Study on Rural Residents’ Satisfaction with the Clean Energy Heating Program in Northern China—A Case Study of Shandong Province

Abstract

The Chinese government announced the Clean Energy Heating Program in northern China in 2017, promoting clean energy for residents’ winter heating. The key difficulty of implementing this policy initiative lies in rural areas. This research hence focuses on evaluating the implementation of this policy in rural areas. Rural residents who directly benefit from, and are integrally involved in, the implementation process in Shandong Province were surveyed to evaluate their satisfaction with this policy. In order to identify their satisfaction indicators and obstacle factors, a TOPSIS obstacle model adjusted by entropy weight was developed. An evaluation system of the indicators of residents’ satisfaction with the policy was developed and converted into a questionnaire. The designed questionnaire was distributed to 341 rural residents in Jinan, Zibo and Heze in Shandong province. Data analysis suggests that, at the fiscal subsidies level, supporting infrastructure, technical supports and support organizations are four important factors affecting rural residents’ satisfaction. The key obstacle factors identified include technical support, supporting infrastructure, the operation subsidies level, heating cost, period of subsidies and achieved temperature. Corresponding suggestions for further clean energy heating policy design and implementation in rural areas in northern China are provided.

1. Introduction

It is a worldwide challenge to meet the increasing energy needs caused by improved human living standards, while protecting the environment and maintaining sustainability at the same time. Such a challenge is particularly significant for China as one of the largest developing countries. In northern areas of China, the outdoor temperature averages between −10 °C and 5 °C, which persists over 3–6 months every year. Given the low temperature, heating in winter is essential for residents in both urban and rural areas. Currently, winter heating is mainly provided by coal, which causes significantly increased pollutant emissions and severe haze [1,2]. Frequent and persistent haze has generated attention among the rest of society. In 2017, four ministries, including the Ministry of Ecology and Environment, jointly issued the Air Pollution Prevention and Control Work Plan for the Beijing—Tianjin–Hebei region and surrounding areas. The work plan promotes the Clean Energy Heating Program (hereafter referred to as CEHP) in the two selected municipalities—Beijing and Tianjin, and 26 cities including Taiyuan, Shijiazhuang, Jinan, Zibo and Heze. Clean energy heating means providing heating by utilizing clean energies such as natural gas, electricity, geothermy, biomass and solar power. The aim of this program is to reduce energy consumption and pollutant emissions caused by utilizing coal for winter heating. In 2018, the Notice on Carrying Out the Pilot Work of Central Government Financial Support for Clean Energy Heating in Winter in Northern China was announced, promising that the central government will provide financial support for clean energy heating renovation in both urban and rural areas in the selected administrative cities. The key point of this policy is to promote heating provided by natural gas and electricity, replacing coal, and is also known as the Double Energy Substitution Policy [3]. This policy aims to increase the proportion of winter heating provided by clean energy in northern China to 70% by 2021. For the selected administrative cities, the aim is to achieve heating in urban areas fully provided by clean energy, while 60% of heating in rural areas is already provided by clean energy. The CEHP has deeply influenced millions of households’ livelihoods, attracting attention from the whole society. To date, the CEHP has been implemented for 4 years. It is necessary to evaluate the performance of this policy in order to provide suggestions for future related policy initiatives. The CEHP covers both urban and rural areas, among which rural areas represent the main obstacle to sustainable heating and hence require special attention. The energy consumption caused by heating in rural areas in northern China accounts for 12% of total building energy consumption in China [4]. In rural areas in northern China, almost all households use decentralized winter heating, and the heating equipment is usually inefficient. Moreover, the thermal characteristics of residential building envelopes are poor, leading to poor indoor thermal environments. Therefore, the scale of the task implied by clean energy heating renovation in these rural areas is immense [5,6]. In addition, economic income in these rural areas is generally low, and the cost of clean energy heating is relatively high for low-income rural residents. Since the task of establishing building energy efficiency in these rural areas has only recently started, rural residents’ awareness of energy efficiency is weak. Considering this, clean energy heating renovation in rural areas needs more policy promotion and fiscal support [7,8,9]. Rural residents are sensitive to related policies and normally choose their means of heating according to these policies. For example, Liu et al. [10] found that some rural residents change back to coal heating because of their perception that fiscal subsidies will soon end [10]. Therefore, rural residents’ satisfaction with the policy is critical for CEHP’s sustainable influence.

The existing literature on clean energy heating demonstration projects and policy can be classified into four categories: (1) the first area of focus explores how demonstration projects promote related technology innovation. For example, Harborne and Hendry studied clean energy heating demonstration projects and their influence on related innovation in Europe and Japan [11]; (2) the second area of focus is mainly concerned with on how demonstration projects promote the industrial applications of related technology. For example, Buijs and Silvester investigated how clean energy heating demonstration projects promoted the related technology application in residential buildings in the Netherlands [12]. Bell and Lowe, as well as Tommerup et al., explored how clean energy heating demonstration projects facilitated applications of clean energy heating technologies in the UK and Denmark [13,14]; (3) the third category primarily focused on investigating how to improve the effectiveness of demonstration projects. In terms of clean energy heating in China, most of the related existing literature has focused on feasibility analyses of a particular clean energy source. For example, Song et al. [15] explored whether corn-straw-densified fuel could be utilized to provide heating in rural areas [15]; Wang et al. [16] investigated the feasibility of utilizing a geothermal heat pump in rural areas [16]; Xu et al. [17] studied the suitability of various clean energy heating technologies; while Zhang et al. [18] studied the road map for clean energy heating in inner Mongolia, China [17,18]; (4) the fourth area of focus mainly studieses clean energy heating policy implementation, such as Liu et al. [10], Yin et al. [19], who adopted a government perspective to explore, focusing on whether given goals and tasks have been achieved [10,19]. Only a few of them have studied policy implementation from the residents’ perspective, which significantly influences policy performance. Song et al. [20] studied the influence of operation subsidy on the rural residents’ willingness to pay and found that the level of operation subsidy has a direct influencestraight effect on the rural residents’ affordability [20]. Factors such as revenue level, technology application, technology attitude and, outcome affect residents’ willingness and behaviors to varying degrees [21,22,23]. According to the theory of citizen-led governments, the main function of governments is to fulfil the citizens’ needs, positioning citizens in the center [24]. In other words, in order to evaluate policy performance, it is essential to investigate the citizens’ evaluation and satisfaction with the policy implementation [25]. Gong et al. [26] evaluated how rural residents’ perceived fiscal subsidy influenced their willingness to pay for clean energy heating. By using a Customer Satisfaction Index, Xu and Ge evaluated the effectiveness and sustainability of the clean energy heating policy in Hebei, China. These two studies provide significant insights into the rural residents’ attitude towards the CEHP [27]. However, a more comprehensive evaluation of rural residents’ satisfaction is needed. Therefore, the current research aims to evaluate the CEHP implementation performance from rural residents’ perspectives by a case study of Shandong Province. Specifically, the second section explains the three selected pilot cities Jinan, Zibo and Heze in Shandong province and their local policies and practices. This will be followed by justifying the adjusted TOPSIS model, which is used to survey and analyze the rural residents’ satisfaction with CEHP and identify the main obstacles. Corresponding policy recommendations will be given in the conclusion section, providing references for further clean energy heating policies.

Therefore, the current research aims to evaluate CEHP implementation performance from rural residents’ perspective by a case study of Shandong Province. Specifically, the second section explains the three selected pilot cities Jinan, Zibo and Heze in Shandong province and their local policy and practices. This will be followed by justifying the adjusted TOPSIS model, which is used to survey and analyze rural residents’ satisfaction with CEHP and identify key obstacle factors. Corresponding policy recommendations will be given in the conclusion section, providing references for further clean energy heating policy.

2. Background Information of Shandong Province

2.1. Winter Heating Situation in Shandong Province

Shandong Provinceprovince is located in the air pollution transmission channel in the Beijing-Tianjin-Hebei area which, hence, is a representative province in Northern China. Shandong is classified into the Cold Zone, hence heating is essential to residents in winter. Winter heating in Shandong is mainly provided by coal. Winter heating in Shandong consumes around 43 million tons of coal, among which 10 million tons are consumed in rural areass. By 2017, the building area in rural area in Shandong is 1.335 billion m², in which only 0.241 billion m² is heated by clean energy. The rate of heating provided by clean energy is clearly too low, similar to other provinces in northern China.

In rural areas of Shandong, buildings are distributed dispersedlydesperately. Most residential buildings are bungalows in soilearth structure or brick and concrete structure, without external thermal insulation system. Hence, the thermal insulation properties of the bungalows are not good, the energy consumption is high, and the thermal comfort level is low. These residential buildings are heated dispersedlydesperately. For 79% of the total building area in rural areas, winter heating is provided by coal stoves, heated brick beds or air conditionersconditioning [28]. Only 21% of the total building area is heated by clean energy such as natural gas, electricity or biomass energy.

In terms of resident’s income levels, average disposable income for rural residents in Shandong is 17,775 yuan in 2019. This number varies in different cities according to its economic situation. The average disposable income for rural residents in Jinan is 19,454 yuan, which is 19,916 yuan in Zibo and 14,176 yuan in Heze [29]. The three selected cities are representative to evaluate the clean energy heating policy performance.

2.2. Fiscal Supports for CEHP in the Selected Pilot Cities

CEHP is implemented in an up-down way in China. Specifically, the Chinese central government listed the key tasks and goals and set performance assessment goals for local governments. Correspondingly, local governments are required to make detailed work plans to guide peasant households to renovate the heating system. In other words, the clean energy heating renovation work in China is mainly driven by the government’s promotion, through providing fiscal subsidies and administrative orders. In many waysTo many points, this could be viewed as a reform on the supply side of winter heating. The Chinese government aims to establish a clean energy heating market and cultivate residents’ new clean energy consumption habits.

The Chinese government established an up-down fiscal support system to implement CEHP in pilot cities. Specifically, the central government, provincial governments and municipal governments provided certain portions for the Program from 2017 to 2019. The central government provides fiscal support according to the scale of pilot citiescity, 0.7 billion yuan for capital municipal government per year and 0.5 billion yuan for municipal government every year. Local governments provide funding for unfulfilled needs such as clean heating equipment purchase and installation and subsidies for energy costs. Seven cities in Shandong, including Jinan, Zibo, and Heze, are selected as pilot cities for CEHP. In this study, Jinan, Zibo and Heze, which are representative of pilot cities in terms of economic conditions and locations, are selected. The local governmentsThe three selected pilot cities, Jinan, Zibo and Heze developed different standards and methods for the three selected pilot cities, Jinan, Zibo and Heze for subsidies (see

Table 1. Subsidies for Replacing Coal by Natural Gas in the Three Pilot Cities.

	Jinan	Zibo	Heze
Equipment	Gas boiler	Gas boiler	Gas boiler
Subsidy standards	(1) Provide 2000 yuan for equipment for each household. (2) Provide operative subsidy, 1 yuan for per m3 natural gas for each household, up to 1200 yuan. (3) Apply the lowest price level for natural gas bill in heating period.	(1) Provide money that equals to 70% of equipment price, up to 2700 yuan. (2) Provide operative subsidy, 1 yuan for per m3 natural gas for each household, up to 1200 yuan.	(1) Provide 4000 yuan for equipment for each household. (2) Provide operative subsidy, 1 yuan for per m3 natural gas for each household, up to 1000 yuan.
Fiscal support methods	(1) Local governments purchase equipment and transfer equipment subsidy to suppliers. (2) Provide operative subsidy in advance of the heating period by purchasing natural gas in advance for each household.	(1) Local governments purchase equipment and transfer equipment subsidy to suppliers. (2) Provide operative subsidy for each household according to natural gas consumption at the end of a heating period.	(1) Local governments purchase equipment and transfer equipment subsidy to suppliers. (2) Provide operative subsidy for each household according to natural gas consumption at the end of a heating period.
Period	Heating periods from 2017–2023	Heating periods from 2017–2023	Heating periods from 2017–2020

and

Table 2. Subsidies for Replacing Coal by Electricity in the Three Pilot Cities.

	Jinan	Zibo	Heze
Equipment	Air source heat pump, domestic electric boiler, heating cable, carbon crystal heating plate	Regenerative electric heating equipment, air source heat pump	Regenerative electric heating equipment, air source heat pump
Subsidy standards	(1) Provide 2000 yuan for equipment for each household. (2) Provide operative subsidy, 0.2 yuan for per KWH electricity for each household, up to 1200 yuan. (3) Apply the lowest price level for electricity bill in a heating period.	(1) Provide allowance which equals to 85% of equipment price, up to 5700 yuan. (2) Provide operative subsidy, 0.2 yuan for per KWH electricity for each household, up to 1200 yuan.	(1) Provide allowance which equals to 50% of equipment price, up to 4000 yuan. (2) Provide 50 KWH electricity for free and 0.3 yuan for per KWH electricity beyond 50 KWH for each household, up to 1000 yuan.
Fiscal support methods	(1) Local governments purchase equipment and transfer equipment subsidy to suppliers. (2) Provide operative subsidy in advance of the heating period by purchasing electricity in advance for each household.	(1) Local governments purchase equipment and transfer equipment subsidy to suppliers. (2) Provide operative subsidy for each household according to natural gas consumption at the end of a heating period.	(1) Local governments purchase equipment and transfer equipment subsidy to suppliers. (2) Provide operative subsidy for each household according to natural gas consumption at the end of a heating period.
Period	Heating periods from 2017–2023	Heating periods from 2017–2023	Heating periods from 2017–2020

).

3. Research Method

3.1. TOPSIS Method Adjusted by Entropy Weight

TOPSIS method is also known as the solution distance method for pros and cons. The TOPSIS method analyzes the distance between ideal goals and limited solutions and sorts accordingly for the purpose of multi-objective decision making. Positive ideal solution and negative ideal solution are two key concepts of the TOPSIS method [30]. This method requires one to establish evaluation indicators system and collect corresponding data. Indicators are evaluated by calculating distance between its value to positive and negative ideal solutions through given formulas. For the sake of the current research, the calculation formula is adjusted by the entropy weight [31]. This is for improving the resolution of evaluated indicators as the entropy weight reflects relative intensities of competition between factors. If the participants’ evaluation of one particular factor differs greatly, it means that rural residents’ levels of satisfaction with one particular factor differ greatly. Hence, this particular factor should be given more weights and deserves more investigation. Therefore, the identified key indicators and obstacle factors have a greater impact on the rural residents’ evaluation of CEHP’s implementation and deserve more investigation and discussion for further development.

By utilizing the entropy weight, the indicator selection bias caused by nuance-patent difference could be avoided. Hence, the accuracy of multi-objective evaluation and decision-making could be improved by the calculation of entropy weights of evaluated indicators [32]. Specifically, the collected raw data of indicators will be adjusted by following four steps: standardize data, calculate indicators’ objective weights by entropy weight method, construct entropy weight decision-making matrix, and calculate ideal solution and proximity. These four steps will be explained in detail in the following paragraphs.

3.1.1. Standardize Data

The evaluation indicator value will be standardized by the extremum standardization method for the purpose of reflecting the relative position of the actual value of an evaluation indicator in the indicator system. The formula is as follows:

$$Y_{ij}=(X_{ij}-X_{min})/(X_{max}-X_{min})$$

(1)

$X_{ij}$ is the average of the actual values of indicator j in city i. $X_{max}$ is the maximum of the actual values, $X_{min}$ is the minimum of the actual value. $Y_{ij}$ will be used to establish the decision-making matrix.

3.1.2. Calculate Indicators’ Objective Weight by the Entropy Weight Method

In the TOPSIS method, indicators’ weight affects final marks for the evaluated object. Hence, in order to achieve objectivity, the weights should be calculated by the objective calculation method [33]. For the sake of this research, the indicators’ weight is calculated by the entropy weight on the basis of raw value from data collection. The greater the difference between indicator’s entropy weight value, the more information that this indicator could express [34]. Suppose $p_{rj}$ is the actual value for indicator j given by person r in a selected region, $e_j$ is the entropy value of indicator j, $u_j$ is the entropy weight of indicator j. The calculation process is as follows:

$$f_{rj}={\displaystyle \frac{p_{rj}}{{\sum }_{r=1}^k{p}_{rj}}},(r=1,2,\dots ,k)$$

(2)

$$e_j=-{\displaystyle \frac{1}{lnk}}{\sum }_{r=1}^k{f}_{rj}ln{f}_{rj}$$

(3)

$$u_j={\displaystyle \frac{1-e_j}{m-{\sum }_{j=1}^m{e}_j}},(j=1,2,\dots ,m)$$

(4)

3.1.3. Construct Entropy Weight Decision-Making Matrix

Calculate the indicators’ entropy weight, $U=(u_1,u_2,\dots ,u_j)$.

Then, the normalized decision-making matrix will be constructed as follows:

$$M=Y\times U=\left[\begin{array}{cccc}{v}_{11}& v_{21}& \cdots & v_{i1}\\ v_{12}& v_{21}& \cdots & v_{i2}\\ \cdots & \cdots & \cdots & \cdots \\ v_{1j}& v_{2j}& \cdots & v_{ij}\end{array}\right]$$

(5)

3.1.4. Calculate the Ideal Solution and Proximity

The positive ideal solution $M^{+}$ and the negative ideal solution $M^{-}$ will be calculated as follows:

$$M^{+}=\left\{max{v}_{ij}|i=1,2,\dots ,n;j=1,2,\dots ,m\right\}=\left\{v_1^{+},v_2^{+},\dots v_m^{+}\right\}$$

(6)

$$M^{-}=\left\{min{v}_{ij}|i=1,2,\dots ,n;j=1,2,\dots ,m\right\}=\left\{v_1^{-},v_2^{-},\dots v_m^{-}\right\}$$

(7)

This will be followed by calculating proximity between the indicators’ value and the positive ideal solution and the negative ideal solution by Euclidean distance. The distance between the indicators’ value and the positive ideal solution $D^{+}$ mean proximity between the indicators’ value and the optimal target. The smaller the $D^{+}$, the closer that the rural residents’ evaluation of implementation of CEHP is to the positive ideal solution, and the better the policy’s performance is. The distance between indicators’ value and negative ideal solution $D^{-}$ mean proximity between the indicators’ value and the worst target. The smaller the $D^{-}$, the closer that the rural residents’ evaluation of the CEHP implementation is to the negative ideal solution, and the worse the policy performance is. $D^{+}$ and $D^{-}$ will be calculated as follows:

$$D^{+}=\sqrt{{\sum }_{j=1}^m{(v_{ij}-v_j^{+})}^2}$$

(8)

$$D^{-}=\sqrt{{\sum }_{j=1}^m{(v_{ij}-v_j^{-})}^2},(i=1,2,\dots ,n;j=1,2,\dots ,m)$$

(9)

Proximity will be calculated through the following formula:

$$T_i={\displaystyle \frac{D_i^{-}}{D_i^{+}+D_i^{-}}},(i=1,2,\dots ,n)$$

(10)

When $0\le T_i\le 1$, the bigger $T_i$ is, the higher that the rural residents evaluate the performance of CEHP. When $T_i=1$, the surveyed rural residents are satisfied with all CEHP in city i. When $T_i=0$, the surveyed rural residents give the lowest value to the performance of CEHP in city i. The $T_i$ is classified into four categories to reflect the surveyed residents’ satisfaction of CEHP (as

Table 3. Categories of Proximity.

Policy Performance	Poor	Average	Good	Excellent
Proximity	0.00–0.30	0.31–0.60	0.61–0.80	0.81–1.00

shows).

3.2. Obstacle Indicator Identification Model

The surveyed rural residents’ evaluation data of CEHP in their lived cities will be further analyzed to identify obstacle factors. Obstacle factors will be identified for the purpose of identifying ways to improve the residents’ satisfaction with CEHP. In addition to entropy weight $U_j$, factor’s deviation $I_j$ and obstacle degree $O_j$ will be analyzed. As explained in cities preceding section, $u_j$ refers to the weight of factor j’s contribution to the main goal. $I_j$ refers to the deviation between factor’s value to the ideal goal of the CEHP, which is the difference between 100% and the standardized values of a specific factor. $O_j$ refers to the degree of $I_j$’s influence on rural residents’ satisfaction with CEHP. The higher the value of $O_j$, the higher that factor j influences rural residents’ satisfaction, factor j hence is the more important obstacle factor.

$$O_j={\displaystyle \frac{I_j{u}_j}{{\sum }_{j=1}^m{I}_j{u}_j}}(j=1,2,\dots ,m)$$

(11)

3.3. Evaluation Indicator System

As stated in the preceding section, the current research evaluates the performance of CEHP by investigating the rural residents’ satisfaction of it. It is hence necessary to establish an evaluation indicator system of the rural residents’ satisfaction. Xu and Ge argued that factors such as the heating temperature, sense of fairness in policy, air quality, subsidy levels affect residents’ satisfaction [27]. Gong et al. pointed out that perceived fiscal subsidies influence rural residents’ acceptance of clean energy heating, as well as heating temperature, air quality and subsidy levels [26]. Skjevrak and Sopha agreed that heating temperature and indoor air quality affect resident’s satisfaction [35]. Liu et al. argued that the better the understanding of renewable energy the resident has, the better the resident recognizes the renewable energy policies [9]. Based on the above analysis, 60 rural households in three villages were then interviewed to collect and identify key aspects that rural residents in Shandong used to evaluate CEHP. The research group set the following interview outline: ‘which aspects of CEHP you are/are not satisfied with’, ‘which aspects of CEHP are important to you’, ‘which aspect of CEHP needs to be improved in your opinion’. The collected data were analyzed and key indicators for performance evaluation were identified. Three key aspects, fiscal subsidies, policy implementation support, and project results were identified as key components of the evaluation indicator system (as shown in

Table 4. Evaluation Indicator System of CEHP.

Evaluation Aspects	Evaluation Indicators	Symbol	Degree of Satisfaction (5-1)
Satisfaction with fiscal subsides	clean energy heating equipment which can obtain fiscal subsidies	S1	Very satisfied to very unsatisfied
	level of subsidies for equipment purchase	S2	Very satisfied to very unsatisfied
	level of subsidies for equipment operation	S3	Very satisfied to very unsatisfied
	types of subsidies for equipment purchase	S4	Very satisfied to very unsatisfied
	types of subsidies for equipment operation	S5	Very satisfied to very unsatisfied
	length of subsidies providing period	S6	Very satisfied to very unsatisfied
Satisfaction with policy implementation supports	Organizational support	G1	Very satisfied to very unsatisfied
	Technical support	G2	Very satisfied to very unsatisfied
	Infrastructure support	G3	Very satisfied to very unsatisfied
Satisfaction with project effect	Heating temperature	R1	Very satisfied to very unsatisfied
	Heating cost	R2	Very satisfied to very unsatisfied
	Indoor environment	R3	Very satisfied to very unsatisfied
	Applicability	R4	Very satisfied to very unsatisfied

).

In respect of fiscal subsides, six indicators were selected to measure residents’ satisfaction. These are clean pieces of energy heating equipment which can obtainfiscal subsidies, subsidy levels for equipment purchases and operation, subsidy types, subsidy length providing period. Clean energy heating equipment includes gas boiler, air source heat pump, domestic electric boiler, heating cable, carbon crystal heating plate, etc. The subsidy levels refer to the number of subsidies that the government provides for equipment purchase and operation. Types of subsidies refer to cash or direct natural gas or electricity subsidies. The length of subsidy providing period refers to the length of the period that the government promises to provide subsidies for. The central government promised to provide fiscal support for three yearsand then local governments decided the length of subsidy providing period according to the local situation. For policy implementation supports aspect, three indicators were selected to measure the residents’ satisfaction. These are organizational support, technical support, and infrastructure support. Organizational support refers to the special organizations established to take responsibility of implementing CEHP. Technical support refers to the technical support for equipment selection, installation and operation. Infrastructure support refers to pipe network for natural gas, electricity supplies, and related debugging facilities. For the project effectaspect, four indicators were selected to measure resident’s satisfaction, heating temperature, heating cost, indoor environment quality, and applicability. Heating temperature refers to the achieved temperature through using clean energy heating equipment. Heating cost refers to the cost that residents need to pay for heating equipment purchase and operation. Indoor environment quality refers to the indoor sanitary condition and air quality by using clean energy heating equipment. Applicability refers to the convenience and safety of clean energy heating equipment.

On the basis of the developed evaluation indicator system, a questionnaire was designed. The questionnaire also includes rural household’s the basic information of rural residents, such as gender, age, disposable annual income level, clean energy heating type, and floor area. The Likert scale of five was used for rural residents to measure their satisfaction with indicators. Residents could give 5–1 to measure their perceptions from very satisfied to very unsatisfied.

3.4. Data Collection

A questionnaire survey was conducted in a rural areas in Jinan, Zibo, Heze during February and March 2020. Questionnaires were distributed to rural residents who have adopted clean energy heating system for over one year and have received government subsidies. 109 valid questionnaires from 12 villages in 3 towns in Jinan, 120 valid questionnaires from 15 villages in 5 towns in Zibo and 112 valid questionnaires from 14 villages in 6 towns in Heze were collected. The selected villages are evenly distributed across the three cities and locations of surveyed towns have been marked in Figure 1.

In total, 341 valid questionnaires were collected. The number of rural residents in Jinan in 2020 was 2,442,425, that is, 1,210,500 in Zibo, and 4,334,114 in Heze. The percentages of selected participants in total rural population in three cities are not high, which is understandable considering difficulties of conducting the survey during the COVID-19 pandemic. However, the surveyed samples cover all types of gender, age, types of clean energy for heating and levels of floor area. The disposable annual income per household of selected residents also covers different levels, which is consistent with the different economic situation in the three selected cities. The basic information of surveyed samples has been demonstrated in

Table 5. Basic Information of Surveyed Rural Households.

Types		Jinan		Zibo		Heze
		Number	Proportion	Number	Proportion	Number	Proportion
Gender	Male	63	57.80%	75	62.50%	64	57.14%
	Female	46	42.20%	45	37.50%	48	42.86%
Age	Under 40	29	26.61%	32	26.67%	23	20.54%
	41–60	58	53.21%	63	52.50%	65	58.04%
	61 and above	22	20.18%	25	20.83%	32	28.57%
Disposable annual income per household	Under 30,000	7	6.42%	8	6.67%	12	10.71%
	30,001–40,000	29	26.61%	38	31.67%	57	50.89%
	40,001–50,000	44	40.37%	46	38.33%	24	21.43%
	50,001–60,000	18	16.51%	20	16.67%	12	10.71%
	70,000 and above	11	10.09%	8	6.67%	7	6.25%
Type of clean energy for heating	electricity	63	57.80%	76	63.33%	71	63.39%
	Nature gas	46	42.20%	45	37.50%	41	36.61%
Floor area for heating	Under 30 m2	17	15.60%	18	15.00%	23	20.54%
	31–60 m2	38	34.86%	42	35.00%	45	40.18%
	61–90 m2	37	33.94%	40	33.33%	34	30.36%
	91–120 m2	12	11.01%	12	10.00%	8	7.14%
	120 m2 and above	5	4.59%	8	6.67%	2	1.79%

. Hence, the surveyed samples are representative.

4. Results and Discussion

4.1. Descriptive Analysis

A descriptive analysis was conducted to analyze the collected data firstly. As shown in

Table 6. Descriptive Analysis of Surveyed Samples’ Satisfaction with CEHP.

		Jinan		Zibo		Heze
		Average	Standard Deviation	Average	Standard Deviation	Average	Standard Deviation
Satisfaction with fiscal subsides	Range	3.239	1.193	3.283	1.265	3.304	1.272
	Equipment subsidies level	3.404	0.954	3.675	0.963	3.536	1.065
	Operation subsidies level	2.89	0.916	2.867	1.328	2.741	1.419
	Equipment subsidies types	3.349	1.142	3.242	1.257	3.178	1.076
	Operation subsidies types	3.569	1.390	2.950	1.201	3.170	0.939
	Length of subsidies period	3.165	1.469	3.075	1.397	2.571	1.299
Satisfaction with policy implementation supports	Organizational support	3.312	1.445	3.092	1.092	3.205	1.148
	Technical support	3.128	1.292	3	1.489	2.92	1.459
	Infrastructure support	3.018	1.186	2.975	1.498	2.768	1.382
Satisfaction with project effect	Heating temperature	3.083	1.148	2.833	1.252	2.839	1.277
	Heating cost	2.807	1.198	2.475	1.181	2.509	1.099
	Indoor environment	3.358	1.032	3.359	1.235	3.455	1.2
	Applicability	3.606	1.106	3.475	1.061	3.554	1.047
Average value		3.225	1.19	3.1	1.248	3.058	1.206

, in the three selected cities, average values of the 13 evaluation indicators are greater than 3, which is 3.225 in Jinan, 3.100 in Zibo, and 3.058 in Heze respectively. It could be interpreted that rural residents are generally satisfied with CEHP. It has also been noticed that the average values of different indicators in one city are different, while the average values of the same indicator in the three selected cities are different. The reason for differences will be analyzed in the second step, details of which will be given in the next section.

4.2. Establish Standardized Decision-Making Matrix and Calculate Positive and Negative Ideal Solution

By using Formulas (1)–(4), each indicator’s standardized data, entropy value and entropy weight were calculated, the results of which are listed in

Table 7. Evaluation Indicators’ Entropy Value and Weight.

Indicator	Standardized Data			Entropy Value			Entropy Weight
	Jinan	Zibo	Heze	Jinan	Zibo	Heze	Jinan	Zibo	Heze
S1	0.560	0.571	0.576	0.9853	0.984	0.9833	0.0701	0.0645	0.0699
S2	0.601	0.669	0.634	0.9914	0.993	0.9902	0.0410	0.0301	0.0409
S3	0.473	0.467	0.435	0.9776	0.976	0.9709	0.1073	0.0985	0.1215
S4	0.587	0.561	0.545	0.9859	0.983	0.9874	0.0674	0.069	0.0529
S5	0.642	0.488	0.543	0.9815	0.982	0.9909	0.0886	0.0754	0.0382
S6	0.541	0.519	0.393	0.9749	0.977	0.9722	0.1202	0.0936	0.1161
G1	0.578	0.523	0.551	0.9893	0.986	0.9861	0.0513	0.0559	0.0583
G2	0.532	0.500	0.480	0.9799	0.973	0.9725	0.0959	0.1122	0.1150
G3	0.505	0.494	0.442	0.9829	0.972	0.9724	0.0818	0.1164	0.1156
R1	0.521	0.458	0.460	0.9837	0.979	0.9774	0.0779	0.0878	0.0945
R2	0.452	0.369	0.377	0.9795	0.976	0.9802	0.0980	0.0973	0.0827
R3	0.590	0.590	0.614	0.9894	0.986	0.9869	0.0506	0.0593	0.0550
R4	0.652	0.619	0.639	0.9895	0.990	0.9906	0.0500	0.0402	0.0395

. By using Formulas (5)–(7), indicators’ weighted decision-making matrix was developed, which was followed by calculating indicators’ positive and negative ideal solutions (see

Table 8. Evaluation Indicators’ Weighted Value, Positive and Negative Ideal Solution.

Indicator	Weighted Value			Positive Ideal	Negative Ideal
	Jinan	Zibo	Heze	Solution	Solution
S1	0.0392	0.0368	0.0402	0.0402	0.0368
S2	0.0246	0.0201	0.0259	0.0259	0.0201
S3	0.0620	0.0460	0.0529	0.0620	0.0460
S4	0.0396	0.0387	0.0288	0.0396	0.0288
S5	0.0569	0.0368	0.0207	0.0569	0.0207
S6	0.0650	0.0486	0.0456	0.0650	0.0456
G1	0.0242	0.0292	0.0321	0.0321	0.0242
G2	0.0510	0.0561	0.0552	0.0561	0.0510
G3	0.0413	0.0575	0.0511	0.0575	0.0413
R1	0.0406	0.0402	0.0435	0.0435	0.0402
R2	0.0443	0.0359	0.0312	0.0443	0.0312
R3	0.0298	0.0350	0.0337	0.0350	0.0298
R4	0.0326	0.0249	0.0252	0.0326	0.0249

).

In the three selected cities, some of the evaluation indicator’s value became a positive ideal solution or a negative ideal solution. Among 10 evaluation indicators, positive ideal solution of six indicators came from Jinan, positive ideal solution of one indicator came from Zibo while, four indicators’ positive ideal solution came from Heze, which shows that the performance of CEHP in Jinan gains the best recognition. Surveyed rural residents in Zibo and Heze have higher satisfaction with subsidy range, equipment subsidy level and organizational support. Hence, all of the three selected cities have space to improve the implementation of CEHP.

4.3. Proximity Analysis

By using Formulas (8)–(10), the proximity between evaluation indicator’s weighted value and positive ideal solution in three selected cities was calculated (see

Table 9. Obstacle Factors and Corresponding Satisfaction Degree of CEHP.

Region	Jinan	Zibo	Heze
Subsidy range	6.88%	7.28%	8.28%
Equipment subsidies	3.64%	3.98%	5.33%
Operation subsidies	10.08%	9.09%	10.87%
Equipment subsidy types	6.19%	7.65%	5.92%
Operation subsidy types	7.06%	7.27%	4.27%
Length of Subsidy period	12.28%	9.60%	9.38%
Organizational support	6.02%	5.78%	6.61%
Technical support	10.00%	11.09%	11.36%
Infrastructure support	9.03%	11.37%	10.51%
Heating temperature	8.32%	7.96%	8.94%
Heating cost	11.98%	7.10%	6.42%
Indoor environment	4.63%	6.92%	6.94%
Applicability	3.88%	4.92%	5.19%

). It has been found that the proximity of Jinan is the highest, which is 0.6230.896; the corresponding performance evaluation level is “good”. It could be argued that the surveyed residents’ satisfaction of CEHP implementation in Jinan is the best among the three selected cities. The proximity of Zibo is 0.4350.461, which is classified into the ‘average’ category. The proximity of Heze is the lowest, which is 0.3690.192 and is categorized into the ’average’ categorypoor. Although the three selected cities have adopted related policies, significant differences in the surveyed rural residents’ satisfaction of CEHP in three selected cities have been observed. The subsidy level in Jinan is not higher than that in Zibo and Heze, but subsidy types and policy implementation support in Jinan won more suveyed residents’ recognition and satisfaction. In other words, policy implementation support also largely influences the surveyed rural residents’ satisfaction.

4.4. Obstacle Factors Analysis

By using the developed obstacle model, surveyed rural residents’ evaluation of CEHP was analyzed to identify obstacle factors of residents’ satisfaction. The detailed results have been presented in Table 8. For the three studied cities, technical support, infrastructure support and length of subsidy period are the three leading obstacle factors. In addition, operation subsidy level and clean energy heating cost are also important obstacle factors, the obstacle degrees of which exceed 9%.

The obstacle degree of technical support in the three studied cities is the highest, 10.00% in Jinan, 11.09% in Zibo and 11.36% in Heze. This shows that the technical support provided by local governments is not satisfying in the surveyed rural residents’ opinions. This could be attributed to the local government officials’ lack of systematic planning for clean energy heating technical routes. It was noticed that the local government simply choose between ‘replacing coal by natural gas’ and ‘replacing coal by electricity’, without giving sufficient consideration of local resources, infrastructures and rural residents’ financial capability. In fact, in rural areas of northern China, biomass resources such as crop straw are affluent and can provide a large amount of energy for heating. However, biomass resources have not been fully utilized. As Song et al. pointed out, only a few heating projects were powered by biomass resources [36]. Moreover, equipment options provided by local governments are limited, which results in low flexibility. In addition, technical support for heating equipment operation is also not sufficient, which causes bad clean energy heating experience for the surveyed rural residents.

The obstacle degree of infrastructure support in the three studies regions is the second highest, 9.03% in Jinan, 11.37% in Zibo and 10.51% in Heze. This shows that infrastructure for natural gas and electricity supply for the surveyed rural residents’ winter heating are not sufficient. This is consistent with the ‘natural gas shortage’ and ‘electricity shortage’ phenomenon identified by Liu et al., who further argued that shortage is caused by the lack of consideration of local natural gas and electricity supply situation when the local work plan was initiated [10]. Along with the implementation of CEHP, the demand for natural gas increased dramatically. However, unstable sources of natural gas plus insufficient gas storage facilities and pipelines caused the natural gas shortage. Moreover, natural gas supply infrastructure in rural areas in Shandong is significantly insufficient, which caused more difficulty for natural gas supply. A similar case was found in electricity shortage. In order to fulfil the increased electricity demand caused by clean energy heating, the distribution of variable capacity per rural household should be increased to 5–7 KVA. However, the increased electricity demand only happens in winter, which is an intermittent load. Operating period of the increased electricity demand caused by winter heating is less than 1000 hours per year, which cannot make the lowest requirement of economic and technical indicators for distribution network renovation. In short, the economic benefit of distribution network renovation is very small. This could be viewed as one key reason why existing power distribution network renovation projects still cannot fulfil the increased demand caused by CEHP.

The obstacle degrees of the operation subsidy level in the three studied regions are 10.08% in Jinan, 9.09% in Zibo, and 10.87% in Heze. The obstacle degrees of heating cost in the three studied regions are 11.98% in Jinan, 7.10% in Zibo and 6.42% in Heze, respectively. This shows that although they received operation subsidies from government, the surveyed rural residents are still unsatisfied with the cost of clean energy heating. This concurs with Wang et al.’s observation that the increased cost by the clean energy heating equipment purchase and operation hinders the clean energy heating from widespread application [37]. This is consistent with the argument that the promotion of clean energy is highly dependent on government subsidies, and residents’ willingness to use clean energy is also largely influenced by the subsidy level and financial situation of the residents [9,26,27]. According to the collected data, the main heating fuel in Jinan, Zibo and Heze is scattered coal and the average cost of winter heating per household per year is around 1500 yuan. However, the average cost of winter heating by clean energy is generally over 3000 yuan, including both natural gas and electricity. The subsidies in the three selected regions are 1200 yuan per household, which means that rural households still need to pay the extra heating cost. It is essential to estimate rural residents’ affordable cost of heating [37]. In the three studied regions, the annual disposable income of 70% of rural households is between 30,000 and 50,000 yuan. According to Fan, household’s reasonable expenditure of fuel to keep comfort temperature should not exceed 5% of their disposable income [7]. Hence, the reasonable expenditure for winter heating in the three studied cities should not exceed 2500 yuan per year. Therefore, the actual expenditure of clean energy heating, 3000 yuan per household per year, as mentioned above, exceeds the reasonable expenditure for rural residents. It could be argued that the subsidies for clean energy heating are essential to relieve rural residents ofthe increased heating cost. This is why the surveyed rural residents are sensitive to the length of subsidy period, the obstacle degrees of which, in the three selected cities, exceed 9%. Correspondingly, the obstacle degree of length of subsidy period in Heze is the highest, 11.13%, compared to Jinan (9.37%) and Zibo (9.11%). Hence, it is urgent to reduce the cost of clean energy heating and figure out stable fiscal sources for subsidies. Otherwise, once subsidies suspend or end; it is highly likely that the rural residents will use coal for winter heating again as Xu et al. argued.

The obstacle degree of heating temperature in the three studied cities is 8.32% in Jinan, 7.96% in Zibo and 8.94% in Heze, respectively. This shows that the surveyed rural residents are not satisfied with the indoor temperature achieved by clean energy heating. It is clear that the purpose of CEHP is not only for energy saving and emission reduction, but also for improving thermal comfort for rural residents in northern China. However, in most rural residential buildings in Shandong, there are no insulated rooves and outer walls, and the wooden or aluminum windows are single-layered. As a result, buildings cannot be sealed well, thermal insulation performance is poor [38]. Considering this, the indoor temperature in rural residential building is difficult to increase. The poor thermal insulation performance reduces the indoor temperature provided by the clean energy heating, which influences the rural residents’ satisfaction with CEHP [27]. Hence, it is necessary to promote energy-saving renovation in existing rural residential buildings, while promoting clean energy heating. Currently, in rural residential buildings, the energy-saving renovation promotion is very slow, because of the rural residents’ low willingness and lack of funds. During the 13th Five Year Plan Period, no specific funding was initiated for energy-saving renovation, except the limited number provided by CEHP for the selected cities. The provided funding for energy-saving renovation is on a smaller scale, compared to the funding for energy-generating side. The provided funding for the energy-saving renovation of rural residential buildings is much less [36]. The local governments paid little attention to energy-saving renovation, mainly focusing on clean energy heating renovation in their work plan for CEHP. The limited work for energy-saving renovation is mainly for urban residential buildings during the CEHP implementation process. Most rural residential buildings have not been renovated for energy-saving, which should be paid more attention in further work.

5. Conclusions and Policy Recommendations

This research evaluated the implementation of CEHP in rural areas in northern China by investigating the rural residents’ satisfaction with CEHP. Through using the TOPSIS method adjusted by entropy weight, the collected data of rural residents’ satisfaction evaluation were analyzed to identify key indicators and obstacle factors. It has been found that the surveyed rural residents are generally satisfied with CEHP in Jinan, Zibo and Heze. However, differences exist between these three selected cities. Surveyed residents’ evaluation of CEHP in Jinan achieves “good”, while that in Zibo and Heze is “average” and that in Heze is poor. Obstacle factors have been identified, including technical support, infrastructure support, clean energy heating cost, subsidy levels and subsidy period and heating temperature. Underlying reasons for these obstacle factors have been discussed, on the basis of which suggestions for further clean energy heating in rural areas will be given in this section.

5.1. Formulate Technological Path That Adapts for Local Conditions

In order to overcome the obstacle factor of technical support, central government and local governments have to formulate technical routes that are adaptive to local conditions. Particularly, local governments should take advantage of the space to interpret and translate the central government’s policy suitable for local rural areas. Local governments should conduct investigations on basic characteristics of winter heating including geographical locations, resources, energy supply system, and energy consumption habits. On the basis of this information, local governments should also consider available policy resources to develop the adaptive local work plan of clean energy heating. Thirdly, the developed clean energy heating work plan should also be in compliance with the city development plan, the environment protection plan and energy development plan.Decide energies for winter heating according to village’s local conditions. Fourthly, technical routes to clean energy heating in different rural areas should be considered according to local resources, energy supply infrastructure, economic affordability, and environmental carrying capacity.

5.2. Improve Clean Energy Supply Infrastructure

Clean energy supply infrastructure should be considered and improved during the period of clean energy promotion. Natural gas supply sources should be confirmed before promoting CEHP and making the related work plan. Local governments should promote natural gas supply pipe construction in the rural–urban continuum. Supply pipe for natural gas, LNG, CNG, should also be conducted in rural areas that are economically affordable and practical. In terms of electricity supply, local governments should improve investment in electricity supply infrastructure construction and enhancement in rural areas, so that infrastructure could meet the needs of increased electricity caused by electricity heating in winter. According to the economic situation, electricity supply infrastructure in key villages should be enhanced, while electricity supply infrastructure should be available in all poor villages.

5.3. Refine Fiscal Subsidy Policy

In terms of the identified obstacle factors of heating cost, fiscal policy is suggested in the following three ways. Firstly, the subsidy level for different technical routes deserves more consideration, given that the commercialization degree of different technical routes varies. For centrally heated buildings, clean energy heating is relatively easy to commercialize. Hence, to reduce subsidy demands, the focus of policy making should lie in attracting the social capital and developing commercial models. Secondly, instead of one single subsidy level, different subsidy levels should be designed according to the income situation of different households.More operation subsidies should be provided to low-income households. Thirdly, develop long-term business models. (1) Clean energy heating renovation projects in large-scale regions should be packed to attract firms. Clean energy heating renovation and operation in a town or even a county could provide more profits and also reduce operation cost for involved firms. (2) The price mechanism of electricity and natural gas should be refined to reduce the clean energy heating cost. For areas suitable for heating by electricity, local governments should refine peak-valley time-of-use tariff and tiered prices for electricity mechanism and encourage market transactions of electricity. Local governments should encourage centrally heated rural residential communities to use regenerative electric boilers, which take advantage of peak-valley time-of-use tariff. For areas suitable for heating by natural gas, local governments should also establish and refine tiered price for natural gas and seasonal difference price. (3) Given that carbon emission could be significantly reduced by adopting clean energy heating, carbon trading and emission trading could also be applied to gain certain capital to provide further subsidies for clean energy heating application.

5.4. Improve Thermal Comfort Level of Rural Residential Buildings

In terms of heating temperature of rural residential buildings, energy-saving renovation should be promoted along with clean energy heating promotion. Currently, most rural residential buildings in Northern China have not had their building envelope renovated, leading to poor thermal insulation. In order to promote energy-saving renovation, standard and guidance of rural residential building planning and designing are recommended. Cost-effective energy-saving plans such as paste insulation board on the gable, updating windows into energy-saving windows and improving leakproofness between windows and walls, installing wind bucket or curtain, etc. are suggested. The Chinese Government is also suggested to initiate fiscal funding to encourage rural residents to conduct energy-saving renovation.

The Clean Energy Heating Program is playing a significant role in the rural living environment improvement initiative. Promotion of clean energy heating is of particular significance for narrowing the gap between urban and rural infrastructure of energy supply and public service and for improving environmental quality of rural areas. The proposed policy suggestions could provide meaningful references to future clean energy heating policies.