Evaluating the Impact of Green Manufacturing on Corporate Resilience: A Quasi-Natural Experiment Based on Chinese Green Factories

Abstract

Corporate resilience, a critical metric assessing firms’ capacity to withstand risks, recover rapidly, and maintain growth in dynamic environments, has garnered increasing attention from academia and industry. This study employs China’s Green Factory certification policy within its green manufacturing system as a quasi-natural experiment, utilizing a multi-period difference-in-differences (DID) model to evaluate the impact of green manufacturing implementation on corporate resilience. Results confirm that Green Factory certification significantly enhances firms’ resilience. Mechanism analyses identify three reinforcing pathways: alleviating financing constraints, optimizing resource allocation efficiency, and fostering green technological innovation. Heterogeneity analyses reveal more pronounced effects among heavily polluting industries, firms with low reputations, and those with higher levels of managerial myopia. Furthermore, the certification exhibits significant spillover effects, transmitting resilience improvements to industry peers and geographic clusters. This research expands the theoretical boundaries of corporate resilience literature while offering practical implications and empirical evidence for enterprises undergoing green manufacturing transitions.

1. Introduction

In this era of volatility, uncertainty, complexity, and ambiguity (VUCA), numerous changes such as unexpected environmental events, geopolitical conflicts, and trade barriers have posed unprecedented threats to the survival and development of enterprises [1,2,3]. There are numerous cases of companies perishing due to external crises. For instance, during the 2008 economic crisis, Washington Mutual collapsed due to a deposit run and asset depreciation. Similarly, the Chinese business travel giant Tengbang International (original A-share stock code: 300178) was delisted in 2022 after being severely impacted by the COVID-19 pandemic. It can be seen that in a business environment characterized by volatility, uncertainty, complexity, and ambiguity, continuously enhancing one’s own resilience is the core proposition for enterprises to achieve sustainable development at the corporate level, rather than blindly expanding and pursuing short-term scale benefits. Resilience serves as a buffer for enterprises to cope with crises, reducing financial uncertainty [4]. Enterprises with greater resilience are more likely to seize new opportunities during crises [5], which in turn facilitates their post-crisis recovery [6]. In view of this, identifying the influencing factors of corporate resilience and then constructing effective strategies to enhance it are urgent theoretical and practical issues. On the other hand, the construction of a green manufacturing system has become an important engine and key path for promoting sustainable development globally. In China, driven by the “dual carbon” goals, the construction of a green manufacturing system, represented by Green Factory certification (For a detailed introduction to China’s Green Factory policy, please refer to Appendix A of this paper), is continuously deepening. As of 31 March 2025, China’s Ministry of Industry and Information Technology (MIIT) has announced 6429 green factories, basically covering all provinces, municipalities, and autonomous regions across the country, and the number of selections is generally increasing. However, in the face of the vigorous wave of green manufacturing, we are particularly interested in whether “green factories” that focus on environmental attributes and production process sustainability can cultivate their “shock resistance” to enhance resilience while actively fulfilling social responsibilities, thereby truly achieving a seamless integration of corporate objectives and social goals.

According to the resource-based view (RBV), an enterprise’s competitive advantage originates from its ownership of valuable, rare, inimitable, and non-substitutable (VRIN) resources [7]. The Green Factory certification acquired by enterprises strongly embodies these VRIN attributes, thereby qualifying as a strategic organizational resource. Its value is manifested through multi-faceted enhancement of corporate competitiveness, resource integration capacity, and sustainable development capabilities. China’s Green Factory certification is government-led, providing significant policy support for certified enterprises in terms of financial subsidies, tax reductions, special green approvals, and financial credit. Previous works in the literature have shown that being selected as a “Green Factory” significantly alters a company’s resource endowment structure and strategic decision-making behaviors [8,9]. This provides a favorable environment for us to delve into the potential impact of green manufacturing on corporate resilience. We posit that Green Factory certification enhances corporate resilience through three theoretical mechanisms: First, a financing constraint mitigation mechanism grounded in Resource Dependence Theory [10] and Information Asymmetry Theory [11], facilitated through direct policy dividends and government-endorsed signaling effects; Second, a resource allocation efficiency improvement mechanism relying on Resource Orchestration Theory [12], achieved via optimized allocation of production resources and human capital structure upgrading; Third, a green innovation promotion mechanism based on Dynamic Capability Theory [13], constructed through full-chain standard backpressure and policy-driven sustained R&D investment. Based on data from A-share listed companies from 2008 to 2023, this paper examines whether and through which paths Green Factory certification enhances corporate resilience. The empirical results show that Green Factory certification strengthens corporate resilience by mitigating financing constraints, improving resource allocation efficiency, and promoting green innovation. Moreover, in heavily polluting enterprises, low-reputation enterprises, and enterprises with higher levels of managerial myopia, the promoting effect of Green Factory certification on corporate resilience is more significant. More importantly, Green Factory certification has spillover effects, effectively promoting the overall improvement of corporate resilience in the region and the entire industry.

The incremental contributions of this paper are mainly reflected in the following aspects: First, it enriches the research on the influencing factors of corporate resilience from the macro-policy perspective. Existing literature primarily analyzes the influencing factors of corporate resilience from the internal strategic behaviors of enterprises (such as digital transformation, innovation, ESG performance) [14,15,16,17,18,19,20,21,22], managerial traits (such as narcissism, greed, overconfidence) [23,24,25], as well as external factors such as social trust [26], supply chain leadership [27], and analyst attention [28]. However, limited attention has been paid to the potential impact of policy environments on corporate resilience. This paper confirms the positive enabling effect of green manufacturing, a highly contemporary macro-strategy, on corporate resilience. The research findings also provide an internal driving force for enterprises to proactively accelerate their green transformation. Secondly, this paper further enriches the literature on the micro-level utility of green manufacturing policies. Most existing studies have primarily focused on the environmental benefits [8,9,29,30,31] and social benefits [32] of green manufacturing, while lacking in-depth exploration of its micro-level impacts. This paper reveals the influence mechanism of Green Factory certification on corporate resilience, providing strong theoretical support and empirical evidence for interpreting the micro-economic value of green development. Finally, this paper captures the spillover effects of Green Factory certification from both industry and spatial dimensions, further enriching the research literature on the economic consequences of green manufacturing policies and offering meaningful policy references for promoting green development strategies.

2. Theoretical Analysis and Research Hypotheses

Corporate resilience refers to the ability of an enterprise to withstand shocks and maintain or improve its operations under adverse conditions [33]. The resource endowments possessed by the enterprise, its ability to allocate resources, and its innovation capabilities are key factors influencing resilience. According to RBV, a firm’s competitive advantage stems from its possession of VRIN resources [7]. Shaping this competitive advantage is crucial for enterprises to cope with and adapt to external crises, as well as enhance their own resilience. Prior literature on organizational resilience has indicated that resilience depends on the ability to rapidly respond to environmental signals and flexibly allocate organizational resources [4]. By maintaining flexibility in resource allocation and moderate redundancy, enterprises can better address crises and recover more quickly from them [34,35]. Innovation helps break traditional mindsets and rigid organizational structures, enabling organizations to better respond to changes in internal and external environments. Williams and Shepherd (2016) also noted that creative actions can help organizations respond quickly to crises [36]. Based on this, this paper mainly analyzes the impact of Green Factory certification on corporate resilience through three paths: financing constraints, resource allocation, and green innovation.

First, obtaining Green Factory certification can alleviate financing constraints for enterprises and subsequently enhance their resilience. The high standards of Green Factory certification imply significant compensation, providing certified enterprises with notable financing advantages. On the one hand, Green Factory certification has a direct incentive effect on enterprise financing. According to the Resource Dependence Theory, organizations cannot be self-sufficient, and their survival and development highly depend on external key resources such as capital, technology, talent, and policy legitimacy [10]. In order to encourage enterprises to actively pursue green manufacturing transformation, certified green factories can enjoy policy dividends such as special construction funds, industrial transformation and upgrading funds, government procurement orders, as well as preferential guarantees and credit support from financial institutions under the credibility of the government (Please refer to the Notice on the Development of a Green Manufacturing System (2016), issued by MIIT of the People’s Republic of China [37]). This directly increases the “disposable” funds of enterprises and effectively alleviates their financing constraints [8]. On the other hand, the certification creates an indirect signaling effect. Enterprises that have obtained Green Factory certification can rely on the government endorsement of this demonstration title to gain external financing convenience [8]. Based on information asymmetry theory [11], external stakeholders—as information-disadvantaged parties—often interpret policy favors as indicators of strong development potential and harmonious government–enterprise relations, thereby granting the firm greater trust and support. This includes lower debt financing costs [9,38,39], improved commercial credit access [40], and enhanced equity financing availability [41,42]. Furthermore, as a voluntary environmental regulation tool, obtaining Green Factory certification is equivalent to publicly declaring the enterprise’s commitment to environmental responsibility, which aligns with the current green preferences of investors, making it easier to attract external financing opportunities [32]. The alleviation of financing pressures not only helps enterprises avoid the survival risks caused by fund shortages during unexpected crises but also provides support for their emergency response and strategic adjustments with abundant fund guarantees, which is crucial for enhancing corporate resilience [18,43].

Second, Green Factory certification prompts firms to improve resource allocation efficiency, thereby enhancing resilience. The Resource Orchestration Theory emphasizes that effective resource management and scientific allocation are structural pillars for creating competitive advantages [12]. This helps enterprises maintain excellent crisis response capabilities in complex and volatile environments. In order to consolidate their exemplary status, enterprises awarded Green Factory status need to continuously invest to ensure the continuity of green production activities, and are therefore more likely to actively promote the internalization of external resources and the reconfiguration of corporate resources. On one hand, environmental regulations force firms to optimize existing resource allocation [44,45,46,47]. The Green Factory certification process explicitly requires enterprises to standardize energy and resource inputs, reduce pollution emissions, and synergistically achieve land use intensification, cleaner production, waste recycling, and low-carbon energy use. These mandatory regulations can guide enterprises to abandon outdated production and operation structures, promptly phase out high-pollution and high-energy-consuming equipment, allocate production resources to cleaner equipment and green products, optimize investment layout to improve capacity utilization rates, and enhance resource allocation efficiency, thereby laying a solid foundation for enterprises to withstand external shocks and achieve long-term stable development. On the other hand, Green Factory certification will assist enterprises in optimizing their human resource structures. Human capital is a prerequisite and core element for fostering new development momentum and enhancing corporate resilience [48]. Due to continuous environmental protection pressures, enterprises that have achieved Green Factory status not only focus on providing green skills training for existing employees but are also more willing to recruit high-caliber external talent to promote the complementarity between human capital and green capital, thereby improving production efficiency [49,50]. Simultaneously, the reputation effect of “green factories” can signal positive information to external stakeholders that “the firm is a leader in green development” and “prioritizes R&D innovation”, enhancing external talents’ positive perception of the firm’s image, thereby further accelerating talent agglomeration.

Finally, being awarded Green Factory status can promote corporate green innovation and thereby enhance corporate resilience. The Green Factory evaluation aims to drive enterprises toward green transformation and upgrading, which undoubtedly creates a strong reverse-pressure mechanism, compelling firms to fully elevate their green innovation capabilities. On the one hand, the evaluation process systematically examines the entire production cycle of applicant enterprises, including infrastructure, management systems, energy and resource inputs, product quality, production processes, and environmental performance. These evaluation criteria effectively drive applicant enterprises to optimize their production processes and increase investments in green technology R&D. It is important to note that the Green Factory certification is not a “one-time achievement”. After being selected, enterprises are subject to random inspections, regular reviews, and corresponding information disclosure obligations. If an enterprise no longer meets the evaluation criteria, it will be removed from the Green Factory list. This dynamic and stringent post-certification regulatory mechanism urges enterprises to continuously improve their green innovation performance after being selected. On the other hand, for enterprises awarded Green Factory status, direct policy incentives such as R&D subsidies and tax preferences, as well as indirect benefits from enhanced reputation like increased stock prices and strengthened social trust, can effectively fill the funding gaps in their green technology R&D processes and provide strong support for improving their green innovation levels [31]. Furthermore, based on the Dynamic Capabilities Theory, enterprises need to continuously adjust and reshape their core competitiveness to flexibly respond to complex external environmental changes [13]. As a core component of dynamic capabilities, innovation helps enterprises adapt to environmental changes by reconstructing resources, serving as a key driver in shaping corporate resilience. The enhancement of green innovation capabilities equips certified enterprises with greater proactive technological preparedness and acute risk perception capabilities, enabling them to strategically deploy defensive measures prior to external shocks and effectively mitigate the adverse impacts of sudden crises [51]. Meanwhile, the endowment advantages accumulated through green innovation also help awarded enterprises quickly adjust their production processes, develop alternative solutions, and explore new markets after experiencing crises, thereby enhancing corporate flexibility [52]. The dual improvements in stability and flexibility will further contribute to the enhancement of corporate resilience [53].

Based on the aforementioned analysis, we propose the following hypotheses.

H1. 

Green Factory certification may enhance corporate resilience.

H2. 

Green Factory certification may alleviate corporate financing constraints, thereby enhancing corporate resilience.

H3. 

Green Factory certification may improve corporate resource allocation efficiency, thereby enhancing corporate resilience.

H4. 

Green Factory certification may promote corporate green innovation, thereby enhancing corporate resilience.

The theoretical framework of this paper is illustrated in Figure 1.

3. Research Design

3.1. Data Sources

Considering data availability, this paper selects Chinese A-share listed companies from 2008 to 2023 as the initial sample. The data are processed as follows: (1) excluding ST and PT firms; (2) excluding financial industry companies; (3) removing samples with missing key variables; (4) winsorizing continuous variables at the 1% and 99% percentiles. The final sample contains 39,596 valid observations from 4712 listed companies.

The original data on corporate resilience come from the CSMAR database. The Green Factory certification data were manually collected from the Green Factory lists published by China’s MIIT. These data were then processed using business registration information and matched with A-share listed companies to identify green factories affiliated with listed companies. Other data are sourced from the CSMAR and CNRDS databases.

3.2. Model Design and Variable Definitions

This paper considers the Green Factory certification implemented by China’s MIIT since 2017 as an exogenous policy shock. If a sample company has a subsidiary factory that has been selected into the Green Factory list, then the company is categorized as the treatment group; otherwise, it is categorized as the control group. Considering the varying certification times across different batches of green factories, the years in which the treatment group samples were included in the Green Factory list differ, making it unsuitable for the traditional single-point difference-in-differences (DID) model. Therefore, this paper constructs a multi-period DID model for testing. For details, refer to Model (1).

$$Resilienc{e}_{i,t}={\alpha }_0+{\alpha }_{1}GreenFactor{y}_{i,t}+{\alpha }_{2}Control{s}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(1)

In this model, i and t represent individual firms and years, respectively. Resilience_i,t represents the resilience level of listed firm i in year t. GreenFactory_i,t indicates whether sample firm i was selected into the Green Factory list in year t. Controls_i,t represents a set of other relevant control variables for sample firm i in year t. μ_i and γ_t represent firm fixed effects and year fixed effects, respectively, and ε_i,t is the random error term. α₁ is the coefficient of primary interest. If α₁ is significantly greater than 0, it suggests that being awarded Green Factory certification can effectively enhance corporate resilience. Additionally, to obtain more reliable results, all regressions in this paper employ cluster-robust standard errors at the firm level.

Dependent Variable: Corporate resilience (Resilience). As a dynamic capability to withstand shocks, maintain, and improve operations [33], corporate resilience is characterized by latency, path dependence, and multidimensionality [54], which makes it challenging to observe and measure directly. From an outcome-based perspective, certain scholars operationalize corporate resilience measurement by evaluating two interrelated dimensions across the long-term horizon: risk resistance capacity and post-crisis recovery capability. For instance, Markman and Venzin (2014) operationalize corporate resilience measurement using the standard deviation of firms’ long-term equity returns relative to industry peers, positing that greater resilience correlates with superior long-term financial performance and reduced volatility [55]. Ortiz-de-Mandojana and Bansal (2016) adopt a multi-indicator approach, incorporating financial volatility, sales growth rates, and survival rates to assess organizational resilience [4]. Building on this foundation, Zhang et al. (2023) [56], Luo et al. (2024) [57], and Zhang et al. (2024) [17] refine the measurement framework by employing financial volatility and long-term growth potential as dual dimensions of resilience assessment. Other scholars have focused on the immediate response and recovery capacity of firms under specific external shocks. For example, certain scholars have examined exogenous shocks such as the 2008 global financial crisis and the 2020 COVID-19 pandemic, operationalizing corporate resilience measurement through the magnitude and duration of stock price declines, coupled with the speed of post-crisis recovery [24,28,53]. In alignment with the research context of this paper and referencing mainstream measurement approaches for corporate resilience in the existing literature, we draw on the methodologies proposed by Ortiz-de-Mandojana and Bansal (2016) [4], Zhang et al. (2023) [56], Luo et al. (2024) [57], and Zhang et al. (2024) [17]. This study conceptualizes high corporate resilience as a dual-dimensional construct encompassing low volatility and high growth potential, employing stock price volatility and performance growth rates as proxy indicators for risk resistance capacity and post-crisis recovery capability, respectively. We measure resistance and growth using stock price volatility and performance growth rate, respectively. Specifically, stock price volatility is the standard deviation of a firm’s monthly stock returns over one year, with a smaller value indicating lower firm volatility. Performance growth rate is the cumulatively calculated sales growth rate of a firm over 3 years, with a higher value indicating stronger long-term growth capacity. Finally, the entropy weighting method is applied to the above two indicators to obtain the comprehensive score of corporate resilience (Resilience). To facilitate observation, the resilience score is multiplied by 100. Additionally, in the robustness test, this paper also measures corporate resilience from the two dimensions of loss degree and recovery capacity, using the outbreak of the COVID-19 pandemic (2020) within the study interval as a shock event.

Independent Variable: Green Factory certification (GreenFactory). During the study window (2008–2023), China’s MIIT announced eight batches of Green Factory lists. Based on this, we define GreenFactory as follows: if listed firm i is selected into the Green Factory list in year t, the variable is assigned a value of 1 in year t and all subsequent years; otherwise, it is assigned a value of 0.

Control Variables: With reference to the relevant literature [20,39,58] and considering the actual situation of firms, the following control variables are selected:

From the financial perspective, firm size reflects corporate resource endowment and risk resistance foundation, as larger enterprises may inherently possess greater resilience. We control for firm size (Size) using the natural logarithm of employee count. The debt-to-asset ratio (Lev), calculated as total liabilities over total assets, indicates financial leverage and debt repayment capacity. Moderate leverage enhances resource integration capacity and resilience, while excessive debt increases repayment pressure and liquidity constraints, thereby reducing resilience. Profitability stability, measured by return on assets (Roa) as net profit over total assets, strengthens resource accumulation for crisis response. Cash flow ratio (Cashflow), defined as net cash flow from operating activities divided by operating revenue, reflects short-term liquidity and shock absorption capacity. Fixed asset ratio (Fixed), calculated as net fixed assets over total assets, indicates asset structure rigidity—high fixed asset proportions may limit adaptability to environmental changes, constraining resilience improvement.

From the corporate governance perspective, board size (Board), measured as the natural logarithm of board members, reflects decision-making diversity and efficiency. Appropriately sized boards provide comprehensive risk management strategies critical for sustainable resilience. Independent director ratio (Indep), calculated as independent directors over total board members, measures board supervision independence. Independent directors mitigate managerial myopia and promote long-term resilience-building initiatives. The largest shareholder’s ownership (Top1), measured as the proportion of shares held by the largest shareholder, reflects equity concentration and shareholder–firm interest alignment. Major shareholders with high stakes are more motivated to implement risk management mechanisms. The separation of control rights and ownership rights (Separation), quantified as the difference between the ultimate controller’s control rights and ownership rights, is controlled to mitigate agency conflicts arising from principal-agent issues that may deter long-term resilience investments.

From the operational and institutional perspectives, management expense ratio (Mfee), calculated as management expenses divided by operating revenue, reflects organizational efficiency. Streamlined management reduces coordination costs in green factory implementation and enhances resource allocation efficiency. Institutional investor ownership (Inst), measured as the proportion of shares held by institutional investors, introduces professional oversight that facilitates sustainable development practices and resilience enhancement. State ownership (Soe), a dummy variable (1 for state-owned enterprises, 0 otherwise), captures institutional advantages such as preferential access to low-interest loans and government subsidies for green initiatives. Listing age (Listage), measured as the natural logarithm of years since IPO, reflects market experience and resource accumulation. Mature firms may possess established crisis response mechanisms but face potential organizational inertia, necessitating control for this temporal effect.

The definitions of the aforementioned variables are presented in

Table 1. Definition of variables.

Variable Type	Variable Name	Symbol	Variable Measurement
Dependent Variable	Corporate Resilience	Resilience	Entropy-weighted score of stock price volatility and performance growth rate.
Independent Variable	Green Factory certification	GreenFactory	Dummy variable that equals 1 if the firm is certified as a “Green Factory” in year t, and 0 otherwise.
Control variables	Firm size	Size	Natural logarithm of the number of employees.
	Asset–liability ratio	Lev	Total liabilities divided by total assets at fiscal year-end.
	Return on assets	Roa	Net profit divided by total assets
	Cash flow ratio	Cashflow	Net cash flow from operating activities divided by operating revenue.
	Fixed asset ratio	Fixed	Net fixed assets divided by total assets at period-end.
	Board size	Board	Natural logarithm of number of directors on the board.
	Independent directors	Indep	Proportion of independent directors.
	Largest shareholder ownership	Top1	Percentage of shares held by the largest shareholder.
	Separation rate of ownership and control rights	Separation	Difference between the ultimate controller’s control rights and ownership rights in the listed firm.
	Management expense ratio	Mfee	Management expenses divided by operating revenue.
	Institutional investor ownership	Inst	Percentage of shares held by institutional investors.
	Nature of property rights	Soe	Dummy variable that equals 1 for state-owned enterprises, and 0 otherwise.
	Firm listing age	Listage	Natural logarithm of years since IPO.

.

4. Empirical Results and Analysis

4.1. Descriptive Statistics

Table 2. Descriptive statistics.

Variable	Sample	Average	Standard Deviation	Minimum	Maximum
Resilience	39,596	56.70	0.802	27.41	99.82
GreenFactory	39,596	0.083	0.277	0	1
Size	39,596	7.620	1.274	4.357	11.08
Lev	39,596	0.444	0.204	0.062	0.916
Roa	39,596	0.030	0.067	−0.277	0.196
Cashflow	39,596	0.090	0.195	−0.759	0.709
Fixed	39,596	0.215	0.161	0.002	0.698
Board	39,596	2.282	0.253	1.609	2.890
Indep	39,596	0.382	0.074	0.250	0.600
Top1	39,596	0.340	0.148	0.090	0.740
Separation	39,596	4.755	7.390	0	28.81
Mfee	39,596	0.087	0.0742	0.008	0.469
Inst	39,596	44.30	24.06	0.435	90.87
Soe	39,596	0.385	0.487	0	1
Listage	39,596	2.421	0.613	1.099	3.434

presents the descriptive statistics of the main variables in this paper. The mean value of Resilience is 56.7, with a maximum of 99.82 and a minimum of 27.41, and a standard deviation of 0.802. This indicates that the resilience levels of the sampled enterprises exhibit significant disparities, with the overall analysis indicating substantial scope for improvement in their resilience capabilities. The mean value of GreenFactory is 0.0834, indicating that approximately 8.34% of the total sample firms have been awarded Green Factory status. The statistical values of other variables are generally consistent with existing research (The skewness and kurtosis statistics for all variables are detailed in Appendix B).

4.2. Baseline Regression

As shown in Appendix C, the variance inflation factor (VIF) values of all explanatory variables in Model (1) are less than 10, indicating no severe multicollinearity among the explanatory variables in the main model. The benchmark regression results in

Table 3. Baseline regression results.

Variable	(1)	(2)	(3)
	Resilience	Resilience	Resilience
GreenFactory	0.1307 ***	0.0349 ***	0.0253 ***
	(12.8778)	(3.5299)	(2.6269)
Size			0.0316 ***
			(8.2129)
Lev			0.0266 *
			(1.7569)
Roa			0.3087 ***
			(9.3941)
Cashflow			−0.0247 ***
			(−3.3935)
Fixed			−0.0347 *
			(−1.6498)
Board			−0.0076
			(−1.0592)
Indep			0.0013
			(0.0606)
Top1			−0.0016
			(−0.0514)
Separation			0.0003
			(0.6907)
Mfee			−0.2758 ***
			(−8.6678)
Inst			0.0009 ***
			(4.3334)
Soe			−0.0117
			(−1.1573)
Listage			0.0028
			(0.2584)
_cons	56.6702 ***	56.6782 ***	56.4253 ***
	(20,927.6009)	(68,662.3811)	(1264.8777)
Observations	39,596	39,596	39,596
Adj R2	0.0160	0.3971	0.4091
Year	No	Yes	Yes
Id	No	Yes	Yes

Notes: Values in parentheses are standard errors of clustering at the firm level; *** and * denote significance at the 1% and 10% levels, respectively. No is without fixed effects and Yes is with fixed effects.

show that when bidirectional fixed effects and control variables are not controlled in Column (1), the regression coefficient of the independent variable (GreenFactory) is 0.1307, which is significant at the 1% level. Furthermore, after sequentially controlling for bidirectional fixed effects and adding control variables in Columns (2) and (3), the regression coefficients of the independent variable (GreenFactory) become 0.0349 and 0.0253, respectively, and remain significant at the 1% level. From an economic perspective, the coefficient of 0.0253 in Column (3) suggests that, on average, enterprises certified as Green Factories have a resilience score 0.0253 points higher than that of non-certified ones, equivalent to a 3.15% (0.0253/0.802) difference relative to the standard deviation. These findings suggest a statistically significant positive relationship between Green Factory certification and corporate resilience, with the observed association being consistent with the relationship postulated in Hypothesis H1.

In addition, the direction and significance of the regression coefficients of the control variables are in line with theoretical predictions. For instance, both Size and Roa are significantly positively correlated with Resilience, indicating that the larger the company and the stronger its profitability, the more prominent its ability to resist external shocks and recover and grow after shocks. Other results are also basically consistent with existing relevant studies, once again supporting the rationality and reliability of the model construction and variable selection in this paper.

4.3. Robustness Test

4.3.1. Parallel Trend Test

The premise of using the difference-in-differences (DID) model is to satisfy the parallel trend assumption, which means that the treatment group and the control group should exhibit consistent trends in the absence of policy interventions. To this end, this paper refers to Beck et al. (2010) [59] to construct Model (2).

$$Resilienc{e}_{i,t}={\alpha }_0+{\displaystyle {\sum }_2^5{\beta }_{k}Pre\text{\_}{k}_{i,t}+{\displaystyle {\sum }_0^5{\beta }_{k}After\text{\_}{k}_{i,t}+{\alpha }_{1}Controls}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}}$$

(2)

Among them, Pre_k and After_k are the dummy variables before and after the Green Factory evaluation, respectively. Specifically, if a treated enterprise is in the k-th year before (after) being awarded the Green Factory status, Pre_k (After_k) is set to 1; otherwise, it is 0. For all control group enterprises, both Pre_k and After_k are 0. Given the long time span of the study, we categorize samples of treated enterprises that are more than 5 years before or after the Green Factory evaluation (k > 5) into the fifth period (k = 5). Meanwhile, to avoid multicollinearity, we use the period 1 year before the Green Factory evaluation (Pre_1) as the base period and exclude it from the regression. The meanings of other variables are the same as those in Model (1). The results in Figure 2 show that there is no systematic difference in corporate resilience between the treated group and the control group in the periods before the Green Factory evaluation (Pre_5 to Pre_2), and the parallel trend assumption holds. Furthermore, the subsequent impact of policies remains significant in the four periods after the Green Factory evaluation (After_1 to After_4). This indicates that with the continuous promotion of Green Factory construction and the implementation of related supporting policies, the Green Factory evaluation has increasingly enhanced corporate resilience, further supporting the research hypothesis H1 of this paper.

4.3.2. Placebo Test

To further mitigate the interference of other unobservable factors with the research results, this paper draws on the research by Callaway and Sant’Anna (2021) [60] by constructing a pseudo-Green Factory evaluation variable (GreenFactory_f) by randomly assigning treatment groups and shock years, and substituting it into Model (1) for regression testing. To enhance the credibility of this test, this paper repeats this process 1000 times. Figure 3 and Figure 4 show the distribution of the coefficients of GreenFactory_f and their t-values, respectively. Figure 3 illustrates that the regression coefficient curve of the pseudo-Green Factory evaluation follows a normal distribution, concentrated around 0, and all are smaller than the estimated coefficient value (0.0253) of the real benchmark regression in this paper. Figure 4 shows that the t-values of most estimated coefficients are less than 1.65, meaning they are not statistically significant. In conclusion, unobservable factors do not have a fundamental impact on the results of this paper.

4.3.3. Propensity Score Matching and Entropy Balancing Matching

There may be some systematic differences between the treatment group and the control group, which can affect enterprises’ attitudes towards and willingness to participate in Green Factory evaluations, potentially leading to self-selection bias. To mitigate this interference, on the one hand, this paper employs Propensity Score Matching-Difference-in-Differences (PSM-DID) for testing. Specifically, we use all the control variables in Model (1) as covariates to perform 1-to-3 nearest neighbor matching for the treatment and control groups. After passing the balance test, the successfully matched samples are substituted into the benchmark regression Model (1) for testing. In columns (1) and (2) of

Table 4. PSM-DID and EBM-DID.

Variable	PSM-DID		EBM-DID
	(1)	(2)	(3)
	Resilience	Resilience	Resilience
GreenFactory	0.0304 ***	0.0211 **	0.0213 **
	(2.8750)	(2.0591)	(2.0577)
Size		0.0458 ***	0.0416 ***
		(7.4967)	(6.1284)
Lev		0.0436 **	0.0559 **
		(1.9614)	(2.1657)
Roa		0.4646 ***	0.5363 ***
		(7.7490)	(8.2196)
Cashflow		−0.0239	−0.0368 **
		(−1.6348)	(−2.2370)
Fixed		−0.0252	0.0069
		(−0.8493)	(0.2064)
Board		−0.0018	0.0060
		(−0.1645)	(0.5328)
Indep		−0.0088	−0.0083
		(−0.2994)	(−0.2639)
Top1		0.0090	−0.0094
		(0.2054)	(−0.1994)
Separation		0.0001	−0.0000
		(0.1234)	(−0.0363)
Mfee		−0.5994 ***	−0.6068 ***
		(−8.7098)	(−7.8852)
Inst		0.0010 ***	0.0009 ***
		(3.7357)	(3.4743)
Soe		−0.0260 *	−0.0291 *
		(−1.6722)	(−1.6662)
Listage		−0.0037	0.0027
		(−0.2430)	(0.1540)
_cons	56.7036 ***	56.3298 ***	56.3239 ***
	(41,285.7996)	(880.7769)	(818.2251)
Observations	24,301	24,301	39,596
Adj R2	0.3872	0.4041	0.4109
Year	Yes	Yes	Yes
Id	Yes	Yes	Yes

Notes: Values in parentheses are standard errors of clustering at the firm level; ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Yes is with fixed effects.

, the Green Factory evaluation remains significantly positively correlated with corporate resilience, once again confirming the reliability of Hypothesis H1. On the other hand, to reduce the issue of sample loss caused by PSM, we draw on the research by Hainmueller (2012) [61] and adopt Entropy Balancing Matching-Difference-in-Differences (EBM-DID) for testing. Using all the control variables in this paper as covariates, we adjust and match the first, second, and third moments of the covariates of the treatment and control groups. After passing the balance test, we re-run the regression. In column (3) of Table 4, GreenFactory is significantly positive at the 5% level, further supporting the research conclusion that Green Factory evaluations enhance corporate resilience.

4.3.4. Lagged Explanatory Variables and Control Variables

Considering that the effect of being awarded a Green Factory certification on enhancing corporate resilience may not be immediate, and that high-resilience enterprises may be more likely to obtain Green Factory evaluations, there may be a potential reverse causality issue in this study. Therefore, drawing on the approach of Nie and Wang (2025) [9], we use lagged explanatory variables to conduct robustness tests. Furthermore, we simultaneously lag control variables for testing. Column (1) in

Table 5. Other robustness tests.

Variable	Resilience
	(1)	(2)	(3)	(4)
	Lagged Explanatory Variables and Control Variables	Adding Control Variables	Changing Fixed Effects	Excluding Interference from Other Policies
L.GreenFactory	0.0301 ***
	(2.8404)
GreenFactory		0.0305 **	0.0252 ***	0.0221 **
		(2.4831)	(2.5840)	(2.2816)
L_Size	−0.0018
	(−0.4095)
L_Lev	0.0613 ***
	(3.9515)
L_Roa	0.5230 ***
	(15.7364)
L_Cashflow	0.0064
	(0.8431)
L_Fixed	0.0003
	(0.0134)
L_Board	0.0110
	(1.3351)
L_Indep	0.0228
	(0.9181)
L_Top1	−0.0847 ***
	(−2.6148)
L_Separation	0.0004
	(0.7290)
L_Mfee	−0.1163 ***
	(−3.4548)
L_Inst	0.0009 ***
	(4.2842)
L_Soe	−0.0027
	(−0.2400)
L_Listage	0.0230 *
	(1.8792)
Size		0.0314 ***	0.0318 ***	0.0319 ***
		(7.5002)	(8.6509)	(8.3047)
Lev		0.0268	0.0236	0.0286 *
		(1.5153)	(1.5198)	(1.8886)
Roa		0.3123 ***	0.3024 ***	0.3018 ***
		(8.5387)	(9.0850)	(9.1955)
Cashflow		−0.0246 ***	−0.0242 ***	−0.0242 ***
		(−3.0315)	(−3.2383)	(−3.3352)
Fixed		−0.0261	−0.0340 *	−0.0368 *
		(−1.0881)	(−1.6476)	(−1.7659)
Board		−0.0050	−0.0062	−0.0080
		(−0.6137)	(−0.8523)	(−1.1165)
Indep		−0.0009	−0.0025	0.0012
		(−0.0369)	(−0.1188)	(0.0583)
Top1		−0.0634 *	−0.0161	−0.0024
		(−1.7604)	(−0.4993)	(−0.0766)
Separation		0.0005	0.0004	0.0003
		(0.9508)	(0.7300)	(0.6910)
Mfee		−0.2765 ***	−0.2727 ***	−0.2809 ***
		(−8.3152)	(−8.8467)	(−8.8129)
Inst		0.0009 ***	0.0008 ***	0.0009 ***
		(4.3813)	(4.4203)	(4.3816)
Soe		−0.0138	−0.0084	−0.0116
		(−1.4109)	(−0.9765)	(−1.1480)
Listage		−0.0015	0.0055	0.0039
		(−0.1219)	(0.5084)	(0.3623)
GDP_Growper		0.0020
		(0.0915)
CM2025				−0.0130 *
				(−1.6979)
EI				−0.0319 ***
				(−3.5542)
ETL				0.0503 ***
				(5.2876)
_cons	56.5673 ***	56.4471 ***	56.4231 ***	56.4214 ***
	(1133.2634)	(1193.5241)	(1338.4887)	(1267.5966)
Observations	34,191	33,426	39,596	39,596
Adj R2	0.4083	0.4235	0.4124	0.4097
Year	Yes	Yes	Yes	Yes
Id	Yes	Yes	Yes	Yes
Ind × Pro	No	No	Yes	No

Notes: Values in parentheses are standard errors of clustering at the firm level; ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. No is without fixed effects and Yes is with fixed effects.

shows that the estimated coefficient of the first-order lagged term of Green Factory evaluation (L.GreenFactory) is 0.0301, which is significant at the 1% level. This rules out the endogenous issue of reverse causality and indicates that there is indeed a certain lag in the empowerment of corporate resilience by Green Factory evaluations. The research conclusion of this paper remains robust.

4.3.5. Adding Control Variables and Changing Fixed Effects

On the one hand, given the differences in functional positioning, resource endowments, and environmental characteristics across regions [62], the evaluation criteria for green factories may exhibit regional differentiation features. Therefore, we further control the potential impact of the regional-level per capita GDP growth rate (GDP_Growper) on this study. On the other hand, to further mitigate the endogenous issues caused by omitted variables, this section adds industry–province interaction fixed effects on the basis of controlling for enterprise and year fixed effects in the benchmark regression. The aim is to further control for certain confounding factors that change with the interaction between industry and province but are difficult to observe, so as to more accurately identify the enhancement effect of Green Factory evaluations on corporate resilience. The regression results of the above two methods are shown in columns (2) and (3) of Table 5. GreenFactory and Resilience still show a significant positive correlation, once again confirming Hypothesis H1.

4.3.6. Excluding Interference from Other Policies

During the study period, corporate resilience may also be affected by other policies, which may “crowd out” the policy empowerment effect of Green Factory evaluations. Based on this, this paper compiles three representative policies that overlap with the sample interval: the “Made in China 2025” pilot demonstration cities (CM2025), central environmental protection inspections (EI), and the Environmental Protection Tax Law (ETL), and incorporates them into the benchmark regression for control (The construction of the three policy variables is described as follows: (1) “Made in China 2025” (CM2025): Samples from pilot and demonstration cities (e.g., Ningbo and 11 other cities, as well as four city clusters, including a cluster of five cities in southern Jiangsu) designated by the MIIT are assigned to the treatment group, while samples from other regions serve as the control group. A time dummy variable is constructed, with 2017—the year when the policy was proposed—serving as the cutoff year. Finally, an interaction term is generated to create the policy proxy variable CM2025 for “Made in China 2025”. (2) Central Environmental Protection Inspection (EI): The treatment and control groups are defined based on whether an area underwent the Central Environmental Protection Inspection. A time dummy variable is constructed using the year of inspection, and an interaction term is generated to create the policy proxy variable EI for the Central Environmental Protection Inspection. (3) Environmental Protection Tax Law (ETL): Following the approach of Liu et al. (2022) [63], a time dummy variable is constructed with 2018—the year when the Environmental Protection Tax Law came into effect—serving as the cutoff year. The treatment and control groups are determined based on whether an enterprise belongs to heavily polluting industries. An interaction term is generated to create the policy proxy variable ETL for the Environmental Protection Tax Law). The results in column (4) of Table 5 show that after controlling for the impact of other potential policies, Green Factory evaluations can still help enhance corporate resilience.

4.3.7. Adjusting the Sample Period

The study interval in this paper covers three major systemic unexpected crisis events: the 2008 financial crisis, the 2015 stock market crash, and the COVID-19 pandemic that broke out in early 2020. These events may trigger extreme fluctuations in corporate resilience and interfere with the research conclusions. To address this, we re-run the benchmark regression after separately and simultaneously excluding the periods related to the above events. As shown in

Table 6. Adjusting the sample period.

Variable	Resilience
	(1)	(2)	(3)	(4)
	Excluding the Financial Crisis	Excluding the Stock Market Crash	Excluding the COVID-19 Pandemic	Excluding All the Aforementioned
GreenFactory	0.0231 **	0.0279 ***	0.0501 ***	0.0503 ***
	(2.4287)	(2.9519)	(2.7321)	(2.8387)
Size	0.0311 ***	0.0309 ***	0.0351 ***	0.0323 ***
	(7.3723)	(8.1648)	(7.9609)	(6.3297)
Lev	0.0259	0.0249	0.0645 ***	0.0687 ***
	(1.6029)	(1.6066)	(3.4504)	(3.2744)
Roa	0.3114 ***	0.3183 ***	0.3373 ***	0.3848 ***
	(9.1449)	(9.6806)	(8.9648)	(9.8736)
Cashflow	−0.0251 ***	−0.0207 ***	−0.0305 ***	−0.0256 ***
	(−3.2166)	(−2.8327)	(−3.7613)	(−2.7876)
Fixed	−0.0504 **	−0.0264	−0.0548 **	−0.0607 **
	(−2.2182)	(−1.2322)	(−2.3137)	(−2.1884)
Board	−0.0072	−0.0059	−0.0036	0.0011
	(−0.9940)	(−0.7915)	(−0.4042)	(0.1135)
Indep	0.0004	−0.0066	0.0156	0.0010
	(0.0170)	(−0.3077)	(0.6090)	(0.0385)
Top1	−0.0006	−0.0034	0.0180	−0.0023
	(−0.0162)	(−0.1057)	(0.4883)	(−0.0526)
Separation	0.0002	0.0003	0.0008	0.0005
	(0.2910)	(0.5253)	(1.4786)	(0.7813)
Mfee	−0.2905 ***	−0.2700 ***	−0.2622 ***	−0.2655 ***
	(−8.7426)	(−8.4576)	(−7.0913)	(−6.7097)
Inst	0.0009 ***	0.0010 ***	0.0010 ***	0.0012 ***
	(4.1716)	(4.8045)	(4.3620)	(4.8961)
Soe	−0.0127	−0.0154	−0.0243 *	−0.0359 **
	(−1.2193)	(−1.5406)	(−1.7114)	(−2.1714)
Listage	0.0084	−0.0040	−0.0097	−0.0070
	(0.7414)	(−0.3747)	(−0.6908)	(−0.4493)
_cons	56.4289 ***	56.4615 ***	56.3805 ***	56.4408 ***
	(1161.0657)	(1278.4231)	(1075.0248)	(948.3058)
Observations	37,121	37,264	24,984	20,177
Adj R2	0.4057	0.3712	0.4835	0.4194
Year	Yes	Yes	Yes	Yes
Id	Yes	Yes	Yes	Yes

Notes: Values in parentheses are standard errors of clustering at the firm level; ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Yes is with fixed effects.

, the Green Factory evaluation remains significantly positively correlated with corporate resilience, indicating that Hypothesis H1 is robust.

4.3.8. Changing the Explained Variable

Drawing on the research by DesJardine et al. (2019) [53] and Sajko et al. (2021) [24], we focus on specific crises to capture corporate resilience. Specifically, this paper uses the global public health emergency of COVID-19 as the research context, selects the period from 2019 to 2022 as the research window, and measures enterprise resistance and recovery capacity by using the magnitude of stock price decline (Res_dec) caused by the pandemic and whether the stock price has recovered to the pre-crisis peak level (Res_rec) to gauge the level of corporate resilience. Model (3) is the formula for calculating the magnitude of stock price decline (Res_dec) of enterprises under the impact of the pandemic.

$$Res\text{\_}de{c}_{i,t}={\displaystyle \frac{P_{i,t}-P_{i,pre}}{P_{i,pre}}}$$

(3)

Res_dec_i,t is the magnitude of stock price decline for enterprise i in year t, representing the enterprise’s resistance. P_i,t and P_i,pre represent the lowest stock price of enterprise i in year t after the pandemic and the highest stock price before the pandemic (2019), respectively. The larger the value of Res_dec, the smaller the negative impact of the pandemic on the enterprise, the stronger the resistance, and the stronger the resilience.

At the same time, we construct a dummy variable (Res_rec) to measure the enterprise’s recovery capacity based on whether the highest stock price of the enterprise after the pandemic exceeds the highest stock price before the pandemic. Finally, Res_dec and Res_rec are substituted into the benchmark regression Model (1) for testing. In

Table 7. Replacing the dependent variable.

Variable	(1)	(2)	(3)	(4)
	Res_dec	Res_dec	Res_rec	Res_rec
GreenFactory	0.0413 **	0.0350 **	0.0474 *	0.0490 *
	(2.2341)	(1.9784)	(1.6473)	(1.7614)
Size		0.0637 ***		−0.0665 ***
		(4.3981)		(−2.7575)
Lev		−0.0129		0.2232 ***
		(−0.2537)		(2.6938)
Roa		0.5475 ***		0.8096 ***
		(9.4146)		(6.6597)
Cashflow		0.0355 **		−0.0445
		(2.2637)		(−1.2134)
Fixed		−0.3805 ***		−0.4715 ***
		(−5.1409)		(−3.7157)
Board		0.0158		0.0176
		(1.3012)		(0.6752)
Indep		0.0270		0.0838
		(0.6894)		(1.0506)
Top1		−0.2355 ***		−0.4587 ***
		(−2.5867)		(−2.6819)
Separation		−0.0022 *		−0.0014
		(−1.7248)		(−0.7150)
Mfee		−0.1763 **		−0.2674
		(−2.0931)		(−1.2532)
Inst		0.0049 ***		0.0056 ***
		(8.5997)		(5.5743)
Soe		−0.0556 **		−0.1079 **
		(−2.0353)		(−2.0400)
Listage		−0.4052 ***		−0.7680 ***
		(−8.5769)		(−8.0257)
_cons	−0.3747 ***	0.0900	0.5279 ***	2.8703 ***
	(−116.4119)	(0.5681)	(105.3209)	(9.5629)
Observations	9764	9764	9764	9764
Adj R2	0.6007	0.6294	0.4662	0.4858
Year	Yes	Yes	Yes	Yes
Id	Yes	Yes	Yes	Yes

Notes: Values in parentheses are standard errors of clustering at the firm level; ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Yes is with fixed effects.

, GreenFactory is significantly positively correlated with both Res_dec and Res_rec, indicating that enterprises awarded Green Factory certification experience smaller stock price declines and their stock prices are more likely to recover after the pandemic, which once again supports the main research conclusion that Green Factory evaluations can enhance corporate resilience.

4.3.9. Heterogeneous Treatment Effect Test for Multi-Time-Point DID

Under two-way fixed effects (TWFE), the estimation results of the multi-time-point difference-in-differences (DID) model may have the problem of negative weight heterogeneity of treatment effects, leading to estimation bias [64,65]. Therefore, this paper uses the decomposition method proposed by De Chaisemartin and D’Haultfoeuille (2020) [64] to further diagnose the robustness of the estimated values of the multi-time-point DID Model (1). The results show that among all 3225 weights, there are 3042 positive weights and 183 negative weights. The sum of positive weights is 1.0049, and the sum of negative weights is −0.0049. The proportion of negative weights is very small, which to a certain extent indicates that the heterogeneous treatment effects have no substantive impact on the estimation results. The research conclusion of this paper is robust.

4.4. Impact Mechanism Testing

Based on the aforementioned theoretical analysis, obtaining a Green Factory certification helps alleviate corporate financing constraints, improve resource allocation efficiency, and promote green innovation, thereby exerting an enabling effect on corporate resilience. To verify these logical linkages, we draw on the research by Wen and Ye (2014) [66] to construct Models (4) and (5) for mechanism testing.

$$M_{i,t}={\phi }_0+{\phi }_{1}GreenFactor{y}_{i,t}+{\phi }_{2}Control{s}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(4)

$$Resilienc{e}_{i,t}={\theta }_0+{\theta }_{1}GreenFactor{y}_{i,t}+{\theta }_2{M}_{i,t}+{\theta }_{3}Control{s}_{i,t}+{\mu }_i+{\gamma }_t+{\epsilon }_{i,t}$$

(5)

M represents the proxy variables for the three mechanisms of financing constraints, resource allocation, and green innovation. The meanings of the remaining symbols are consistent with those in the aforementioned model.

4.4.1. Alleviating Financing Constraints

As previously mentioned, Green Factory certification may alleviate financing constraints through both direct incentives and indirect signaling, thereby enhancing corporate resilience. To test the above theoretical logic, we employ the KZ index [67] to measure financing constraints and perform regressions on Models (4) and (5) to examine the mediating effect of financing constraints in the process through which Green Factory certification influences corporate resilience. The coefficient of GreenFactory in column (1) of

Table 8. Mechanism test.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	KZ	Resilience	TFP	Resilience	GreenInno	Resilience
GreenFactory	−0.1763 ***	0.0209 **	0.0804 ***	0.0152 *	0.0617 ***	0.0251 ***
	(−4.4709)	(2.1807)	(5.1762)	(1.6534)	(2.6899)	(2.6197)
KZ		−0.0221 ***
		(−16.6784)
TFP				0.1318 ***
				(19.1801)
GreenInno						0.0084 **
						(2.1869)
Size	−0.1123 ***	0.0289 ***	0.0119	0.0303 ***	0.0393 ***	0.0315 ***
	(−5.2705)	(7.4033)	(0.9622)	(8.0948)	(5.4463)	(8.1622)
Lev	5.6616 ***	0.1540 ***	0.5511 ***	−0.0437 ***	0.0159	0.0259 *
	(52.1774)	(8.8696)	(12.6267)	(−2.9057)	(0.5074)	(1.7089)
Roa	−5.7938 ***	0.1845 ***	1.0061 ***	0.1816 ***	0.1029 *	0.3071 ***
	(−28.6357)	(5.7671)	(14.1452)	(5.6170)	(1.8432)	(9.3348)
Cashflow	−3.8180 ***	−0.1097 ***	−0.0195	−0.0225 ***	−0.0413 ***	−0.0246 ***
	(−46.4422)	(−12.8901)	(−1.0352)	(−2.8651)	(−3.1802)	(−3.3595)
Fixed	2.8092 ***	0.0259	−1.0100 ***	0.0973 ***	−0.0398	−0.0345
	(19.4213)	(1.1884)	(−14.6224)	(4.5486)	(−0.8882)	(−1.6240)
Board	−0.0258	−0.0081	0.0485 ***	−0.0141 **	−0.0215	−0.0079
	(−0.6475)	(−1.1184)	(3.4425)	(−1.9787)	(−1.2920)	(−1.0958)
Indep	0.1127	0.0045	−0.0335	0.0027	0.0128	−0.0005
	(0.9973)	(0.2197)	(−0.8437)	(0.1321)	(0.2703)	(−0.0258)
Top1	−0.4400 **	−0.0103	−0.1680 **	0.0267	−0.0906	−0.0019
	(−2.4646)	(−0.3289)	(−2.1027)	(0.8919)	(−1.2834)	(−0.0593)
Separation	0.0000	0.0003	−0.0010	0.0005	0.0017 *	0.0003
	(0.0113)	(0.6880)	(−0.8460)	(1.0833)	(1.6800)	(0.6426)
Mfee	0.3949	−0.2677 ***	−4.7657 ***	0.3415 ***	−0.2460 ***	−0.2765 ***
	(1.4396)	(−8.4885)	(−39.3979)	(7.4327)	(−3.5111)	(−8.5995)
Inst	−0.0042 ***	0.0008 ***	0.0046 ***	0.0003	−0.0009 **	0.0009 ***
	(−3.7046)	(3.9743)	(9.5951)	(1.4033)	(−2.2436)	(4.3743)
Soe	0.0180	−0.0111	−0.0862 ***	0.0012	0.0195	−0.0123
	(0.2990)	(−1.1035)	(−3.3336)	(0.1267)	(0.8585)	(−1.2168)
Listage	1.1217 ***	0.0297 ***	0.1305 ***	−0.0152	−0.0509 *	0.0046
	(16.3623)	(2.7299)	(4.7138)	(−1.4392)	(−1.7094)	(0.4269)
_cons	−2.8033 ***	56.3587 ***	6.4621 ***	55.5724 ***	0.2680 ***	56.4208 ***
	(−11.1727)	(1239.4499)	(57.0335)	(831.9848)	(2.6465)	(1257.5594)
Observations	38,889	38,889	38,629	38,629	39,330	39,330
Adj R2	0.7874	0.4156	0.8820	0.4306	0.6517	0.4095
Year	Yes	Yes	Yes	Yes	Yes	Yes
Id	Yes	Yes	Yes	Yes	Yes	Yes

Notes: Values in parentheses are standard errors of clustering at the firm level; ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Yes is with fixed effects.

is significantly negative at the 1% level, indicating that the financing environment of enterprises has been effectively improved after being awarded the Green Factory certification. Furthermore, the coefficient of KZ in column (2) is significantly negative at the 1% level, while the coefficient of GreenFactory is significantly positive at the 5% level, suggesting that Green Factory certification can enhance corporate resilience by alleviating financing constraints, thus verifying Hypothesis H2.

4.4.2. Improving Resource Allocation

This paper posits that the high standards, stringent requirements, and long-term oversight associated with Green Factory certification motivate enterprises to continuously optimize their production portfolio and factor allocation, thereby helping them better withstand and adapt to external environmental changes and enhance their resilience. In this study, total factor productivity (TFP) estimated using the Olley–Pakes (OP) method [68] is employed as a proxy variable for enterprise resource allocation efficiency. Models (4) and (5) are utilized to examine the mediating effect of enterprise allocation efficiency in the process through which Green Factory certification influences corporate resilience. The coefficient of TFP in column (3) of Table 8 is significantly positive at the 0.01 level, indicating that Green Factory certification indeed promotes the improvement of enterprise resource allocation efficiency. The coefficient of TFP in column (4) of Table 8 remains significantly positive at the 0.01 level, implying that Green Factory certification can indeed enhance corporate resilience by improving resource allocation efficiency, thus supporting Hypothesis H3.

4.4.3. Promoting Green Innovation

After being awarded Green Factory certification, enterprises are subject to stringent environmental regulations and long-term oversight, compelling them to increase their investment in green R&D and enhance their level of green innovation. This process may foster stronger crisis resilience and growth recovery capabilities, demonstrating a high level of corporate resilience. Drawing on the research by Liu et al. (2025) [8], this paper measures the green innovation level of enterprises (GreenInno) using the natural logarithm of the number of green patent applications plus one. The coefficient of GreenFactory in column (5) of Table 8 is significantly positive at the 1% level, indicating that being awarded Green Factory certification promotes the enhancement of enterprises’ green innovation levels. The coefficient of GreenInno in column (6) is significantly positive at the 0.05 level, suggesting that Green Factory certification does enhance corporate resilience by promoting green innovation, and Hypothesis H4 is thereby verified.

5. Further Research

5.1. Heterogeneity Analysis

5.1.1. Pollution Attribute

Compared with non-heavily polluting enterprises, heavily polluting enterprises that actively participate in the evaluation and are selected as Green Factories can send stronger positive signals to the outside world, such as their willingness for green transformation, competitive advantages over peers, and green development potential. This enables them to obtain more positive external resources such as funding, social recognition, and environmental reputation. Furthermore, heavily polluting enterprises awarded Green Factory certification usually face higher levels of supervision and more frequent inspections. Therefore, under the dual influence of resource incentives and policy supervision, heavily polluting enterprises are more motivated to accelerate green transformation and upgrading and respond to the requirements of Green Factory construction, making green manufacturing demonstrate more prominent policy effectiveness. In this paper, the full sample is divided into two subgroups: heavily polluting and non-heavily polluting enterprises. The grouped regression results in columns (1) and (2) of

Table 9. Heterogeneity analysis.

Variable	Resilience
	Pollution Attribute		Reputational Pressure		Managerial Myopia
	(1)	(2)	(3)	(4)	(5)	(6)
	Heavily Polluting	Non-Heavily Polluting	High Reputational Pressure	Low Reputational Pressure	High Managerial Myopia	Low Managerial Myopia
GreenFactory	0.0404 ***	0.0090	0.0432 ***	0.0139	0.0439 ***	0.0139
	(2.8641)	(0.6765)	(2.6276)	(1.5290)	(3.6702)	(1.0574)
Size	0.0264 ***	0.0298 ***	0.0428 ***	0.0251 ***	0.0295 ***	0.0335 ***
	(4.0246)	(7.1931)	(7.2678)	(5.9732)	(5.6039)	(6.1136)
Lev	0.0066	0.0420 **	0.0679 ***	0.0115	0.0137	0.0457 **
	(0.2175)	(2.3065)	(2.6012)	(0.6899)	(0.6936)	(2.1197)
Roa	0.3532 ***	0.2855 ***	0.4778 ***	0.0599 *	0.3055 ***	0.2978 ***
	(5.4225)	(7.5048)	(8.9122)	(1.7422)	(7.0135)	(6.1390)
Cashflow	−0.0431 **	−0.0220 ***	−0.0242 **	−0.0341 ***	−0.0208 **	−0.0302 **
	(−2.1937)	(−2.7754)	(−1.9702)	(−4.0218)	(−2.1131)	(−2.4892)
Fixed	−0.0027	−0.0331	−0.0473	−0.0265	−0.0336	−0.0083
	(−0.0792)	(−1.2988)	(−1.4018)	(−1.0860)	(−1.2462)	(−0.2698)
Board	0.0082	−0.0120	0.0010	−0.0037	−0.0071	−0.0045
	(0.6247)	(−1.4043)	(0.0800)	(−0.4667)	(−0.7166)	(−0.4066)
Indep	0.0348	−0.0080	−0.0278	0.0033	0.0012	−0.0026
	(0.8190)	(−0.3345)	(−0.7525)	(0.1409)	(0.0420)	(−0.0814)
Top1	0.0482	−0.0236	−0.0724	0.0960 ***	−0.0124	0.0009
	(0.7682)	(−0.6661)	(−1.5360)	(2.6064)	(−0.2809)	(0.0229)
Separation	0.0003	0.0002	0.0006	−0.0004	0.0005	0.0000
	(0.3922)	(0.3394)	(0.7822)	(−0.7529)	(0.7213)	(0.0729)
Mfee	−0.2149 ***	−0.2948 ***	−0.2357 ***	−0.2879 ***	−0.2908 ***	−0.2747 ***
	(−2.8692)	(−8.7909)	(−4.6763)	(−7.4545)	(−7.3808)	(−5.5986)
Inst	0.0005	0.0007 ***	0.0013 ***	0.0003	0.0009 ***	0.0009 ***
	(1.5773)	(3.5051)	(4.3203)	(1.5134)	(3.5627)	(3.0251)
Soe	0.0089	−0.0080	−0.0309 **	−0.0088	−0.0074	−0.0179
	(0.4822)	(−0.8597)	(−2.3489)	(−0.8714)	(−0.5053)	(−1.3115)
Listage	0.0159	−0.0021	0.0421 **	0.0034	−0.0003	0.0103
	(0.7662)	(−0.1695)	(2.1421)	(0.3042)	(−0.0207)	(0.6648)
_cons	56.3732 ***	56.4716 ***	56.2068 ***	56.4788 ***	56.4519 ***	56.3800 ***
	(736.6603)	(1166.1551)	(759.6760)	(1210.8096)	(962.5342)	(855.1982)
Difference between groups (p-Value)	0.000		0.000		0.000
Observations	11,515	28,081	19,610	18,449	19,383	19,944
Adj R2	0.3731	0.4358	0.4208	0.4391	0.4119	0.4171
Year	Yes	Yes	Yes	Yes	Yes	Yes
Id	Yes	Yes	Yes	Yes	Yes	Yes

Notes: Values in parentheses are standard errors of clustering at the firm level; ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Yes is with fixed effects.

show that the regression coefficient of Green Factory certification is only significantly positive in the heavily polluting enterprise group, indicating that the effect of Green Factory certification on enhancing the resilience of heavily polluting enterprises is stronger.

5.1.2. Reputational Pressure

As a highly influential “green business card” in the new era, Green Factories have become a catalyst for comprehensively promoting corporate green philosophies and profoundly enhancing corporate images. Compared with enterprises with a good reputation, enterprises with a poor reputation may be more urgently in need of this “green business card” for reputation repair, in order to attract the attention and favor of stakeholders, thereby boosting market confidence and overcoming current difficulties. Considering that these enterprises have stronger motivations and clearer goals, we hypothesize that the enabling effect of Green Factory certification on corporate resilience may be more evident in these samples. Drawing on the research by Long et al. (2024) [69], this paper measures reputational pressure using the natural logarithm of the number of annual negative media reports on enterprises and divides the full sample into two groups: high reputational pressure and low reputational pressure, based on the industry’s annual median. The grouped regression results in columns (3) and (4) of Table 9 show that Green Factory certification only significantly enhances corporate resilience in the high reputational pressure group, supporting the above inference. This indicates that being awarded Green Factory certification indeed has an additional reputation repair effect on enterprises with a poor image, thereby presenting a stronger resilience enhancement effect.

5.1.3. Managerial Myopia

Managerial cognition significantly shapes behavioral decision-making and becomes internalized within corporate strategic practices. Myopic managers tend to prioritize short-term gains over long-term development, undermining the formation of sustainable competitive advantages essential for navigating the VUCA business environment. For firms awarded green factory certification, the attendant policy incentives and resource advantages may compensate for resilience deficits arising from managerial myopia, thereby generating more pronounced resilience-enhancing effects. To empirically validate this proposition, we adopt Zhang et al. (2023) [70] methodology by measuring managerial myopia through the short-sighted word frequency ratio in firms’ annual report Management Discussion & Analysis (MD&A) sections. A higher ratio indicates greater managerial myopia. We then categorize the full sample into high-myopia and low-myopia subgroups using industry-year median thresholds. The grouped regression results presented in Table 9, Columns (5) and (6), reveal that Green Factory certification significantly enhances corporate resilience only within the high-myopia subgroup. This finding underscores the capacity of green manufacturing initiatives to effectively compensate for cognitive limitations and associated negative externalities within corporate leadership.

5.2. Spillover Effect Test

The policy incentives and significant reputational boost brought by Green Factory certification are highly attractive to enterprises, strongly motivating them to actively participate in the competition for Green Factory certification. However, compared with enterprises that successfully make it onto the demonstration list, those that fail to be selected or do not participate in the evaluation often view this as a development threat and a warning signal of their own development lag. Based on this, these enterprises are likely to develop a stronger willingness for green development, actively explore innovative ideas, and accelerate the pace of green product R&D and production, hoping to bridge the gap with certified enterprises and catch up and surpass them in the green development race. Therefore, Green Factory certification can further stimulate “green involution” behavior among enterprises in the same industry and region, giving rise to environmental protection races between industries and regions [71], and ultimately exerting spillover effects on corporate resilience. To verify this logic, drawing on the research by Leary and Roberts (2014) [72], this paper replaces the explained variables in Model (1) with the average resilience of other enterprises in the same industry (Res_spillInd) and the average resilience of other enterprises in the same city (Res_spillCity). The regression coefficients of GreenFactory in

Table 10. Spillover Effect Test.

Variable	(1)	(2)	(3)	(4)
	Res_spillInd	Res_spillInd	Res_spillCity	Res_spillCity
GreenFactory	0.0090 ***	0.0072 **	0.0059 **	0.0059 **
	(2.7823)	(2.2602)	(2.0121)	(2.0259)
Size		0.0009		−0.0002
		(0.6605)		(−0.1669)
Lev		0.0114 **		0.0012
		(1.9623)		(0.2183)
Roa		0.1012 ***		0.0047
		(8.3332)		(0.4401)
Cashflow		0.0058 **		−0.0018
		(2.0351)		(−0.7886)
Fixed		0.0278 ***		−0.0187 **
		(3.2255)		(−2.3767)
Board		−0.0015		−0.0009
		(−0.5636)		(−0.3756)
Indep		0.0107		−0.0116
		(1.3448)		(−1.5680)
Top1		−0.0189 *		−0.0070
		(−1.7493)		(−0.6886)
Separation		0.0002		0.0001
		(0.9623)		(0.3873)
Mfee		−0.0066		0.0143
		(−0.5366)		(1.1680)
Inst		0.0001		0.0000
		(1.0895)		(0.3911)
Soe		0.0004		−0.0069 *
		(0.1066)		(−1.8863)
Listage		−0.0079 *		−0.0012
		(−1.9399)		(−0.3087)
_cons	56.6805 ***	56.6795 ***	56.6811 ***	56.6977 ***
	(208,803.0667)	(3243.2786)	(231,498.0972)	(4107.1628)
Observations	39,520	39,520	37,439	37,439
Adj R2	0.7174	0.7190	0.6996	0.6997
Year	Yes	Yes	Yes	Yes
Id	Yes	Yes	Yes	Yes

Notes: Values in parentheses are standard errors of clustering at the firm level; ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. Yes is with fixed effects.

are all significantly positive, indicating that the enabling effect of Green Factory certification on corporate resilience is not limited to the certified enterprises themselves but can effectively radiate to other enterprises, demonstrating industry and spatial spillover effects. This further supports the scientific validity and necessity of green manufacturing system construction.

6. Conclusions, Suggestions, and Prospects

6.1. Main Conclusions

In the increasingly complex, uncertain, and dynamic external environment, corporate resilience has become crucial for enterprises to withstand shocks and achieve long-term survival. Meanwhile, global sustainable development has emerged as the trend of the times, and accelerating green manufacturing transformation is an inevitable choice for enterprises to adapt to this trend. Based on this, starting from Green Factory certification, a key initiative in the construction of China’s green manufacturing system, this paper utilizes data from A-share listed companies from 2008 to 2023 to examine the impact, action paths, and spillover effects of green manufacturing on corporate resilience. The main research conclusions are as follows: (1) Being awarded Green Factory certification can significantly enhance corporate resilience; (2) Green Factory certification exerts an enabling effect on corporate resilience by alleviating financing constraints, improving resource allocation efficiency, and promoting green innovation; (3) The effect of Green Factory certification on enhancing corporate resilience is more pronounced in heavily polluting industries, firms with low reputations, and those with higher levels of managerial myopia; (4) Green Factory certification has spillover effects, further incentivizing other enterprises in the same industry and city to collaboratively enhance their resilience levels. This paper broadens the research horizon on how green manufacturing policies can empower micro-enterprises and points out directions for enterprises to withstand external shocks and enhance their resilience.

6.2. Suggestions

To facilitate the synergistic development of green manufacturing and corporate resilience, we propose the following suggestions.

Firstly, promote the global unification of green manufacturing certification standards and establish a comprehensive and diversified incentive mechanism. International organizations such as the United Nations Industrial Development Organization (UNIDO) can collaborate with governments worldwide to develop a globally applicable green manufacturing certification standards system. Clear and quantifiable requirements should be set for core indicators such as energy utilization efficiency, pollutant emission control, and green supply chain management to eliminate differences and barriers between standards in different countries and regions, providing enterprises with clear and consistent directions for their efforts and promoting the coordinated development of global green manufacturing. Simultaneously, globally unified certification standards will also assist enterprises in obtaining broader and more diversified incentives. On the one hand, governments and international financial institutions can create exclusive green financing channels for enterprises that meet the unified standards, offering green credit products with preferential interest rates and relaxed credit limits to address the funding shortages of enterprises’ green manufacturing projects and help them expand their green production scale. On the other hand, green manufacturing certification issued based on unified standards will also serve as a “green business card” for enterprises in the global market. Certified enterprises should be able to enjoy tariff reductions, priority customs clearance, and other facilitation policies in international trade, enhancing their products’ international competitiveness and expanding overseas market share, thereby achieving economies of scale and synergistic effects in green manufacturing on a global scale, forming a virtuous cycle, and continuously promoting enterprises to increase their investment and innovation in the field of green manufacturing.

Secondly, promote industry collaboration and regional linkage in green manufacturing. Encourage various industries to establish green manufacturing alliances on a global scale, formulating green self-regulatory guidelines and collaborative development plans within the industry. Member enterprises of the alliances can achieve mutual assistance and jointly enhance their green manufacturing levels and corporate resilience by sharing resources such as green technologies, raw material procurement channels, and market information. Simultaneously, the alliances can communicate and coordinate with governments and international organizations on behalf of the industry to strive for a more favorable policy environment. Actively promote regional green development cooperation. Governments, enterprises, and scientific research institutions in neighboring regions should strengthen cooperation to jointly promote green infrastructure construction, green industrial chain building, and ecological environmental governance within the region: for example, by establishing a cross-regional green logistics system, optimizing resource allocation, reducing enterprise operating costs, and achieving an overall improvement in the resilience of enterprises within the region, forming a regional collaborative development model for green manufacturing that can be replicated and promoted globally.

Thirdly, strengthen the integration of corporate resilience management systems with green manufacturing. International organizations, governments, and industry alliances can guide enterprises to establish a corporate resilience assessment system that incorporates green manufacturing indicators. Regularly and quantitatively assess enterprises in terms of resource flexibility (such as the diversification of raw material supply and the stability of green energy acquisition), operational flexibility (the adaptability of green production processes to market fluctuations and environmental policy changes), and corporate resilience (the ability of green innovation culture to respond to internal changes and external shocks) during the green manufacturing process. Based on the assessment results, enterprises can adjust their green manufacturing strategies in a timely manner, optimize resource allocation, and ensure that they maintain a high level of resilience in a complex and dynamic environment. Simultaneously, enterprises should actively cultivate a resilience culture oriented towards green manufacturing. Integrate green values into corporate strategies, organizational structures, and employee daily behavior codes, so that all employees deeply understand the importance of green manufacturing for the long-term survival and development of the enterprise. In the face of external shocks, this culture can prompt the enterprise to quickly form a cohesive force, actively utilize the advantages of green manufacturing, such as the stability of the green supply chain and the differentiated competitiveness brought by green innovation, to effectively resist risks and enhance corporate resilience.

6.3. Prospects

This paper has unveiled the significant role of China’s Green Factory certification policy in enhancing corporate resilience, yet acknowledges certain limitations. Firstly, due to disparities in policies across different countries and regions, the applicability of the conclusions drawn in this study may be somewhat constrained. Future research could broaden its scope by investigating green manufacturing policies in various national and regional contexts, thereby enhancing the generalizability of the findings. Secondly, in the VUCA era, supply chain collaboration is crucial for enterprises to survive and withstand shocks, and future research can further extend to the enabling effects of green supply chains on corporate resilience.