Acceptability toward Policy Mix: Impact of Low-Carbon Travel Intention, Fairness, and Effectiveness

Abstract

Policymakers are implementing a combination of various travel demand management (TDM) policies to effectively reduce carbon emissions. Widespread acceptance of these policy mixes by residents plays a pivotal role in ensuring their successful implementation. Drawing upon the value–belief–norm (VBN) theory and policy-specific beliefs (perceived freedom, perceived fairness, and perceived effectiveness), this study evaluates the impact of low-carbon travel psychology and policy characteristics on the acceptability of a policy mix formed by the combination of mandatory measures and guiding measures within TDM. The empirical analysis leverages survey data obtained from 839 participants and employs a structural equation model. The results indicate that low-carbon travel willingness, perceived fairness, and perceived effectiveness significantly influence acceptability. Consequently, fostering low-carbon travel willingness among residents is essential for enhancing overall policy acceptability. Moreover, policy mixes that encompass incentives for the development of new energy vehicles (NEVs) are generally considered the most acceptable among residents due to their non-restrictive impact on private car usage. Given this analysis, policymakers should consider designing policy mixes that balance acceptability and effectiveness, such as incorporating policies for the development of NEVs, to address traffic pollution.

1. Introduction

In 2021, China was the world’s largest emitter of carbon dioxide (CO₂), accounting for approximately 33% of global emissions [1]. Notably, the transport sector in China contributed 901 million tons of carbon dioxide emissions, accounting for 10% of the country’s total emissions [2]. In view of transportation’s significant impact on carbon emissions, China has placed particular emphasis on this area in its pursuit of achieving carbon neutrality [2,3]. As part of its efforts to mitigate traffic pollution, the Chinese government has introduced a series of travel demand management (TDM) policy instruments, including the promotion of new energy vehicles (NEVs), driving restrictions, and public transportation priority. However, the effective implementation of TDM policy hinges not only on governmental endeavors but also on the support and cooperation of residents, making their backing crucial for success.

Currently, policymakers are striving to integrate a variety of TDM policy instruments to achieve the goal of carbon neutrality in China. Numerous major cities have selectively adopted mandatory measures to regulate private car ownership and usage, including purchase restrictions, driving restrictions, and parking fees. Meanwhile, guiding measures have been introduced to improve alternative travel options [4], including improving public transport services, promoting NEVs, and upgrading transport infrastructure. These policies are bundled to affect public travel behavior [5]. Existing studies, which mainly focus on the effectiveness of a single TDM policy, have shown that mandatory measures could effectively alleviate traffic congestion and pollution in the short-term [6,7,8,9,10,11,12,13]. However, residents’ support rate of mandatory measures is low due to the restrictions imposed on travelers’ behavior [14]. Even if guiding measures are typically well accepted by residents, it is difficult to significantly alleviate traffic pressure in the short-term [15,16,17]. Considering the fact that each of these policies has its own limitations and obstacles [18,19], if mandatory measures and guiding measures are combined to form a TDM policy mix (short for policy mix), it would provide a measure to presumably solve issues with conflicting policy effectiveness and public support. However, it is challenging to analyze the potential synergistic effects of these measures in the policy mix due to the complex interactions between different measures within a policy mix [20]. Nevertheless, studying the effect of a policy mix can offer a forward-looking prediction for future transportation policies, including potential policy mix scenarios and the societal, economic, and environmental impacts these scenarios might bring. Therefore, it is a critical task to study a policy mix that deserves more attention from policymakers and researchers.

The complexity of a policy mix increases the difficulty in assessing its acceptability. It has been widely recognized that low public acceptance is an obstacle to policy implementation [21,22], especially when policies result in an increased cost of using cars [23]. The acceptability of mandatory measures among travelers is generally low, but it is still necessary to control the number of cars [21]. Travelers have a higher acceptance of guiding measures, which are expected to change the travel habits of travelers in the long-term. However, implementing guiding measures requires more money and time, which can make it challenging to improve traffic conditions in the short-term. To overcome the limitations of these two types of measures and enhance the acceptability of less popular measures, policy mixes become essential. However, assessing the acceptability of a policy mix is complicated because the acceptability varies depending on its specific combinations and intensities, which necessitates an analytical framework for measuring the effective paths and key factors that influence the acceptability of a policy mix. Therefore, by exploring concepts in environmental psychology and perceived policy features, our study develops a model for assessing the acceptability of policy mixes, which contributes to providing policymakers with diverse ideas and directions and informing the development of effective and acceptable policies aiming to reduce emissions and control the number of cars. In addition, empirical research utilizing real survey data is employed for analyzing the practical impact of the acceptability of policy mixes, which contributes to providing the government with a scientific basis for policy formulation and enhancing the policy’s feasibility.

Environmental psychology plays a significant role in influencing the acceptability of policy [9,23,24,25]. The Chinese government has recognized the importance of providing education and guidance on low-carbon travel to enhance residents’ willingness to adopt sustainable modes of transportation. However, Steg and Vlek [26] have pointed out that the willingness to engage in low-carbon travel may have varying impacts on the acceptance of policies aiming to reduce carbon emissions. Despite the importance of this topic, there are limited empirical studies analyzing the impact of low-carbon travel willingness on the acceptability of a policy mix [23]. To date, some studies have focused on the relationship between travel psychology and travel mode [10,27,28,29]. In fact, it is clear that the challenges facing low-carbon travel transitions are related to both low-carbon travel psychology and the acceptability of TDM policy. This study goes beyond merely understanding residents’ attitudes toward low-carbon travel; it delves deeper into how their low-carbon travel willingness affects their acceptance of a policy mix. For this purpose, this paper integrates low-carbon travel willingness, policy acceptability, and individual psychological factors. It specifically analyzes the impact of low-carbon travel willingness on the acceptability of policy mixes while recognizing that individual environmental psychology serves as a crucial intrinsic motivator for travelers to adhere to policy mixes. This approach provides a novel and comprehensive perspective for understanding and improving the residents’ attitudes toward policy mixes.

This paper is organized as follows. Section 2 provides a summary of previous relevant studies on the acceptability of policy mix. Section 3 presents the hypotheses and conceptual model used in this study. Section 4 describes the data collection process and provides an overview of the sample characteristics. Section 5 discusses the results of the structural equation model. Lastly, the conclusion section offers a comprehensive summary and interpretation of the findings.

2. Literature Review

2.1. Policy Mix

Policy mix has become more important recently because of the challenges associated with achieving multiple policy objectives with a single policy [30,31,32,33,34,35]. There are various definitions of policy mix which depend on the specific research problem, but they generally revolve around the policy process and specific goals [36,37,38]. In this context, the policy mix can be defined as the presence of several policies implemented within the same country or region, all aiming to achieve the same social objectives [35].

Previous studies have classified TDM policies into “push” and “pull” categories [24,39,40]. Specifically, push measures primarily involve restrictions on travelers’ behavior, directly limiting car usage. Meanwhile, pull measures entail providing new and attractive travel modes to reduce reliance on fossil-fuel-powered cars. Policy mixes that balance both push and pull measures have been demonstrated to effectively promote sustainable transportation choices [41,42]. Many researchers have successfully employed the push–pull measure combination to examine residents’ travel mode preferences [36,39] and the acceptability of policy mixes [23]. Therefore, the classification and integration of push and pull measures offer an effective approach to study a TDM policy mix.

2.2. Acceptability

The acceptability of a TDM policy is a significant factor influencing its implementation [43,44,45,46,47]. A TDM policy with a higher acceptability tends to be implemented more effectively and can be enforced for a longer time [14,48]. However, mandatory TDM policies may encounter strong resistance from travelers, making their implementation particularly challenging. For instance, in Edinburgh, UK, a road user charging scheme was proposed in 2005 for a referendum, but it was rejected by a majority of three to one, resulting in the abandonment of the scheme [14]. In this sense, it is clearly shown that low acceptability has become an obstacle to the implementation of a TDM policy.

The acceptability of a TDM policy is a crucial topic that has been extensively studied in the literature. The majority of research on the acceptability of TDM policies has focused on the impact of sociodemographic factors, including but not limited to age, gender, region, and income [42,49,50]. Additionally, the perception of the public regarding policy fairness, freedom, and effectiveness significantly influences their willingness to accept these policies [9,23,24]. For instance, road charging policies may be viewed as impinging on the fairness and freedom of residents, resulting in reduced acceptability and potentially hindering policy implementation [51]. Conversely, fair and effective policies promoting electric vehicles can encourage residents to adopt sustainable transportation options [52]. Furthermore, the impact of environmental protection psychology on policy acceptability has been widely discussed. Researchers have applied the value–belief–norm (VBN) theory to demonstrate that individuals with a higher environmental awareness are more likely to support TDM policies, thereby contributing to the reduction in pollution in the transportation system [9,24,25]. However, most of these studies analyze the acceptability of individual policies rather than policy mixes. Assessing the acceptability of a policy mix in practice poses a major challenge due to its complex internal structure [5,53]. Therefore, to make contributions to this field, this paper seeks to examine the acceptability of a TDM policy mix, based on actual TDM policies in China.

2.3. Low-Carbon Travel Willingness

Policy mixes that guide travelers toward low-carbon travel options can contribute to the goal of reducing pollution in the transport system [45]. Current research on low-carbon travel predominantly focuses on travel behavior and psychology, revealing that TDM policy, familiarity with transportation, and time perception may affect the choice of low-carbon travel modes [13,54,55]. Furthermore, the psychology of low-carbon travel encourages travelers to select ecofriendly modes of transportation [56]. Hence, both TDM policies and individual low-carbon travel psychology play pivotal roles in shaping low-carbon travel modes. Moreover, residents’ recognition of TDM policy can lead to a transformation of their environmental protection psychology into actual environmentally friendly behavior [28]. Aligning the policy goal of reducing pollution with residents’ low-carbon travel intentions is likely to enhance travelers’ motivation to comply with a policy, thereby adopting ecofriendly travel modes.

The willingness to engage in low-carbon travel is formed by the values of environmental protection [45]. This willingness serves as the driving force behind travelers’ support for TDM policies. Previous research has indicated that residents with a higher environmental awareness are more likely to support TDM policy, which in turn can facilitate the reduction in pollution in the traffic system [9,24,25]. However, the extent to which individual environmental protection psychology can still be stimulated in the context of complex policy mix has not been fully verified [23]. Therefore, this paper seeks to thoroughly investigate the impact of low-carbon travel willingness on the acceptability of policy mix.

3. Choice of Policy Mix

The Chinese government has made various efforts in TDM policy to address the issue of urban traffic pollution. We examine 10 TDM policies [47,57,58] that are widely adopted in traffic management through a collection of 50 big cities in China (We collected information on the implementation of TDM policies in 50 major cities in China from government official websites, online searches, and the literature. These 50 cities include: Beijing, Shanghai, Guangzhou, Shenzhen, Chengdu, Chongqing, Xi’an, Tianjin, Hangzhou, Jinan, Taiyuan, Changsha, Guiyang, Nanjing, Haikou, Nanning, Zhengzhou, Hefei, Fuzhou, Wuhan, Changchun, Dalian, Nanchang, Lanzhou, Xuzhou, Tangshan, Urumqi, Qingdao, Jilin, Kunming, Baotou, Fushun, Anshan, Qiqihar, Zibo, Kaifeng, Ningbo, Suzhou, Xining, Liaoyang, Shijiazhuang, Handan, Hohhot, Daqing, Wuxi, Zaozhuang, Anyang, Hengyang, Changzhou, and Xiamen.). The red column in Figure 1 represents the number of cities implementing push measures, and the blue column represents the number of cities implementing pull measures. Out of these ten TDM policies, we select the two most widely implemented pull measures, i.e., “public transportation priority” and “promotion of NEVs”, as well as the two most widely implemented push measures, i.e., “driving restriction” and “parking fees”. Finally, we propose the policy mix schemes of our study based on these four TDM policy contents.

Public transportation priority plays a pivotal role in promoting the coordinated development of the urban transportation system [57]. The effective implementation of public transportation priority requires the presence of incentives that support the development of public transportation, including the improvement of service quality [15]. Poor-quality services may discourage passengers from using public transportation, which can render the public transportation priority meaningless. Moreover, studies have demonstrated that a significant enhancement in public transportation services is essential to reduce and ultimately eradicate the significant strain on urban traffic resulting from the growth in car ownership [59]. Thus, we choose the improvement of public transport services as one of the pull measures studied.

The development of NEVs is an important means of solving air pollution [35,37]. To promote the development of NEVs, China has made numerous efforts in green licenses, purchase subsidies, and infrastructure investment [58]. From the experience of the user, the difficulty of electric charging has become an essential obstacle to buying NEVs [60]. Thus, we choose increasing NEV charging piles as the other pull measure studied.

Under the impact of COVID-19, residents perceived that taking public transport would increase the risk of infection [16]. As a result, residents have come to rely more on private cars for travel. Meanwhile, the Chinese government has implemented policies aiming to stimulate consumption in response to the economic downturn, including encouraging residents to purchase cars. For instance, in an effort to expedite the economic recovery in Shanghai, the local government increased the number of fossil fuel vehicle licenses by 40,000 in the second half of 2022, increasing the monthly average number of license plates by about 50% [61] (Shanghai has implemented an auction system for car license plates). As a consequence, the number of fossil-fuel-powered cars is expected to rise, which could lead to greater congestion and potential environmental issues. To quickly and effectively alleviate traffic pressure, several major Chinese cities, including Beijing, have already implemented driving restrictions [62,63]. The successful implementation of these policies serves as a valuable reference for Shanghai in formulating similar measures in comparable situations. Thus, we choose the implementation of driving restrictions as one of the push measures studied.

At present, excess cars and limited parking spaces have become prominent problems in urban transportation development [64]. To address these problems, cities have implemented parking fees in addition to creating more parking spaces [65]. The long-term implementation of parking fees can cut private car travel by 4–22% [66] and CO₂ emissions by 2–13% [67]. Parking fees that stimulate travelers to give up car travel is a feasible way to solve air pollution [35]. Thus, we choose an increase in parking price as the other push measure studied.

In conclusion, this paper chooses the improvement of public transport services, increase in NEV charging piles, implementation of driving restrictions, and increase in parking fees as the specific measures for study (Appendix A shows more details). To ensure a better balance between the effectiveness and acceptability of policy mixes, push measures and pull measures are combined to form the four policy mixes studied (

Table 1. TDM policy mixes.

Policy Mix	Push Measures		Pull Measures
	Implementing Driving Restrictions	Increasing Parking Fees	Improving Public Transport Services	Increasing NEV Charging Piles
P1	√		√
P2	√			√
P3		√	√
P4		√		√

).

4. Hypotheses and Conceptual Model

4.1. Value–Belief–Norm Theory and Acceptability of Policy Mix

The influence of environmental psychology on the acceptability of a policy mix can be measured through the value–belief–norm (VBN) theory. The VBN theory emphasizes that social agents should take personal moral responsibility for the natural environment in an appropriate way [68,69,70]. Ünal et al. [25] provided a more comprehensive explanation of each variable based on Stern’s VBN theoretical framework [70], as shown in Figure 2. The theory is composed of three main variables: values, beliefs, and personal norms. Values, such as pro-environment orientation, play a crucial role in raising residents’ awareness of the negative impact of their behavior on the environment [71,72]. This heightened awareness, in turn, activates a sense of moral obligation to engage in pro-environmental actions, such as embracing low-carbon travel. As a result, residents become more receptive to car use reduction policies [23,25].

In this paper, belief is described by consequence awareness. The VBN theory posits that values indirectly influence behavior through beliefs and personal norms [25]. Values serve as guiding principles in the initial stage, leading to the formation of beliefs [73]. For instance, the value of environmental protection can make residents aware of the negative impact of using cars on the environment, prompting them to take on corresponding responsibilities. Residents with a strong pro-environmental orientation tend to have a heightened awareness of the negative consequences associated with car usage [23]. Therefore, the first hypothesis is:

H1. 

Pro-environmental orientation positively impacts residents’ awareness of consequences.

Low-carbon travel willingness reflects the moral obligation to reduce car use based on personal norms. A sense of responsibility for the negative impact of car use on the environment can contribute to a willingness to support pro-environmental transport policies [74,75]. When residents realize the negative impact of traffic pollution due to car travel, their low-carbon travel willingness can be stimulated. The environmental protection belief of travelers has a positive effect on their behavioral intention regarding low-carbon travel [28]. Therefore, the second hypothesis is proposed as:

H2. 

Awareness of consequences positively impacts residents’ low-carbon travel willingness.

The VBN theory is valuable in reflecting the influence of individuals’ environmental psychology on the acceptability of a TDM policy. By measuring psychological states related to the environment, the VBN theory provides insights into the factors that influence policy acceptability [48]. Personal norms are shaped by personal values of protecting the environment and awareness of environmental issues in the VBN theory. Therefore, various factors in the VBN theory ultimately affect the acceptability of a TDM policy through the behavioral intentions of personal norms. Personal norms can promote low-carbon travel willingness, which can enable individuals to comply with and accept environmentally friendly policies [52]. Furthermore, the willingness to act can prompt users to accept and comply with the policy [9,24]. Therefore, the third hypothesis is:

H3. 

Low-carbon travel willingness positively influences residents’ acceptability of a policy mix.

4.2. Policy-Specific Beliefs and Acceptability of Policy Mix

The acceptability of a policy mix can be influenced by various policy characteristics, particularly policy-specific beliefs, such as perceived freedom, fairness, and effectiveness [24,45,46]. The process of formulating and implementing policies should carefully consider the potential infringement on travelers’ freedom, the fairness of the policy, and the policy’s effectiveness in addressing practical issues.

Perceived freedom and perceived fairness are related, where a greater violation of freedom leads to a perception of greater unfairness of the policy [24,76]. Jakobsson et al. [51] found that perceived fairness had a direct positive correlation with perceived freedom and an indirect association with the acceptability of road pricing. Therefore, the fourth hypothesis is:

H4. 

Perceived freedom positively influences residents’ perceived fairness.

Research on the acceptability of Swiss transport policies [23,24] and the promotion of electric vehicles [52] has demonstrated that the fairer the policy, the more effective it is [73]. Therefore, the fifth hypothesis is set up as:

H5. 

Perceived fairness positively influences residents’ perceived effectiveness.

Perceived freedom is measured by free choice in travel mode and travel time [24]. Enforcing mandatory measures can have a negative impact on public freedom. Studies on road pricing have demonstrated that opposition to a measure is stronger when the measure is perceived as mandatory [51]. Therefore, the sixth hypothesis is:

H6. 

Perceived freedom positively influences residents’ acceptability of a policy mix.

Perceived fairness is evaluated based on the degree of fairness of the policy to oneself and the public [24]. Ensuring fairness is a critical prerequisite for successful policy implementation [51]. Numerous studies have highlighted the importance of perceived fairness in promoting policy acceptability. When residents perceive a policy as fair, they are more likely to accept it, demonstrating the positive impact of fairness on policy acceptance [24,52,76]. Therefore, the seventh hypothesis is:

H7. 

Perceived fairness positively influences residents’ acceptability of a policy mix.

Perceived effectiveness refers to the extent to which residents believe that a policy has achieved its intended goals. Despite the inconvenience that the policy process may cause residents, a high perceived effectiveness can encourage them to tolerate the inconvenience [24]. A TDM policy aims to reduce traffic pollution and congestion. Therefore, the perceived environmental benefits of TDM policies can serve as a crucial motivator for travelers to tolerate the inconvenience of complying with such policies [9,21,53]. Therefore, the eighth hypothesis is:

H8. 

Perceived effectiveness positively influences residents’ acceptability of a policy mix.

Combining the VBN theory with policy-specific beliefs provides a comprehensive framework for studying the acceptability of a policy mix, and has the potential to inform transportation policy formulation and implementation [23,24]. Figure 3 illustrates the hypothesized formative model of the acceptability of a policy mix, along with the proposed relationships between pro-environmental orientation, awareness of consequences, low-carbon travel willingness, freedom, fairness, and effectiveness.

5. Research Methodology

Figure 4 illustrates the detailed process of data analysis in this study. Firstly, we focused on determining the study site, laying the foundation for subsequent research. Next, guided by the research questions, we designed a survey questionnaire covering four main aspects: descriptions of four TDM policy measures, sociodemographic information of the survey participants, the VBN theory, and policy-specific beliefs related to the four policy mixes. The third step involved the data collection, where we chose appropriate methods for data collection and conducted simple statistical analysis on the collected data. Finally, based on the obtained data, we conducted in-depth analysis of the acceptability. The following subsection provides further details on this process.

5.1. Study Site

The survey was conducted in Shanghai, one of the four municipalities in China that are directly controlled by the central government. As a predominant economic center, Shanghai faces challenges associated with a large population, diverse travel needs, and serious traffic congestion and pollution. To address these challenges, the city has implemented various TDM policies, including promoting public transportation, incentivizing the use of NEVs, and improving parking charge mechanisms. The average daily number of public transport passengers was 13,988 million in 2021 [77]. The cumulative promotion of NEVs in Shanghai exceeded 670,000 in 2022 [77]. These policies have contributed to the city’s recognition as a demonstration city for transit metropolis construction in China. Notably, the “White Paper on Shanghai Transportation Development” released in 2013 suggested that driving restrictions were considered based on road congestion and air pollution. The study of TDM policies in Shanghai can provide valuable insights for future transportation strategy and urban development decisions in China.

5.2. Questionnaire Design

The questionnaire (see Appendix B) was designed to gather data from residents in Shanghai on their opinions and beliefs regarding the policy mixes under investigation. The questionnaire is divided into four parts. Part one of the questionnaire describes the four TDM policies investigated (see Appendix A). Part two captures respondents’ sociodemographic information, such as income, education, and age. Part three comprises variables in the VBN theory. Part four comprises variables in the policy-specific beliefs of a policy mix. The variables in the questionnaire mainly refer to the past literature [9,23,24,25,28,47,52,59,78,79,80]. These variables are measured on a 5-point Likert scale ranging from “1 = strongly disagree” to “5 = strongly agree”. Before the extensive investigation, the research team discussed the survey scheme and questionnaire content several times for pretest. Then, we interviewed 20 scholars and colleagues and made further adjustments to the questionnaire based on their feedback and the existing literature [9,23,24,25]. We finalized the questionnaire content and began extensive study.

5.3. Data Collection

The online survey was conducted through the Credamo platform [81]. The platform has a sample base of more than three million and supports feature screening according to sample attributes, such as region, gender, and income, which are designed according to national statistical standards. The Credamo platform provides users with online random sampling. A total of 1180 questionnaires were collected through online surveys, out of which, 839 were considered valid, resulting in an effective response rate of 71.1%. The sociodemographic composition of the sample is presented in

Table 2. Summary of sample demography (n = 839).

Dimension	Category	Frequency (Percentage)	Dimension	Category	Frequency (Percentage)
Gender	Male	374 (44.6%)	Raise children	Yes	325 (38.7%)
	Female	465 (55.4%)		No	514 (61.3%)
Age	$\le 20$	32 (3.8%)	Income	<3000	77 (9.2%)
	21–30	301 (35.9%)		3000–5999	41 (4.9%)
	31–40	486 (57.9%)		6000–9999	414 (49.3%)
	$\ge 41$	20 (2.4%)		10,000–14,999	263 (31.3%)
Education	Junior/senior high school	36 (4.3%)		$\ge$15,000	44 (5.2%)
	Junior college	218 (26%)	Buying a car will choose	fossil-fuel-powered cars	89 (10.6%)
	College	514 (61.3%)		NEVs	750 (89.4%)
	Master or above	71 (8.5%)	Number of bus trips per week	0	161 (19.2%)
Marital status	Married	359 (42.8%)		1–3	331 (39.5%)
	Unmarried	480 (57.2%)		4–6	204 (24.3%)
				$\ge$7	143 (24.3%)

. The 465 females represent 55.4% of our sample. Most respondents (61.3%) had a college education. The survey reveals that 89.4% (n = 750) of respondents showed a willingness to buy NEVs. Furthermore, 39.5% (n = 331) of the respondents reported taking the bus between one and three times per week.

6. Analysis and Findings

This paper examines the acceptability of four policy mixes using the policy mix acceptability model (Figure 3). To evaluate the measurement model, the reliability and validity of the variables used in this study are assessed using Statistical Product and Service Solutions 26 (SPSS 26) [82]. Subsequently, to test the hypotheses, the structural model is assessed using AMOS 24 [83] software.

6.1. Measurement Model Assessment

Before estimating the structural model, it is necessary to assess the reliability and validity of the latent factors by analyzing the list of measurement items.

Table 3. Reliability and validity of the VBN theory.

Construct	Items	Factor Loading	Cronbach’s α	CR	AVE
The VBN theory	ENV	0.738	0.70	0.813	0.5919
		0.791
		0.778
	AC	0.739	0.629	0.7754	0.5352
		0.706
		0.749
	WT	0.774	0.675	0.8033	0.5767
		0.723
		0.780

ENV: Pro-environment orientation; AC: awareness of consequences; WT: low-carbon travel willingness.

shows the factor loadings, Cronbach’s α [84], composite reliability (CR), and average variance extracted (AVE) [85] of the variables in the VBN theory.

Table 4. Reliability and validity of the policy-specific beliefs.

Construct	Items	Factor Loading	Cronbach’s α	CR	AVE
P1	PE	0.820	0.875	0.8585	0.6691
		0.812
		0.822
	PFR	0.793	0.817	0.7674	0.6225
		0.785
	PFA	0.762	0.790	0.7616	0.6151
		0.806
	ACC	0.816	0.766	0.7965	0.6618
		0.811
P2	PE	0.827	0.871	0.8621	0.6757
		0.824
		0.815
	PFR	0.783	0.852	0.7475	0.5969
		0.762
	PFA	0.772	0.818	0.7742	0.6317
		0.817
	ACC	0.805	0.775	0.7787	0.6376
		0.792
P3	PE	0.809	0.860	0.8485	0.6515
		0.772
		0.839
	PFR	0.752	0.834	0.7108	0.5514
		0.733
	PFA	0.820	0.848	0.8082	0.6782
		0.827
	ACC	0.829	0.795	0.8071	0.6765
		0.816
P4	PE	0.835	0.891	0.8715	0.6933
		0.827
		0.828
	PFR	0.760	0.822	0.7121	0.553
		0.727
	PFA	0.829	0.865	0.8088	0.679
		0.813
	ACC	0.845	0.811	0.8136	0.6859
		0.812

PE: Perceived effectiveness; PFR: perceived freedom; PFA: perceived fairness; ACC: acceptability of the policy mix.

shows the factor loadings, Cronbach’s α, CR, and AVE of the variables in the policy-specific beliefs. Table 3 and Table 4 indicate that all the factor loadings exceed 0.7, demonstrating the presence of convergent validity for all the latent variables [15]. The internal consistency reliability of each latent variable is determined by evaluating Cronbach’s α and CR, both of which are found to be greater than 0.6 and 0.7, respectively, indicating good reliability. Convergent validity is assessed by examining AVE, which is greater than 0.5, indicating good validity [17]. Discriminant validity is also verified by Fornell and Larcker’s [86] criteria. The inter-structure correlation between variables is lower than the square root of the AVE of the respective variable, which confirms that the discriminant validity is good (refer to Appendix C).

6.2. Structural Model Assessment

The internal variance inflation factor (VIF) values of all the predictors of the acceptability of policy mixes are less than 5, as shown in

Table 5. Model assessment estimate.

Policy Mix	Items	VIF	R-Square
P1	Low-carbon travel willingness$\to$Acceptability	1.199	0.124
	$\mathrm{Effectiveness}\to$Acceptability	2.277	0.612
	$\mathrm{Freedom}\to$Acceptability	2.846	0.572
	$\mathrm{Fairness}\to$Acceptability	2.916	0.594
P2	Low-carbon travel willingness$\to$Acceptability	1.282	0.178
	$\mathrm{Effectiveness}\to$Acceptability	2.147	0.588
	$\mathrm{Freedom}\to$Acceptability	3.085	0.532
	$\mathrm{Fairness}\to$Acceptability	3.717	0.626
P3	Low-carbon travel willingness$\to$Acceptability	1.313	0.187
	$\mathrm{Effectiveness}\to$Acceptability	2.81	0.648
	$\mathrm{Freedom}\to$Acceptability	2.139	0.453
	$\mathrm{Fairness}\to$Acceptability	2.995	0.661
P4	$\mathrm{Low}-\mathrm{carbon}\mathrm{travel}\mathrm{willingness}\to$Acceptability	1.361	0.225
	$\mathrm{Effectiveness}\to$Acceptability	2.774	0.667
	$\mathrm{Freedom}\to$Acceptability	2.127	0.478
	$\mathrm{Fairness}\to$Acceptability	3.222	0.703

, indicating no multicollinearity between the latent variables in the structural model [49]. The R-squared values of the variables are used to evaluate the structural model. Higher R-squared values for the explanatory variable indicate a higher explanatory suitability for the acceptability of policy mixes, also shown in Table 5 [87].

6.3. Factors Influencing the Acceptability of Policy Mix

The fitting of a structural equation model (SEM) is commonly evaluated using a set of fitting indicators. These indicators only provide the average or overall fitting of the SEM. Therefore, different types of fitting indicators should be used for a complementary description. In this study, we selected the most reported fitting indicators in the literature to ensure the accuracy and precision of the SEM [16]. As shown in

Table 6. Indicators of model fitting of policy mixes.

Fitting Indicators	${\mathit{\chi }}^2$	GFI	RMESA	CFI	TLI	NFI
Standard Range	Smaller and Better	>0.9	<0.08	>0.9	>0.9	>0.9
Overall	748.542	0.915	0.077	0.952	0.941	0.943
P1	466.033	0.944	0.057	0.950	0.940	0.934
P2	505.242	0.940	0.060	0.947	0.936	0.931
P3	553.143	0.935	0.064	0.940	0.928	0.924
P4	631.321	0.927	0.069	0.935	0.921	0.920

, all the indicator values meet the criteria for a good fit, providing compelling evidence of a good fit between the established SEM and the actual survey data.

In this study, we use the theoretical model in Figure 3 to examine the acceptability of the overall policy mix, as shown in

Table 7. Path coefficients of overall policy mix.

Structural Path	Estimate	Std.Error	Decision
$\mathrm{H}1\mathrm{Pro}-\mathrm{env}\mathrm{orientation}\to$Awareness of consequences	0.515 ***	0.059	Supported
$\mathrm{H}2\mathrm{Awareness}\mathrm{of}\mathrm{consequence}\to$Low-carbon travel willingness	0.675 ***	0.082	Supported
$\mathrm{H}3\mathrm{Low}-\mathrm{carbon}\mathrm{travel}\mathrm{willingness}\to$Acceptability	0.046 *	0.022	Supported
$\mathrm{H}4\mathrm{Freedom}\to$Fairness	0.930 ***	0.022	Supported
$\mathrm{H}5\mathrm{Fairness}\to$Effectiveness	0.726 ***	0.018	Supported
$\mathrm{H}6\mathrm{Freedom}\to$Acceptability	−0.008	0.035	Rejected
$\mathrm{H}7\mathrm{Fairness}\to$Acceptability	0.582 ***	0.051	Supported
$\mathrm{H}8\mathrm{Effectiveness}\to$Acceptability	0.280 ***	0.036	Supported

Note. Significance levels indicated with * (*** p < 0.001; * p < 0.05).

and Figure 5. Based on the VBN theory, we find that pro-environment orientation has a significant and positive effect on the awareness of consequences ($\beta =0.515$). The awareness of consequences of travel behavior has a significant impact on low-carbon travel willingness ($\beta =0.678$). Thus, H1 and H2 are supported. Low-carbon travel willingness has a significant and positive impact on the acceptability of the overall policy mix. Thus, H3 is supported. However, the influence coefficient is relatively small ($\beta =0.046$), indicating that it is currently difficult to ensure the acceptability of a policy mix based solely on the low-carbon travel psychology of residents. Therefore, it is essential to strengthen long-term publicity and education on low-carbon travel to improve the acceptability of a policy mix.

In the policy-specific beliefs, perceived freedom significantly affects perceived fairness ($\beta =0.930$), and perceived fairness also significantly affects perceived effectiveness ($\beta =0.726$). These indicate that hypotheses H4 and H5 are supported. Interestingly, the impact of perceived freedom on the acceptability of the overall policy mix contradicts hypothesis H6, indicating that residents feel that a policy mix consisting of push and pull measures limits their freedom of travel. While a policy mix is accepted by residents, it is still viewed as restricting freedom of travel. However, the influence of perceived freedom on the acceptability of a policy mix is not significant. Thus, H6 is rejected. The perceived fairness and perceived effectiveness have a significant influence on the acceptability of the overall policy mix. Thus, H7 and H8 are supported. Specifically, the fairer the policy mix, the more easily it is accepted, and the higher its effectiveness, the more easily it is accepted. The influence coefficient of perceived fairness on the acceptability of a policy mix is 0.582. The influence coefficient of perceived effectiveness on the acceptability of a policy mix is 0.280. This suggests that residents attach a greater importance to perceived fairness than the perceived effectiveness of a policy mix. Therefore, ensuring public support for a policy mix requires paying attention to its fairness.

6.4. Comparison of Acceptability of Different Policy Mixes

We discuss the influence of low-carbon travel psychology and policy characteristics on the acceptability of four policy mixes of different intensities from two aspects of the VBN theory and policy-specific beliefs (

Table 8. Path coefficients of four policy mixes.

Structural Path	P1			P2			P3			P4
	Estimate	S.E.	Decision	Estimate	S.E.	Decision	Estimate	S.E.	Decision	Estimate	S.E.	Decision
$\mathrm{H}1\mathrm{ENV}\to$AC	0.505 ***	0.059	Supported	0.511 ***	0.059	Supported	0.510 ***	0.059	Supported	0.508 ***	0.059	Supported
$\mathrm{H}2\mathrm{AC}\to$WT	0.675 ***	0.082	Supported	0.673 ***	0.081	Supported	0.675 ***	0.082	Supported	0.678 ***	0.082	Supported
$\mathrm{H}3\mathrm{WT}\to$ACC	0.042	0.040	Rejected	0.163 ***	0.042	Supported	0.054	0.040	Rejected	0.084 *	0.038	Supported
$\mathrm{H}4\mathrm{PFR}\to$PFA	0.979 ***	0.042	Supported	0.885 ***	0.033	Supported	0.755 ***	0.033	Supported	0.825 ***	0.036	Supported
$\mathrm{H}5\mathrm{PFA}\to$PE	0.781 ***	0.035	Supported	0.698 ***	0.031	Supported	0.857 ***	0.032	Supported	0.762 ***	0.028	Supported
$\mathrm{H}6\mathrm{PFR}\to$ACC	−0.08	0.226	Rejected	−0.240	0.155	Rejected	−0.050	0.041	Rejected	−0.024	0.041	Rejected
$\mathrm{H}7\mathrm{PFA}\to$ACC	0.642 **	0.247	Supported	0.777 ***	0.184	Supported	0.595 ***	0.103	Supported	0.633 ***	0.076	Supported
$\mathrm{H}8\mathrm{PE}\to$ACC	0.347 ***	0.059	Supported	0.355 ***	0.052	Supported	0.296 ***	0.087	Supported	0.216 ***	0.062	Supported

Note. Significance levels indicated with * (*** p < 0.001; ** p < 0.01; * p < 0.05).

and Figure 6). Our findings reveal that the freedom of the policy-specific beliefs does not significantly affect the acceptability of the four policy mixes (p > 0.05). Thus, H6 is rejected. This is consistent with the conclusion drawn in Table 7.

Residents have the highest acceptability of P₂ in Figure 6b, and the lowest acceptability of P₃ in Figure 6c. P₂ involves implementing driving restrictions and increasing NEV charging piles. NEVs are often exempt from driving restrictions in Chinese cities. Increasing NEV charging piles can incentivize residents to choose NEVs for travel. NEVs effectively replace traditional fuel vehicles and reduce air pollution. P₂ is less restrictive than other policy mixes. Therefore, this policy mix is easily accepted by residents. Policy mix P₃, which increases car parking fees and improves public transport services, is difficult for residents to accept. Increasing parking fees raises the cost of using cars through pricing and improving public transportation services aims to encourage residents to use public transportation. However, residents are still reliant on private cars, and it may not be possible to persuade them to give up their cars in the short-term. Thus, this policy mix faces challenges in achieving its intended goals.

Low-carbon travel willingness does not have a significant impact on the acceptability of either P₁ (Figure 6a) or P₃ (Figure 6c). Thus, H3 for P₁ and P₃ is rejected. Both P₁ and P₃ include a measure to improve public transportation services, suggesting that it may be difficult to stimulate residents’ low-carbon travel willingness by improving public transport. In contrast, the acceptability of P₂ (Figure 6b) and P₄ (Figure 6d) is significantly influenced by three factors: low-carbon travel willingness, perceived fairness, and perceived effectiveness. Both P₂ and P₄ include measures to increase NEV charging piles. This indicates that NEV promotion measures are more effective and acceptable for mitigating air pollution. Axsen et al. [35] similarly concluded that policy mixes that incorporate the development of NEVs are an effective means of addressing air pollution. Hence, including measure to promote NEVs in policy mixes can enhance their acceptability among residents and achieve policy objectives. Policymakers can consider promoting the use of NEVs to achieve environmental protection goals.

In promoting low-carbon travel willingness, policy mix P₂, which includes the implementation of driving restrictions and the increase in NEV charging piles, is widely regarded as the most effective approach. NEVs are usually exempted from the driving restrictions, thereby enhancing residents’ travel flexibility. Additionally, growing environmental awareness lets more residents show a willingness toward environmental protection, and to use NEVs as a way to reduce carbon emissions. Moreover, 89.4% of the respondents indicated their willingness to purchase NEVs in Table 2, which suggests that residents have a positive attitude toward the adoption of NEVs. Increasing the number of charging piles can make the NEVs’ charging easier, reducing residents’ concern about using NEVs. The survey results indicate that even with the implementation of driving restrictions, the degree of restriction on travel with NEVs is relatively low.

Regarding the influence of perceived fairness and perceived effectiveness on the acceptability of policy mixes, P₁ and P₂ are more acceptable than P₃ and P₄. P₁ and P₂ both include driving restrictions, whereas P₃ and P₄ include increases in parking fees. Policy mixes containing an increase in parking fees are judged to be less fair and effective. This suggests that residents are more sensitive to pricing policies. Current driving restrictions typically have no substantial impact on residents who use NEVs or rely on public transportation. Survey data from Table 2 indicate that 89.4% of the respondents expressed a strong willingness to purchase NEVs, and approximately half of the respondents stated they take the bus more than four times a week. This suggests that for the majority of respondents, the effects of the driving restrictions are relatively minor. Therefore, a policy mix that includes measures to restrict driving is considered fair and effective.

7. Conclusions and Policy Implication

Addressing the negative externalities of transportation requires an effective policy mix [35,37]. The success of policy implementation hinges on the attitudes of residents, which are influenced by their environmental psychology. Therefore, this study provides a framework for testing the acceptability of a TDM policy mix based on a questionnaire in Shanghai. Specifically, this paper estimates the impact of low-carbon travel psychology and policy characteristics on the acceptability of four policy mixes of varying intensity: (1) implementing driving restrictions and improving public transport services, (2) implementing driving restrictions and increasing NEV charging piles, (3) increasing parking fees and improving public transport services, and (4) increasing parking fees and increasing NEV charging piles. To estimate the impact of low-carbon travel psychology and policy characteristics on policy acceptability, this study applies the VBN theory and policy-specific beliefs. The results show that the low-carbon travel willingness, perceived fairness, and perceived effectiveness of the policy mix are crucial factors affecting the policy mix’s acceptability. Furthermore, this study finds significant differences in the acceptability of policy mixes with varying intensity. Based on these findings, policymakers can estimate the acceptability of different policy mixes a priori, which could inform the development of effective policies to mitigate air pollution.

The VBN theory sheds light on the relationship between the willingness to engage in low-carbon travel and the acceptability of a policy mix. The results indicate that low-carbon travel willingness has a significant but small impact on the acceptability of certain policy mixes. When policy mixes include measures increasing NEV charging piles, environmental psychology significantly affects their acceptability. Therefore, businesses can consider providing convenient charging facilities in major transportation hubs and commercial areas within the city to encourage more people to purchase and use NEVs. Additionally, the government can incentivize residents to buy NEVs by offering purchase subsidies, tax exemptions, and free parking, alleviating their financial burden and encouraging more individuals to opt for environmentally friendly modes of transportation. However, when policy mixes include measures improving public transport services, low-carbon travel willingness does not affect the acceptability. This suggests that it is challenging to encourage travelers to shift from cars to public transport solely based on their environmental psychology. To tackle this challenge, it is essential to enhance residents’ awareness and education regarding low-carbon travel. This can be achieved through community education and media campaigns, introducing residents to the environmental benefits, health impacts, and their personal role in reducing carbon emissions through low-carbon travel. Furthermore, government and community organizations can collaborate to conduct environmental education activities, aiming to enhance residents’ environmental awareness. The findings imply that policymakers need to carefully design policy mixes to guide travelers toward low-carbon transportation effectively.

Regarding policy-specific beliefs, perceived fairness and perceived effectiveness significantly affect the acceptability of a policy mix, with perceived fairness having the most significant impact. Therefore, when formulating a policy mix, it is crucial to pay special attention to the fairness of mandatory measures within the policies to ensure the overall equity of the policy mix. For instance, regarding parking fees, differentiated charges based on vehicle emissions and usage periods can be considered, ensuring that high-emission vehicles and peak-hour users pay higher fees, thereby achieving fairness. Simultaneously, the government could contemplate implementing differentiated driving restrictions, restricting the movement of high-emission vehicles based on emission standards and tailpipe emissions, while exempting new energy vehicles and low-emission vehicles, encouraging more people to buy ecofriendly vehicles. Furthermore, the government should establish a regular policy evaluation process, closely monitoring the implementation effects of policies. Based on the evaluation results, policy measures should be adjusted promptly to ensure their continued effectiveness and meet the residents’ needs. This study also finds that residents’ perceived effectiveness and perceived fairness have a more positive impact on acceptability when policy mixes include measures increasing NEV charging piles. These policy mixes, which encourage the adoption of NEVs, strike a more moderate balance, allowing residents to continue using their private cars. Therefore, policymakers can consider introducing such moderate policy mixes to ensure their effectiveness and fairness.

This study, examining the acceptability of four policy mixes, revealed that the policy mix of implementing driving restrictions and increasing NEV charging piles is the most acceptable to residents. Under this policy mix, fossil-fuel-powered cars are restricted by the driving restrictions measure, but travelers who are willing to buy NEVs are not restricted from driving. Therefore, implementing a moderate policy mix, such as driving restrictions coupled with incentives for NEVs, appears to be a more practical solution for mitigating urban traffic pollution. The policy mix of increasing parking fees and improving public transport services is the least acceptable. This indicates that the price mechanism of traffic management is challenging for residents to accept. It also suggests that encouraging travelers to shift from cars to public transport might be difficult to achieve in the short-term. Therefore, as the government contemplates reforming the parking fee system, it should simultaneously optimize public transportation services to the fullest extent possible. Specifically, this could involve increasing the number and frequency of public buses to ensure the convenience and comfort of public transportation. Introducing intelligent traffic management systems that provide real-time public transit information can incentivize more people to use public transportation.

There are several areas for potential improvement in this study. Firstly, our focus is solely on the acceptability of the policy mixes. However, research on other aspects of policy mixes, such as cost–benefit analysis and policy efficiency, is also in its infancy. Secondly, although social psychology is a significant factor influencing the acceptability of the policy mix, sociodemographic factors also affect residents’ attitudes toward it. Future studies could consider incorporating sociodemographic characteristics to reveal differences in the acceptability of a policy mix among residents with diverse backgrounds. Lastly, older travelers are underrepresented in the online sample in this study. Future studies could expand the sample by including more offline surveys.