Urban Renewal and Transformation of Residents’ Pro-Environmental Behaviors: Evidence from the Renovation of Old Residential Areas in Chengdu, China

Abstract

Based on research data from the renovation of old residential areas in Chengdu, China, this study provides empirical evidence regarding the causal relationship between urban renewal and the transformation of residents’ pro-environmental behaviors. The research results show a significant enhancement in residents’ engagement in pro-environmental behaviors due to the renovation of old residential areas. Robustness tests based on replacement samples and two-stage difference–difference model also confirm the reliability of the conclusions. The preliminary mechanism test indicates that the mechanism of the impact of renovation of old residential areas on the transformation of residents’ pro-environmental behaviors can be attributed to two main factors. Firstly, improving environmental infrastructure, such as upgrading waste segregation facilities and installing intelligent resource recycling devices, provides residents with better conditions for engaging in pro-environmental behaviors. Secondly, an increase in residents’ environmental awareness has led to more voluntary participation in pro-environmental behaviors. This study proposes that decision makers acknowledge that urban renewal serves as both a means of improving urban infrastructure and as a hands-on environmental education for the entire population. Taking urban renewal as an opportunity to enhance citizens’ awareness of environmental protection will effectively promote the transformation of pro-environmental behaviors and support sustainable social development.

1. Introduction

Since the reform and opening up of China, the development model that emphasizes economic growth but not environmental protection has pushed the carrying capacity of China’s resources and environment to its limits [1,2]. Under the severe form of environmental protection, China must strike a balance between economic development and ecological environmental protection. Since the 18th National Congress of the Communist Party of China (CPC), the CPC Central Committee has elevated the construction of ecological civilization to a more prominent strategic position [3,4]. This shift in priority has been accompanied by a series of new decisions and deployments aimed at advancing the formation of a green and low-carbon lifestyle as a key element of ecological civilization construction. The promotion of such a lifestyle has emerged as an essential focus for the CPC in the contemporary era. The 19th National Congress of the CPC report in October 2017 emphasized the importance of advocating for a simple, moderate, green, and low-carbon lifestyle, while opposing extravagance, wastefulness, and irrational consumption. Subsequently, in October 2022, the CPC reiterated this stance in the report of the 20th National Congress, calling for the promotion of green and low-carbon modes of production and lifestyles. These consistent messages from the CPC leadership reflect a sustained commitment to fostering pro-environmental behaviors and practices among the populace. The CPC Central Committee’s top-level designs have effectively provided strategic guidance for the widespread adoption of a green and low-carbon lifestyle throughout China. By championing sustainable living practices and encouraging eco-friendly choices, the CPC aims to cultivate a more environmentally conscious citizenry and contribute to the broader goal of ecological civilization development [5].

The formation of a low-carbon lifestyle is a social transformation process, which not only requires government institutional design at the macro level, but ultimately needs to be realized in the practice of the public. China can attain a low-carbon lifestyle transformation only through the active participation of the entire society in pro-environmental behaviors. So, what factors will affect the public’s participation in pro-environmental behavior? Many scholars have given answers from different angles. Combing through related studies, they can be roughly categorized into the following aspects: first, socio-demographic characteristics, including gender, age, education level, income, social class status [6,7,8,9,10], etc.; second, situational factors, including policies and regulations, social norms, interpersonal relationships, and rewards and punishments [11,12,13,14], etc.; third, environmental protection awareness factors, including environmental concern, environmental knowledge, environmental emotions, environmental aesthetics [15,16,17,18], etc.

Urban renewal is an inevitable process that cities undergo at a certain stage of development. As cities expand, challenges like congestion, pollution, and high living costs increase in central areas. Urban renewal serves as a response and coping strategy to address the issues of urban sprawl and the deterioration of old urban areas [19]. As an eternal topic in the process of human urbanization, urban renewal has become an important driving mechanism for urban development, and its core concept has gradually evolved from focusing only on physical space to being a multidimensional process of urban protection, restoration, and reuse, covering various aspects such as environmental improvement, improvement of living conditions, improvement of urban functions, and sustainable economic development [20]. Urban renewal does not directly aim at changing residents’ pro-environmental behavioral patterns, and it is difficult to find a direct causal link between the two. In addition, few researchers have provided empirical evidence to support the idea that urban renewal is a causal factor in changing residents’ pro-environmental behaviors.

This study was conducted twice in the urban area of Chengdu City, encompassing 39 old residential areas in the main city area. It conducts empirical analysis based on the cross-sectional data from the first survey and utilizes the OLS model in the benchmark regression. As renovations in the old residential areas are promoted, more old residential areas have completed the renovation one after another. To enhance the robustness of the findings, the study incorporates data from both surveys and employs the two-period difference–difference model in the robustness test. This model is applied to mitigate the impact of individual unobservable heterogeneity and time-related unobservable factors, thereby addressing selection bias and enhancing the accuracy of assessing policy effects.

This study may offer innovative contributions in two areas: First, it goes beyond the prevailing emphasis of existing research, which primarily centers on the enhancement of urban hardware and function through urban renewal. It broadens the scope by examining how urban renewal can foster the adoption of a green lifestyle and promote pro-environmental behavior among residents. Such an expansion not only contributes to a deeper understanding of the determinants of pro-environmental behavior but also enriches the discourse on the relationship between urban renewal and sustainable green development. Secondly, the study delves into the impact of urban renewal on residents’ pro-environmental behaviors by examining the mechanism underlying this relationship. Through a preliminary validation, this study offers practical insights for policymakers to leverage the Urban Renewal Plan strategically, encouraging increased participation of Chinese residents in eco-friendly practices to foster sustainable economic and social development.

2. Research Background and Impact Mechanisms

2.1. Research Background

In November 2021, the General Office of the Ministry of Housing and Urban–Rural Development of China issued the Notice on the First Batch of Urban Renewal Pilot Work, setting out plans to implement Urban Renewal Pilot Work projects in 21 cities nationwide, including Chengdu, over a two-year period [21]. Concurrently, in February 2022, the Chengdu Municipal Government unveiled the “14th Five-Year Plan for the Renovation of Chengdu’s Old Urban Residential Areas”, outlining the comprehensive renovation of old courtyards constructed before the end of 2004 in Chengdu [22]. The ambitious plan aims to complete the renovation of these historic courtyards by the conclusion of 2025, with a focus on transforming approximately 2242 old courtyards, which are home to nearly 233,000 residents. This initiative not only provides a strategic roadmap for the refurbishment of old compounds in Chengdu, but also presents a detailed task breakdown, marking a significant milestone in the city’s urban renewal efforts.

Figure 1 displays the spatial distribution of the initial urban renewal pilot cities in China. Being one of the pioneering pilot cities for urban renewal in China, Chengdu was the first city to propose a specific implementation plan. This plan not only symbolizes the rapid progress of old neighborhood transformation in Chengdu but also sets a benchmark for other cities to emulate. Therefore, Chengdu’s method of renovating old residential areas provides valuable insights for promoting urban renewal nationwide in China.

The renovation of old residential areas in Chengdu has been significantly accelerated with the implementation of the urban renewal pilot program, marking Chengdu as one of the pioneering cities in China. The key objectives of this program encompass a range of crucial tasks. Firstly, safety enhancement is prioritized, focusing on aspects such as power supply, drainage, fire protection, gas, and housing structure. This is aimed at identifying and eliminating safety hazards to fortify the infrastructure and ensure the safety and resilience of residents in these old residential areas. Secondly, efforts are directed towards bolstering the community service functions, which involve installing elevators, establishing convenient commercial service facilities, and enhancing property management services to elevate the overall public service capacity of the community. Lastly, there is a concerted emphasis on optimizing the living environment of the community by introducing facilities for garbage classification, renovating and beautifying exterior walls, enhancing community greening initiatives, and creating greenway spaces for recreational activities. These measures collectively enhance the living conditions in the old residential areas, fostering a heightened sense of satisfaction and well-being among residents. Consequently, the renovation of old residential areas is not only a beloved endeavor among residents but also a project that significantly benefits the whole community.

2.2. Influence Mechanisms

2.2.1. Enhancing Environmental Infrastructure

Pro-environmental behaviors, which encompass various terms such as environmentally responsible behaviors, environmentally friendly behaviors, and green behaviors, are actions aimed at minimizing negative impacts on the environment or engaging in behaviors that contribute positively to environmental conservation [23]. It involves the deliberate choices made by individuals, which are influenced by both self-imposed constraints and societal norms. Despite individuals’ intentions to engage in pro-environmental behaviors, the lack of external conditions may hinder the transformation of these intentions into tangible behaviors [24]. Some empirical studies have shown that improving environmental infrastructure significantly influences residents’ pro-environmental behaviors, such as water conservation [25], and resource recycling [26].

What drives the increase in residents’ pro-environmental behaviors through the renovation of old residential areas? In the implementation plan for the renovation of old residential areas in Chengdu, upgrading the infrastructure facilities’ shortcomings in old residential areas stands out as a crucial task. The continuous enhancement of infrastructure facilities and the improvement of the hardware environment are the most significant impacts of the old residential areas’ renovation project. For example, during the renovation of old residential areas, enhancing and modernizing garbage-classification facilities has enabled residents to participate effectively in proper waste segregation, overcoming the long-standing barrier of inadequate and deteriorated facilities. Another example is that some old residential areas have installed numerous clothing and paper recycling bins to promote the recycling of old items among residents. Moreover, some old residential areas are implementing smart meters for residents to effectively monitor electricity usage and promote energy conservation.

The renovation of old residential areas not only involves upgrading the shortcomings of basic environmental facilities but also promotes the improvement of environmental governance mechanisms and enhances the governance capacity of grass-roots organizations. It encompasses the long-term maintenance of environmental infrastructure to ensure that residents have the necessary hardware to engage in pro-environmental behavior for an extended period.

2.2.2. Raising Residents’ Awareness of Environmental Protection

Environmental protection awareness plays a crucial role in shaping civic values and qualities by influencing individuals’ recognition of environmental issues as well as their support for solving these problems [27]. It is the outcome of extensive processing of environmental information and intricate psychological processes among citizens. This encompasses individuals’ subjective cognitive perceptions and behavioral intentions related to environmental concern, awareness of environmental responsibility, and environmental knowledge. Environmental protection awareness emerges as a decisive factor affecting individuals’ engagement in pro-environmental behaviors [28,29]

How does renovating old residential areas increase residents’ environmental protection awareness? Improving the human environment of old residential areas through the renovation and beautification of facades, enhancing the quality and efficiency of community greening, and adding walking greenways and open spaces is a key priority in the transformation of these areas. When residents experience firsthand the enhancement of the quality of their living environment within the community, they develop a deeper appreciation for the hard-earned improvements, further strengthening their emotional connection and preference for a more sustainable ecological setting, and thereby stimulating environmental protection awareness. Consequently, this impact can lead to a broader influence on residents, encouraging them to engage in various forms of pro-environmental behaviors.

In addition, the renovation of old residential areas motivates local environmental management efforts at the community level. To promote sustained improvement in the quality of the community environment, grass-roots communities are encouraging public participation in community environmental management and long-term maintenance. For instance, certain community residents engage in public service activities, such as environmental cleanups, garbage sorting supervision, and other related tasks. These efforts help to enhance the residents’ awareness and commitment to environmental protection. Moreover, the revitalization of old residential areas serves as a form of community-wide civic environmental education. To advance continual enhancement of the community environment, various promotional activities such as door-to-door outreach, owner communication groups, and public campaigns are organized within the community. These initiatives aim to heighten public environmental awareness, underscore the significance of environmental protection, and promote a sustained improvement in residents’ environmental consciousnesses. Such efforts significantly contribute to augmenting residents’ involvements in pro-environmental behaviors.

3. Description of Data Sources and Variables

3.1. Data Sources Description

The empirical data for this paper is derived from the research data of the National Social Science Foundation of China (NSSFC)’s funded project, titled “Study on the Impact of Lack of Ecological Responsibility on Citizens’ Practice of Green and Low-Carbon Lifestyles and Policy Intervention” [30]. This project’s research concentrates on tracking the evolution of Chinese residents’ eco-responsibility awareness and green practices amidst economic and social changes. The primary objective is to present evidence-based recommendations to policymakers for enhancing green lifestyle initiatives. Concerning this, fieldwork was primarily conducted in Chengdu, where the project team gathered a substantial number of questionnaires for rigorous empirical scrutiny. Specifically, focusing on the revitalization of old residential areas constructed before 2004, the team randomly engaged six–eight residents from each district of Chengdu’s core urban areas (Chenghua District, Qingyang District, Wuhou District, Jinjiang District, Jinniu District) to participate in survey work during October 2022. A total of 1025 validated questionnaires formed the foundational dataset for the empirical analysis in this paper. Subsequently, by the time the first round of questionnaires was collected, 14 old residential areas in Chengdu had completed the renovation process, encompassing 329 respondents. Concurrently, 25 old residential areas were either in progress or had not yet initiated the renovation, containing 696 respondents.

Figure 2 illustrates the regional distribution of renovating old residential area in Chengdu City. The renovation of old residential area in Chengdu City predominantly focuses on the central urban areas, which were developed earlier and are the most concentrated areas of old residential area in the city.

In July 2023, as the renovation of Chengdu’s old residential areas progressed; nearly half of the 25 planned renovation projects were successfully completed. This provides valuable material for observing the evolution of residents’ pro-environmental behaviors during the urban regeneration process. A questionnaire return survey was conducted by the project team to gather insights from first-time residents residing in the residential areas that had not yet undergone renovation. After excluding individuals who declined participation, a valid sample of 488 respondents was obtained during the second callback. Moving forward, the study will subject the data collected from the questionnaire to a robustness test to validate the benchmark regression results.

3.2. Key Variables

3.2.1. Pro-Environmental Behavior Participation Index

Scholars have adopted various approaches to define and measure pro-environmental behaviors, including energy-saving and emission-reduction behaviors [31,32], green consumption [33,34], reduction in wastefulness [35,36], and pro-environmental behaviors in the public sphere [37,38]. To comprehensively measure residents’ participation in pro-environmental behaviors, this study established a scale encompassing 10 categories of daily pro-environmental behaviors.

Table 1. Statistics of residents’ pro-environmental behaviors.

Pro-Environmental Behaviors	Never	Occasionally	Sometimes	Often	Always
Refuse to Use Disposable Products	16.60%	18.07%	35.69%	21.99%	7.65%
Garbage Classification	32.43%	26.29%	12.36%	19.45%	9.47%
Turn off Unused Water and Electricity	0	9.43%	12.43%	29.57%	48.57%
Recycle and Exchange Used Clothes	33.58%	26.01%	16.17%	16.25%	7.99%
Bring Your Own Shopping Bag	13.34%	22.31%	31.19%	18.54%	14.62%
Use Public Transportation	0.64%	8.31%	16.57%	34.83%	39.65%
Appropriate Air Conditioning Temperature Setting	7.07%	17.77%	12.33%	38.17%	24.66%
Purchase Green Products	11.56%	30.64%	39.52%	9.83%	8.45%
Participate in Environmental Volunteering	40.44%	32.38%	9.08%	11.21%	6.89%
Reduce Leftovers and Waste	0.37%	10.6%	17.24%	39.56%	32.23%

illustrates the respondents’ engagement in these 10 categories, allowing them to select their frequency of participation (never, occasionally, sometimes, often, and always). Each option was assigned a value of 1–5 based on the frequency of participation, with higher values indicating greater involvement. In this study, the respondents’ pro-environmental behavior participation index was derived by converting the ten values into a composite score through the equal-weighted weighting method. Subsequently, the deviation was standardized to ensure that all respondents’ composite scores were within the range of [0, 1], which were then transformed into percentage scores.

The respondents’ participation in 10 categories of pro-environmental behaviors is summarized in Table 1. Variations in the level of engagement across different behaviors are evident among the interviewees. Notably, the levels of engagement in certain behaviors, such as refusing to use disposable products, classifying garbage, recycling and exchanging used clothes, purchasing green products, and participating in environmental volunteering, are relatively low. Specifically, the proportion of respondents who “often” or “always” refuse to use disposable products is 29.64%, while the percentage of respondents who “often” or “always” engage in garbage classification is 28.92%. Moreover, 24.24% of respondents “often” or “always” refuse to use disposable products. Additionally, there is a relatively low percentage of respondents who “often” or “always” purchase green products (18.28%), and those who “often” or “always” participate in environmental volunteering (17.9%).

To visually depict differences in residents’ levels of participation in various pro-environmental behaviors, respondents were categorized into two groups based on their frequency of participating in pro-environmental behaviors: those who answered “often” and “always” were classified as the “high-frequency” group, while those who selected other options were placed in the “low-frequency” group. The results in Figure 3 reveal that the level of participation in pro-environmental behaviors is generally low. Specifically, for the six pro-environmental behaviors examined, over 60% of respondents fell into the “low-frequency” group. Notably, the low-frequency group comprised more than 80% of respondents for “purchasing green products” and “participate in environmental volunteering”.

3.2.2. Renovation of Old Residential Areas

In this study, the core explanatory variable is a dummy variable, indicating whether the respondent’s residential area has completed the renovation. Participants residing in the old residential areas that have completed the renovation are placed in the experimental group and assigned a value of 1. Conversely, those living in the old residential areas where the old renovation has not been completed are placed in the control group and assigned a value of 0.

3.3. Control Variables

3.3.1. Characteristic Variables of Old Residential Areas

For this study, the impact of residential characteristics on respondents’ engagements in pro-environmental behaviors was examined by selecting two indicators of residential quality. The indicators chosen were the number of years since the old residential areas were established and the standard of property fees imposed in the old residential areas. These indicators were crucial in assessing the living environment of the old residential areas and their influence on individuals’ pro-environmental behaviors.

3.3.2. Socio-Demographic Variables

In this study, demographic characteristics such as gender, age, and marital status are examined to determine their impact on residents’ pro-environmental behaviors. The observation group consists of female and married respondents who are assigned a value of 1, while the remaining groups form the control group with a value of 0. Moreover, control variables such as years of education, income, social class identity, political profile, and type of residence are introduced in this study. The years of education correspond to the highest level attained, with assigned values as follows: illiteracy (0), elementary school (6), junior high school (9), senior high school (12), university (16), master’s degree (19), and doctor’s degree (22). Social class identity is subjectively evaluated through specific questionnaire items, assigning a value of 1 to 10 based on respondents’ self-perception of their social standing, with higher values indicating a higher perceived social class. The political profile variable distinguishes between CPC members and non-CPC members, with CPC members serving as the observation group and non-CPC members as the reference group. Additionally, the type of residence is categorized as own property or others’ property, with own property designated as the observation group and others’ property as the reference group.

3.3.3. Variables of Environmental Protection Awareness

To control the impact of individual environmental protection awareness factors on respondents’ engagement in pro-environmental behaviors, this study incorporates various dimensions of environmental protection awareness, including environmental knowledge, environmental concern, and environmental responsibility. The study developed an environmental protection knowledge scale comprising 10 items, and respondents were scored 1 for correct answers and 0 for incorrect or unknown responses. The scores for the 10 questions were aggregated and standardized using deviation standardization, resulting in values within the range [0, 1]. Subsequently, these standardized scores were converted into percentage scores to establish the personal environmental knowledge index. In terms of assessing environmental concern, participants were asked about their level of concern regarding environmental issues, each assigned a value from 1 to 5 based on the level of agreement expressed (not at all concerned; not too concerned; somewhat concerned; quite concerned; extremely concerned). Furthermore, respondents were asked about their beliefs regarding citizen responsibility for environmental protection. Responses were rated on a scale from 1 to 5, with values indicating different levels of perceived responsibility towards environmental protection (not at all; a little; moderately; quite a lot; very much).

3.4. Descriptive Statistics of Variables and Tests of Balance

The differences in descriptive characteristics between experimental group and control group are presented in

Table 2. Descriptive statistics of variables.

Variables	Full Sample		Experimental Group		Control Group		Difference
	N	Mean	N	Mean	N	Mean	Mean	T-Value
Pro-environmental Behavior Participation Index	1025	32.976	329	36.562	696	31.281	5.281 ***	2.732
Age of the Old residential areas	1025	29.182	329	28.537	696	29.487	−0.95	−1.053
Property Fee (yuan/m2)	1025	0.445	329	0.504	696	0.417	0.087	0.875
Age	1025	43.679	329	43.908	696	43.571	0.337	−0.227
Female	1025	0.466	329	0.497	696	0.452	0.045	1.390
Married	1025	0.754	329	0.765	696	0.749	0.016	0.437
Years of Education	1025	11.195	329	11.479	696	11.061	0.418 ***	4.087
Self-owned House	1025	0.608	329	0.586	696	0.619	−0.033 *	−1.955
Income (in CNY 10,000)	1025	6.796	329	7.347	696	6.535	0.812 ***	2.903
CPC Member	1025	0.077	329	0.082	696	0.075	0.007	1.329
Social Class Identity	1025	4.439	329	4.507	696	4.407	0.101 ***	3.695
Environmental Knowledge	1025	40.507	329	41.386	696	40.092	1.294	1.133
Environmental Concern	1025	3.225	329	3.476	696	3.106	0.037 ***	3.443
Environmental Liability	1025	3.376	329	3.544	696	3.296	0.248 ***	4.916

Notes: * and *** denote significance at the 10 and 1 percent levels, respectively (two-tailed).

. In terms of residential characteristics, there is no significant difference between the experimental group and the control group in terms of both the years in which the old residential areas were built and property fees. Residents in the experimental group were significantly better than the control group in terms of years of education, income, and subjective social class identity. Additionally, the probability of owning their own home was significantly lower than that of the control group. Furthermore, residents in the experimental group displayed higher levels of environmental concern and environmental responsibility compared to the control group. The differences in 6 out of 13 control variables indicate a potential impact of sample unbalance on the regression results. To address this issue, the study employed outlier removal and difference–difference model in a robustness test in the subsequent section to re-estimate the parameters and enhance the reliability of the baseline regression findings.

4. Empirical Analysis

4.1. Benchmark Regression Methods

The study conducted a benchmark regression analysis using the OLS model to examine the impact of the renovation of old residential areas on residents’ participation in pro-environmental behaviors. The dependent variable was the pro-environmental behavior participation index, with the key explanatory variable being a dummy variable representing whether the old residential areas had completed renovation.

Table 3. Benchmark regression results.

	(1)	(2)	(3)	(4)
Experimental Group	3.1742 ***	3.0113 ***	2.5416 ***	2.2182 ***
	(0.6611)	(0.5134)	(0.4924)	(0.4167)
Age of the Old Residential Areas		−0.6092 ***	−0.3187 **	−0.2027 **
		(0.2077)	(0.1476)	(0.1012)
Property fee (yuan/m2)		4.2186 ***	3.6877 ***	2.7167 **
		(1.2337)	(1.2256)	(1.3229)
Age			−0.1672 ***	−0.1284 ***
			(0.0406)	(0.0319)
Female			0.6294 **	0.5651 *
			(0.3172)	(0.0301)
Married			−0.5767	−0.3729
			(0.7251)	(0.4187)
Years of Education			0.5899 ***	0.4022 ***
			(0.1077)	(0.1023)
Self-owned House			0.9212 ***	0.6071 ***
			(0.2492)	(0.2162)
Income (in millions of dollars)			0.6421 ***	0.3728 ***
			(0.2145)	(0.1252)
CPC Member			4.4199 ***	3.7229 ***
			(1.2267)	(1.1954)
Social Class Identity			0.3207 **	0.2667 **
			(0.1447)	(0.1225)
Environmental Knowledge				0.3157 ***
				(0.0915)
Environmental Concern				0.7109 ***
				(0.1654)
Environmental Liability				0.6422 ***
				(0.1927)
Obs.	1025	1025	1025	1025
R2	0.0481	0.0871	0.1974	0.2215

Notes: Standard errors are presented in parentheses. *, **, and *** denote significance at the 10, 5, and 1 percent levels, respectively (two-tailed).

presents the benchmark regression results. The model in column (1) solely evaluates the effect of the renovation of old residential areas on residents’ pro-environmental behavioral participation. The model in column (2) includes controls for residential characteristics, while the model in column (3) introduces the controls for socio-demographic characteristics. Finally, the model in column (4) incorporates the individual awareness of environmental protection variable. Across all the models, the results consistently indicate that the renovation of old residential areas significantly enhances residents’ engagement in pro-environmental behaviors. For instance, in column (4), the data reveal a substantial increase of 2.2182 points in residents’ pro-environmental behavior participation scores following the renovation of old residential areas, which represents an average improvement of approximately 6.73% when compared to the sample mean of 32.967 in Table 2.

In the benchmark regression results, it was also found that various factors influenced residents’ participation in pro-environmental behaviors. Firstly, residents living in neighborhoods with older ages and lower property fees exhibited lower levels of participation in pro-environmental behaviors. This aligns with the study’s anticipated outcome, indicating that residents residing in neighborhoods with superior habitat quality generally demonstrate higher engagement in pro-environmental practices. Moreover, socio-demographic variables also played a significant role in determining residents’ engagements in pro-environmental activities. Specifically, younger individuals and females demonstrated higher levels of participation in pro-environmental behaviors [39]. Additionally, higher education levels, income levels, ownership of housing, Communist party membership, and social class identity were associated with increased participation in pro-environmental actions. Contrarily, marital status did not have a significant impact on residents’ participation in pro-environmental behaviors. Lastly, environmental protection awareness factors, such as individual environmental knowledge, environmental concern, environmental responsibility, were identified as crucial drivers that significantly enhance residents’ involvements in pro-environmental activities.

This study provides insight into the relative importance of various variables on pro-environmental behavior participation by presenting the standard regression coefficients in

Table 4. Standardized regression coefficients results.

	Unstandardized Coefficients		Standardized Coefficients	T-Value
	B	Std. Err	Beta
Experimental Group	2.2182 ***	0.4167	0.0302 ***	5.3233
Age of the Old Residential Areas	−0.2027 **	0.1012	−0.0844 **	−2.0030
Property fee (yuan/m2)	2.7167 **	1.3229	0.0366 **	2.0536
Age	−0.1284 ***	0.0319	−0.0917 ***	−4.0251
Female	0.5651 *	0.0301	0.0122 *	1.8830
Married	−0.3729	0.4187	−0.0121	−0.8906
Years of Education	0.4022 ***	0.1023	0.1167 ***	3.9316
Self-owned House	0.6071 ***	0.2162	0.0200 ***	2.8080
Income (in millions of dollars)	0.3728 ***	0.1252	0.1085 ***	2.9776
CPC Member	3.7229 ***	1.1954	0.0645 ***	3.1144
Social Class Identity	0.2667 **	0.1225	0.0311 **	2.1771
Environmental Knowledge	0.3157 ***	0.0915	0.1542 ***	3.4503
Environmental Concern	0.7109 ***	0.1654	0.1209 ***	4.2981
Environmental Liability	0.6422 ***	0.1927	0.1033 ***	5.3233
Obs.	1025
R2	0.3015

Notes: Standardized regression coefficients are calculated by multiplying the unstandardized regression coefficient by the standard deviation of the independent variable and dividing it by the standard deviation of the dependent variable. *, **, and *** denote significance at the 10, 5, and 1 percent levels, respectively (two-tailed).

. When analyzing the coefficients, it becomes evident that environmental knowledge, environmental concern, education level, income level, and awareness of environmental responsibility play crucial roles, with their importance decreasing in that order. The influence of renovating old residential areas on residents’ pro-environmental behavior falls slightly below that of social class identity but surpasses the effect of gender on participation in pro-environmental behavior. Further details can be accessed in Table 4.

4.2. Robustness Tests

4.2.1. Excluding Abnormal Samples

Table 2 reveals significant differences between the experimental group and the control group across 6 out of the 13 control variables. This study suggests that the basis of pro-environmental behavioral participation may differ between experimental group and control group due to sample imbalance. The unbalanced sample in this study may be attributed to the influence of the policy’s screening effect on the selection of targets for renovating old residential areas. This effect led to the prioritization of neighborhoods with better living environments and less renovation difficulties for inclusion in the old residential areas’ renovation projects. Consequently, residents in these residential areas typically possess a stronger foundation of environmental behavioral participation. As a result, the sample imbalance may lead to an incorrect attribution of the higher levels of residents’ environmental behavioral participation to the renovation of old residential areas.

To test the robustness of the previous regression conclusions, this study (

Table 5. Robustness test based on excluding abnormal community interference.

	(1)	(2)	(3)	(4)
Experimental Group	1.9168 ***	1.9569 ***	1.8918 ***	1.7706 **
	(0.6344)	(0.6054)	(0.5019)	(0.7167)
Residential Characteristics	YES	YES	YES	YES
Socio-Demographic Characteristics	YES	YES	YES	YES
Environmental Protection Awareness	YES	YES	YES	YES
Obs.	912	983	947	834
R2	0.2281	0.2107	0.2015	0.2169

** and *** denote significance at the 5 and 1 percent levels, respectively (two-tailed).

) conducted sequential reruns of the regression by excluding respondents residing in abnormal old residential areas. Specifically, the model in column (1) excludes respondents living in old residential areas with high-quality school district resources; this is followed by column (2), which excludes respondents living in old residential areas constructed through pooled funds from state-owned units. Subsequently, column (3) excludes respondents residing in the Chengdu City Garbage Classification Demonstration Neighborhoods; finally, column (4) excludes respondents living in neighborhoods falling in the aforementioned categories.

After re-testing the samples with the abnormal old residential areas removed, the study discovered that the coefficient representing the influence of renovating old residential areas on residents’ participation in pro-environmental behaviors exhibited a slight decrease. This suggests that the exclusion of samples from these abnormal neighborhoods did not compromise the robustness of the benchmark regression findings. In essence, the findings reinforced the conclusion that the renovation of residential areas significantly enhances residents’ engagements in pro-environmental behaviors.

4.2.2. Re-Testing Based on Difference–Difference Model

In order to further mitigate sample selection bias from policy screening factors, this study conducted a robustness test using a difference–difference model. In July 2023, residents who resided in old residential areas that had not yet undergone renovation during the initial visit were subsequently interviewed via a web-based questionnaire. The second survey, conducted as part of the follow-up, focused on collecting information specifically pertaining to the residents’ degrees of engagement in pro-environmental practices. Notably, the initial visit had already established the foundational socio-statistical variables, rendering the subsequent survey exclusive to gauging the residents’ environmental behaviors. The study successfully secured a sample of 488 returning respondents during the second visit, providing a robust pool of data for analysis. By leveraging the data collected from both survey instances, the research adopted a standard two-period difference–differences model to evaluate the impact of renovating old residential areas on residents’ participation in pro-environmental behaviors. This analytical approach offers the advantage of accounting for individual unobservant differences and mitigating the influence of time-varying factors, thereby enabling a more precise assessment of the policy effects. Furthermore, employing this model helps diminish potential issues related to selection bias and endogeneity, ultimately enhancing the accuracy of the policy evaluation.

Before conducting the robustness test, this study compared changes in pro-environmental behavioral involvement between the experimental and control groups. The experimental group consisted of residents from areas that had completed renovation at the time of the second visit, while the control group consisted of residents from areas that had not yet completed renovation by the second visit. During the second interview, both groups showed an increase in pro-environmental behavior participation, as demonstrated in Figure 4. Notably, the experimental group exhibited a significantly higher increase in pro-environmental behavior compared to the control group. This initial finding suggests that the renovation of old residential areas potentially contributes to enhancing residents’ pro-environmental attitudes and actions.

The study introduces a time dummy variable, which is assigned a value of 1 if the result of the second visit and 0 if the result of the first visit.

Table 6. Robust test based on two-period difference–difference model.

	(1)	(2)	(3)	(4)
Experimental Group × Time	2.5824 ***	2.3116 ***	2.1239 ***	1.6682 **
	(0.7245)	(0.7019)	(0.6624)	(0.6149)
Experimental Group	0.3976 **	0.3278 **	0.1767	0.1415
	(0.1767)	(0.1502)	(0.1661)	(0.1709)
Time	0.6744 ***	0.5482 ***	0.3852 **	0.2809 **
	(0.1458)	(0.1896)	(0.1769)	(0.1364)
Residential Characteristics	NO	YES	YES	YES
Socio-demographic Characteristics	NO	NO	YES	YES
Environmental Protection Awareness	NO	NO	NO	YES
Obs.	976	976	976	976
R2	0.0719	0.1049	0.2783	0.3583

** and *** denote significance at the 5 and 1 percent levels, respectively (two-tailed).

shows the results of the robustness tests obtained based on the two-period difference–differences model. The result of interest for the study is the coefficient of the interaction term between the core explanatory variables and the time variable, which reflects the policy effect of the renovation of old residential areas in terms of its impact on the pro-environmental behaviors of the residents. The results show that this coefficient is still significantly positive, and the robustness test using two-period difference–differences model does not deviate from the conclusion obtained from the benchmark regression that renovating old residential areas significantly boosts residents’ participation in pro-environmental behaviors.

4.3. Preliminary Study on Impact Mechanisms

Through the empirical analysis, this study determined that the renovation of old residential areas has a positive impact on enhancing residents’ engagements in pro-environmental behaviors. As analyzed earlier, the renovation of old residential areas impacts residents’ pro-environmental behavior participation in two possible ways. Firstly, by enhancing environmental infrastructure, the renovation creates more favorable external conditions that encourage residents to engage in pro-environmental behaviors. Secondly, the renovation contributes to raising residents’ awareness of environmental protection. Which pathway affects residents’ participation in pro-environmental behaviors? Do both of them have an effect? This study offers a preliminary validation.

First, this study argues that if the renovation of old residential areas can elevate participation in pro-environmental behaviors by enhancing the environmental infrastructure within these communities, then its influence on pro-environmental behaviors, particularly those closely tied to improvements in the environmental infrastructure, will be more pronounced. Therefore, this study investigated how renovation of old residential areas affects various forms of pro-environmental behavior. The study resets the assignment of the dependent variables by assigning a value of 1 to individuals who “always” or “often” engaged in certain types of pro-environmental behaviors, representing more frequent participation, and a value of 0 for less frequent participation. Consequently, the PROBIT model was selected for regression analysis, and the explanatory variables chosen aligned with those in the prior section. The regression results in

Table 7. The influence of renovating old residential areas on specific pro-environmental behaviors.

Pro-Environmental Behaviors	Coefficient	Control Variables	Pseudo R2	Obs.
Garbage Classification	0.1742 *** (0.0506)	YES	0.309	1025
Recycle and Exchange Used Clothes	0.1027 *** (0.0328)	YES	0.254	1025
Participate in Environmental Volunteering	0.0618 ** (0.0303)	YES	0.278	1025
Refuse to Use Disposable Products	0.0379 ** (0.0168)	YES	0.231	1025
Turn Off Unused Water and Electricity	0.0517 ** (0.0229)	YES	0.248	1025
Bring Your Own Shopping Bag	0.0278 ** (0.0131)	YES	0.232	1025
Use Public Transportation	0.0133 (0.0519)	YES	0.187	1025
Appropriate Air Conditioning Temperature Setting	0.0419 *** (0.0132)	YES	0.192	1025
Purchase Green Products	0.0167 * (0.0097)	YES	0.204	1025
Reduce Leftovers and Waste	0.0122 ** (0.0059)	YES	0.229	1025

Notes: The estimated coefficients represent the marginal effects derived from the Probit model. *, **, and *** denote significance at the 10, 5, and 1 percent levels, respectively (two-tailed).

show that old residential areas renovation has a greater impact on two specific pro-environmental behaviors, garbage classification, and recycling and exchanging used clothes, compared to other pro-environmental behaviors. Specifically, the renovation of old residential areas has resulted in increased rates of more frequent participation in garbage classification and recycling and exchanging of used clothes by 17.42% and 10.27%, respectively. Concurrently, there has been a 1.22% to 6.18% increase in the rate of more frequent participation in other pro-environmental behaviors.

Why does the renovation of old residential areas lead to increased participation in these two types of pro-environmental behaviors to a greater extent? It can be seen that the level of participation in these two behaviors is closely linked to the enhancement of the related environmental infrastructure. For instance, one notable aspect of this renovation is the refurbishment, expansion, and smart upgrading of garbage-classification facilities. Such upgrades create an improved environment conducive to residents actively participating in garbage classification initiatives. Meanwhile, the addition of waste clothing recycling bins during the renovation of old residential areas, as well as the active maintenance and initiatives of the community and property owners, have created the hardware conditions for residents to participate in the pro-environmental behavior of recycling and exchanging waste clothing. These results demonstrate that the renovation of old residential areas can enhance residents’ engagements in pro-environmental behaviors by enhancing environmental infrastructure. Figure 5 clearly displays the enhanced basic environmental facilities of residents following the renovation of old residential areas through various scenarios.

Can renovating old residential areas increase residents’ engagements in pro-environmental behaviors by raising awareness of environmental protection? To validate this channel, the study investigated how renovating old residential areas affects residents’ awareness of environmental protection in terms of environmental knowledge, concern, and responsibility. The regression results in

Table 8. Influence of the renovation of old residential areas on residents’ environmental protection awareness.

	(1)	(2)	(3)
	Environmental Knowledge	Environmental Concern	Environmental Liability
Experimental Group	0.8268 ***	0.0744 **	0.1244 **
	(0.2446)	(0.0337)	(0.0516)
Residential Characteristics	YES	YES	YES
Socio-demographic Characteristics	YES	YES	YES
Obs.	1025	1025	1025
R2	0.2447	0.2635	0.2105

**, and *** denote significance at the 5 and 1 percent levels, respectively (two-tailed).

reveal a significant positive effect of the renovation of old residential areas on residents’ awareness of environmental protection across different dimensions of individual environmental protection awareness. This indicates that the renovation of old residential areas has the potential to impact residents’ participation in pro-environmental behaviors on a broader scale. A key implication of these findings is that urban renewal presents a significant opportunity to foster the concept of green living among residents through their personal experiences. During urban renewal initiatives, policymakers can leverage improvements in urban habitats as compelling environmental education resources to enhance public awareness of environmental protection. This strategy can effectively encourage individuals to engage in pro-environmental behaviors more spontaneously.

5. Conclusions and Discussion

In March 2021, China’s “14th Five-Year Plan” was released, featuring the explicit inclusion of urban renewal actions among the 102 major projects [40]. This significant decision by China’s central government not only symbolized the elevation of urban renewal to a national strategy but also marked a new chapter in China’s urban and rural development. Chengdu, being one of the pioneering cities for urban renewal initiatives, has identified the renovation of old residential areas as a key focus area and task within this framework. The study conducted research on the renovation of old residential areas in Chengdu’s main urban region, revealing that this process not only enhances the human environment but also plays a role in increasing residents’ engagements in pro-environmental activities to some extent. Through further analysis, it was determined that the positive influence of old residential areas renovation on residents’ pro-environmental behaviors stems from two main sources. Firstly, the enhancement of environmental infrastructure contributes to this impact; secondly, there has been a notable rise in residents’ environmental protection awareness due to the renovation efforts.

This study concludes, based on its findings, that urban renewal serves not only as a means of enhancing urban infrastructure and functionality but also as a pivotal opportunity to reshape the populace’s contemporary living principles and eco-friendly behaviors. A classic example is the “De Ceuvel” project in Amsterdam. In light of this, urban renewal should be leveraged to galvanize the enhancement of both physical and social aspects within old residential areas, thereby cultivating a more habitable environment. Authorities ought to seize the occasion presented by urban renewal to elevate the quality of urban living through initiatives such as environmental enhancement, infrastructure upgrades, and services enhancement in old residential areas. By fostering an environment conducive to pro-environmental actions, a sense of community engagement and contentment can be instilled among residents, leading to a nurturing of affection and concern towards their surroundings and fostering increased participation in pro-environmental behaviors, and thereby facilitating the widespread adoption of sustainable lifestyles. Simultaneously, decision makers should integrate urban renewal efforts with the reinforcement of grassroots community environmental governance structures, clearly defining roles and responsibilities for maintaining environmental quality within communities. By establishing enduring and scientifically sound management frameworks tailored to the nuances of older neighborhoods, the efficacy of renovation efforts can be upheld over time, fostering an environment conducive to sustained pro-environmental practices among residents.

In addition, urban renewal is a multifaceted, long-term social process that impacts an entire population, serving as an extensive “environmental education” platform for the entire community. This process provides a significant window for residents to shape their perspectives on sustainable living. Within the realm of urban renewal, instilling an understanding of ecological importance, environmental accountability, and relevant environmental knowledge can fundamentally bolster citizens’ environmental consciousness, fostering greater engagement in pro-environmental behaviors. Policymakers should capitalize on public involvement in urban renewal initiatives to encourage alignment with heightened environmental consciousness among residents.

Strategies such as integrating environmental education for the whole society into the urban renewal process and using the media to publicize the results of the renovation of old residential areas have collectively enhanced residents’ understanding of environmental protection. These initiatives have played a crucial role in stimulating residents’ awareness of environmental protection and promoting environmentally friendly behaviors. Furthermore, they have also facilitated the dissemination of environmentally friendly lifestyles within the community.

However, this study has limitations, primarily in the following areas. First, this research is constrained by data limitations. Currently, only two periods of observational data have been collected. The renovation of old residential areas is an ongoing process, prompting the need to ascertain whether the positive impact on residents’ pro-environmental behaviors is a short-term or long-term effect. It is essential to determine whether this effect diminishes over time. To shed light on these aspects, continuous tracking of changes in residents’ pro-environmental behaviors is imperative. This ongoing monitoring is central to the study’s future work, focusing on clarifying the sustainability of the renovation’s impact on residents’ behaviors. Secondly, this study highlights that the renovation of old residential areas has the potential to shape residents’ pro-environmental behaviors by enhancing their environmental protection awareness. However, the study was unable to empirically confirm the specific mechanism by which the renovation of old residential areas boosts residents’ environmental consciousness due to limitations in questionnaire design. Thirdly, the main purpose of this study is causal inference. The empirical strategy adopts traditional econometric tools such as OLS, PROBIT, and difference–difference models. Further incorporation of machine learning methods may enable more effective addressing of the “prediction” problem in the future. This is also one of the directions of the research work, as predicting the impact of renovation of old residential areas on residents’ long-term pro-environmental behavior change holds significant importance for policy making.