0:19FINt –1:16FUELt ð0:0636Þ** ð0:0547Þ**

ð0:0553Þ**

ð0:0067Þ**

ð0:5526Þ*

Standard errors in parenthesis (** ¼1% and * ¼ 5% significance level). The result of the cointegration model shows that all the variable except FUEL are significant at 1% significance level. Based on the result, a 1% increase in real GDP per capita leads to a 0.26% increase in the share of renewables in total electricity consumption in China. This is in line with previous studies such as Rafiq et al. [33] for China, Marques et al. [18] for EU member countries and Pfeiffer and Mulder [30]. Economic development enhances the share of renewables in electricity consumption in a number of ways. First, the government would have sufficient resources to invest in environmental protection. Given that the basic development needs have been met to a large extent, the government would be willing to make sacrifice to promote renewable energy while also encouraging energy efficiency. This is currently the situation in China. After over two decades of impressive economic growth growth and development, the Chinese government is taking initiatives and making investments in renewable energy and environmental protection, even at the expense of the macroeconomy. Second, as a result of increase in income level and improvement in living standard, people will demand for environmental protection and will be willing and capable of paying for renewable energy. The current environmental situation in China has attracted public incitement and the public is willing to embrace renewable energy to reduce the severity of air pollution. Trade openness and foreign direct investment have significant negative impact on the share of renewables in total electricity consumption in China. This result confirms earlier study by Pfeiffer and Mulder [31], which find in a study of 108 developing countries that increasing openness delay renewable energy consumption.

D_LNre _ce1 L1 Lnre LD L2D LNrgdppc LD L2D Lnopen LD L2D Lnfin LD L2D Lnfuel LD L2D Lnfdig LD L2D

Coefficient

Std. error

Z

P 4|z|

 0.658170

0.3005342

 2.19

0.029

 0.0350144  0.0457279

0.2833982 0.2607668

 0.12  0.18

0.902 0.861

0.5900659 1.285845

0.7591601 1.00554

0.78 1.28

0.437 0.201

0.1375637 0.0783509

0.2640119 0.1546212

0.52 0.51

0.602 0.612

 0.5804305  0.1896088

0.2523678 0.312194

 2.30  0.61

0.021 0.544

0.2147138 0.006606

2.085874 2.029159

0.10 0.00

0.918 0.997

 0.0021098  0.0046407

0.0729423 0.0707203

 0.03  0.07

0.977 0.948

D_LNre is the dependent variable.

Also Peterson [29], find little evidence on the positive impact of trade and FDI on clean technology. The result also contradicts earlier studies that find positive impact of trade and foreign direct investment on renewable energy [27]. According to the result, a 1% increase in trade openness and foreign direct investment leads to 0.58% and 0.02% reduction respectively in the share of renewables in electricity consumption. China has attracted significant amount of foreign direct investment in the past decades, primarily due to the opening-up reforms and low cost of production. Similarly, the opening-up policy has enabled China to become the largest exporter and one of the top importing countries in the world. However, the influx of foreign direct investment and the surge in trade has contributed significantly to total electricity consumption relative to renewable electricity consumption. This study does not agree with the current opinion that FDI and trade openness are essential for improving the share of renewable-generated electricity in total electricity consumptiom. FDI and trade openness may contribute to renewable energy development through agglomeration of talents and technology transfer, but they may not increase the share of renewables in total electricity consumption. Another point that could explain this is that lack of concern for environmental issues in China in the past decades could have also contributed to FDI and trade as companies rely and thrive heaviliy on cheap and subsidised fossil fuel. This argument is based on Unruh [41] and Perc and Szolnoki [29]. Unruh [41] argue that certain systemic processes could interact to undermine the dissemination of clean energy technologies. Perc and Szolnoki [28]’s evolutionary game theory asserts that economic growth could produce some benefits to investors and the economy, but at the expense of the environment.

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

693

Fig. 3. Plot of actual and fitted values of renewable electricity consumption. Source: Authors' computation.

Roots of the companion matrix

Table 6 Jarque–Bera test test for normality.

1 0.724 0.700

Imaginary

0.5

H0: Residuals are normally distributed H1: Residuals are not normally distributed

0.819 0.679 0.827

0.406

0 0.988

1.000 0.406

-0.5 0.827 0.679

0.700 0.724

0.819

-1 -1

-0.5

0

0.5

Equation

Chi2

Df

prob4 chi2

D_LNre D_LNrgdppc D_LNopen D_LNfin D_LNfuel D_LNfdig ALL

0.908 2.716 1.167 0.150 0.810 1.229 6.981

2 2 2 2 2 2 12

0.63508 0.25719 0.55783 0.92771 0.66686 0.54085 0.85887

1

Real The VECM specification imposes 5 unit moduli Points labeled with their moduli

Fig. 4. Graph of eigenvalue stability test. The VECM specificantion imposes 5 unit moduli.

Table 5 Langrange-multiplier test for serial correlation. H0: there is no autocorrelation at lag order H1: there is autocorrelation at lag order Lag

Chi2

Df

Prob4chi2

1 2

43.0715 40.6315

36 36

0.19440 0.27362

Financial development has a significant positive impact on renewable energy in China but the impact is small. A 1% increase in financial development leads to 0.19% increase in renewable electricity adoption. This corroborates the study by Pfeiffer and Mulder [31] and Brunnschweiler [4]. As the financial sector develops, the capacity to provide credit facilities to finance major projects such as clean energy technological development and other forms of renewable energy technologies increases. The small impact of financial development on renewable energy in China can be explained by two factors. First, the financial sector in China is still regulated to a large extent thereby undermining its capability to finance major projects effectively without government guarantee. Second, the inherent risks involved in the financing of clean energy technology due to uncertainties about future climate policies discourage the financial sector from providing funds to finance clean energy projects. This argument follows earlier study by IEA [15] which posit that financial institutions are unwilling to

Table 7 Result of forecast-error variance decomposition (FEVD). Step

RE

RGDPPC

OPEN

FIN

FUEL

FDIG

1 2 3 4 5 6 7 8 9 10

1 0.793506 0.677307 0.704216 0.762784 0.762744 0.760061 0.750748 0.761468 0.766796

0 0.022272 0.084235 0.067642 0.051344 0.046082 0.053652 0.057037 0.057370 0.054614

0 0.052931 0.049218 0.054122 0.041280 0.039017 0.034368 0.035466 0.032222 0.033088

0 0.116406 0.172606 0.157216 0.127970 0.129022 0.131692 0.138632 0.133043 0.129567

0 0.012862 0.010937 0.012554 0.013389 0.020322 0.017732 0.015886 0.013939 0.014086

0 0.002022 0.005697 0.004249 0.003233 0.002813 0.002495 0.002230 0.001959 0.001850

invest in new energy technologies because of uncertainties in future climate policies and long payback periods. This therefore supports Liming [23] and Wang and Chen [45] findings that innovative financing frameworks, instruments and mechanisms are required to finance renewable energy in China. The share of fossil fuel in total energy consumption has a negative significant impact on renewable electricity adoption in China. A 1% increase in the share of fossil fuel in energy consumption leads to a 1.16% reduction in renewable energy consumption in China. This is in line with the result of Aguirre and Ibikunle [1], Marques et al. [18] and Sovacool [37]. Sovacool [37] argue that the lobby effect of traditional energy sources impede renewable energy while Pfeiffer and Mulder [30] argue that high fossil fuel production appear to delay renewable enegry. This explains the power of the lobby of conventional energy in underming renewable energy. According to Wang et al [44], the rapid growth of fossil fuel capacity is one of the three major constraints to the development of renewable electricity in China.

694

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

irf, LNfdig, LNre

irf, LNfin, LNre

irf, LNopen, LNre

irf, LNre, LNre

irf, LNfuel, LNre

2 1 0 -1 -2

irf, LNrgdppc, LNre

2 1 0 -1 -2 0

5

10

0

5

10

0

5

10

step Graphs by irfname, impulse variable, and response variable Fig. 5. Impulse response function (IRF).

China is the largest primary energy consumer in the world, and over 50% of energy consumption in the country is sourced from fossil fuel. There is high possisbility that investors in the fossil fuel sector would enhance efforts to undermine renewable energy policies. Though China has achieved some fair result in increasing the level of renewable energy in recent years, the lobby of the fossil fuel industry still has strong influence on the development of renwable energy in the country. 4.5. Vector error correction model: short run dynamics The result of the vector error correction model in Table 4 shows that there is a long run causality running from the dependent variables as a whole to renewable electricity consumption. The error correction term (  0.66) is negative and significant which confirms the existence of long run causality. The error term indicates that the speed of adjustment towards long run equilibrium in the system is 66%. In other words, when there is an exogenous shock to the model, the model corrects its disequilibrium by 66% speed of adjustment per year in order to return to the equilibrium. In the short term, only the first lag of financial development has significant impact on renewable electricity consumption. From the result of the short run causality, there is a uni-directional causality from financial development to renewable electricity consumption at 10% significance level, and from renewable electricity consumption to trade openness at 5% level. There is no short run causality between renewable electricity onsumption and real GDP per capita, foreign direct investment and share of fossil fuel consumption in either directions. 4.6. Diagnostic tests We perform a number of test to examine the validity and stability of the model. The tests include goodness of fit to check how related is the actual curve and the fitted curve; eigenvalue stability test to verify the stability of the model; Langrange-multiplier test to check for serial correlation; and Jarque–Bera to determine whether the residual are normally distributed.

4.6.1. Goodness of fit We examine the goodness of fit of the cointegrating equation. The historical data of all the independent variables are substituted into the cointegrating equation. We compare the curve of the actual values of renewable energy with the fitted values. Based on Fig. 3, it is seen that the cointegration equation has a very high goodness of fit. The correlation coefficient between the actual curve and the fitted curve is 0.85, which is very close to 1. 4.6.2. Eigenvalue stability condition for stability test The result of the stability test is shown in Fig. 4. The result shows that except the unit roots assumed by the VECM model itself, all the eigenvalues of the adjoint matrix are smaller than 1, and there is no characteristic root outside of the unit circle in the figure. The result of Fig. 4 implies that the model is stable. 4.6.3. Test for serial correlation Table 5 shows the result of the Langrange-multiplier test for serial correlation. Based on the prob 4chi2 values, the null hypothesis of no serial correlation cannot be rejected. This implies that there is no serial correlation in the model. 4.6.4. Test for normality of residuals The result of the Jarque–Bera test in Table 6 shows that the prob 4 chi2 is more than 0.05 significance level. This implies that the null hypothesis of normally distributed residuals cannot be rejected. In other words, the residuals are normally distributed. 4.7. Impulse response and variance decomposition We examine the impact of shocks on the model using the impulse response function (IRF) and the forecast-error variance decompistion (FEVD). The result is shown in Table 7 and Fig. 5. Based on the result in Table 7, in the first year (step 1), variation in renewable electricity consumption in China is caused by 100% shocks in itself. In the second year, shock in renewable electricity consumption is caused by 79.3% variation in itself, 2.2% variation in real GDP per capita, 5.3% variation in trade openness, 11.6% variation in financial development, 1.3% variation in the “lobby effect”, and 0.2% variation in foreign direct investment. By year 10, shocks

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

in renewable electricity consumption in China will be caused by 76.7% shock in itself, 5.5% shock in real GDP per capita, 3.3% shock in trade openness, 13.0% shock in financial development, 1.4% shock in lobby effect, and 0.2% shock in foreign direct investment. The impulse response reflects the response of renewable energy to one standard deviation shock to the independent variables. The figure indicates that shock to foreign direct investment in China has a temporary effect on renewable electricity consumption in China, as the effect dies out quickly. Similarly, shocks to financial development, trade openness, real GDP per capita and renewable electricity consumption tend to die out in the long term after intial deviations. On the contrary, shocks to the “lobby effect” seems to have a permanent and explosive effect as there is no indication of stabilising to zero after the tenth period.

5. Conclusions and policy implications This study examines the long term determinants of renewable electricity in China using data from 1980 to 2011. A number of studies have been conducted on renewable energy adoption but most of these studies adopt panel data analysis, and focus on the amount of renewable energy. This study differs by focusing on time series analysis of China and examine the determinants of the share of renewables in total electricity consumption. The Johansen cointegration technique and vector error correction model are used to analyse the long run and short run relationship between renewable energy and its influencing factors in China. Based on the result of the analysis, a number of findings emerge. First, real GDP per capita promotes the share of renewables in total electricity consumption in China. This is because as a result of economic development the country has enough financial and human capital to invest in and adopt renewable electricity. Also, due to increase in income and standard of living, the public will be willing to adopt renewables in order to mitigate the air pollution associated fossil-fuel electricity production. Second, foreign direct investment and trade openness undermines the share of renewable electricity in total electricity consumption. FDI and trade openness lead to increase in total electricity consumption relative to renewable electricity consumption. FDI and trade openness may enhance the amount of renewables but may not increase the share of renewables in total electricity consumption. Third, financial development has a positive and significant impact on the share of renewables in electricity consumption but the impact is small. As the financial sector develops, it develops the capacity to finance clean energy technology projects. However, the risks involved in financing renewable energy projects due to uncertainties in future climate policies and long payback period limits the impact of the financial sector on renewable energy development. Fourth, conventional fossil fuel have significant negative impact on renewable energy. The result of the forecasterror variance decomposition show that fluctuations in renewable electricity consumption in China is majorly caused by shocks in itself and financial development. Also, while the impact of shocks to all the variables have a temporay effect, shocks to the lobby effect seems to have a permanent and explosive effect on renewable electricity. Based on the findings of the study, the following policy suggestions are recommended. First, the Chinese government need to give priority to renewable electricity in order to ensure a lowcarbon and sustainable economic development. The current environmentally-harmful economic development model in China should be effectively changed to create way for renewable energy use and a more sustainable model. This could be done by mobilising the resources from economic development towards renewable energy development. Second, while trade openness and foreign direct investment should be encouraged, the Chinese

695

government need to deliberately ensure that foreign investors and manufacturers of export goods in China develop and embrace renewable electricity. Third, the financial sector should be strengthened and supported to improve their capability to finance clean energy technology investments. This could be done by providing government guarantees for projects that promotes the development and dissesmination of renewables. Fourth, the Chinese government should decisively deal with the lobby influence of the fossil fuel industry on renewable energy development. Thus, deliberate measures should be taken to significantly reduce fossil fuel consumption. This could be done by eliminating subsidies to the fossil fuel sector and imposing environmental tax to capture the economic and environmental costs of fossil fuel consumption. This paper examines the factors that drives the development of renewable energy in China. Specifically, it investigates the factors that promotes the share of renewable-generated electricity in total electricity consumption in China. The result of this paper is important for policies that would not only promote the amount of renewable electricity but also the share of renewable-generated electricity in total electricity consumption. However, the limitation of this paper is that it examines renewable energy in China at the aggregate level. Future studies should aim at investigating the drivers of the share of renewable-generated electricity at the provincial and sectoral level in China, given the differences in renewable energy resource endowment and economic development level across provinces in China. Also, future studies can also comprehensively investigate the determinants of different types of renewable energy.

Acknowledgements This paper is supported by Xiamen University-Newcastle University Joint Strategic Partnership Fund, the Grant for Collaborative Innovation Center for Energy Economics and Energy Policy (No. 1260-Z0210011), Xiamen University Flourish Plan Special Funding (No. 1260-Y07200), and the China Sustainable Energy Program (G1506-23315). The authors also acknowledge the initial comments by Dr. John T. Dalton of the Department of Economics, Wake Forest University, Winston-Salem, North Carolina, United States.

References [1] Aguirre M, Ibikunle G. Determinants of renewable energy growth: a global sample analysis. Energy Policy 2014;69:374–84. [2] Beckman J, Borchers A, Stenberg P. The determinants of on-farm renewable energy adoption. Presentation at the Agricultural and Applied Economics Association’s 2011 AAEA & NAREA Joint Annual Meeting. Pittsburgh, Pennsylvania. July 24-26; 2011. [3] Bird L, Bolinger M, Gagliano T, Wiser R, Brown M, Parsons B. Policies and market factors driving wind power development in the United States. Energy Policy 2005;33:1397–407. [4] Brunnschweiler CN. Finance for renewable energy: an empirical analysis of developing and transition economies. Environ Dev Econ 2010;15(3):241–74. [5] Carley S. State renewable energy electricity policies: an empirical evaluation of effectiveness. Energy Policy 2009;37:3071–81. [6] Chang TH, Huang CM, Lee MC. Threshold effect of the economic growth rate on the renewable energy development from a change in energy price: evidence from OECD countries. Energy Policy 2009;37:5796–802. [7] Chien T, Hu JL. Renewable energy: an efficient mechanism to improve GDP. Energy Policy 2008;36:3045–52. [8] Damijan JP, Knell M, Majcen B, Rojec M. The role of FDI, R&D accumulation and trade in transferring technology to transition countries: Evidence from firm panel data for eight transition countries. Econ Syst 2003;27:189–204. [9] Dickey DA, Fuller WA. Distribution of the estimators for autoregressive time series with a unit root. J Am Stat Assoc 1979;74:427–31. [10] Engle RF, Granger CWJ. Co-integration and error correction: representation, estimation and testing. Econometrica 1987;55:251–76. [11] Eyraud L, Wane A, Zhang C, Clements B. Who’s going green and why? Trends and determinants of green investment. IMF Working Paper No WP/11/296; 2011.

696

B. Lin et al. / Renewable and Sustainable Energy Reviews 55 (2016) 687–696

[12] Fengqi Z. Development of China renewable energy. Renew Energy 1996; 9(1):1132–7. [13] Gan L, Eskeland G, Kolshus. H. Green electricity market development: lessons from Europe and the US. Energy Policy 2007;35:144–55. [14] Hunag MY, Alavalapati J, Carter D, Langholtz M. Is the choice of renewable portfolio standards random? Energy Policy 2007;35:5571–5. [15] IEA. Reducing greenhouse gas emissions: the potential of coal. Paris: Coal Industry Advisory Board, International Energy Agency; 2005. [16] Johansen S, Juselius K. Maximum likelihood estimation and inferences on cointegration with applications to the demand for money. Oxf Bull Econ Stat 1990;52(2):169–210. [17] Johnstone N, Hascic I, Popp D. Renewable energy policies and technological innovation: Evidence based on patent counts. Environ Resource Econ 2010;45:133–55. [18] Marques AC, Fuinhas JA, Manso JRP. Motivations driving renewable energy in European countries: a panel data approach. Energy Policy 2010;38:6877–85. [19] Martinot E. Power sector restructuring and environment: trends, policies and GEF experience. Washington, DC: Global Environment Facility; 2002. [20] Menz F, Vachon S. The effectiveness of different policy regimes for promoting wind power: experience from the States. Energy Policy 2006;34:1786–96. [21] Mitchell C, Sawin J, Pokharel GR, Kammen D, Wang Z, Fifita S, et al. Policy, financing and implementation. In: Edenhofer O, Pichs-Madruga R, Sokona Y, Seyboth K, Matschoss P, Kadner S, Zwickel T, Eickemeier P, Hansen G, Schlomer S, von Stechow C, editors. IPCC special report on renewable energy sources and climate change mitigation. New York: Cambridge University Press; 2011. [22] Moomaw W, Yamba F, Kamimoto M, Maurice L, Nyboer J, Urama K, Weir T. Introduction. In: Edenhofer O, Pichs-Madruga R, Sokona Y, Seyboth K, Matschoss P, Kadner S, Zwickel T, Eickemeier P, Hansen G, Schlomer S, von Stechow C, editors. IPCC special report on renewable energy sources and climate change mitigation. New York: Cambridge University Press; 2011. [23] Liming H. Financing rural renwable energy: a comparison between China and India. Renew Sustain Energy Rev 2009;13:1096–103. [24] Lin B, Xie C. Reduction potential of co2 emissions in China’s transport industry. Renew Sustain Energy Rev 2014;33:689–700. [25] Narayan PK, Smyth R. Energy consumption and real GDP in G7 countries: new evidence from panel cointegration with structural breaks. Energy Econ 2008;30:2331–41. [26] Omojolaibi JA. Financing the alteratives: renewable energy in the Nigerian economy. In: Adenikinju A, Iwayemi A, Iledare W, editors. Green energy and energy security: options for Africa. Ibadan: Atlantis Books; 2012. [27] Omri A, Nguyen DK. On the determinants of renewable energy consumption: international evidence. Energy 2014;72:554–60. [28] Perc M, Szolnoki A. Coevolutionary games – a mini review. Biosystems 2010;99:109–25. [29] Peterson S. Greenhouse gas mitigation in developing countries through technology transfer: a survey of empirical evidence. Mitig Adapt Strat Global Change 2007;13(3):283–305. [30] Pfeiffer B, Mulder P. Explaining the diffusion of renewable energy technology in developing countries. Energy Econ 2013;40:285–96.

[31] Popp D, Hascic I, Medhi N. Technology and the diffusion of renewable energy. Energy Econ 2011;33(4):648–62. [32] Rafiq S., Alam K., 2010. Identifying the determinants of renewable energy consumption in leading renewable energy investor emerging countries. Paper presented at the 39th Australian Conference of Economists, held 27-29 September, in Sydney. [33] Rafiq S, Bloch H, Salim R. Determinants of renewable energy adoption in China and India: a comparative analysis. Applied Econ 2014;46(22):2700–10. [34] Sadorsky P. Renewable energy consumption, CO2 emissions and oil prices in the G7 countries. Energy Econ 2009;31:456–62. [35] Sadorsky P. Renewable energy consumption and income in emerging economies. Energy Policy 2009;37:4021–8. [36] Saibu OM, Omoju O, Nwosa PI. Trade openness and the dynamics of unemployment and poverty incidence in Nigeria: A multivariate cointegration analysis. Niger J Econ Soc Stud 2012;54(3):367–88. [37] Sovacool B. Rejecting renewables: the socio-technical impediments to renewable electricity in the United States. Energy Policy 2009;37:4500–13. [38] Sinani E, Meyer KE. Spillovers of technology transfer from FDI : the case of Estonia. J Comp Econ 2004;32:445–66. [39] SSDN, IDDRI. Pathways to deep decarbonisation – 2014 interim report. New York: Sustainable Development Solutions Network (SDSN) and Institute for Sustainable Development and International Relations (IDDRI), Columbia University; 2014. [40] Stadelmann M, Castro P. Climate policy innovation in the South – domestic and international determinants of renewable energy policies in developing and emerging countries. Global Environ Change 2014;29:413–23. [41] Unruh GC. Understanding carbon lock-in. Energy Policy 2000;28:817–30. [42] Vachon S, Menz F. The role of social, political, and economic interests in promoting state green electricity policies. Environ Sci Policy 2006;9:652–62. [43] Van Rooijen S, van Wees M. Green electricity policies in the Netherlands: an analysis of policy decisions. Energy Policy 2006;34:60–71. [44] Wang F, Yin H, Li S. China’s renewable energy policy: commitments and challenges. Energy Policy 2010;38:1872–8. [45] Wang Q, Chen Y. Barriers and opportunities of using the clean development mechanism to advance renewable energy development in China. Renew Sustain Energy Rev 2010;14:1989–98. [46] Wang Y. Renewable electricity in Sweden: An analysis of policy and regulations. Energy Policy 2006;34:1209–20. [47] Wustenhagen R, Bilharz M. Green energy market development in Germany: Effective public policy and emerging customer demand. Energy Policy 2006;34:1681–96. [48] YeY.China’s renewableenergy: boom, bust andtransformation.TheChinaMonitor. 67. Stellenbosch:CentreforChineseStudies,StellenboschUniversity;2011.p. 4–9. [49] Zhao Z, Zuo J, Fan L, Zillante G. Impacts of renewable energy regulations on the structure of power generation in China – a critical analysis. Renew Energy 2011;36(1):24–30.