The Impact of Green Finance Policies on Corporate Debt Default Risk—Evidence from China

Abstract

As global climate change issues have become increasingly severe, green finance has gained widespread attention from governments and financial institutions as a crucial tool for promoting sustainable development. This paper explores the impact of green finance reform pilot zones on corporate debt default risks based on a difference-in-differences model. We found that green finance policies significantly increase corporate debt default risks by exacerbating financing constraints and reducing stock liquidity. A heterogeneity analysis revealed that polluting enterprises, non-state-owned enterprises, and companies in the Eastern region are more susceptible to the impacts of this policy. This paper suggests that the government should formulate differentiated green finance policies tailored to different types of enterprises and regional characteristics.

1. Introduction

With the growing severity of global climate change, green finance has emerged as a pivotal tool for promoting sustainable development, garnering widespread attention from governments, financial institutions, and international organizations worldwide [1,2,3,4]. As nations strive to meet the targets set by the Paris Agreement and the United Nations Sustainable Development Goals (SDGs), green finance has become a cornerstone of global efforts to combat environmental degradation and facilitate the transition to a low-carbon economy. By channeling capital toward environmentally friendly projects—such as renewable energy, energy efficiency, and pollution control—green finance not only supports ecological preservation but also fosters economic resilience and innovation.

Globally, initiatives like the European Union’s Sustainable Finance Action Plan, the Network for Greening the Financial System (NGFS), and the Green Climate Fund underscore the critical role of financial systems in addressing climate challenges. These efforts highlight the increasing recognition that sustainable finance is not merely an environmental imperative but also a financial one as it seeks to align economic growth with long-term ecological stability. However, while the benefits of green finance in driving environmental and economic transformation are well documented, its potential unintended consequences, particularly on corporate financial stability, remain underexplored.

As the world’s largest developing country and a major contributor to global carbon emissions, China has been at the forefront of green finance innovation. In recent years, China has implemented a series of policies and regulatory frameworks to promote green finance, including the establishment of green credit guidelines, green bond standards, and pilot zones for green finance reform. These measures have significantly accelerated the flow of capital toward sustainable projects, positioning China as a global leader in green finance development. Nevertheless, the implementation of these policies may also introduce new risks, particularly for corporations navigating the transition to greener practices. For instance, firms in high-pollution industries or those undergoing green transformations often face heightened financial pressures due to increased compliance costs, technological investments, and regulatory uncertainties [5,6].

In recent years, with the deepening of green finance reforms, enterprises, especially those in high-polluting industries and those facing greater financing constraints, have experienced increased financial pressure [7,8]. Green finance policies aim to drive these enterprises to undergo green transformation, but due to the multifaceted factors involved in the transformation process, such as technological innovation, capital investment, and market risks, these enterprises may face a higher risk of debt default [9]. In particular, against the backdrop of financing difficulties and technological uncertainties, some companies may experience exacerbated financial pressure, increasing the likelihood of debt default. Therefore, understanding the impact of green finance policies on corporate debt default risks, especially the heterogeneity across different types of enterprises and regions, is a key part of evaluating the effectiveness of green finance policies.

This study aims to investigate the impact of green finance reform pilot areas on corporate debt default risk, focusing on how green finance policies function across different types of enterprises and regions. By analyzing pollution-intensive enterprises, non-pollution enterprises, and state-owned enterprises, this study reveals the heterogeneous responses of different types of enterprises after policy implementation. Furthermore, through regional classification analysis, the study explores the different impacts of green finance policies on enterprises in Eastern and Central–Western regions, providing more targeted theoretical support for policy formulation.

The primary innovation of this study lies in its focus on the potential financial risks associated with green finance policies, particularly their impact on corporate debt default risks. While the existing literature extensively explores the positive role of green finance in promoting environmental protection, resource efficiency, and low-carbon transition, there is relatively less attention paid to its potential unintended consequences [10], especially the possible impacts on corporate financial stability. Previous research primarily emphasizes how green finance supports sustainable development by directing capital toward environmentally friendly projects [11,12] but largely overlooks the financial pressures that companies, particularly those in high-pollution industries, may face during the transition process.

This study addresses this gap by examining how green finance policies, while driving environmental improvements, may simultaneously increase corporate debt default risks. For instance, companies undergoing green transformation often face significant challenges such as high costs of technological innovation, uncertain market returns, and stringent regulatory requirements [13], which may exacerbate financial pressures. By shifting the focus from the benefits of green finance to its risks, this study provides a more balanced perspective on the overall impact of these policies, contributing to a more comprehensive understanding of their effects on corporate financial health.

The structure of this paper is as follows: Section 2 presents a theoretical analysis; Section 3 outlines the research methods and data sources; Section 4 analyzes the specific impact of green finance policies on corporate debt default risks; Section 5 provides policy recommendations based on the research findings and research prospects.

2. Theoretical Analysis and Research Hypotheses

The policies in green finance reform pilot areas typically encourage companies to invest in green projects by offering tax incentives, government subsidies, and green bonds [14,15]. These green projects usually involve fields such as renewable energy, energy efficiency improvement, carbon reduction, and environmental protection. While these projects have long-term positive impacts in terms of environmental and social benefits, they generally require a longer investment cycle to realize financial returns [16,17]. This phenomenon is not unique to China; similar trends have been observed globally. For instance, studies on green finance initiatives in the European Union (EU) and the United States have highlighted the long gestation periods and high upfront costs associated with green projects, which often necessitate substantial debt financing [18]. Therefore, companies typically need substantial capital investment to support the initiation, development, and implementation of these projects, which leads to a strong demand for debt financing [19].

However, the funding requirements for green projects are not only large in amount but also involve long-term capital commitments [19,20]. When implementing these green projects, companies often face two types of risks: first, technological risks, particularly in areas like green energy technologies and carbon capture, where the maturity and practical application of the technology can be highly uncertain, and second, policy risks, where changes in policies and regulations may affect the return rates of green projects. These risks are not confined to China. For example, international research on green energy projects in Germany and the UK has shown that technological uncertainties and policy shifts can significantly impact project viability and financial returns [21]. In particular, in cases where policy encouragement and regulatory enforcement are inadequate, the returns on investments may fall short of expectations.

As companies’ debt burdens increase, financial leverage rises, and the pressure to repay debt also grows [22]. In the face of project return uncertainties and policy changes, companies’ cash flows may fail to generate as expected, which can lead to deteriorating financial conditions [23]. This pattern is consistent with findings from international studies. For instance, research on green finance in emerging markets, such as India and Brazil, has demonstrated that the high leverage associated with green projects often exacerbates default risks, particularly when policy support is inconsistent [24]. When a company’s cash flow is unable to meet debt repayment obligations, it may face liquidity problems [25] and even default. Therefore, the implementation of green finance reform pilot area policies will not only drive companies to increase their debt financing efforts but also increase the risk of debt default. Based on this, the following research hypothesis is proposed:

H1: 

Green finance reform pilot zone policies will increase the debt default risk of enterprises in pilot zones.

Although the policies in green finance reform pilot areas provide companies with a wide range of green financing products and financial support, in practice, companies often face significant financing constraints [26]. This is partly due to the long return cycles and high uncertainty within green projects. While green projects have long-term positive environmental effects, their economic benefits tend to be delayed, especially when technological innovation and market application are still immature. As a result, companies find it difficult to quickly realize financial returns from green projects [27]. For example, studies on green finance in the EU have highlighted similar financing constraints, where the long-term nature of green investments often conflicts with the short-term profit expectations of investors [28,29].

Companies typically rely on traditional business models and profit sources to sustain their operations in the short term [30]. However, the long-term investment returns from green projects often conflict with these short-term profit models. This causes companies to struggle in securing quick capital inflows through green project financing when they face urgent funding needs [31,32]. As a result, companies tend to rely more on debt financing to fill the funding gap. However, increased debt financing raises the company’s financial leverage, further increasing debt repayment pressure [1,33,34]. For instance, research on green finance in the United States has shown that companies often resort to debt financing to bridge the gap between long-term green investments and short-term liquidity needs, which in turn heightens default risks [25].

At the same time, due to the long-term nature and uncertainty of green projects, traditional financing channels and creditors may question the company’s ability to repay its debt, leading to higher financing costs [35,36,37]. Companies may face higher risk premiums during the financing process, exacerbating financing constraints. To cope with these pressures, companies might resort to more debt financing. This financing dilemma is unlikely to be alleviated in the short term, thereby increasing the risk of default [38]. Based on this, the following research hypothesis is proposed:

H2a: 

Enterprises in green finance reform pilot zones may experience increased debt default risk, as green finance policies can exacerbate financing constraints by increasing the uncertainty and long-term nature of green project investments.

Companies in green finance reform pilot areas often require substantial capital investment to support green projects, and the capital demands and liquidity issues of these projects pose significant challenges for businesses [39,40]. Green projects typically take several years to yield the expected environmental benefits and financial returns, which results in lower liquidity for the stocks associated with these projects. Stock liquidity is generally understood as the ability of investors to quickly buy or sell a company’s assets, and low liquidity means that a company’s stock is not easily traded or [10] valued in the market, thus affecting the company’s ability to raise funds through equity financing [40,41].

The diversity, technological complexity, and regional differences in green projects result in a low level of standardization, which creates significant barriers for investors to understand and assess these projects [42,43]. This, in turn, affects the market’s pricing of these companies. In the stock market, a lack of liquidity means that investors are unable to quickly adjust their portfolios, especially when a company faces urgent funding needs. In such cases, shareholder and investor support may not meet expectations. This low liquidity not only limits the company’s ability to raise funds through capital markets but may also make it difficult for the company to resolve funding shortages through equity financing [44].

Furthermore, low liquidity can lead to a depressed market valuation for the company’s stock, which affects the company’s market confidence [44,45]. In such situations, companies rely more on debt financing to meet their funding needs. The increased reliance on debt financing inevitably leads to higher financial leverage, thereby raising the risk of debt default. Due to low stock liquidity, the company’s financing costs rise, which in turn increases the risk of debt default. For example, research on green finance in Scandinavia has shown that low stock liquidity often exacerbates financing challenges, leading to higher default risks for companies engaged in green projects [10]. Based on this, the following research hypothesis is proposed:

H2b: 

Enterprises in green finance reform pilot zones may face higher debt default risk, as the low liquidity of stocks associated with green projects can limit their ability to raise funds through equity markets, forcing them to rely more on debt financing.

3. Research Design and Sample Selection

3.1. Sample Selection and Data Sources

This study extracts data from A-share listed companies from 2010 to 2023, excluding ST companies, banks, insurance companies, and other financial sector listed companies, as well as companies with missing data. The data types involved are mainly divided into two categories: one includes the basic financial data of the companies, and the other includes the debt default risk data of the companies. The data are sourced from CNRDS and CSMAR, with a total of 33,600 valid samples. The study period from 2010 to 2023 is selected for several reasons. First, 2010 marks a significant milestone in China’s capital market development as it saw the launch of margin trading and short-selling mechanisms, which substantially impacted corporate financing behavior and debt risk dynamics. Second, this period provides a sufficiently long time series to capture evolving trends in corporate debt default risk, particularly in the context of China’s economic transition and regulatory reforms. Third, the inclusion of recent data up to 2023 ensures that this study reflects the latest developments in the post-pandemic economic environment, characterized by heightened financial volatility and increased default risks. Additionally, the choice of pilot zones, such as those under the China (Shanghai) Pilot Free Trade Zone and other regional reform initiatives, is based on their representativeness and policy relevance. These zones serve as testing grounds for financial and regulatory innovations, experiencing unique debt risk dynamics due to preferential policies, cross-border financing, and experimental regulatory frameworks. By focusing on companies within these zones, this study can isolate the effects of policy interventions and regional economic characteristics on corporate debt default risk while also ensuring a diverse sample of companies across industries and ownership types, enhancing the robustness and generalizability of the findings.

3.2. Model Design and Variable Definitions

3.2.1. Model Design

Based on the above research hypotheses, this study sets the listed companies in the green finance reform pilot zones—namely Zhejiang, Guangdong, Xinjiang, Guizhou, and Jiangxi—as the treatment group, marked as 1; the remaining companies are marked as 0, forming the control group. The specific model for the difference-in-differences (DID) method is constructed as follows:

$$\begin{array}{c}{RC}_{i,t}=\alpha +\beta Trea{t}_{i,t}\times Pos{t}_{i,t}+\delta X_{i,t}+{\gamma }_i+{\upsilon }_t+{\epsilon }_{i,t}\end{array}$$

(1)

In this model, i represents the listed company, and t represents time. ${RC}_{i,t}$ indicates the debt default risk of company i in year t. $Trea{t}_i$ is a location dummy variable, with listed companies in the five green finance reform pilot zones being marked as 1, and the rest marked as 0. $Pos{t}_{i,t}$ is a time dummy variable, where it is marked as 1 for years after the pilot policies were implemented and as 0 for before that period. $Trea{t}_{i,t}\times Pos{t}_{i,t}$ represents the net effect of the policy. $X_{i,t}$ represents the control variables. At the same time, individual fixed effects ${\gamma }_i$ and time fixed effects ${\upsilon }_t$ are controlled for. ${\epsilon }_{i,t}$ is the random disturbance term. The coefficient of Treati × Posti,t denoted as β, which measures the impact of the pilot policy implementation on the debt default risk of enterprises. If $\beta$ is significantly positive, it indicates that the implementation of the pilot policy increases the debt default risk of enterprises.

3.2.2. Variable Definition

Dependent Variable

Referring to the bankruptcy probability model proposed by Ohlson to measure the business operational risk [46], this paper selects the company’s debt default risk as the dependent variable as follows:

$$\begin{array}{c}OScore=-1.32-0.407Size+6.03Lev-1.43WCTA+0.0757CLCA-2.37ROA\\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}-1.83FUTL+0.285INTWO-1.72OENEG-0.521CHIN\end{array}$$

(2)

$$\begin{array}{c}CHIN={\displaystyle \frac{\left(\begin{array}{cc}N{I}_{t^{+}}& -N{I}_{t-1}\end{array}\right)}{\left(\begin{array}{cc}|N{I}_{t^{+}}& -\left|N{I}_{t-1}\right|\end{array}\right)}}\end{array}$$

(3)

$$\begin{array}{c}{\displaystyle \frac{{\mathrm{R}\mathrm{C}=\mathrm{e}}^{0-scors}}{\left(1+{\mathrm{e}}^{0-scors}\right)}}\end{array}$$

(4)

The definitions of each indicator are as follows: Size represents the firm’s scale, Lev refers to the financial leverage, and ROA indicates the return on assets. WCTA is the ratio of working capital to total assets, while CLCA is the current ratio, which measures the proportion of current assets to current liabilities. FUTL represents the ratio of operating cash flow to total liabilities. INTWO is a dummy variable; if a firm’s net profit is negative for two consecutive years, the value is 1; otherwise, it is 0. OENEG is another dummy variable; if a firm’s liabilities exceed its assets in the current year, the value is 1, otherwise, it is 0. NI stands for net profit. The $OScore$ index is used to measure a company’s bankruptcy risk. The higher the value, the greater the risk the company faces.

Explanatory Variables

In this study, the listed companies in Jiangxi, Xinjiang, Guizhou, Guangdong, and Zhejiang are designated as the treatment group, marked as 1, while the listed companies in other regions are considered as the control group, marked as 0. The time period after the policy pilot in the five provinces, i.e., after 2017, is marked as 1, and the period before the policy pilot, i.e., before 2017, is marked as 0.

Control Variables

Referring to existing research, variables such as enterprise size, leverage ratio, proportion of tangible assets, cash flow, cash holdings, and return on total assets were selected for control [47]. The mechanism variables, financing constraints, and stock liquidity are represented by the KZ index and Zeros index, respectively. The aforementioned data are sourced from the CNRDS and CSMAR databases.

4. Empirical Analysis

4.1. Descriptive Statistics

The descriptive statistics of the data used in this paper are shown in

Table 1. Variable definition table.

Variable Type	Variable Name	Definition	Symbol
Dependent Variable	Financial Risk of Firm	Financial risk of firm	RC
Core Explanatory Variable	Financial Reform	1 if year of establishing financial reform pilot zone or later, otherwise 0	Treat × Post
Mechanism Testing Variables	Financing Constraints	Financing constraints	KZ
	Stock Liquidity	Stock liquidity	Zeros
	Firm Size	Natural logarithm of total assets	Size
	Firm Leverage	Total liabilities/total assets in annual report	Lev
Control Variables	Tangible Asset Ratio	Net fixed assets/total assets in annual report	Fix
	Cash Flow	Net cash flow from operating activities/total assets in annual report	Cf
	Cash Holdings	(Trading financial assets + cash)/total assets in annual report	Cash
	Return on Assets	Total equity/total assets in annual report	Roe

. It can be observed that there are significant differences in debt default risk across different firms. Additionally, for the main control variables, notable differences are also evident between listed companies. The descriptive statistical results of the data are shown in

Table 2. Descriptive statistics of key variables.

Variable	Sample Size	Mean	Standard Deviation	Minimum	Maximum
RC	33,600	0.002	0.043	0.001	0.115
Treat × Post	33,600	0.162	0.137	0	1
Size	33,600	20.120	1.339	20.000	24.451
Lev	33,600	0.457	0.119	0.109	0.896
Fix	33,600	0.220	0.094	0.012	0.687
Cash	33,600	0.063	0.101	0	0.866
Roe	33,600	0.351	1.744	0.186	1.145
Cf	33,600	0.221	0.383	0	1

. For example, in terms of financial risk, the minimum value is 0.001, the maximum value reaches 0.135, and the standard deviation is 0.013.

4.2. Main Regression Results

The benchmark regression results based on model (1) are shown in

Table 3. Main regression results.

	RC	RC	RC	RC
	(1)	(2)	(3)	(4)
Treat × Post	0.003 ***	0.002 ***	0.002 **	0.003 **
	(3.372)	(2.603)	(2.405)	(2.334)
Size		−0.006 ***	−0.005 ***	−0.003 ***
		(−10.101)	(−9.803)	(−13.030)
Lev		0.030 ***	0.029 ***	0.018 ***
		(13.804)	(13.387)	(11.519)
Fix		0.003	0.002	0.004 ***
		(1.274)	(1.117)	(5.344)
Cf		−0.013 ***	−0.013 ***	−0.015 ***
		(−5.425)	(−5.689)	(−7.741)
Q		−0.001	−0.001	−0.001 **
		(−1.195)	(−1.502)	(−2.276)
Cash		0.000 **	0.001	0.000 ***
		(2.189)	(1.600)	(4.821)
Roe		−0.002 **	−0.002 **	−0.002 ***
		(−2.161)	(−2.447)	(−3.075)
Constant	0.0040 ***	0.1160 ***	0.1010 ***	0.0560 ***
	(8.694)	(9.491)	(9.096)	(13.319)
N	33,600	33,600	33,600	33,600
Individual	Yes	Yes	Yes	No
Year	Yes	Yes	No	Yes
R-squared	0.006	0.163	0.155	0.143

Note: *** p < 0.01, ** p < 0.05, with t-values in parentheses.

. In columns (1) and (2), the regression results are presented before and after introducing control variables, respectively. A comparison shows that, regardless of whether control variables are included, the coefficient of Treat × Post is consistently significantly positive. This indicates that the green finance pilot policy has increased the debt default risk of firms. Therefore, hypothesis H1 is validated.

After considering individual fixed effects (column 3) and time fixed effects (column 4), the interaction term for the policy’s net effect remains significantly positive. The coefficient of Treat × Post is still significantly positive at the 5% level, confirming that the conclusion of H1 remains unchanged.

4.3. Mechanism Test

The results of the mechanism test in this paper are shown in

Table 4. The results of the mechanism.

Variables	KZ	Zero
	(1)	(2)
Treat × Post	0.104 **	−0.002 **
	(2.462)	(−2.113)
Size	−0.565 ***	−0.000 ***
	(−18.786)	(−7.829)
Lev	7.106 ***	0.011 ***
	(42.243)	(7.193)
Fix	1.678 ***	0.010 ***
	(8.253)	(5.583)
Cf	−5.067 ***	−0.001
	(−2.599)	(−0.614)
Q	−0.102	−0.020 ***
	(−0.861)	(−3.128)
Cash	−1.074 ***	−0.003 *
	(−8.461)	(−1.945)
Roe	−0.070	−0.009 ***
	(−0.448)	(−2.967)
Constant	10.555 ***	0.074 ***
	−16.392	−8.681
N	33,600	33,600
Individual	Yes	Yes
Year	Yes	Yes
R-squared	0.579	0.244

Note: *** p < 0.01, ** p < 0.05, and * p < 0.1, with t-values in parentheses.

. This study uses the absolute value of the KZ index to measure financing constraints. The larger the KZ index, the more severe the debt financing constraints faced by the firm [48]. In

Table 5. Test results for polluting vs. non-polluting firms.

	Polluting Firms	Non-Polluting Firms
	(1)	(2)
Treat × Post	0.004 ***	0.001
	(4.701)	(1.642)
Size	−0.003 ***	−0.003 ***
	(−5.372)	(−11.352)
Lev	−0.027 **	0.018 ***
	(−2.369)	(9.651)
Fix	0.004	0.003 ***
	1.551	3.447
Cf	−0.006 *	−0.015 ***
	(−1.696)	(−6.851)
Q	0.001	−0.001 ***
	(−0.356)	(−2.852)
Cash	0.001	0.003 ***
	(0.326)	(3.891)
Roe	−0.051 ***	−0.002 ***
	(−5.069)	(−3.306)
Constant	0.114 ***	0.061 ***
	(8.201)	(11.722)
N	9372	24,228
Individual	Yes	Yes
Year	Yes	Yes
R-squared	0.3125	0.1703

Note: *** p < 0.01, ** p < 0.05, and * p < 0.1, with t-values in parentheses.

, the coefficient of Treat × Post in column (1) is significantly positive, indicating that the implementation of the green finance pilot policy has increased the financing constraints of firms. Additionally, Li found that alleviating financing constraints can improve the availability of funds [49], which in turn reduces the debt default risk. Therefore, firms in green finance reform pilot areas face an increased debt default risk due to the channel of increased financing constraints.

This study adopts the method proposed by He for measuring stock liquidity through the Zeros indicator in the CSMAR database [50]. The calculation formula is the number of days with zero returns within a year divided by the total trading days in that year. A larger Zeros indicator indicates higher stock liquidity for listed companies. In Table 5, the coefficient of Treat × Post in column (2) is significantly negative, suggesting that the green finance pilot policy has reduced the stock liquidity of firms. This reduction may lead to a decreased trading volume, resulting in inefficient capital allocation. Additionally, insufficient capital supply can affect the firm’s cash flow, further increasing the risk of debt default. Since Yu found a negative correlation between stock liquidity and debt default risk [51], firms in green finance reform pilot areas face increased debt default risk through the channel of reduced stock liquidity.

4.4. Heterogeneity Test

4.4.1. Heterogeneity Analysis of Polluting vs. Non-Polluting Firms

In this section, we analyze the heterogeneity between polluting and non-polluting firms. By distinguishing firms based on their pollution levels, we aim to understand whether the green finance pilot policy has different impacts on the debt default risks of polluting versus non-polluting firms. This analysis helps to identify whether the policy’s effects vary according to the environmental footprint of the firms, providing insights into how green finance reforms may influence firms with differing environmental impacts.

In the regression results, the coefficient for the Treat × Post interaction term in column (1) is 0.004, and it is statistically significant at the 1% level. This result indicates that the green finance pilot policy has had a significant positive impact on the debt default risk of polluting companies, meaning that after the policy was implemented, the debt default risk of polluting companies increased. This could be due to the increased financial pressure on polluting companies as a result of the green finance reform, particularly in terms of the financing requirements for green projects and the technological uncertainties associated with them. As these companies face higher financial leverage and greater debt repayment pressure, their default risk is elevated.

However, in column (2), the coefficient for Treat × Post is not statistically significant at the 10% level, indicating that the green finance pilot policy does not have a significant effect on the debt default risk of non-polluting companies. This result suggests that the debt default risk of non-polluting companies is not significantly affected by the green finance reform. One possible explanation is that non-polluting companies typically operate in industries with lower environmental compliance pressures and more stable financial conditions. As a result, they are less affected by the financing constraints and regulatory changes introduced by the green finance reform. Additionally, non-polluting companies may already align with the policy’s objectives, such as promoting sustainable development and reducing carbon emissions, making it easier for them to access green financing without significant disruptions to their operations or financial structures. Furthermore, these companies may benefit from the policy through improved access to green financial instruments, such as green bonds or low-interest loans, which can offset any potential negative impacts on their debt risk.

4.4.2. Heterogeneity Analysis of State-Owned vs. Non-State-Owned Enterprises

To further explore the impact of the green finance pilot policy on different types of enterprises, this study classifies local state-owned enterprises (SOEs) and Central SOEs as “state-owned enterprises” and categorizes other firms as “non-state-owned enterprises”. The results are shown in

Table 6. Test results for state-owned vs. non-state-owned enterprises.

Variables	State-Owned Enterprises (1)	Non-State-Owned Enterprises (2)
Treat × Post	0.001	0.004 ***
	(0.317)	(5.386)
Size	−0.005 ***	−0.003 ***
	(−5.956)	(−10.59)
Lev	0.027 ***	0.021 ***
	(7.914)	(9.423)
Fix	0.002	0.002
	(0.646)	(1.403)
Cf	−0.011 ***	−0.015 ***
	(−5.937)	(−6.478)
Q	−0.002	−0.001
	(−1.621)	(−1.590)
Cash	0.001	0.003 ***
	(0.617)	(4.076)
Roe	−0.006 ***	−0.002 ***
	(−3.015)	(−2.968)
Constant	0.100 ***	0.072 ***
	(5.670)	(10.709)
N	12,511	21,089
Individual	Yes	Yes
Year	Yes	Yes
R-squared	0.2243	0.164

Note: *** p < 0.01, with t-values in parentheses.

.The sample size for state-owned enterprises is 12,511, while the sample size for non-state-owned enterprises is 21,089. By conducting heterogeneity tests on these two types of enterprises, we are able to more deeply analyze the differential effects of green finance policies in different ownership contexts. This analysis provides a more nuanced perspective and reveals how green finance reforms produce different outcomes for state-owned versus non-state-owned enterprises.

In the regression results presented in column (1), the coefficient for the Treat × Post interaction term is not significant at the 10% level, indicating that the green finance pilot policy does not have a significant impact on the debt default risk of state-owned enterprises. This result may be attributed to the unique advantages and policy support that state-owned enterprises typically enjoy. State-owned enterprises often have strong government backing, and particularly in the early stages of green finance policy implementation, the government may prioritize supporting these enterprises through favorable loans, subsidies, and other forms of assistance. As a result, state-owned enterprises are more likely to obtain funding easily under the green finance reform, which reduces their debt default risk. Furthermore, state-owned enterprises usually have more stable cash flows and can sustain operations throughout the long investment horizon of green projects, making them less sensitive to debt default risk. Therefore, the green finance pilot policy does not appear to have a significant impact on the debt default risk of state-owned enterprises, possibly due to their stronger financial stability and government support.

In contrast, in column (2), the coefficient for Treat × Post is 0.004, and it is statistically significant at the 1% level, indicating that the green finance pilot policy has significantly increased the debt default risk for non-state-owned enterprises. This result likely reflects the financing constraints and market uncertainties faced by non-state-owned enterprises when implementing green finance policies. Compared to state-owned enterprises, non-state-owned enterprises generally face a more challenging financing environment, particularly when the long-term financial returns from green projects are not immediately apparent. As a result, these enterprises often struggle to secure sufficient capital from the capital markets. While the green finance pilot policy provides some policy support and financing channels, non-state-owned enterprises typically face higher capital costs and greater financing difficulties. These enterprises also encounter higher levels of technological and market risks when implementing green projects. These factors may lead non-state-owned enterprises to rely more heavily on debt financing, thereby increasing their financial leverage and, consequently, their debt default risk.

4.4.3. Heterogeneity Analysis of Eastern, Central, and Western Regions

To further explore the impact of the green finance pilot policy on enterprises from different regions, this study categorizes companies based on their geographical locations. Specifically, companies from regions such as Beijing, Tianjin, Hebei, Liaoning, and similar areas are classified as “Eastern enterprises”, while companies from regions like Shanxi, Jilin, Heilongjiang, and similar areas are classified as “Central enterprises” and “Western enterprises” based on their geographic location. The sample sizes are as follows: Eastern enterprises include 11,471 samples, Central enterprises have 5684 samples, and Western enterprises comprise 3934 samples. By examining the impact of the green finance reform in these different regions, we can assess whether the policy’s effectiveness varies geographically and whether regional economic characteristics influence how companies experience the effects of the reform.

In

Table 7. Test Results for Eastern, Central, and Western firms.

Variables	Eastern	Central	Western
	(1)	(2)	(3)
Treat × Post	0.003 ***	−0.001	−0.002
	(3.103)	(−0.417)	(−0.184)
Size	−0.004 ***	−0.008 ***	−0.007 ***
	(−7.743)	(−6.170)	(−5.956)
Lev	0.006	0.043 ***	0.039 ***
	(1.113)	(8.877)	(6.943)
Fix	0.001	0.01 *	0.001
	(0.629)	(1.678)	(0.124)
Cf	−0.012 ***	−0.007	−0.018 ***
	(−9.123)	(−1.387)	(−3.068)
Q	−0.001	−0.006 ***	−0.005 ***
	(−0.228)	(−4.014)	(−2.981)
Cash	0.001	0.002	−0.001
	(1.184)	(0.835)	(−0.471)
Roe	−0.017 ***	−0.006 ***	−0.007 ***
	(−3.947)	(−4.634)	(−3.621)
Constant	0.089 ***	0.158 ***	0.152 ***
	(9.705)	(5.587)	(5.804)
N	16,471	11,423	5706
Individual	Yes	Yes	Yes
Year	Yes	Yes	Yes
R−squared	0.232	0.286	0.217

Note: * p < 0.1, *** p < 0.01, with t-values in parentheses.

, columns (1), (2), and (3) show the test results for Eastern, Central, and Western companies, respectively. The results indicate that for Eastern enterprises, the coefficient of the Treat × Post interaction term is significantly positive, suggesting that the green finance pilot policy has increased the debt default risk of companies in the Eastern region. This outcome aligns with expectations, as enterprises in these more economically developed regions are more likely to be directly impacted by reform policies. The Eastern region, characterized by higher levels of industrialization, urbanization, and economic activity, is often at the forefront of policy implementation and regulatory enforcement. The green finance pilot policy may have intensified the pressure on firms in this region to adopt environmentally sustainable practices, leading to increased costs associated with green investments, technological upgrades, and compliance with stricter environmental standards. These additional financial burdens, coupled with the higher competition and market volatility typical of Eastern regions, could exacerbate the challenges of securing sufficient financing for green projects, thereby elevating debt default risks. Furthermore, the Eastern region’s advanced financial markets and greater reliance on external financing may amplify the sensitivity of firms to changes in financing conditions, making them more vulnerable to the policy’s impact.

In contrast, the coefficients for Central and Western enterprises in columns (2) and (3) are not statistically significant at the 10% level, suggesting that the green finance pilot policy has not had a significant impact on the debt default risk of companies in these regions. This finding reflects the regional disparities in economic development, industry structure, and policy implementation. The Central and Western regions, which are generally less economically developed compared to the Eastern region, may have a different industrial composition, with a higher proportion of traditional industries and fewer large-scale enterprises engaged in green projects. As a result, companies in these regions may face less immediate pressure to transition to green practices or comply with stringent environmental regulations. Additionally, the less developed financial markets and limited availability of green financing instruments in Central and Western regions could reduce the exposure of firms to the risks associated with green finance reform. The slower adoption of green financial policies and weaker enforcement mechanisms in these areas may also contribute to the lack of significant impact on debt default risks.

Moreover, the varying levels of government support and infrastructure development across regions could play a role in shaping the policy’s effectiveness. In Eastern regions, where government support for green initiatives and infrastructure for green projects are more advanced, firms may face higher expectations and greater scrutiny, leading to increased financial pressures. In contrast, Central and Western regions may benefit from more gradual policy implementation and targeted support for transitioning industries, which could mitigate the immediate financial risks for firms. These regional differences highlight the importance of tailoring green finance policies to local economic conditions and ensuring that firms in less developed regions receive adequate support to manage the transition to sustainable practices.

4.5. Robustness Test

4.5.1. Replacing the Measurement Model

To verify the robustness of the impact of the green finance pilot policy on corporate bankruptcy risk, this study further uses the debt-to-equity ratio (D/E) as an alternative dependent variable for a regression analysis. The results are shown in

Table 8. Regression results with changed dependent variables.

Variables	D/E (1)
Treat × Post	0.098 ***
	2.732
Size	−2.025 ***
	(−2.933)
Constant	46.708 ***
	−3.086
N	33,600
Controls	Yes
Individual	Yes
Year	Yes
R-squared	0.362

Note: *** p < 0.01, with t-values in parentheses.

. The debt-to-equity ratio is a key financial metric for assessing a company’s financial leverage and debt-servicing capacity. A higher debt-to-equity ratio typically indicates greater financial pressure and an increased risk of bankruptcy. Therefore, by analyzing changes in the debt-to-equity ratio, we can further validate whether the impact of the green finance pilot policy on corporate financial health is consistent with changes in bankruptcy risk.

In this robustness check, we continue to use the difference-in-differences (DID) model, treating the implementation of the green finance pilot policy as the “treatment” effect, and incorporate corporate debt-to-equity ratio data in the regression analysis. In the regression, the coefficient of the Treat × Post interaction term is the key parameter of this study. According to the regression results, the coefficient of the Treat × Post interaction term is 0.098, which is statistically significant at the 1% level, indicating that the green finance pilot policy significantly increased the debt-to-equity ratio of firms in the pilot regions.

4.5.2. Propensity Score Matching (PSM)

This study employs the Propensity Score Matching (PSM) method to address the endogeneity between the green finance reform policy and corporate debt default risk, further analyzing the net effect for the treatment group firms. Firm size, leverage ratio, tangible asset ratio, cash flow, cash holdings, and return on total assets are chosen as matching variables for the PSM process. The regression results are shown in

Table 9. Regression results for PSM-DID.

Variables	RC (1)
Treat × Post	0.003 ***
	2.784
Constant	0.110 ***
	−9.65
N	33,570
Controls	Yes
Individual	Yes
Year	Yes
R-squared	0.178

Note: *** p < 0.01, with t-values in parentheses.

, where the coefficient of Treat × Post is significantly positive, indicating that the main conclusions of this study still hold.

4.5.3. Parallel Trend Test

The parallel trend assumption suggests that before the implementation of the policy, the experimental group (listed companies in pilot provinces) and the control group (listed companies in non-pilot provinces) exhibited a similar trend in terms of debt default risk. After the implementation of the policy, this similar trend is disrupted. To test whether a parallel trend exists, this study constructs the following model for empirical testing.

The test results for the parallel trend assumption are illustrated in Figure 1. The vertical axis represents the coefficient values of θj within the 95% confidence interval, while the horizontal axis shows the years before and after the establishment of the green finance reform pilot zones. Before the implementation of the pilot policy, the estimated coefficients for the interaction term were not significant; however, after the policy was introduced, the coefficients of the interaction term generally showed significance.

4.5.4. Sample Adjustment

To further validate the robustness of the research findings, we made an adjustment to the sample size, specifically by removing the samples from the municipalities directly under the central government to test their impact on the regression results. The results are shown in

Table 10. Sample adjustment.

Variables	RC (1)
Treat × Post	0.002 **
	2.031
Constant	46.708 ***
	−3.086
N	28,400
Controls	Yes
Individual	Yes
Year	Yes
R-squared	0.362

Note: *** p < 0.01, ** p < 0.05, with t-values in parentheses.

. China’s municipalities (such as Beijing, Shanghai, Tianjin, Chongqing, etc.) typically have higher economic development levels and relatively stable financial market environments. As a result, enterprises in these regions may differ significantly from enterprises in other provinces in terms of financing ability, market conditions, and policy support. To eliminate these potential regional biases, we chose to exclude enterprises from the municipalities directly under the central government from the sample to examine whether this adjustment would cause significant changes in the effect of the green finance pilot policy on corporate bankruptcy risk.

After making this adjustment, we re-executed the Difference-in-Differences (DID) model to test the impact of the green finance pilot policy on the bankruptcy risk of enterprises in the remaining sample. The results show that after removing the municipal samples, the coefficient of the Treat × Post interaction term remains significant and passes the test at the 1% significance level. This indicates that even after excluding the samples from the municipalities directly under the central government, the impact of the green finance pilot policy on corporate bankruptcy risk remains significant. This suggests that the inclusion of enterprises from the municipalities directly under the central government in this study’s sample did not significantly affect the regression results, further validating the robustness of the research conclusions.

4.5.5. Lagged Effect Test

To further verify whether there is reverse causality in the impact of the green finance reform pilot policy on corporate debt default risk, this study introduces the lagged debt default risk (RC⁻¹) as part of the robustness check. The results are shown in

Table 11. Sample adjustment.

Variables	RC−1 (1)
Treat × Post	0.002 **
	2.321
Constant	39.748 ***
	−3.094
N	31,051
Controls	Yes
Individual	Yes
Year	Yes
R-squared	0.292

Note: *** p < 0.01, ** p < 0.05, with t-values in parentheses.

. Reverse causality may cause changes in debt default risk before and after policy implementation that are not driven by the policy itself, but rather by the financial condition of firms affecting the implementation of the policy. Therefore, incorporating lagged debt default risk helps capture the historical impact of debt default risk, reducing the interference from reverse causality. The regression results show that the coefficient of the Treat × Post interaction term is 0.002, which is statistically significant at the 1% level, indicating that the green finance reform pilot policy significantly increased corporate debt default risk. By introducing the lagged variable, we ensure that the estimation of the policy effect is not influenced by reverse causality, confirming that the impact on debt default risk is due to the policy itself rather than the reverse influence of corporate financial conditions on policy implementation.

5. Research Conclusions and Policy Recommendations

The study presented in this paper shows that green finance reform policies have exacerbated corporate debt default risks to some extent, particularly for polluting firms and those facing greater financing constraints. Polluting firms have experienced increased financing pressures and greater uncertainty regarding technological transformation following policy implementation, leading to a significant rise in their default risk. This can be attributed to several underlying factors. First, polluting firms face substantial costs to meet stricter environmental standards, including investments in cleaner technologies and production processes. These costs can strain their financial resources, especially for firms already operating with limited liquidity. Second, as green finance policies emphasize environmental performance, polluting firms may suffer from reputational damage, which can further restrict their access to financing. Investors and financial institutions may perceive these firms as higher-risk borrowers, leading to higher borrowing costs or even credit rationing. Third, variations in policy enforcement across regions may exacerbate the challenges faced by polluting firms. In regions with stricter enforcement, firms may face more immediate and severe financial pressures, while in regions with more lenient enforcement, the transition may be slower but still fraught with uncertainty.

In contrast, non-polluting and state-owned enterprises are less affected by policies, indicating heterogeneous responses across different types of firms within the green finance reform. Non-polluting firms typically have lower compliance costs and are better positioned to benefit from green finance incentives, such as preferential loans or tax breaks. State-owned enterprises, on the other hand, often have stronger financial backing and greater access to government support, which mitigates the risks associated with the transition to greener practices.

Furthermore, the regional classification analysis revealed the varying impacts of green finance policies across different areas. In the Eastern region, firms face a significantly higher default risk, which may be linked to stricter environmental regulations and increased market competition. The Eastern region, being more economically developed, often implements environmental policies more rigorously, placing additional financial and operational burdens on firms. Additionally, the higher level of market competition in this region may amplify the financial risks for firms undergoing green transformation, as they must balance compliance costs and maintaining competitiveness. Meanwhile, firms in the Central and Western regions are less affected, likely due to smaller firm sizes and lower pressures related to the implementation of green projects. These regions may also benefit from more gradual policy implementation and greater government support aimed at promoting regional economic development.

Based on these findings, this paper offers the following policy recommendations: First, the government should adopt differentiated green finance policies tailored to the characteristics of different types of enterprises and regions, avoiding a one-size-fits-all approach. For polluting firms and those with greater financing constraints, policies should provide targeted financial support, such as green credit lines, low-interest loans, and financing guarantees, to alleviate the financial pressure they face during the transformation process. For example, establishing a dedicated green transition fund could help these firms access the capital needed for technological upgrades and compliance with environmental standards.

Second, for firms in the eastern region with high default risks, it is essential to strengthen financing support for green projects and implement measures to mitigate the financial risks associated with the green transition. Specific tools, such as green bonds or public–private partnerships, could be leveraged to mobilize additional resources for green investments. Additionally, the government could introduce risk-sharing mechanisms, such as credit insurance or loan guarantees, to reduce the financial burden on firms and encourage broader participation in green initiatives.

Lastly, in the Central and Western regions, the promotion of green finance policies should focus on the design of incentive mechanisms to enhance firms’ green financing capacity and promote their green transformation. For instance, the government could offer tax incentives or subsidies for firms that adopt green technologies or achieve specific environmental performance targets. Regional green finance pilot programs could also be established to test and scale innovative financing models tailored to the unique economic and environmental conditions of these regions.