What affect energy poverty in China? A path towards sustainable development

Abstract Despite the crucial role of investment in energy under public–private partnership (I.E.P.P.P.) in abating environmental pollution and reducing energy poverty (E.P.), the existing literature offers less information about the nexus between I.E.P.P.P. and E.P. In order to identify the E.P. gap based on accessibility, affordability, and availability dimensions, this study investigates the factors influencing E.P., and examines the impact of I.E.P.P.P., globalisation (G.L.O.), output (G.D.P.), risk, technological innovation (T.I.) and renewable energy consumption (R.E.C.) on E.P. in China during the period of 1990 to 2019. The causal relationship between E.P. with its determinants is also examined. Utilising fully modified ordinary least squares (F.M.O.L.S.) econometric approach, we find that investment in energy with a public–private partnership, T.I., and gross domestic product (G.D.P.) bridge the gap for E.P., whereas R.E.C., composite risk index (C.R.I.), and G.L.O. increase the E.P. gap in China. In addition, frequency Domain Causality test reveals that unidirectional causation from I.E.P.P.P., G.D.P., T.I., G.L.O., risk, and R.E.C. to E.P. in the short run to long run.


Introduction
Many worldwide development challenges regarding climate change, inequality, and poverty have relationship with energy.Energy poverty (E.P.) is one of the concerning issues of the current era.International organisations identify E.P. as inaccessibility and difficulty in affording energy services or new kinds of energy in modern times (International Energy Agency [IEA, 2019;UNDP, 2010).As the household barely receives the essential energy resources, the I.E.A. ( 2019) describes E.P. E.P. usually happens with low-income households who have less ability to afford renewable energy resources.The unavailability of modern energy in developing countries leads to health and environmental issues.Moreover, Energy-poor economies seek to minimise energy scarcity through fossil-fuel subsidies.World Bank declared that the scarcity or inaccessibility of renewable energy leads to relying on fossil solid fuels which pollute indoor air and endanger human health (World Bank, 2017).

Overview of energy situation in China
As China has undergone strong economic growth over the past more than 40 years, China is the World's leading economy.China has organised market-oriented energy measures to provide vital energy to its industrial and household sectors.China has attained 99% of electrification since 2009 and aims to acquire 100% electricity access for its households to minimise E.P. (World Bank, 2017).The Figure 1 indicates that 98.39% of electricity is accessible to the population of China in 2018, which shows the high energy availability to its households.However, electric power losses are about 6.45% in transmission and distribution, which is less effective because the populace has a high electricity supply.China has attained high electricity access percentage, while the households rely on solid fuels such as coal and firewood (Tang & Liao, 2014), which is alarming for health and the environment (World Bank, 2017).China's technology implementation is essential for reducing energy scarcity, health, and climate change risks.About 51.88% of its population have access to clean fuels and technologies for cooking.

Consumption of modern energy in China
China's overall energy consumption is enormous as energy use in terms of kg of oil equivalent per capita was 2216.28 in 2018 and 767.00 in 1990.In comparison, its renewable energy consumption (R.E.C.) scores poorly and reduced from 34.08% in 1990 to 12.23% in 2018 (see Figure 2).As mentioned above, China's efficiency in electrification still lacks renewables.China's electricity consumption per person has surpassed the average per capita of upper-middle-income countries, creating E.P. in China (World Bank, 2014).The fossil fuel subsidy to lessen E.P. may be the reason behind this renewable deficiency (One Earth, 2019).Moreover, renewable energy is the least affordable for low-income households as solid fuels are readily available at less cost.These consequences have affected the competency of R.E.C. in China.China is now implementing subsidies for renewable energy reforms to reduce energy scarcity and collaborating with the clean environment.Above all, installing solar panels on the roof of low-income households may alleviate E.P. and income poverty in China (One Earth, 2019).

Energy supply in China
China is counted as the leading energy supplier and has an ample energy supply.The total primary energy supply measured in tons of oil equivalent is from 873.64 million to 3076.09 million in 30 years.However, private participation is essential for E.P. alleviation by cutting down its costs.In China, private energy investment is substantial, and the average investment was 2178.97 million dollars annually during the past 30 years (World Bank, 2018).Figure 3 shows on and off investments in energy by private participation.It is observed that high private investments in energy occurred in some years, such as 1997, 1999, 2003, 2013 and 2016.In contrast, it is recorded as very low in 2010 and 2018.Still, China has maintained its private participation in energy at the admired level to combat E.P. The past study favours that private investment can reduce costs, maximise benefits, and open up new projects and promotions in large projects like renewable systems (Bhide & Monroy, 2011).Energy consumption in other sectors as the ratio of total energy consumption was 53.06% in 1990, and this ratio had started to go down afterwards.It had a slight uprising interval during the period of 2012 and 2018, stable at above 25% and recorded 28.67% in 2018.However, in Figure 4, the overall average percentage of energy consumption in other sectors was about 37.11% from 1990 to 2018.
Besides, energy import had also proliferated, witnessing an average of above 14% during the period of 2015-2019 and 14.83% in 2018 (see Figure 3).However, in Figure 4, the average net energy import is about 5.98% during the period of between 2012 and 2018.China has started implementing renewable systems to enhance R.E.C. so as to clean the environment and reduce energy scarcity.The average consumption  of renewable energy is 21%, and the total consumption of energy is about 1438 kg of oil equivalent (World Bank, 2018).The public-private investment in energy has shown many fluctuations in the past 30 years (see Figure 5).Its stable input can reduce energy scarcity, but need government supports.However, a massive reduction in public-private energy investment in the last three years appeared in Figure 5.The previous study favours the public-private investment providing essential energy to the energy-poor (Abbas et al., 2020).
This study aims to identify E.P. gap based on accessibility, affordability and availability dimensions.The key emphasis is on reducing energy shortages in China.Hence, we investigate the factors influencing E.P. in the case of China over the period of 1990 to 2019.The impact of investment in energy under public-private partnership (I.E.P.P.P.), globalisation (G.L.O.), output (gross domestic product [GDP]), risk (C.R.I.), technological innovation (T.I.) and R.E.C. on E.P. is examined.Moreover, this study examines the causal relationship between E.P. with its determinants such as public-private partnership (I.E.P.P.P.), G.L.O., G.D.P., risk (C.R.I.), T.I. and R.E.C. for China.Thus, in terms of energy scarcity in China, it is worthwhile to assess renewable energy effects.Besides, the study would have policy consequences for the public-private investment in energy and renewable energy, rendering it more valuable for E.P. alleviation.
Hence, this study makes several interesting contributions to the existing body of literature on environmental sustainability.First, the empirical contribution of this work lies in using comprehensive proxies for E.P. gap.This study uncovers the identification of E.P. gap based on accessibility, affordability, and availability dimensions in the case of China, which has never been identified before.This approach enables us to estimate different aspects of E.P. Most of the previous studies on E.P. suffer from narrow definition of E.P. Second, the regression model of this endeavour departs from the previous studies, which did not consider the role of G.L.O., risk and T.I. in the examination of the nexus between investment in energy and the E.P. gap.We make an important contribution to environmental sustainability by offering the first study on the impact of investment in energy on E.P. in case of China.Third, from a practical perspective, the study highlights factors which affect China's E.P. gap with the significance of benefiting the Chinese authorities to devise strategy for bridging the E.P. gap.
The remainder of the study is structured as follows.Section 2 provides a review on relevant literatures.Section 3 discusses methodology.Section 4 reports and discusses the main results and Section 5 contains our conclusions and relative policy implications.

Literature review
Although E.P. is one of the big issues in developing economies, there is no standard definition of E.P., with the conditions different from region to region.In assessing the E.P. threshold, no widely used approaches exist.Researchers thus prefer to use various metrics to construct the assessment method.The I.E.A. describes the incompetence of access to modern, clean energy, electricity or conventional biomass as E.P. (IEA, 2002), which is extensively accepted.Bhide and Monroy (2011) agreed with the I.E.A. statement and discussed E.P. as the central issue.Management is incompetent to access unrealistic goals like implementing electricity policies and renewable energy technologies and coping with E.P.However, in rural areas, the households relying on solid fuels such as wood and some other biomass fuels for energy supply is then evidence of E.P. Simultaneously, the energy shortage has been an issue in rural areas for both households and industries for years, as the local demand increases and the rural areas cannot meet the increased energy demand (Kaygusuz, 2011).Sesan (2012) stated that E.P. is inaccessible to modern energy sources for households in developing countries and their corresponding dependence on solid cooking biofuels.E.P. is an issue with household economic circumstances, life, and energy prices (EPEE, 2006;Liddell et al., 2011).It has been found that the low household income intensifies E.P. (Healy & Clinch, 2004).However, in a description of E.P. or fuel poverty, basic living energy prices are above the social average, and residual income is below the authorised or professional economic poverty line (Hills, 2011).The households' incomes and the percentage spent on accessing energy or fuels are associated with the budget standard approach.The fuel poverty approach is directed to households that can bear their fuel cost less (Moore, 2012).Li et al. (2015) analysed rural household energy consumption and renewable energy systems in Zhangziying town of Beijing, and found that coal covers the most significant share of household energy usage and the requirement for household heating in winter.Renewables, such as biogas and straw gas, have simultaneously substituted coal, firewood, and L.P.G., shielding the energy scarcity and saving the atmosphere.However, a modern form of renewable energy often has good efficiency, sustainability, adequacy, and convenient access.Anyway, affordability comes first.The households are unable to afford expensive modern energy sources, and E.P. harms households' living standards.In India's urban areas, E.P. is exacerbated by the deficiencies in modern cooking and illumination fuels.Moreover, the energy scarcity is subdivided into three groups; transitional and moderate energy poor, and extremely poor in energy to access the exact energy deprivation.The findings revealed the shortfall in modern cooking and illumination fuels and the intensity of E.P. are among the extremely poor.However, the extreme energy-poor is amongst the entire population who is unable to afford the high energy costs (Nathan & Hari, 2020).
The developing countries lack research and development, technology and innovation, and other modern system aspects.The emerging nations are trying to employ renewable resources to meet sustainable energy and cover E.P., but it is hard to afford for struggling economies whose household incomes are low.The reason why reliance on solid energy is most significant lies in it covers household energy demand for living with less cost.China relies heavily on solid energy even though its rural area attained 100% electrification in 2014.Statistics indicate that 490 million Chinese citizens rely heavily on solid energy, coal and firewood for household living, i.e., cooking and heating (Tang & Liao, 2014).A nationwide household survey in 2016 shows that 30% of Chinese households depend on solid fuels.In contrast, that percentage in rural area of China is much higher (Xie & Hu, 2014).However, China still has E.P. problems in rural and urban households.The low-income households accept the contemporary energy forms but are unable to afford high energy costs (Yang & Mukhopadhaya, 2017).According to a survey, 46% of households lack modern energy, and a substantial number of households are deemed less competitive for energy prices and fall in E.P. Central China has a high E.P. rate at contemporary energy prices, while there is E.P. in Western China.Lin and Wang (2020) emphasised low-income households are suffering from E.P. and need to enact policies.Social and economic factors, such as house size, family size, schooling, occupation, and household head also affect energy scarcity.The family head is critical for getting over E.P. since men usually earn more than women.The house's ownership status is also essential for E.P. because rental status takes a more substantial and significant effect on E.P. (Abbas et al., 2020).
Household's cooking and heating reliance on solid fuel has enhanced the risk of respiratory disease and deaths in low-income countries (World Health Organization (WHO, 2007a(WHO, , 2007b).E.P. is not explicitly linked with mortality but triggers associated illnesses that reflect the detrimental health consequences of E.P. (Wilkinson et al., 2007).According to previous studies, E.P. has a major impact on human being health.Solid fuel reliance has exaggerated anthropogenic pollutants and induced indoor air contamination, responsible for inadequate health standards (Lacey et al., 2017).Unreliable indices of E.P., incapability, and inadequate access to renewable energy sources damage emerging economies.China's household reliance on solid fuels such as coal and firewood is high, resulting in indoor air pollution and affecting household health (Zhang et al., 2019).China has established a market-driven modern energy source but has not solved the E.P. crisis yet.Solid or non-renewable sources still plays a significant role in China.Other determinants, including household wages, health, and environmental measures not synchronised with E.P., need further investigation in China's context.
Most industrialised economies have taken important steps to cope with environmental and economic challenges.These steps include the transformation of industrial structure from non-renewable to renewable energies (Huang et al., 2022).For such transformation, I.E.P.P.P. is considered as crucial.Also, investment in energy under I.E.P.P.P. plays a critical role in abating environmental pollution and reducing E.P.
However, the existing literature offers but less information about the nexus between investment in energy under I.E.P.P.P. and E.P.This study adds something new to the existing literature by investigating the impact of I.E.P.P.P., along with G.L.O., GDP, risk, T.I. and R.E.C. on E.P. in case of China.To our knowledge, no study till now has been conducted to examine these important determinants of E.P.

Theoretical framework
Figure 6 presents the conceptual basis for the nexus proposed in this paper.The main aim of this study is to find out the impact of investment in energy under I.E.P.P.P. on E.P. in case of China.I.E.P.P.P. is the energy sector reform which would improve environmental quality by reducing carbon emissions.In green energy finance, the I.E.P.P.P. is a useful aspect as it overwhelms the contraction in public financing and enables energy facilities production.Hence, we expect a negative impact of I.E.P.P.P. on E.P.
Following Wang et al. (2022), this study uses T.I. as important explanatory variable in the model of E.P.. T.I. is the modern and revamped process to increase productivity that advances the standard of living and can strengthen long-term economic prosperity.T.I. in the energy sector can augment energy resources, enhance energy services quality, provide environment-friendly energy, and tackle E.P. by covering economic costs.Renewable energy as carbon-neutral forms obtained from renewables are naturally replenished on a human scale.
Following Zhao et al. (2022), this study uses G.L.O. as an important explanatory variable in the model of E.P. G.L.O.involves expanding movements and interactions of human beings, products and resources, assets, technology and cultural norms worldwide.One of G.L.O.'s consequences is it facilitates and strengthens the relations between distinct regions and communities worldwide.The World Trade Organization (WTO) defined G.L.O.into two elements.First, foreign boundaries open quickly to transfers of products, resources, finance, citizens, and thought.Second, reforms to national and international structures and policies encourage or foster those flows.G.L.O.measures include both de jure and de facto.Hence, we expect that G.L.O. will lead to reducing E.P. in China.
Following Wang et al. (2022), this study uses renewable energy as important explanatory variable in the model of E.P. Renewable energy is renowned as green energy or sustainable energy.Renewable energy reforms can hinder the quality of the atmosphere and also can reduce E.P. Hence, we expect that R.E.C. enhances E.P. in China.
To create an index to determine the extent of E.P., we constructed the 'weighted average of the four E.P. indicators' (W.A.E.P.I.).We used three different weight sets (W 1 , W 2 and W 3 ) to represent different scenarios (Table 1).
Three weight sets are used to measure E.P. We construct W.A.E.P.I. as follows: where n ¼ Normalised indicator (Figure 7).We obtained the Composite Energy Poverty Index for identifying gap (C.E.P.I.) in the following way: C.E.P.I. for a country ¼ 100 -W.A.E.P.I.
Considering accessibility (i.e., W 1 ), the numbers are bigger for all years and have significantly improved from 2001.
With equal weight to all the parameters (i.e., W 2 ), the E.P. index reduces to a deficient number.Equal weight exaggerates the E.P. situation.With unequal weightage (i.e., W 3 ), we gave more importance to accessibility but did not ignore the role of affordability and availability, thus arriving at a balanced index (Figure 8).
With either of the weights, the most energy poor is in 1990 while the least is in 2019.Comparing these years, all terms of energy availability, accessibility, and affordability have improved.

Model specification
In Equation ( 1), E.P. is measured through three different composite energy poverty (C.E.P.) indexes for identifying gaps.All three models are given below as:  Source: Calculated by the authors.
In Equations ( 2) to (4) CEP 1 , CEP 2 and CEP 3 index for E.P. gap measuring accessibility, availability and affordability gaps, respectively.Definitions and sources of each variable are provided in Table 2  or mixed integrated variables ought to arise.The unit root test is also central for verifying not series I(2) or higher.So with two structural breaks, we use the latest unit root assessment (Narayan & Popp, 2010).This root unit structural break test is compatible and differentiates the actual dates of splits.This newest unit root test for structural breaks includes two models to validate stationarity orders: model 1 is intercept-related and facilitates structural intercept breaks.In contrast, model 2 permits intercept and slope structural breaks.Below are the two model equations for the Narayan and Popp (2010) unit root test: Model 1 Model 2 Here, X t represents all the included variables, and D with (X t , X t-i ) shows the first difference with e it error term in both model's equations.
P P i¼1 d i DX t-i is the dependent variable's first difference lagged to eradicate the serial correlation.Model 1 equation reflects dummy variables in intercept as DU 1 and DU 2 progress in time T B1 , T B2 to hook promising structural breaks.Model 2 equation imitates the dummy variables as DU i and DT i , i ¼ 1,2 progressing in time T B1 , T B2 to permit intercept and slope structural breaks.However, the unit root null hypothesis alongside the unit root alternative hypothesis refers to the t-statistics of X t-1 .

Cointegration tests
3.3.2.1.Maki (2012) cointegration test.In contrast to earlier Gregory and Hansen (2009) and Hatemi-J (2008), Maki (2012) is ideally used for cointegration while structural breaks are valued.Maki's procedure of cointegration permits multiple structural breaks, permitting up to five unforeseen structural breaks in cointegration processing.We also progress the regime approach consists of level shift with trend and regime shift with trend.This approach sanctions structural breaks in levels (t) and regressors (d) in level shift with trend which also refers to regime shift, while regime shift approach with trend has structural breaks (q), levels (t) and regressors (d).Furthermore, the existing test survives on the basis of the null hypothesis (support no cointegration existence) and the alternative hypothesis (support cointegration existence).The equation of Maki cointegration test with regime approach is as follows: Equation of Level Shift with Trend Here, Y t is a dependent variable, d t shows the series of regressors, and t specifies the time.However, structural breaks denoted by q, levels with t, and e t with disturbance term.Z it value is 0 if time periods are less than structural breaks periods (t < TB i ), while Z it value is 1 if time periods are greater than structural breaks periods (t > TB i ).This Maki (2012) model will identify the structural breaks and cointegration among variables.
3.3.2.2.Bayer-Hanck cointegration test.This Byer-Hanck (2013) cointegration test is a joint integration process in which integration of series should be I(1) (Bayer & Hanck, 2013).In the collaborative integration process, it includes the outcomes of EG (Engle & Granger, 1987), J (Johansen, 1991), Bo (Boswijk, 1994) and Ba (Banerjee et al., 1998).The critical value should be less than F-statistics (F-stat > Critical value) to repudiate the null hypothesis of no cointegration.There are two steps of this cointegration test; step 1 contains EG-J values, and step 2 consists of EG-J-Bo-Ba values to deny the null hypothesis. Step-1 Step-2 Here, the consistent parameter indicates b i Ã , while the endogenous variable in the transformed form refers to y Ã .However, the adjustment parameter of autocorrelation is denoted by d and T is for the time period.Equation ( 7) is employed to cover biased endogeneity and serial correlation problems.

Frequency domain causality
Finally, we evaluate the causation interaction and assess parameter stability models using the Breitung and Candelon (2006) spectral granger-causality test.Equated with the traditional Granger causality test, this analysis's measurement approach enables the estimation of the response variable with specific time frames.This indicates that there are historical shifts in policy intervention, which means policy actions will be enacted across historical shifts.However, the method only applies to a finite time scale, so infinite horizon models are not anticipated.The equation of frequency domain causality is the following equation: In Equation ( 8), the linear constraint leads to a null hypothesis M y!x x ð Þ ¼0.The evaluated parameters, though, are a's and b's in time t.The lag is symbolised by p, and an error term by e t .

Results
This section discusses the outcomes based on renewable energy and public-private energy investments to limit China's E.P.However, G.L.O., T.I., and GDP also reduce E.P.This study also contains the composite risk, including economic, financial, and political, all matters in energy generation to limit China's E.P. To evaluate these variables, the first step is unit root testing to measure the integration order of variables, leading us to further long-run investigation.The shock in the economy, such as economic and financial shock, is observed by structural break in the Narayan  Table 3 shows the outcomes of the Narayan and Popp (2010) unit root test.This latest unit root test has variable integration order and two structural breaks.I.E.P.P.P. is integrated at the first difference I(1) with 2000 and 2008 structural breaks.G.L.O.Table 4 consists of two cointegration tests, i.e.Maki (2012) and Bayer-Hanck (2013) cointegration test.Maki cointegration test allows cointegration to accept the alternative hypothesis and permits multiple structural breaks.In Table 4 of Maki (2012) cointegration analysis, cointegration exists for all C.E.P.'s in both level shifts with trend and regime shift with trend.CEP 1 , CEP 2 and CEP 3 are cointegrated at 1% in both level shifts with the trend and regime shift with trend.The findings also permit structural breaks of 1996, 2001and 2008for CEP 1 , 1993, 1998and 2008for CEP 2 , 1998, 2005and 2008 for CEP 3 .After this, Bayer-Hanck (2013) cointegration test is applied, which is the combined effect of EG (Engle & Granger, 1987), J (Johansen, 1991), Bo (Boswijk, 1994) and Ba (Banerjee et al., 1998).In the first step of EG-J, the F-statistics values of CEP 1 CEP 2 and CEP 3 are greater than 5% critical value and comprehensively reject the null hypothesis of no cointegration.In the second step of EG-J-Bo-Ba, the F-statistics significant values are greater than 5% critical value, repudiate the null hypothesis, and admit the existence of cointegration in all CEP 1, CEP 2 and CEP 3 .The results of both Maki (2012) and Bayer-Hanck (2013) expose cointegration, which permits the long run existence and refers to F.M.O.L.S. for the long run estimation.
The long run results of determinants of E.P. have been listed in Table 5.The results suggest that I.E.P.P.P., T.I., GDP, R.E.C., G.L.O. and C.R.I. are important variables influencing E.P. The overall findings of F.M.O.L.S. have shown that GDP, T.I.  1996-2001-20081993-1998-20081998-2005-2008Bayer-Hanck Test (2013) EG-J EG-J-Bo-Ba EG-J EG-J-Bo-Ba EG-J EG-J-Bo-Ba The coefficient value of the investment in energy by private-public participation (I.E.P.P.P.) for CEP 1 is À0.002, CEP 2 À0.000 and CEP 3 is À0.001, which indicates the negative influence on E.P. The improved public-private participation in energy investment has resolved China's E.P.This analysis is identical to previous research; energy investment by private-public participation provided the elemental energy requirements to the energy poor (Abbas et al., 2020).
In modern society, clean energy needs leading economies to protect the environment and limit E.P. Renewable energy refers to clean energy, and its implementation is China's leading reform to eliminate energy scarcity.The coefficient value of R.E.C. for CEP 1 is 0.087, for CEP 2 0.021 and for CEP 3 is 0.049, which positively influenced E.P.These findings support the earlier findings of Zhao et al. (2022) and Wang et al. (2022).
China's T.I. is essential to minimise the risk of energy shortages and climate change.Access to renewables and technologies for cooking contributes to their communities, comprising about 51.88% of the population (World Bank, 2018).The coefficient of T.I. for CEP 1 is À0.062, CEP 2 is À0.011 and CEP 3 is À0.029, negatively influencing E.P.These findings support the earlier findings of Zhao et al. (2022) and Wang et al. (2022).
China is now the global leader of economic interaction and G.L.O., and is seeking to extend its operations across the globe.One example is the Belt and Road Initiatives, which focuses on connecting a wide range of nations to the regions and is one of the factors for the rise in G.L.O.China and G.L.O., which has been going on for a long time, are no strangers.Chinese markets are developing new investment prospects, and the global economy appears to be turning its attention to China.The shift towards China, however, induces energy scarcity.The coefficient values of G.L.O.towards the composite index for E.P. are CEP 1 0.002, CEP 2 0.000, and CEP 3 0.001, which increase E.P. in China.China is in the midst of electrifying and introducing clean energies, and the energy shortages still take considerable time to be alleviated.In this situation, it's too early to anticipate G.L.O.'s assenting to limit China's E.P.These findings support the earlier findings of Zhao et al. (2022).
The coefficients of C.R.I. towards the composite index for E.P. are À0.051 (CEP 1 ), À0.007 (CEP 2 ) and À0.020 (CEP 3 ), which suggest that GDP has a substantial and detrimental effect on China's E.P. The results have shown that GDP has performed its role in limiting E.P. in China's long-term.These findings support the earlier findings of Zhao et al. (2022) and Shahbaz and Feridun (2012).
The GDP for CEP 1 , CEP 2 and CEP 3 are 0.001, 0.0002 and 0.0004 positively influencing China's E.P. in the long run.This C.R.I. refers to economic, financial and political risks that illustrate less effectiveness in limiting E.P. The collective participation of the financial, economic, and political sectors is essential to initiate new ventures.Suppose any sector is less competitive, or in doubt, it is not easy to progress.In this scenario, China's financial, economic, and political sectors seemed reluctant to make any mutual risks to ensure minimal E.P. Abbas et al. (2020) manifested economic factors seemed effective in reducing E.P.
Table 6 explores the findings of the Breitung and Candelon (2006) frequency domain causality test.The frequency-domain causality test is proposed to inspect the causality relations from GDP, T.I., G.L.O., C.R.I., I.E.P.P.P., and R.E.C. to composite indexes for E.P. (CEP 1 , CEP 2 , CEP 3 ).The findings are shown in long (x i ¼ 0.05), medium (x i ¼ 1.50), and short run (x i ¼ 2.50).The indicators, GDP, T.I., G.L.O., C.R.I., I.E.P.P.P. and R.E.C., examined the unidirectional causality to CEP 1 for the short, medium, and long run.The unidirectional causality from such variables to CEP 1 applies to ÃÃÃ , ÃÃ and Ã , with a significance of 1%, 5% and 10%.In CEP 2 and CEP 3 , the unidirectional causation assessed from GDP, T.I., G.L.O., C.R.I., I.E.P.P.P. and R.E.C. to CEP 2 and CEP 3 significantly in the short, medium, and long run.The Breitung-Candelon Frequency Domain Test reveals that all short-, medium-, long-term indicators significantly cause E.P. The study focuses greatly on energy investment with public-private participation, renewable energy and C.R.I. comprising financial, economic and political risks, which induced E.P. in the short, medium, and long run.

Discussions
It is evident from the results that investment in energy with P.P.P. model reduce E.P. in China.China has established a modern enterprise system to collectively participate in the government and private sectors to initiate new ventures, which has achieved three public-private participation phases.The first phase began in the 1990s with the development of power stations, and the second phase began in 2004 with a landmark issuance.However, the third and transformative stage, energy and technical advancement with public-private participation, began in 2013.To settle energy scarcity, private and public sectors jointly participate in China's market-based energy initiative (Wang et al., 2020).During the period of 1990-2019, there have been many public-private participation fluctuations, but the operative projects have perceived a reduction in energy shortages (World Bank, 2018).
GDP is negatively related to E.P. in China.In recent decades, China has witnessed a high GDP, and China's strong economic growth is taking part in the energy-generating process.China has organised a market-oriented energy system to provide energy for its household and industry sectors.In 2009, China obtained 99% electrification and in 2013 reached 100% electricity (Lin & Wang, 2020;World Bank, 2017).
China's renewable energy enactments are reluctant to limit E.P. in the long-term.It may be of inefficiency and high prices of renewable energy in China, some shreds of evidence have exposed the reasons for ineffective renewables.The I.E.A. declares that the management is incompetent to access renewables' unrealistic targets (Bhide & Monroy, 2011).China relies heavily on solid fuels, which are less expensive and readily accessible, although renewables are not feasible (Tang & Liao, 2014;Xie & Hu, 2014).Extreme poor and low-income households can't afford high energy prices (Nathan & Hari, 2020;Yang & Mukhopadhaya, 2017).Moreover, the subsidy on fossil fuels to lessen E.P. is also a hurdle behind renewables inefficiency (One Earth, 2019).Forty-six per cent of households lack modern energy like renewable energy, because of high energy rates (Lin & Wang, 2020).
Technological advancements have been affirmative to limit the E.P. in China for a long period.Technological advances in electricity contribute to reducing energy deficiency.Nevertheless, electric vehicles are one of the finest T.I.s that seek to reduce E.P. and environmental hazard.However, families with low wages cannot have the affordability of renewable energies and modern technologies.Implementing solar panels for low-income communities aims to alleviate E.P. and income poverty (One Earth, 2019).

Conclusion
This study focuses on public-private participation in energy investment and R.E.C. to alleviate China's E.P. We also reveals the C.R.I. based on financial, economic and political risks to limit E.P. The China-based data from 1990-2019 is taken for variables such as composite indexes for E.P. (CEP 1 , CEP 2 , CEP 3 ), GDP, T.I., G.L.O., C.R.I., I.E.P.P.P. and R.E.C.However, the composed data source is 'World Bank, World Development Indicators'.The empirical model incorporates various econometric techniques such as Narayan and Popp (2010) structural breaks unit root test, Maki (2012) and Bayer-Hanck (2013) cointegration tests, and F.M.O.L.S.Moreover, the Breitung and Candelon (2006) frequency domain causality test is applied for short, medium, and long run causation to E.P. in China.The findings reveal that Narayan and Popp (2010) unit root test has the same integration order of variables at the first difference I(1) with two structural breaks and leads to cointegration existence and F.M.O.L.S. in the long run.Maki (2012) have shown cointegration for CEP 1, CEP 2, and CEP 3 .The combined effect cointegration test named Bayer-Hanck (2013) comprehensively rejected the null hypothesis of no cointegration in both EG-J and EG-J-Bo-Ba steps and admitted the cointegration existence in all CEP 1 CEP 2 and CEP 3 .Both cointegration tests, Maki (2012) and Bayer-Hanck (2013), permit long-run existence.
The F.M.O.L.S. method shows that GDP, T.I., and I.E.P.P.P. negatively influence E.P. in China.However, R.E.C., G.L.O., and C.R.I. have failed to limit China's E.P. in the long run.The study greatly focuses on R.E.C. because of China's investment in renewables projects.However, the inefficiency and other factors are reluctant to limit China's E.P. in the long term.The study also emphasises China's financial, economic, and political risks, seemed unenthusiastic to make any mutual risks to ensure minimal E.P. Nevertheless, renewables and C.R.I. positively contribute to E.P. in China.In contrast, I.E.P.P.P. limits China's energy shortages in the long run.The frequency-domain causality test shows unidirectional causation from GDP, T.I., G.L.O., C.R.I., I.E.P.P.P. .andR.E.C. to CEP 1 , CEP 2 and CEP 3 in the short-, medium-, and long-term.
On the basis of the findings of this study, we suggest that China should subsidise renewables to limit E.P. and ineffectiveness.Renewables and updated technologies should be low costs for low-income households as they cannot afford the high cost of living.The installation of solar panels, and electrification in low-income households, perhaps alleviate E.P.Moreover, China should improve its efficiency in electrification, which still lacks renewables.Reducing the E.P. will help the policy makers to devise strategies for improving environmental quality.
Anyway, this study only focuses on the China economy.Future research can be carried out to replicate these findings across different OECD countries.Furthermore, this study is limited to examining important variables' linear impact on E.P. A future study may be carried out to examine the interaction effect of I.E.P.P.P. and G.L.O. in affecting E.P.

Disclosure statement
No potential conflict of interest was reported by the authors.

Figure 8 .
Figure 8. Composite Energy Poverty Index for Gap (CEPI) for China.Source: Calculated by the authors.

Figure 7 .
Figure 7. Weighted Average Energy Poverty Index for China.
and Popp unit root test.The structural breaks of 2000 and 2008 for I.E.P.P.P. indicate economic reforms started of 2000 and 2008 respectively.During these periods, China invested billions of yuan to strengthen the energy sector.Similarly, structural breaks of 2001 and 2009 for G.L.O.indicate China's joining in the W.T.O. in 2001 and financial crises in 2009.The structural break of 1998 represents the financial crises during the period.The structural break of 2013-2014 represents the oil shock.The structural break for 1992, 2008 and 2010 represent market based economic reforms implemented by the Chinese government in 1992, the global financial crisis of 2007-2008 and the stimulus package announced by the government of China in 2010.
is integrated at level I(0), but more significance at first order and integrated order is taken I(1) with structural breaks 2001 and 2009.T.I. is integrated at the first difference I(1), and its structural breaks are 2003 and 2007.R.E.C., Composite risk index (C.R.I.), and Economic growth as GDP are integrated at a level but more effective at first difference.Integration order is taken at I(1) with structural breaks 2002 and 2012 for R.E.C., 2002 and 2005 for C.R.I., 2001 and 2008 for GDP.However, a composite index for E.P. CEP 1 , CEP 2, and CEP 3 are integrated at first difference I(1) with structural breaks 1998 and 2013 for CEP 1 , 2005 and 2008 for both CEP 2 and CEP 3 .The Narayan and Popp (2010) unit root test has shown the same orders of integration I(1) among variables, which led this research to follow the cointegration testing and F.M.O.L.S. for the long run.

Table 1 .
Weights for Energy Poverty Index.

Table 2 .
Definitions and sources of data.
Source: Calculated by the authors.
Critical value at 5% for each test is 10.352 and 19.761, respectively.Note: Ã , ÃÃ and ÃÃÃ represent significance level 10%, 5% and 1%, respectively.Source: Calculated by the authors.and I.E.P.P.P. have reduced E.P. in China in the long run.Nevertheless, China's E.P. has been further enhanced by its C.R.I., R.E.C. and G.L.O.