Climate-Resilient City Construction and Firms’ ESG Performance: Mechanism Analysis and Empirical Tests

Abstract

This study investigates how climate-resilient city construction (CRCC) influences the Environmental, Social, and Governance (ESG) performance of Chinese listed firms, employing a difference-in-differences (DID) model with firm-year data from 2012 to 2023. The empirical results demonstrate that CRCC exerts a significant positive effect on firms’ ESG performance, with particularly pronounced improvements in the environmental and social dimensions. The mechanism analysis reveals that strengthening government environmental guidance and stimulating firms’ environmental response strategies are the key channels via which CRCC improves firms’ ESG performance. The heterogeneity tests show more pronounced effects for the central–eastern regions, state-owned firms, non-regulated industries, and non-heavily polluting sectors. A further analysis indicates that better ESG performance drives firms to increase their environmental investment, upgrade their value chains, and enhance new quality productive forces. This study extends the framework of ESG determinants by integrating climate adaptation policies, offering insights for urban climate governance and firms’ low-carbon transitions.

1. Introduction

Global climate change has reshaped humans’ living environments at unprecedented rates since the beginning of the 21st century [1]. The Sixth Assessment Report of the IPCC highlighted that the accelerated warming of the climate system has increased the frequency of extreme weather events by 1.7 times. In the face of this severe situation, climate-resilient city construction (CRCC) has become a core strategy of the international community for addressing systemic risks. The United Nations’ 2030 Agenda for Sustainable Development lists “making cities inclusive, safe, resilient and sustainable” as Goal 11 and promotes the formulation of international climate adaptation standards, such as ISO 14090: 2019 [2]. China’s policy framework for CRCC began with the 2013 National Strategy on Climate Change Adaptation, followed by the 2016 Action Plan for Urban Adaptation. By 2024, 39 pilot cities had adopted measures, such as green infrastructure and disaster early-warning systems, to enhance resilience [3].

Firms, as micro-entities in urban economic systems, have their operational activities closely linked to urban climate resilience. Extreme climate events can increase the average annual profit volatility of firms by 12–18%, through channels such as supply chain disruptions and asset damage [4]. At the same time, the rise of the Environmental, Social, and Governance (ESG) concept has reconstructed firms’ value assessment paradigm [5,6,7,8]. Data from the UNPRI shows that in 2023, the global ESG asset management scale exceeded USD 40 trillion, accounting for 36% of the total assets of institutional investors. Driven by China’s “dual-carbon” goals, the ESG rating system has been deeply embedded in the capital market, forming a complete ecosystem encompassing information disclosure, green finance, and regulatory incentives [9]. However, while the existing literature has explored internal ESG drivers, such as digital transformation [10] and executive characteristics [11], it has overlooked exogenous policy shocks, particularly China’s urban resilience-related policies. The international evidence has shown that urban resilience policies shape firms’ sustainability practices [12].

The existing literature has extensively explored CRCC and firms’ ESG performance separately. The CRCC research has focused on construction pathways [13,14,15], the evaluation of construction levels [16,17], and the impact of construction [18,19]. Scholars have pointed out that CRCC should be tailored to the local conditions, fully consider the strategic position of cities and encourage the full participation of the public [20], build intelligent disaster response technology systems and intelligent disaster response social systems [21], establish scientific evaluation systems [22], and provide timely summaries of pilot experiences and deepen international cooperation [23]. In terms of firms’ ESG performance, the research has mainly focused on the evaluation of, the influencing factors on, and the impacts of firms’ ESG performance. A firm’s ESG performance is regarded as an important indicator for measuring the degree of the firm’s sustainable development, and its influencing factors comprise the executive’s characteristics [24], the board of director’s characteristics [25], its compensation policies [26], the firm’s digital transformation [27], policies and regulations [28], and public environmental concerns [29]. Notably, no work has bridged CRCC and ESG performance mechanistically—a critical gap given that urban resilience policies may generate spillover effects on firms’ sustainability, as observed in the EU’s urban adaptation initiatives [30] and Singapore’s green infrastructure programs [31]. This omission risks undervaluing CRCC’s role in accelerating low-carbon transitions.

Based on this, this study uses data from Chinese A-share listed firms from 2012 to 2023 and applies empirical research methods to thoroughly analyze the impact of CRCC on firms’ ESG performance and its internal mechanisms. Compared with the existing research, the novel contributions of this study are mainly reflected in the following three aspects: First, this study takes the CRCC pilot policy as the starting point to explore its micro-level impact on firms’ ESG performance, offering a new angle for understanding the link between climate adaptation strategies and firms’ social responsibility. Second, using the difference-in-differences (DID) method, it verifies CRCC’s positive effect on firms’ ESG performance, especially in the environmental and social dimensions. The mechanism tests reveal the key transmission path between government environmental guidance and firms’ response strategies, providing empirical support for CRCC’s influence mechanism. Third, the findings provide new strategies for firms to improve their ESG performance through CRCC, offer a scientific basis for government policies promoting urban green transformations, and contribute to sustainable economic and social development.

The subsequent content in this article is arranged as follows: The second part presents a literature review and theoretical analysis; the third part elaborates on the methodology, including the variable selection, data sources, and model settings; the fourth part presents the empirical results and discussion; the fifth part conducts expansion research; and the sixth part summarizes the research conclusions and puts forward targeted suggestions.

2. Literature Review and Theoretical Analysis

2.1. The Impact of CRCC on Firms’ ESG Performance

Against the backdrop of global climate change, climate risks have become one of the key risks that firms must identify and prevent in their operations. The relevant research shows that climate change can have a negative impact on firms’ performance [32]. Its action path primarily weakens the profitability of firms by disrupting their production and supply chain systems and changing the market demand for products [33]. Since the pilot city construction work began in 2017, CRCC, as an important measure for addressing climate change, has gradually become the focus of the government’s environmental protection efforts. However, its micro-level impacts on firms remain underexplored. The existing studies on CRCC have predominantly focused on urban-scale outcomes. CRCC strengthens urban infrastructure resilience through measures such as flood control systems and drought-resistant facilities, enhancing cities’ capacities to mitigate extreme weather impacts [18]. At the same time, it actively promotes low-carbon and energy-saving technologies and models, such as green buildings and green transportation, driving cities towards a green and low-carbon development direction.

From the perspective of firm behavior analyses, on the one hand, firms will improve their environmental protection performance in response to external regulatory pressure [34]. As a highly restrictive environmental protection measure implemented by the government, CRCC places higher requirements on firms in terms of the application of green environmental protection technologies, the implementation of pollutant emission standards, and the improvement of energy utilization efficiency, creating external pressure to prompt firms to strengthen their environmental protection awareness and improve their performance in fulfilling their environmental protection responsibilities. On the other hand, based on the goals of maximizing shareholder returns and sustainable development, firms will take the initiative to assume their social responsibilities and environmental protection obligations, paying attention to resource allocation efficiency and technological innovation; adopting intelligent and green technical means and production models; and promoting their own transformation towards greater intelligence, greenness, and high-end characteristics [35].

Based on the above analysis, this study proposes Hypothesis 1.

H1. 

CRCC can significantly improve firms’ ESG performance.

2.2. The Influence Mechanism of CRCC on Firms’ ESG Performance

The government’s environmental attention is mainly reflected in the following three dimensions. Firstly, cities that have implemented the CRCC policy have exhibited increased initiative in adopting climate governance measures [36] and have strengthened stakeholder interactions between government departments and firms [37], which in turn has affected firms’ ESG performance. Furthermore, the relevant policy tools, including the promulgation of China’s Environmental Protection Tax Law [38,39] and the implementation of pilot emission trading schemes [10], have been confirmed to play a role in improving firms’ ESG performance.

Secondly, the CRCC policy aimed to strengthen the climate change adaptation infrastructure and enhance the climate resilience of the pilot cities by 2020. This could reduce firms’ exposure to climate-related risks, boost their expectations and confidence, and improve firms’ ESG performance by alleviating financing constraints [40]. Additionally, the CRCC policy could help maintain stability over an extended period, which would help lower the operational costs and risks for firms [4], optimize resource allocation [41], enhance sustainable development capabilities [42], and ultimately elevate firms’ ESG performance [43].

Finally, the CRCC policy tends to encourage firms to disclose data on their environmental protection performance and carbon emissions, thereby compelling them to attach greater importance to their ESG performance. Drawing from a sample of China’s top 100 firms participating in the Carbon Disclosure Project, Li et al. (2018) observed that the firms showed a stronger propensity to disclose their carbon-related information when subjected to pressures stemming from informal environmental legitimacy [44].

Regarding the government’s environmental regulatory intensity, the CRCC policy aims to enhance regulatory stringency and strengthen environmental protection supervision [34]. This enables the prompt detection and correction of firms’ environmental violations, thereby driving firms to innovate to develop clean energy production technologies [45]. Concurrently, the policy establishes higher environmental protection standards for firms, prompting them to increase their investments in environmental protection for their production and operational processes and enhance their environmental protection capabilities. Such a sequence of effects ultimately generates a positive impact on firms’ ESG performance [46].

Based on the above analysis, this study proposes Hypothesis 2.

H2. 

CRCC can improve firms’ ESG performance by increasing the government’s environmental attention and strengthening the intensity of environmental regulations.

In terms of firms’ awareness of climate risks, the CRCC policy prompts firms to attach significance to climate risks and heighten their awareness of such risks. During decision-making processes, firms will fully consider the implications of climate risks and continuously upgrade their internal governance. This, in turn, will enable them to mitigate financing constraints and improve their ESG performance [47]. Firms with a stronger awareness of climate risks tend to maintain their emotional stability when responding to climate change. They can drive improvements in ESG performance by increasing their growth option value and Tobin’s Q [48]. Additionally, the CRCC policy can maintain relative stability, which helps reduce investors’ pessimism, enhances their ability to perceive climate risks, and thereby boosts firms’ ESG performance [42].

In terms of firms’ green innovation capabilities, the CRCC policy acts as an external policy-driven pressure, serving to raise firms’ environmental standards and foster their green innovation capacities. This, in turn, enables firms to demonstrate enhanced environmental protection performance and social responsibility outcomes. Policies, such as pollution charges and environmental protection subsidies [37], tax policies [49], and environmental protection taxes [38,50], have all been shown to contribute to the improvement of firms’ green innovation capabilities. The stability inherent in policies like CRCC can strengthen firms’ adaptive capacities for responding to climate change by boosting their green innovation capabilities [41], thereby enhancing firms’ ESG performance.

Based on the above analysis, this study proposes Hypothesis 3.

H3. 

CRCC can enhance firms’ ESG performance by raising their climate risk awareness and strengthening their green innovation capabilities.

3. Methodology

3.1. Data

This study selected data for Chinese listed firms on Shanghai and Shenzhen A-share markets from 2012 to 2023 as research samples, which were processed as follows: (1) exclude listed firms marked as ST, *ST, and PT during the sample period; (2) exclude financial industry firms; (3) exclude firms with significant missing values for key variables; (4) apply 1% two-tailed winsorization to all continuous variables to mitigate impact of extreme values. After these procedures, balanced panel dataset of 15,888 firm-year observations was obtained. ESG scores and ratings data used in this study were sourced from Wind database, while Bloomberg ESG disclosure data for robustness tests were sourced from Bloomberg Terminal. Firms’ green patent data were retrieved from CNRDS database, and other financial data were obtained from CSMAR database. Government environmental attention data were derived from municipal government work reports through text analysis, and city-level data were extracted from China Urban Statistical Yearbook.

3.2. Variable Definitions

3.2.1. Explained Variable

Consistent with Wang (2023) and Xiao et al. (2024), this study uses the Huazheng ESG rating score to quantify the firms’ ESG performance [51,52]. We prioritize the Huazheng over the alternatives (e.g., Bloomberg) due to its superior granularity in capturing the China-specific ESG contexts, including the regulatory compliance metrics and environmental disclosure practices mandated by Chinese authorities. This alignment with local standards increases the validity of our sample of Chinese firms [53]. The ESG score ranges from 0 to 100, with higher scores indicating better ESG performance. The Huazheng ESG ratings are divided into nine tiers based on the comprehensive scores, descending from AAA, AA, A, BBB, BB, B, CCC, CC, to C. In the empirical model, the ESG scores are primarily used to reflect the firms’ ESG performance, while robustness tests are also conducted using the Huazheng ESG ratings and Bloomberg ESG rating indices.

The variations in the Huazheng ESG scores and their three-dimensional components for the listed firms during the sample period are presented in Figure 1. As shown, before the implementation of CRCC in 2017, the ESG scores, along with their environmental (E) and social (S) sub-dimensions, remained relatively stable, while the governance (G) sub-dimension displayed a notable downward trend; conversely, following the 2017 policy implementation, the ESG, E, and S scores all demonstrated significant upward trajectories, whereas the G scores continued their gradual decline. This preliminary analysis suggests that CRCC has exerted a discernible positive influence on firms’ ESG performance.

3.2.2. Explanatory Variable

As a key initiative of local governments to proactively address climate change and enhance urban resilience, CRCC has shown remarkable effectiveness at promoting efficient resource utilization, strengthening ecological adaptability and driving green and low-carbon development [15]. In 2017, the National Development and Reform Commission (NDRC), the Ministry of Housing and Urban–Rural Development (MOHURD), and other agencies jointly promulgated the “Notice on Launching CRCC Pilot Programs,” selecting 28 cities nationwide as CRCC pilot sites. The pilot city selection was exogenous to the firms’ characteristics, as the 2017 NDRC/MOHURD notice prioritized regional climate vulnerability and administrative capacity over economic factors. To further address potential selection bias, we employ Propensity Score Matching (PSM), confirming the comparability between the treatment and control groups.

Drawing on this policy context, this study uses the exogenous shock of the CRCC pilot designation (policy) as a proxy variable for measuring urban climate resilience. Specifically, if a firm is headquartered in a CRCC pilot city, it is assigned a value of 1 for 2017 and subsequent years; otherwise, it is assigned a value of 0.

3.2.3. Control Variables

Following prior studies [54,55,56], this study incorporates the following control variables: firm size ($size$), firm age ($firmage$), leverage ratio ($lev$), cash flow ratio ($cashflow$), growth rate ($growth$), board size ($board$), ownership concentration ($top1$), and proportion of independent directors ($indep$). The variable definitions are detailed in

Table 1. Variable definitions.

Variable	Symbol	Definition
ESG performance	$esg$	Huazheng ESG score
Environmental behavior	$eb$	E score
Social behavior	$sb$	S score
Governance behavior	$gb$	G score
Pilot policy	$policy$	Construction of climate-resilient cities
Company size	$size$	ln (total assets)
Firm age	$firmage$	ln (current year − establishment year + 1)
Asset liability ratio	$lev$	Total liabilities/total assets
Cash flow ratio	$cashflow$	Net cash flow/total assets
Revenue growth rate	$growth$	Current year’s operating income/previous year’s operating income − 1
Board size	$board$	ln (number of board members)
Ownership concentration	$top1$	Number of shares held by largest shareholder/total number of shares
Proportion of independent directors	$indep$	Number of independent directors/number of directors
Government environmental attention	$encon$	Environmental keyword frequency in municipal governmental reports
Environmental regulation intensity	$regu$	Environmental regulation intensity index
Firms’ climate risk awareness	$clico$	Climate-term frequency in annual reports
Green innovation	$grein$	Number of green invention patents obtained

.

3.2.4. Mediating Variables

Government Environmental Attention ($encon$): Drawing on the methodology of Zhang et al. (2024), this study used the frequency of environmental keywords as a proxy for local government environmental attention [18]. Specifically, we collected municipal government work reports from 2012 to 2023 and identified the keywords associated with environmental governance and green ecology. These keywords serve as a measure of the policy-level focus, reflecting the emphasis and investment orientation of local governments. Using text analysis, we calculated an indicator that captures both the intensity of the environmental attention and the policy priorities within environmental affairs.

Environmental Regulation Intensity ($regu$): Following the approach of Hu et al. (2022), this study constructed a comprehensive environmental regulation intensity index using three representative pollution indicators: industrial wastewater discharge, industrial sulfur dioxide emissions, and industrial smoke and dust emissions [57]. A higher index value indicates stricter environmental regulations and stronger constraints on firms’ environmental behaviors.

Firms’ Climate Risk Awareness ($clico$): Adopting the method of Wang and Hui (2024), this study developed a climate risk terminology dictionary through a statistical analysis of the extended climate-related vocabulary [58]. The firm-level risk perception, measured using the frequency of these climate-related terms in firms’ annual disclosures, served as the core metric. We calculated the occurrence of the dictionary terms across the full annual reports to quantify the firms’ climate risk awareness, where higher values indicate a greater recognition of the climate risks and a more systematic integration of climate factors into business strategies.

Green Innovation ($grein$): Green innovation is operationalized as the natural logarithm of the number of independent green invention patents granted to a firm in the current year. This metric reflects a firm’s capability for green technological innovation, with higher values indicating stronger performance in this domain.

3.3. Empirical Model

To examine the impact of CRCC on firms’ ESG performance, the following econometric model is constructed:

$$es{g}_{it}={\alpha }_0+{\alpha }_{1}polic{y}_{it}+{{\displaystyle \sum }}_{j=2}^J{\alpha }_j{X}_{it}^j+{\omega }_i+{\omega }_h+{\omega }_t+{\epsilon }_{it}$$

(1)

In this study, the dependent variable is the ESG performance, denoted as $es{g}_{it}$, which measures the ESG performance of firm $i$ in year $t$ using the Huazheng ESG score. This score ranges from 0 to 100 points. The core independent variable is $polic{y}_{it}$, an indicator variable that takes a value of 1 if firm $i$’s city is selected as a CRCC pilot in year t or in any subsequent year, and 0 otherwise. $X_{it}$ represents a set of control variables that may influence a firm’s ESG performance. The model incorporates the firm’s fixed effects (${\omega }_i$), the industry’s fixed effects (${\omega }_h$), and the year’s fixed effects (${\omega }_t$), while ${\epsilon }_{it}$ captures the stochastic error term. The parameter ${\alpha }_1$ is the key coefficient of interest, reflecting the impact of the CRCC initiatives on the ESG performance. A significantly positive ${\alpha }_1$ indicates that the CRCC initiatives enhance a firm’s ESG performance.

As shown in

Table 2. Descriptive statistics.

Variable	Obs.	Mean	Std. Dev.	Minimum	Maximum
$esg$	15,888	73.746	5.034	44.670	92.930
$eb$	15,888	61.492	7.685	31.450	92.300
$sb$	15,888	75.051	9.585	0.000	100.000
$gb$	15,888	79.486	6.570	19.600	96.670
$policy$	15,888	0.063	0.243	0.000	1.000
$size$	15,888	22.654	1.347	19.585	26.440
$firmage$	15,888	2.980	0.326	1.609	3.638
$lev$	15,888	0.435	0.200	0.035	0.925
$cashflow$	15,888	0.050	0.065	−0.199	0.266
$growth$	15,888	0.133	0.356	−0.653	3.808
$board$	15,888	2.141	0.200	1.609	3.065
$top1$	15,888	0.343	0.150	0.076	0.758
$indep$	15,888	0.377	0.056	0.109	0.600

, the descriptive statistics reveal substantial standard deviations in the ESG scores, with a 48.260-point gap between the minimum and maximum values. This indicates significant heterogeneity in the firms’ ESG performance. The relatively small standard deviations of the other variables suggest that these measures are within normal ranges, implying the model is not unduly influenced by extreme observations.

4. Results and Discussion

4.1. Benchmark Results

This study employs the DID model to estimate the impact of CRCC on firms’ ESG performance.

Table 3. Benchmark regression results.

Variable	ESG Score			Environmental (E)	Social (S)	Governance (G)
	(1)	(2)	(3)	(4)	(5)	(6)
policy	0.781 ***	0.514 ***	0.391 **	0.992 ***	0.665 *	−0.091
	(4.28)	(2.75)	(2.06)	(3.55)	(1.81)	(−0.35)
size		1.197 ***	1.446 ***	1.739 ***	1.796 ***	1.249 ***
		(13.99)	(15.60)	(12.75)	(10.04)	(9.77)
firmage		−1.169 ***	0.674	0.676	−0.095	−0.341
		(−5.00)	(1.22)	(0.83)	(−0.09)	(−0.45)
lev		−4.038 ***	−4.328 ***	−1.762 ***	0.422	−9.576 ***
		(−11.94)	(−12.51)	(−3.46)	(0.63)	(−20.05)
cashflow		−0.786	−0.809	1.199	−0.574	−2.214 ***
		(−1.38)	(−1.43)	(1.44)	(−0.52)	(−2.83)
growth		−0.320 ***	−0.239 ***	−0.608 ***	0.265	−0.189
		(−3.67)	(−2.72)	(−4.70)	(1.56)	(−1.56)
board		0.686 **	0.288	−1.199 **	0.950	0.654
		(1.97)	(0.83)	(−2.35)	(1.42)	(1.37)
top1		1.615 ***	1.330 **	0.341	−0.898	3.548 ***
		(2.93)	(2.40)	(0.42)	(−0.84)	(4.63)
indep		2.982 ***	2.603 ***	−0.656	2.100 ***	4.520 ***
		(7.13)	(6.26)	(−1.07)	(2.62)	(7.87)
Constant	73.697 ***	39.096 ***	30.780 ***	24.768***	24.797 ***	37.012 ***
	(2323.45)	(15.53)	(7.80)	(4.27)	(3.26)	(6.80)
Firm fe	Yes	Yes	Yes	Yes	Yes	Yes
Year fe	No	No	Yes	Yes	Yes	Yes
Industry fe	No	No	Yes	Yes	Yes	Yes
N	15,888	15,888	15,888	15,888	15,888	15,888
R2	0.0013	0.0253	0.0599	0.1413	0.0988	0.1231

Notes: () represents the corresponding t-statistic. *** p ≤ 0.01; ** 0.01 < p ≤ 0.05; * 0.05 < p ≤ 0.1. Columns (4)–(6) report disaggregated effects on E/S/G dimensions.

presents the benchmark regression results. Columns (1)–(3) show that the policy coefficient is significantly positive across the models, even after adding the control variables and the firm/year/industry fixed effects. This confirms Hypothesis 1. Notably, the control variables, such as the firm size and ownership concentration, correlate positively with the ESG, while the leverage shows a negative relationship.

Columns (4)–(6) in Table 3 disaggregate the policy effects across the environmental (E), social (S), and governance (G) sub-dimensions. The results show significant positive impacts on the E and S performance, but no significant effect on the G performance. This finding aligns with Hypothesis 1, which theorizes that CRCC exerts dual pressures—regulatory compliance and strategic adaptation—to enhance ESG performance. The significant effects on the environmental (E) and social (S) dimensions specifically validate the theoretical argument that CRCC’s focus on climate resilience directly drives firms to improve their resource efficiency and stakeholder engagement. Conversely, the governance (G) metrics are less responsive to city-level interventions, as they hinge on internal organizational reforms requiring longer-term institutional adjustments rather than external policy shocks [59].

4.2. Robustness Checks

4.2.1. Parallel Trends Test

A fundamental premise for obtaining accurate causal identification using the DID model is that the treatment and control groups should satisfy the parallel trends assumption prior to policy implementation; in other words, it is necessary to verify whether the treatment and control groups exhibit identical ESG change trends in the absence of policy shocks. Drawing on the approach proposed by Guo et al. (2024) to avoid collinearity issues in the estimation, Figure 2 presents the parallel trends test with the year preceding the policy implementation (k = −1) as the baseline period [60].

As shown in Figure 2, before the implementation of the CRCC policy pilot, the difference in the ESG performance between the treatment group and the control group was not significant, indicating that the model conforms to the premise assumption of a parallel trend. After the implementation of the policy, the difference between the treatment group and the control group is very significant, suggesting that the ESG performance level of the firms in the treatment group is significantly higher than that of the firms in the control group after the policy implementation; therefore, the parallel trend test is passed, which further confirms the rationality of the empirical results based on the DID model.

4.2.2. Placebo Test

To eliminate the possibility of bias impacts on the results of the benchmark regression caused by some unobservable individual factors that change over time and location, this study further employs the approach of Chetty et al. (2009), and randomly generates policy pilot cities and repeats the benchmark regression 500 times to conduct a placebo test [61]. In this test, if the estimated coefficient is significantly different from 0, it indicates that the influence of unobservable factors that cannot be ignored has been omitted by the benchmark model. Figure 3 shows the kernel density function distribution diagram of the estimated coefficients using the random sampling method. The coefficient estimates of the randomly generated policy pilot cities approximately follow a normal distribution with a mean of 0, while the estimated value of the benchmark regression coefficient (which is 0.391) of CRCC is significantly different from 0. The placebo test is passed, indicating that the results of the benchmark regression are less affected by unobservable factors and have strong credibility.

4.2.3. PSM-DID Test

Given that CRCC is not completely exogenous, there may be a problem of sample selection bias; that is, it cannot be guaranteed that the individual characteristics of the treatment group and the control group are the same before the implementation of the policy. Therefore, this study further uses the PSM-DID method to conduct a robustness test. Specifically, eight control variables are taken as the covariates, and the propensity score matching method is used to conduct nearest neighbor matching between the pilot cities and the non-pilot cities; that is, a logit regression is performed on the samples to obtain the propensity score values. As shown in Figure 4, the deviation between the treatment group and the control group after matching is significantly reduced, indicating that the matching effect is good and the balance test is passed. There is no systematic deviation between the treatment group and the control group after matching, indicating that the results of the benchmark regression are reasonable and effective. The results of the estimation using the benchmark regression model for the matched samples are shown in column (1) of

Table 4. Robustness tests.

Variable	(1)	(2)	(3)	(4)	(5)	(6)	(7)
policy	0.412 **	0.072 **	1.197 ***	0.375 *	0.319	0.244	0.408 **
	(2.16)	(2.11)	(3.19)	(1.65)	(1.61)	(0.97)	(2.15)
policy−1							0.228 *
							(1.93)
Constant	30.711 ***	−3.624 ***	−7.894	26.419 ***	30.815 ***	30.858 ***	30.843 ***
	(7.75)	(−5.07)	(−1.18)	(4.53)	(7.81)	(7.82)	(7.81)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Firm fe	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year fe	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Industry fe	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	15,852	15,888	8074	11,916	15,888	15,888	15,888
R2	0.0601	0.0618	0.5898	0.0458	0.0598	0.0597	0.0602

Notes: () represents the corresponding t-statistic. *** p ≤ 0.01; ** 0.01 < p ≤ 0.05; * 0.05 < p ≤ 0.1; policy₋₁ refers to the introduction of other relevant policy variables, which is the “Three-Year Action Plan for Winning the Battle for a Blue Sky”, publicly released by the State Council in 2018.

. It can be observed that after eliminating the potential problem of sample selection bias, the coefficient of “policy” is significantly positive, which is consistent with the results of the benchmark regression and further confirms that the results are robust.

4.2.4. Other Robustness Tests

(1)

Replacing the explained variable. By using different measurement indicators to represent the explained variable, the robustness of the benchmark regression results can be confirmed if there is no obvious change in the regression results. Referring to the research by Bai et al. (2024), the Huazheng ESG rating and the Bloomberg ESG rating are, respectively, adopted, and a re-estimation is carried out after replacing the ESG score in the benchmark regression [53]. As shown in the results in columns (2) and (3) of Table 4, the coefficient of “policy” is still significantly positive, which confirms the robustness of the benchmark regression results.

(2)

Adjusting the window period. To ensure the timeliness of the implementation of the pilot policy, this study adjusts the time window of the samples to the period from 2014 to 2022 to reduce the influence of long-term factors. As can be seen from the results in column (4) of Table 4, the coefficient of “policy” is still significantly positive, which further supports the robustness of the benchmark regression results.

(3)

Counterfactual test. One of the preconditions for using the DID model is that the treatment group and the control group are comparable; that is, without the implementation of the pilot policy, the ESG performance of the treatment group and the control group will not show significant differences over time. Therefore, drawing on the approach of Shi et al. (2018) [62], the implementation time of CRCC is advanced to 2014 and 2016, respectively, to construct dummy variables for the false policy pilot time. Then, the estimation is carried out according to the benchmark regression model. As can be seen from the results in columns (5) and (6) of Table 4, when the pilot policy is advanced to 2014 and 2016, respectively, the coefficients of “policy” are not significant. Therefore, from a counterfactual perspective, it is further verified that the ESG performance of the firms in the pilot areas has not been affected by other unknown factors, indicating that the pilot policy has indeed significantly improved the ESG performance level of the firms.

(4)

Excluding the impacts of other policies. Another potential reason for the instability of the regression results is that during the sample period, the ESG performance level of the firms may have also been influenced by other similar policies or policies implemented during the same period, so that the regression results do not represent the net effect of this pilot policy [60]. In 2018, the State Council issued the “Three-Year Action Plan for Winning the Battle for a Blue Sky”, aiming to significantly reduce the total emissions of major air pollutants and jointly reduce greenhouse gas emissions, which may have had a direct impact on the ESG performance level of firms. In order to eliminate the influence of this policy, this study includes the DID dummy variable of the “Three-Year Action Plan for Winning the Blue Sky Defense War”, namely ${\mathrm{policy}}_{-1}$, in the benchmark regression. As can be seen from the results in column (7) of Table 4, the coefficient of “policy” remains significantly positive, which once again confirms the robustness of the benchmark regression results.

4.3. Mechanism Test

We investigate two potential channels through which CRCC influences the firms’ ESG performance. The first is government environmental guidance, including government environmental attention and environmental regulation intensity. The second channel is the firms’ environmental response strategies, specifically including the firms’ climate risk awareness and green innovation. To verify the above logic, this paper draws on the practices of [63,64], and we use a two-step regression method. In the first step, we test whether CRCC affects the mechanism variable. In the second step, we test the impact of the predicted value of the mechanism variable on the firms’ ESG performance. Specifically, the following two-stage regression model is constructed.

$$channe{l}_{it}={\beta }_0+{\beta }_{1}polic{y}_{it}+{{\displaystyle \sum }}_{j=2}^J{\beta }_j{X}_{it}^j+{\mu }_i+{\mu }_h+{\mu }_t+{\epsilon }_{it}$$

(2)

$$es{g}_{it}={\gamma }_0+{\gamma }_1\widehat{channe{l}_{it}}+{{\displaystyle \sum }}_{j=2}^J{\gamma }_j{X}_{it}^j+{\phi }_i+{\phi }_h+{\phi }_t+{\epsilon }_{it}$$

(3)

where $channe{l}_{it}$ is the potential mediating variable, specifically including the government’s environmental attention ($enc{o}_{it}$) and environmental regulation intensity ($reg{u}_{it}$), and the firms’ climate risk awareness ($clic{o}_{it}$) and green innovation ($grei{n}_{it}$). $\widehat{channe{l}_{it}}$ is the predicted value of the mechanism variable generated in the first-step regression in Equation (2). First, we regress Equation (2) and test whether the coefficient ${\beta }_1$ of the CRCC variable $polic{y}_{it}$ is significant; if it is significant, it indicates that CRCC can affect the mechanism variable. Second, we incorporate the predicted value of the mechanism variable $\widehat{channe{l}_{it}}$ into Equation (3) for a further regression. ${\gamma }_1$ is the impact coefficient of the predicted value of the mechanism variable $\widehat{channe{l}_{it}}$ on the firms’ ESG performance. If ${\gamma }_1$ is significant, it indicates that the mechanism variable $channe{l}_{it}$ plays an important role in the process by which CRCC promotes firms’ ESG performance.

As shown in

Table 5. The mechanism test of government environmental guidance.

Variable	Government Environmental Attention	ESG Score	Environmental Regulation Intensity	ESG Score
	(1)	(2)	(3)	(4)
policy	0.001 ***		0.057 ***
	(8.85)		(5.81)
$\widehat{encon}$		3.774 **
		(2.08)
$\widehat{regu}$				0.096 **
				(2.08)
Control variables	Yes	Yes	Yes	Yes
Firm fe	Yes	Yes	Yes	Yes
Year fe	Yes	Yes	Yes	Yes
Industry fe	Yes	Yes	Yes	Yes
N	15,888	15,888	15,888	15,888
R2	0.2227	0.0581	0.0929	0.0581

Notes: () represents the corresponding t-statistic. *** p ≤ 0.01; ** 0.01 < p ≤ 0.05.

, the pilot policy elevated the firms’ ESG performance through strengthened government environmental guidance—specifically by elevating governmental environmental attention and regulatory intensity. These results provide empirical support for Hypothesis 2. The significant coefficients for government environmental attention and regulation intensity validate the theoretical mechanism. CRCC prompts governments to issue environmental incentives, such as subsidies and stricter standards, which in turn pressure firms to improve their ESG practices. This aligns with the theoretical argument proposed in Section 2.2 that policy-driven environmental guidance acts as a critical mediator.

Similarly,

Table 6. The mechanism test of firms’ environmental response strategies.

Variable	Firms’ Climate Risk Awareness	ESG Score	Green Innovation	ESG Score
	(1)	(2)	(3)	(4)
policy	0.042 **		0.027 *
	(2.13)		(1.78)
$\widehat{clico}$		0.133 **
		(2.08)
$\widehat{grein}$				0.204 **
				(2.08)
Control variables	Yes	Yes	Yes	Yes
Firm fe	Yes	Yes	Yes	Yes
Year fe	Yes	Yes	Yes	Yes
Industry fe	Yes	Yes	Yes	Yes
N	15,888	15,888	15,888	15,888
R2	0.3540	0.0581	0.0335	0.0581

Notes: () represents the corresponding t-statistic. ** 0.01 < p ≤ 0.05; * 0.05 < p ≤ 0.1.

supports Hypothesis 3, demonstrating that CRCC improves the ESG performance by stimulating firms’ environmental response strategies. The positive effects on firms’ climate risk awareness and green innovation confirm that CRCC motivates firms to internalize climate risks, adopt green technologies, and align their strategies with sustainable development.

4.4. Heterogeneity Analysis

4.4.1. Location Characteristics

There is an imbalance in the economic development among the regions in China, with significant differences in multiple dimensions, such as industrial structure, firm characteristics, resource endowment, and climate environment [65]. Based on this, this study categorizes the provinces where the firms are located into the central–eastern region and the western region according to the classification criteria of the National Bureau of Statistics.

Based on the results in columns (1) and (2) of

Table 7. (a) Regional/ownership heterogeneity. (b) Industry heterogeneity.

(a)
Variable	Location Characteristics		Property Rights
	(1)	(2)	(3)	(4)
policy	0.426 *	−0.102	0.492 **	0.118
	(1.83)	(−0.28)	(2.02)	(0.39)
Constant	30.632 *** (7.47)	38.326 *** (3.80)	27.190 *** (5.19)	32.062 *** (6.96)
Control variables	Yes	Yes	Yes	Yes
Firm fe	Yes	Yes	Yes	Yes
Year fe	Yes	Yes	Yes	Yes
Industry fe	Yes	Yes	Yes	Yes
N	13,668	2220	7692	8196
R2	0.0586	0.0951	0.0689	0.0787
(b)
Variable	Regulated Industry		Industry Attributes
	(1)	(2)	(3)	(4)
policy	0.388	0.410 *	0.245	0.432 **
	(1.18)	(1.75)	(0.55)	(2.06)
Constant	37.454 *** (6.92)	25.356 *** (5.53)	48.733 *** (6.74)	25.085 *** (5.97)
Control variables	Yes	Yes	Yes	Yes
Firm fe	Yes	Yes	Yes	Yes
Year fe	Yes	Yes	Yes	Yes
Industry fe	Yes	Yes	Yes	Yes
N	6161	9727	3438	12,450
R2	0.0596	0.0700	0.0761	0.0657

Notes: () represents the corresponding t-statistic. *** p ≤ 0.01; ** 0.01 < p ≤ 0.05; * 0.05 < p ≤ 0.1.

a, the regression coefficient for the central–eastern region is significantly positive, while that for the western region is not significant. This regional heterogeneity aligns with the theoretical perspective on institutional environment and resource endowment. CRCC’s effectiveness depends on local governments’ regulatory capacities and firms’ adaptive resources. The central–eastern region’s higher economic development and denser firm networks enable stronger policy implementation and faster firm responses.

In contrast, the western regions’ focus on economic growth over environmental goals, combined with its weaker institutional capacity, limits the impact of CRCC. This finding empirically validates the theoretical hypothesis that policy effectiveness varies with regional development stages.

4.4.2. Ownership Structure

The firms are divided into state-owned firms (SOEs) and non-state-owned firms based on their ultimate ownership structure recorded in the CSMAR database. SOEs are defined as firms where the state holds a controlling stake, reflecting the leading role of the state-owned economy in China’s development and its mandate to lead on climate action and ESG responsibilities. Non-state-owned firms encompass private, foreign-invested, and mixed-ownership firms with non-state controlling shareholders, capturing the diversity of ownership structures in the market. This classification aligns with the state-owned sector’s strategic role in driving sustainable development, enabling a comparative analysis of ESG performance across ownership types.

As can be seen from the results in columns (3) and (4) of Table 7a, the pilot policy significantly improved the ESG performance level of the state-owned firms, while its impact on the non-state-owned firms is not obvious. A possible reason behind this is that the state-owned firms have strong capabilities in aspects such as technology, scale, talent reserve, and climate risk management. They can not only effectively mitigate the negative impacts brought about by climate change and improve their own green production capabilities, but also quickly respond to and implement the requirements of the pilot policy [66], thus improving their ESG performance level.

In contrast, the non-state-owned firms incur higher costs to mitigate the losses caused by climate change; moreover, they have certain gaps in their technology, funding, talent, and climate risk management compared with state-owned firms. This has affected the speed of their green transformation to a certain extent; thus, the impact of the pilot policy on their ESG performance was not significant.

4.4.3. Degree of Industry Regulation

As part of China’s low-carbon transformation, the government has implemented a series of policies and adopted regulatory measures of varying degrees for multiple industries, which in turn have influenced the market structure and the effective allocation of resources. Therefore, in this section, the industries to which the firms belong are further divided into two major categories based on the Industry Classification Guidelines for Listed Companies issued by the China Securities Regulatory Commission (CSRC). The regulated industries include mining, petroleum processing, coking, nuclear fuel processing, chemical raw materials and chemical products manufacturing, and other similar sectors. The unregulated industries comprise all the remaining sectors not categorized as regulated.

As can be seen from the results in columns (1) and (2) of Table 7b, the pilot policy had a significant effect on improving the ESG performance level of the unregulated industries, but no obvious impact on the ESG performance level of the regulated industries. This may be because most of the regulated industries are monopolistic industries, with high barriers to market entry, which reduces the intensity of market competition and results in a market structure with low competitiveness. In this situation, the demand for firms’ self-adjustment is relatively small, and the phenomenon of soft budget constraints inhibits the utilization efficiency of funds and innovation ability [67]. Therefore, the impact of the pilot policy on the ESG of the regulated industries was relatively small.

In industries with a low degree of regulation, the market environment is more complex and changeable. The pilot policy significantly affected the ability of firms to respond to and adapt to climate change, prompting firms to continuously innovate and thus improve their ESG performance levels; specifically, the pilot policy provided firms in unregulated industries with a clear policy orientation and incentive mechanism, encouraging these firms to increase their motivation and direct more resources to actions to address climate change and improve their ESG performance level, when faced with the complex market environment during the implementation of CRCC.

4.4.4. Degree of Industry Pollution

Due to differences in the levels of pollution of different industries, there are significant differences in aspects, such as the adaptability of each industry when facing the impacts of climate change, the pressure of pollution reduction and carbon emission reduction, and the ESG performance level [68]. Therefore, this study further divides the industries to which the firms belong into two major categories: heavy pollution industries and non-heavy pollution industries.

Based on the results in columns (3) and (4) of Table 7b, the pilot policy had no significant impact on the ESG performance level of the firms in heavy pollution industries, but it did improve the ESG performance level of the firms in non-heavy pollution industries. The reason for this is that the operating income of the firms in heavy pollution industries primarily comes from production using non-green technologies, and the degree of pollutants or carbon emissions in their production processes is relatively high. When adapting to climate change, they not only face the high cost of updating a large amount of technical equipment, but also face a higher risk of asset stranding. This leads to greater financial pressure and a low carbon transformation pressure on firms, resulting in a “crowding-out effect” of resources, which reduces the resources that firms invest in the ESG field, thereby hindering the improvement of their ESG performance levels.

In contrast, firms in non-heavy pollution industries have a stronger ability to adapt to climate change themselves, and their financial pressure and low-carbon transformation pressure are relatively low. There is no “crowding-out effect” of resources in resource allocation; therefore, they have more resources to improve their ESG performance level during the implementation of CRCC.

5. Expansion Analysis

The existing research has established that ESG performance serves as a strategic driver for firms’ sustainable development; yet, its specific mechanisms for fostering environmental investment, value chain transformation, and productivity enhancement remain underexplored. Firms with strong ESG performance tend to accumulate intangible resources, such as reputational capital and stakeholder trust, which can be leveraged to secure low-cost financing for environmental initiatives [69]. ESG-aligned practices enable firms to legitimize their operations within regulatory and societal frameworks, reducing transaction costs and facilitating value chain collaboration [70]. Additionally, the stakeholder theory posits that prioritizing environmental and social responsibilities can enhance employee engagement and consumer loyalty, directly contributing to productivity gains [71]. These theoretical perspectives form the basis for examining whether CRCC-induced ESG improvements trigger a virtuous cycle of sustainable growth.

Building on the theoretical argument that ESG performance fosters sustainable competitiveness,

Table 8. Expansion analysis.

Variable	Environmental Investment	Value Chain Upgrading	New Quality Productive Forces
	(1)	(2)	(3)
esg	0.031 *	0.088 ***	0.095 ***
	(1.76)	(3.26)	(13.05)
Constant	−0.233 *	−0.902 ***	−0.629 ***
	(−1.80)	(−4.40)	(−11.42)
Control variables	Yes	Yes	Yes
Firm fe	Yes	Yes	Yes
Year fe	Yes	Yes	Yes
Industry fe	Yes	Yes	Yes
N	10,703	15,888	15,612
R2	0.0690	0.0987	0.1609

Notes: () represents the corresponding t-statistic. *** p ≤ 0.01; * 0.05 < p ≤ 0.1.

shows that improved ESG scores significantly promote environmental investment, value chain upgrading, and the development of new quality productive forces. This empirically validates the theoretical mechanism whereby ESG performance reduces financing costs and drives green innovation, as posited in Hypotheses 2 and 3. The positive effect on the green total factor productivity specifically supports the theory that ESG integration optimizes resource allocation and talent attraction, reinforcing the feedback loop among CRCC, ESG, and sustainable growth.

The contribution of the ESG performance of firms to their environmental protection investment is achieved through two key channels: carbon emissions and financing costs. Specifically, the better the ESG performance of a firm, the more likely it is to actively adopt more clean energy technologies, thus further improving its energy utilization efficiency and achieving carbon emission reduction targets. Carbon emission reduction not only brings environmental benefits but also aligns with policy orientation, which enables firms to obtain higher returns on their environmental protection investments and thus improve the efficiency of their environmental protection efforts. At the same time, superior ESG performance by a firm can win the support of relevant stakeholders, effectively alleviate financing constraints, and reduce a firm’s equity and debt financing costs [72,73], thereby attracting more investments and providing financial guarantees for expanding the scale of the firm’s environmental protection investments.

The ESG performance of a firm contributes to the upgrading of the firm’s value chain through two main paths: green research and development innovation upstream of the value chain, and green demand by consumers downstream of the value chain. Specifically, superior ESG performance will prompt a firm to increase the quantity and quality of their green innovation, thereby forming an innovative competitive advantage [74] and enhancing its influence on the value chain [75]. This, in turn, will expand market sales of green products and create a market advantage. This is not only helpful for reducing the difference in green productivity between the firm and its upstream and downstream firms, but also for reducing the cost difference of entering the value chain. The reduction in this difference makes the firm’s position in the supply chain more stable and flexible, reduces its dependence on a few key upstream and downstream firms, and effectively reduces the degree of concentration of the supply chain. With greater discourse power in the supply chain, the firm can better integrate resources, obtain more added value, and promote the upgrading of its value chain.

The ESG performance of firms contributes to the new quality productive forces of firms mainly through the influencing path of the green total factor productivity. On the one hand, the better the ESG performance of firms, the more effectively they can alleviate the financing constraints they face, thereby securing more financing for their green transformation and upgrading of their production technologies. When considering the aspects of capital and technology, energy conservation, consumption reduction, and the optimized allocation of resources [76] can improve the green total factor productivity, thereby enhancing the level of a firm’s new quality productive forces. On the other hand, the higher S dimension in the ESG performance means that a firm performs better in terms of social responsibility; for example, it provides employees with a good working environment, a complete career development plan, a reasonable salary and welfare system, etc., which can attract and retain excellent talent. Excellent talent brings advanced concepts and technologies, which can stimulate employees’ innovation awareness and work efficiency [9], improving the green total factor productivity from a talent perspective, and ultimately promoting the improvement of the firm’s new quality productive forces.

These findings highlight that ESG performance drives environmental investment, value chain upgrading, and new quality productive forces. Such insights directly inform policy recommendations. They underscore CRCC’s role in fostering a virtuous cycle of sustainable firms development, where policy-induced ESG improvements translate into tangible economic and environmental benefits. This demonstrates how enhancing ESG performance through CRCC can create a positive feedback loop for firms, linking environmental responsibility with tangible developmental gains.

6. Conclusions and Policy Recommendations

This study employs a DID approach to examine the impact of CRCC on the ESG performance of Chinese listed firms from 2012 to 2023. The key finding are as follows: CRCC significantly enhances the overall level of firms’ ESG performance, with especially prominent improvements in the environmental and social dimensions. Its impact remained robust after multiple tests. The mechanism analysis reveals that CRCC promotes ESG performance through a twofold pathway: by strengthening the orientation of government environmental policies and stimulating the environmental strategy responses of firms. Based on the heterogeneity analysis, firms in the central and eastern regions, state-owned firms, firms in non-regulated industries, and firms in non-heavily polluting industries have demonstrated more positive ESG performance as a result of CRCC. The extended test further shows that the superior ESG performance of firms can not only attract more environmental protection investments but also drive the upgrading of the value chain, thereby enhancing the new quality productive forces of firms. Based on the above research conclusions, this study proposes the following countermeasures and recommendations.

First, policymakers should systematically expand the CRCC pilot program to broader regions. Our analysis confirms that CRCC implementation significantly elevates firms’ ESG performance, particularly in the environmental and social dimensions. Scaling up this initiative to cities facing high climate vulnerability would amplify these benefits while advancing national carbon neutrality goals. Success stories from the existing pilots cases, such as infrastructure retrofits that lower firms’ climate risk exposure, should be documented to guide their replication.

Second, CRCC policies must reflect the observed heterogeneity in the firms’ responses. The limited ESG improvements in the western regions, non-state-owned firms, and heavily polluting industries require compensatory interventions. Local governments could introduce green technology subsidies for small and medium-sized firms or tax incentives for polluting firms undertaking low-carbon transitions. These measures would directly address the financial constraints inhibiting ESG progress in these sub-groups.

Third, effective CRCC implementation demands coordinated action among government, financial institutions, and firms. Since both regulatory pressure and firms’ climate strategies drive ESG gains, regulators should strengthen environmental mandates while incentivizing private sector initiatives. Financial institutions might tie capital access to ESG metrics, leveraging our finding that strong ESG performers achieve lower financing costs. Firms themselves should prioritize the climate response measures identified as critical transmission channels in this study, particularly green innovation and supply chain resilience investments.