1. Background

global municipal solid waste (MSW) generation levels are estimated to increase to approximately 2.2 billion tons per year by 2025 [1]. Consequently, the traditional waste management system has encountered difficulties in disposing of large volume of MSW [1], which has led to inadequate disposals and has become a severe problem in many developing countries. China surpassed the U.S as the world’s top MSW producer in 2004. Now, China’s MSW generation is growing at an annual rate of 8–9% [1]. From 2010 to 2019, MSW production in Beijing rose from 6.35 million tons per year to 10.11 million tons, averaging 27,700 tons per day, which is 1.2 kg per person per day. With a series of policies, regulations, and measures, the composition structure of MSW management technology has changed in proportion; in 2020, Beijing reached 87% of MSW recycling (43% by incineration, 45% by landfill, 12% by composting) [2]. In addition, the main problem of MSW management in China is that the MSW composition is complicated. The average water content of MSW in Beijing reaches 50.19%, and this mixed MSW makes it difficult to sort and recycle [2]. The water contained in MSW not only pollutes recyclables in MSW, but also brings difficulties and increases the cost of MSW disposal, collection, transfer, transportation, and treatment. Therefore, the pre-sorting and treatment of MSW becomes a necessary condition for incineration [2]. In this case, effective MSW management has become necessary, not only from the human health perspective, but also from the aspect of environmental concerns [3]. Effective MSW management approaches such as smart MSW sorting systems [3] have been proposed to minimize the harmful effects of inadequate disposal in urban areas. However, in China, there is a lack of a uniform standard for MSW management charges, with local governments specifying the charges within administrative districts, one of which is levied incidentally through utility charges such as water, gas, and electricity [4]. Specifically, the payments for MSW management have normally been extracted from utilities expenses by local residents, which has become a possible cause of conflict regarding payments between residents and the local government [4]. The intelligent garbage sorting system (IGSS) is a widely used smart MSW sorting approach and is an effective way to improve urban MSW management and increase the productivity of cleaning contractors [5]. Compared with the current MSW treatment methods mainly performed in Beijing city, it can improve the efficiency of waste incineration treatment by substantially improving the efficiency of recycling and separation [5]. The MSW sorting features of IGSS are one of the most practical and economical methods and can reduce the costs of unsynchronized disposal by 80−90% [5]. In that regard, health-hazardous factors in MSW such as germs and other substrates can be substantially reduced [5]. In addition, the pre-sorting of unburnable MSW (metals, glass, etc.) can significantly reduce the costs of the second separation [5]. In this study, willingness to pay (WTP) is defined as the willingness of local residents to pay for IGSS, which is influenced by their thoughts and perceptions. Cases focusing on the WTP for IGSS have been frequently discussed in previous research and have played an important role in improving MSW management in developed countries [6,7]. In Europe, different countries have regulations on MSW separation and recycling, but there is no uniform standard for MSW management (it mainly depends on the consciousness of local residents) [8,9]. In addition, in Germany, the recycling of certain bottles can be rewarded in supermarket recycling devices (Similar to a type of IGSS), which improves the motivation of local residents for MSW separation [10]. In Japan, local residents sort MSW by different colored bags, and different levels of fees are associated with sizes and categories, which significantly increases the treatment capability of IGSS [11,12]. Compared with Europe and Japan, MSW management began late in China, with a larger amount of MSW. In this entry, investigating local residents’ WTP for MSW management will contribute to the development of government policies on MSW management.

2. Demographic Characteristics of the Respondents in Questionnaire and WTP

The social background of the respondents included gender, age, education level, household size, and monthly income per person (Shown in Table 1). Most of the respondents were in the 18 to 55 age range (94.2%). 52.1% of the respondents were male, slightly higher than the percentage of females. As for the education level, more than 75% of respondents have a bachelor’s degree or higher. More than 75% of the respondents had a monthly income of 3000–Yuan and 12% of them who earn more than 10,000 yuan a month. Regarding the family size, 73% of the families were composed of two to five family members.

Table 1. Individual demographic characteristics.

Item	Response	Frequency	Percentage
Gender	Male	162	52.1
	Female	149	47.9
Age	18–25	99	31.8
	26–35	99	31.8
	36–45	47	15.1
	46–55	48	15.4
	>55	18	5.8
Education Level	Elementary school & High school	74	23.8
	Bachelor	205	65.9
	Master’s degree	32	10.3
Household Size	2 or 3	113	36.3
	4 or 5	114	36.7
	>5	84	27
Monthly income per people	1000–3000	79	25.4
	3001–5000	71	22.8
	5001–8000	60	19.3
	8001–10,000	64	20.6
	>10,001	37	11.9

The result of WTP is presented in Table 1. As the price increases, the percentage of responses agreeing to pay decreases gradually. Nearly 30% of respondents in the pre-survey and final questionnaire refused to pay. Among those who refused to pay, the top three reasons were: payments can be diverted, unable to pay currently, and the MSW management is the responsibility of the government. Respondents who chose the option such as “I don’t have the ability to pay for the fund” and “I don’t think the MSW management worth that much” are “real zero responses” (Shown in Figure 1). The survey on social background found that respondents with higher incomes were more likely to pay. (β = 0.49, p < 0.05)

Figure 1. The motivation for the zero responses.

3.2. Measurement and Structural Model

The structural model of the raw TPB model was estimated using the maximum likelihood method. In the first step, we tested the plausibility and reliability of the measurement model by confirmatory factor analysis (CFA), then estimated the variable structure and correlations of the current model. To ensure convergence and discriminability, and measurement reliability, we conducted confirmatory factor analysis (CFA), including mainly the components of AT, SN, and PBC (Shown in Table 2). The results showed that the model data were within the plausibility interval (Chi-Square = 247.1, GFI = 0.808, CFI = 0.867, NFI = 0.804, RMSEA = 0.096). All four variable components were included and tested. The standard regression coefficients of AT, PBC, and EC in 0.01 level. The standard regression coefficient of the subject norm in the 0.05 level. All the scales achieved internal consistency.

Table 2. Distribution of responses.

WTP	10	20	30	Total
Positive	239	226	145	610
Negative	72	32	81	185
Protest zero	53	53	53	159
Total	311	311	311	933

The validity of the questionnaire data is corroborated by the fact that the AVE values in Table 3 are all higher than 0.5, according to reference [40], if the squared correlation coefficients of the different constructs are smaller than the AVE of each construct, then the discriminant validity can be confirmed.

Table 3. Reliability and CFA for the extended TPB model.

Scales	Mean (s. d.)	β	CR	AVE
Attitude			0.83	0.77
I think paying for MSW management is very positive	4.04 (1.11)	0.485
I think paying for MSW management is a responsibility	3.96 (1.15)	0.563
I think paying for MSW management is a pro-environmental behavior	3.96 (1.12)	0.584
Subject Norm			0.79	0.59
I think people who are close to me will pay for MSW management	3.83 (1.20)	0.668
I think people who are close to me will support the action of paying for MSW management	3.72 (1.24)	0.627
I think people who are close to me will support me paying for MSW management	3.77 (1.20)	0.494
Perceived Behavioral Control			0.87	0.74
I think my payment will improve the urban environment	3.97 (1.04)	0.466
It is not difficult for me to pay for MSW management	3.81 (1.27)	0.485
I think I have time, money, and resources to contribute to the MSW management.	3.67 (1.26)	0.593
Environment-concern
I care about urban environmental issues very much	3.86(1.23)	0.475	0.82	0.71
I think I will reduce other expenses for urban environment improvement	3.84(1.19)	0.693

Mean (s. d): Standard deviation. β: factor loading CR (composite reliability); AVE (average variance extracted).

A discriminant validity test of the scale was performed. According to reference [40], if the squared correlation coefficients of the different constructs are smaller than the AVE of each construct, the discriminant validity can be confirmed, as is shown in Table 4, the correlations between the factors of the variables in the new extended TPB model. The high correlations between AT, SN, and PBC show profound evidence of validity. Our objective was to discover whether the TPB model, in the context of an integrated framework for understanding consumers’ WTP and behavior [41] could also assess willingness towards pro-environmental behaviors that encompass “AT, SN, PBC”. These are the determinants for the WTP for IGGS for MSW management in two successive questionnaires; although a 5-point scale was used in the Likert scale data statistics, it is still reliable for the applicability of the model (Chi-Square = 262.8, GFI = 0.808, CFI = 0.834, NFI = 0.787, RMSEA = 0.094), and all structural coefficients were statistically persuasive (p < 0.01). According to the result, AT (β = 0.573, p < 0.01) and PBC (β = 0.692, p < 0.01) affecting respondents’ WTP. So, Hypothesis 1 and Hypothesis 2 can be accepted, and Hypothesis 3 was rejected.

Table 6. The scales’ discriminant validity.

Title 1	1	2	3	4
1. Attitude	0.77
2. Subjective norms	0.48 **	0.59
3. Perceived behavioral control	0.45 **	0.50 ***	0.74
4. Environment-concern	0.42 ***	0.42 **	0.42 ***	0.71

** p < 0.05, *** p < 0.01.

At the same time, the influence of SN on AT (β = 0.762, p < 0.01) and PBC (β = 0.800, p < 0.01) are confirmed, so Hypothesis 4 and Hypothesis 5 can be accepted. (Shown in Figure 2)

Figure 2. Raw TPB model for WTP. β represents standard regression weight.

The fit measure of the extended model can be accepted (Chi-Square = 243.8, GFI = 0.838, CFI = 0.834, NFI = 0.787, RMSEA = 0.094), and most of the structural coefficients are significant (p < 0.01).

When the facts of PBC and EC were compared, PBC had the most significant effect on WTP (β = 0.692, p < 0.01) followed by the EC (β = 0.594, p < 0.01). Hypothesis 3 and Hypothesis 9 were validated. The positive effect relationship between EC and AT (β = 0.382, p < 0.05), SN (β = 0.610, p < 0.01), and PBC (β = 0.341, p < 0.05) can be verified, thus allowing Hypothesis 6, Hypothesis 7, and Hypothesis 8 to be accepted.

通过之前的推测，我们假设了环境关注对 IGGS 支付意愿的间接影响。图 1显示 AT、SN 和 PBC 在 EC 和 WTP 之间起中介作用。然而，Likert 量表的 Sobel 检验的回归系数表明所有间接影响在本问卷中都不成立。因此，我们可以得出结论，该 IGSS 的 WTP 中的 AT (β = -0.31)、SN (β = -0.36) 和 PBC (β = -0.12) 没有间接影响。（如图3所示）

图 3. WTP 的扩展 TPB 模型。β 代表标准回归权重。

显然，如果北京居民认为他们有额外的资源来为城市生活垃圾管理做出贡献，他们会对环境行为做出更积极的反应。因此，有必要加强那些已经认为自己能够为解决这个问题做出贡献的人的积极信念，改变那些认为自己目前没有相应资源的人的消极信念。这也表明政府行政部门在激励人们或居民支付方面发挥着至关重要的作用。

This entry is adapted from the peer-reviewed paper 10.3390/su132413902