The Effect of Financial Mismatch on Corporate ESG Performance: Evidence from Chinese A-Share Companies

Abstract

This study examines the effect of financial mismatch on corporate ESG performance in the context of China’s developmental strategy and its dual-carbon goals. Using panel data for Chinese A-share firms spanning 2009–2023 and employing fixed-effects regression models, we find that financial mismatch significantly weakens ESG performance. Further analysis reveals that this negative effect mainly operates through three channels: increased financing constraints, weakened internal control quality, and reduced innovation capability. The results remain robust across a series of alternative specifications and sensitivity tests. This study contributes to the literature by identifying financial mismatch as a key determinant of ESG outcomes and by clarifying the mechanisms through which it exerts influence. From a practical perspective, the findings suggest that alleviating financial mismatch by fostering patient capital, improving internal governance structures, and supporting firms’ green and sustainable investments is essential for enhancing corporate ESG performance and achieving China’s dual-carbon targets.

1. Introduction

Against the backdrop of China’s high-quality development agenda and its dual-carbon objective, corporate ESG (environmental, social, and governance) performance has increasingly become a key indicator for assessing sustainable competitiveness (H. Wang, 2023). The advancement of green finance and the achievement of national strategic goals, such as carbon peak and carbon neutrality, require a deeper understanding of the factors influencing corporate ESG performance and the mechanisms they operate through. Most studies, however, focus on the economic consequences of ESG practices—such as their effects on financial performance, financing costs, and long-term value creation—while paying relatively little attention to the drivers behind ESG implementation. In particular, a significant research gap exists on how financial mismatch shapes corporate ESG behavior (Ma et al., 2025). In recent years, China has pledged to achieve carbon peak by 2030 and carbon neutrality by 2060. These “dual-carbon” targets place considerable pressure on enterprises to strengthen their ESG practices, particularly in terms of undertaking green investment and low-carbon transformation projects. Meeting such requirements often demands stable, long-term financial support. However, financial mismatch—which arises from the inefficient allocation of credit resources—may prevent firms from obtaining the patient capital needed for these initiatives. As a result, financial mismatch can not only limit firms’ ESG improvement but also pose an obstacle to achieving China’s dual-carbon goals.

Financial mismatch is a prominent issue in China because banks monopolize financial resources and adopt risk-averse lending preferences that favor the strong (P. Zhang & Ma, 2012). This behavior hinders a firm’s ability to engage in ESG initiatives. Since ESG strategies often entail resource occupation, cost increases, and the potential short-term deterioration of operational performance (Schuler & Cording, 2006), firms require not only adequate capital reserves but also sustained external long-term financial support. Thus, ‘patient capital,’ with its long-term investment orientation and low sensitivity to short-term fluctuations—comports with the long-term-development nature of ESG practices (S. Li & Wen, 2025).

However, for a considerable period, financial resources in China tended to flow toward high-return, low-value-added sectors such as real estate and capital markets, thereby constraining a firm’s capacity to pursue green development and fulfill CSR (corporate social responsibility) (He et al., 2019). Although improvements in green finance policies and ESG rating systems have enabled financial institutions to help optimize resource allocation and facilitate environmentally friendly transformation and inclusive growth (S. Zhang et al., 2024), financial mismatch still negatively affects ESG performance through mechanisms such as intensified financing constraints (Z. Zhang & Deng, 2022). Moreover, the ESG performance of banks may exert external supervisory effects on a firm’s ESG engagement (Houston & Shan, 2022).

This study utilizes financial mismatch as an analytical lens to examine its influence on the ESG performance of A-share companies in China. We investigate internal transmission mechanisms such as the amplification of financing constraints, the deterioration of internal control quality, and the weakening of innovation capacity, as well as heterogeneous effects across firms. We additionally explore how firm-level characteristics such as lifecycle stage, ownership type, and industry attributes moderate the relationship between financial mismatch and ESG performance.

This study’s contributions are twofold. Theoretically, we extend existing research by addressing the underexplored question of how financial mismatch affects corporate ESG performance and by clarifying the mechanisms through which this effect operates. Practically, the findings have implications for both regulators and firms. For policymakers and financial institutions, they highlight the importance of alleviating financial mismatch to promote sustainable development and support China’s dual-carbon goals. For corporate managers, the results point to strategies that can be adopted, such as strengthening governance structures and securing patient capital to finance long-term green investment.

2. Theoretical Framework and Research Hypotheses

2.1. Mechanism Analysis of the Effect of Financial Mismatch on Corporate ESG Performance

Financial mismatch inhibits corporate ESG performance through three channels: capital structure distortion, governance deterioration, and financial market segmentation.

First, firms typically match financing instruments with the characteristics of their assets. However, given the chronic shortage of long-term capital in the financial markets, firms are often compelled to finance long-term investments with short-term debt, resulting in maturity mismatches (Bai et al., 2016). Such distortions in capital structure lead to high liquidity and debt-servicing pressure, which incentivizes managers to prioritize projects with rapid cash-flow returns, thereby cutting back on ESG investments that have long payback periods. A high level of financial mismatch also obscures the market-governance function of debt financing and weakens the role of asset characteristics in capital structure decisions (Paroutoglou et al., 2022), leading to imbalanced capital structures—that is, distorted proportions between equity and debt. Firms with high asset specificity—which would ideally rely more on equity to hedge against debt risk—might instead overleverage owing to the relatively easy, low-cost access to misallocated debt financing. Consequently, financial flexibility declines, leaving firms with limited capacity to undertake long-term investments, such as ESG initiatives. In addition, financial mismatch, when coupled with ownership structures that reduce capital allocation efficiency, lowers investment efficiency and weakens a firm’s attentiveness to long-term issues such as ESG (Shao, 2010b).

Second, financial mismatch exacerbates principal–agent problems by encouraging management to prioritize short-term gains over long-term value creation (Huang & Ritter, 2009). When mismatch intensifies financing constraints, it undermines the independence of supervisory boards, weakening their capacity to oversee sustainability-related decision-making. This can also incentivize rent-seeking and interest entanglements that further lower the priority given to ESG (Krueger et al., 2020). Moreover, financial mismatch interferes with the self-regulating function of capital markets, reinforces short-termism, and diminishes long-term strategic investment (Hong et al., 2019). Distorted resource allocation results in the overfinancing of speculative short-term activities and underfinancing of long-term value-creating investments (Cao et al., 2019). In response to market preferences, firms are pressured to pursue short-term performance and reduce investment in environmental governance.

Third, a distinctive feature of financial mismatch in China is the segmented, misaligned allocation of financial resources across ownership types, industries, and regions, i.e., the financial market segmentation effect (X. Lu, 2008). While similar segmentation can be observed in other institutional contexts, it is particularly pronounced in China due to its dual-track financial system. State-owned enterprises (SOEs), benefiting from implicit government guarantees and soft budget constraints, often secure credit at lower costs (Lin & Li, 2004). Non-SOEs, who historically faced systemic financing barriers, had to prioritize short-term survival and thus reduce long-term investments such as ESG. Although recent policy initiatives have increased support for non-SOEs, alleviating some financing disadvantages, disparities still remain in practice. Financial mismatch also contributes to the long-standing issue of overcapacity in heavy-polluting, high-energy industries, which continue to receive financing owing to institutional advantages (Qian & Dai, 2023). Furthermore, local governments often restrict the cross-regional flow of capital (Lan et al., 2024), and public environmental awareness is much higher in eastern and central China than in the western regions (Tao et al., 2024), resulting in regional imbalances in ESG investment. Enterprises in the financially developed eastern regions tend to have more adequate funding and invest more in ESG while those in the central and western regions—owing to financing difficulties—are more likely to neglect ESG initiatives.

Based on the above, we propose Hypothesis 1:

H1. 

A high level of financial mismatch inhibits the improvement of corporate ESG performance.

2.2. How Financial Mismatch Affects Corporate ESG Performance

2.2.1. Financing Constraint Mechanism

High financial mismatch intensifies external financing constraints, leaving firms with insufficient affordable capital for long-term projects such as ESG. When companies must rely on short-term, high-cost debt owing to a lack of long-term financing, they face liquidity pressures and may forego/scale back investments with long payback periods (e.g., environmental initiatives). In other words, financial mismatch raises the effective cost of capital and tightens credit availability (Z. Zhang & Deng, 2022), directly crowding out funds that could be used for ESG activities. Thus, greater financial mismatch should lead to more severe financing constraints, hindering a firm’s ability to invest in sustained ESG improvements. Thus, we propose the following:

H2. 

A high level of financial mismatch suppresses corporate ESG performance by increasing financing constraints.

2.2.2. Internal Control Quality Mechanism

Internal control is also critical for sound operations, risk mitigation, and the pursuit of long-term strategic objectives (Fan & Xiao, 2014). An effective internal control system establishes clear roles and responsibilities, strengthens supervision, and promotes the efficient flow of information within the organization. Such features enable management to undertake stronger environmental protection efforts, proactively fulfill CSR, and improve corporate governance quality (Tang et al., 2021; Z. Li et al., 2020). However, high financial mismatch can erode a firm’s internal governance capacity and, in turn, lower ESG performance.

Given financial mismatch, firms may divert attention toward obtaining external funding, thereby reducing the resources allocated to internal supervision, information circulation, and risk assessment (L. Zhang & Fan, 2022). In scenarios where rent-seeking is prevalent, some firms may deliberately weaken internal control to engage in speculative or high-risk financial activities (F. Lu & Yao, 2004). Additionally, financial mismatch can lead to either excessive or insufficient capital, both of which undermine internal control. Surpluses may encourage managerial complacency and inefficiencies, whereas deficits may keep firms from investing in the systems and procedures needed to strengthen governance (Ding et al., 2022). Lacking proper internal monitoring, firms with weak internal control are more prone to environmental violations and CSR lapses. Conversely, strong internal control systems support corporate compliance and accountability by ensuring risk-management processes are in place, which in turn enhances ESG outcomes (Guo et al., 2023). We therefore propose the following:

H3. 

A high level of financial mismatch deteriorates corporate ESG performance by weakening internal control quality.

2.2.3. Innovation Capability Mechanism

Technological innovation is a resource-intensive, high-risk, long-term endeavor that contributes to improving environmental performance, developing sustainable products, and achieving long-term corporate value (Chen et al., 2006). A high degree of financial mismatch undermines a firm’s innovation capability by increasing the cost of external financing, which weakens the ability to sustain R&D investment (J. Li et al., 2023). When financial resources are disproportionately channeled into low-risk, low-innovation sectors, firms with high innovation potential are often deprived of sufficient capital, resulting in inadequate R&D expenditure (Z. Song et al., 2011).

In addition, financial mismatch tends to create volatility in working capital, forcing firms to divert funds away from long-term innovation projects to maintain short-term operations (Brown et al., 2012). This reallocation of financial resources erodes innovation capacity. A decline in innovation capacity compromises ESG performance, particularly in the environmental dimension, which relies heavily on technological advancement and sustained innovation investment. Green innovation enables firms to reduce emissions, enhance resource efficiency, and improve environmental outcomes (Mohammadi et al., 2023). Furthermore, limited innovation capacity constrains a firm’s ability to implement new CSR initiatives. Green technological development has been empirically shown to improve environmental outcomes, boost operational efficiency, and reduce long-term risks (Carrión-Flores & Innes, 2010). Firms with stronger innovation capacity are more likely to generate sustainable development strategies and demonstrate superior performance across all ESG dimensions.

These observations suggest that effective financial resource allocation is essential for firms to maintain adequate innovation investment and improve ESG performance. Thus, the following hypothesis is proposed:

H4. 

A high level of financial mismatch suppresses corporate ESG performance by hindering the development of innovation capability.

2.3. Summary

Financial mismatch affects ESG performance via three underlying mechanisms: capital structure distortion, governance deterioration, and financial market segmentation. Operationally, financial mismatch impedes ESG improvement through three pathways: increasing financing constraints, weakening internal control quality, and inhibiting innovation capability. Figure 1 illustrates this conceptual framework.

3. Materials and Methods

3.1. Sample Selection and Data Sources

We selected nonfinancial A-share companies in China spanning 2009–2023 as the research sample. Following the literature, several data-cleaning steps were applied to ensure the reliability and validity of the dataset:

• Exclude firms from the financial and real estate sectors;
• Remove firms designated as ST or *ST;
• Exclude insolvent firms or observations with negative total assets;
• Eliminate samples with missing or abnormal key variables;
• Winsorize all continuous variables at the 1st and 99th percentiles.

After screening, we obtained a final unbalanced panel dataset comprising 32,600 firms and 41,311 firm–year observations.

Data on corporate ESG performance were derived from the Huazheng ESG Rating System. We obtained the core variable of financial mismatch, along with the control variables, from the iFind and CSMAR databases. Internal control quality is measured using the Internal Control Index provided by the DIB database. Innovation capability is assessed using manually collected patent data from the China National Intellectual Property Administration. While patents are a widely used indicator of innovation, critics argue that dominant firms may strategically accumulate patents to maintain market power rather than foster genuine innovation (Standing, 2017). We acknowledge this limitation, but still adopt patents as a proxy due to their availability and established use in empirical research.

3.2. Variable Definitions

3.2.1. Dependent Variable

Based on the literature (Qi & Guo, 2024), we construct a proxy for corporate ESG performance using the rating scores from the Huazheng ESG Rating System. Specifically, the nine-tier ESG ratings, ranging from C to AAA, are converted into numerical values ranging from 1 to 9 in ascending order (C = 1, CC = 2, CCC = 3, …, up to AAA = 9). A higher assigned score indicates better ESG performance.

3.2.2. Independent Variable

Financial mismatch (Finmis). Financial mismatch is typically reflected in a firm’s inability to access financing at a reasonable cost in capital markets (Chari et al., 2007). Following Ting (Shao, 2010a), we measure financial mismatch by calculating the proportion of interest expense in financial costs relative to total liabilities net of accounts payable. The larger the absolute value of this ratio, the higher the degree of financial mismatch faced by the firm. The formula is

Financial mismatch = |[Interest expense/(Liabilities − Industry average interest rate)]/Industry average interest rate|.

(1)

3.2.3. Control Variables

To minimize the potential for omitted-variable bias, we follow the literature (Han & Li, 2020) and include a comprehensive set of firm-level control variables, as shown in

Table 1. Variable Names, Symbols, and Calculation Methods.

	Variable Symbol	Variable Definition	Construction Method
Dependent Variable	ESG	Corporate ESG performance	Huazheng ESG ratings from AAA to C are assigned values from 9 to 1, with higher values indicating better corporate ESG performance
Independent Variable	Finmis	Financial mismatch	Financial mismatch = |[interest expense/(liabilities − industry average interest rate)]/industry average interest rate|
Mechanism Variable	FC	Financing constraint	Whited and Wu (2006) index (WW index), calculated based on firm-level financial indicators; higher values indicate stronger external financing constraints
	ICQ	Internal control quality	Internal Control Index from the DIB database; higher values indicate more effective and comprehensive internal control systems
	Innov	Innovation capability	Weighted number of patents, with invention patents weighted as 3, utility model patents as 2, and design patents as 1; natural logarithm of (weighted total + 1)
Control Variable	Size	Company size	Natural logarithm of total assets at the end of the year
	Lev	Leverage ratio	Total liabilities divided by total assets, multiplied by 100%
	Cash	Cash holding ratio	(Cash and short-term investments) divided by total assets
	Age	Firm age	Ln (current year − year of firm registration + 1)
	Growth	Growth rate	((Current revenue − previous revenue)/previous revenue) × 100%
	Roa	Return on assets	Net income for the current year divided by total assets at year-end
	Iot	Institutional ownership	Number of shares held by institutional investors divided by total number of shares
	Dual	Duality	Dummy variable equal to 1 if the CEO and board chair are the same person and 0 otherwise
	Indirect	Proportion of independent directors	Number of independent directors divided by total number of board members
	Top1	Top shareholder ownership	Shareholding ratio of the largest shareholder

, including firm size, leverage ratio, cash holdings, firm age, growth rate, return on assets, institutional ownership, CEO–chairperson duality, proportion of independent directors on the board, and the shareholding ratio of the largest shareholder. Additionally, we control multidimensional fixed effects at the year–industry–firm level. To account for potential heteroskedasticity, we conduct formal tests, which confirm its presence. Accordingly, all regressions report robust standard errors, ensuring the validity of the statistical inference.

3.3. Model Specification

To regress the effect of financial mismatch on corporate ESG performance, we construct the following model:

$${ESG}_{i,t}={\alpha }_0+{\alpha }_1{Finmis}_{i,t}+{\alpha }_2\sum {Control}_{i,t}+{\phi }_i+{\mu }_t+{\gamma }_j+{\epsilon }_{i,t}$$

(2)

Here, the dependent variable represents the ESG performance of firm i in year t, and the key independent variable denotes the level of financial mismatch faced by firm i in year t. A set of control variables is included: firm size (Size), leverage ratio (Lev), cash holdings (Cash), firm age (Age), growth rate (Growth), return on assets (Roa), institutional ownership (Iot), CEO–chairperson duality (Dual), proportion of independent directors (Indirect), and ownership of the largest shareholder (Top1). The model also incorporates firm-fixed effects to control for unobserved time-invariant firm characteristics, year-fixed effects to account for macroeconomic and policy shocks, and industry-fixed effects to control for sector-specific heterogeneity. Subscripts i, t, and j represent firm, time (year), and industry, respectively. Finally, ${\epsilon }_{i,t}$ denotes the random error term.

3.4. Descriptive Statistics

Table 2. Descriptive Statistics of Key Variables.

Variable	Mean	Std. Dev.	Median	Min	Max	Sample Size
Finmis	0.676	0.576	0.587	0.011	3.627	41,311
ESG	4.125	0.881	4.000	2.000	6.000	41,311
Size	22.128	1.28	21.936	19.745	26.131	41,311
Lev	0.41	0.207	0.399	0.05	0.924	41,311
Cash	0.049	0.068	0.048	−0.154	0.244	41,311
Age	2.91	0.345	2.944	1.792	3.555	41,311
Growth	0.153	0.372	0.098	−0.533	2.238	41,311
Roa	0.042	0.067	0.041	−0.231	0.23	41,311
Iot	0.434	0.249	0.447	0.004	0.915	41,311
Dual	0.298	0.457	0.000	0.000	1.000	41,311
Indirect	37.617	5.312	36.360	33.33	57.14	41,311
Top1	33.942	14.831	31.650	8.434	74.018	41,311

presents the descriptive statistics of the key variables. The mean value of financial mismatch (Finmis) is 0.676, with a standard deviation of 0.576, indicating that most sample firms experience a certain degree of financial mismatch and that the extent of mismatch varies significantly across firms. In addition, the mean value of corporate ESG performance, which is constructed based on the assigned numerical values of Huazheng ESG ratings, is 4.125, with a standard deviation of 0.881. The ESG scores range from 2 to 6, suggesting that the average ESG performance of the sample firms falls between B and BB ratings. The considerable variation and low overall rating levels imply that there is considerable room for improvement in corporate ESG performance.

4. Results

4.1. VIF Test

We conduct multicollinearity tests on the main independent variables prior to the regression analysis The results indicate that all variance inflation factors (VIFs) are below the threshold value of 10, suggesting that multicollinearity is not a serious concern. In addition, we use the Hausman test to determine the appropriate model specification. The test yields a p-value of 0, leading to the rejection of the null hypothesis of the random effects model, thereby supporting the use of a fixed-effects model for the subsequent regression analysis.

4.2. Baseline Regression

Table 3. Baseline Regression Results (Full Sample).

Variable	(1)	(2)	(3)
	ESG	ESG	ESG
Finmis	−0.0725 *** (−7.5617)	−0.0637 *** (−7.0463)
L. Finmis			−0.0512 *** (−5.3487)
Size		0.2495 *** (16.7025)	0.2626 *** (15.8706)
Lev		−0.8728 *** (−16.1315)	−0.8304 *** (−13.9744)
Cash		−0.2355 *** (−3.1671)	−0.1905 ** (−2.3211)
Age		−0.1422 (−1.3402)	−0.0684 (−0.5570)
Growth		−0.0997 *** (−8.8133)	−0.0817 *** (−6.7015)
Roa		0.4203 *** (4.4577)	0.2236 ** (2.1979)
Iot		−0.1154 * (−1.9045)	−0.1520 ** (−2.3132)
Dual		−0.0056 (−0.3179)	−0.0011 (−0.0588)
Indirect		0.0087 *** (5.9441)	0.0080 *** (5.1269)
Top1		0.0027 *** (2.6881)	0.0018 * (1.7307)
_cons	4.3806 *** (26.0418)	−0.5943 (−1.3925)	−1.2750 ***
			(−2.6383)
N	41,311	41,311	35,859
R2	0.027	0.065	0.061
Control	No	Yes	Yes
Year	Yes	Yes	Yes
Industry	Yes	Yes	Yes
Company	Yes	Yes	Yes

Note: *** p < 0.01, ** p < 0.05, * p < 0.10; t-statistics in parentheses.

(Column 1) presents the baseline regression using the full sample, controlling for industry, firm, and year-fixed effects. The coefficient of financial mismatch is −0.0725 and is statistically significant at 1%, indicating a significant negative effect on corporate ESG performance. In column (2), after incorporating control variables, the coefficient decreases slightly to −0.0637 but remains negative and significant at 1%, demonstrating the robustness of the result. Column (3) uses the lagged value of the financial mismatch variable to address potential reverse causality and to test for lagged effects. The coefficient remains negative (−0.0512) and passes the 99% confidence level.

Taken together, regardless of whether control variables are included or lagged variable treatment is applied, financial mismatch exhibits a significant inhibitory effect on corporate ESG performance.

4.3. Robustness Tests

4.3.1. Alternative Measurement of the Dependent Variable

To verify the robustness of the regression results, we follow Gao Jieying et al. (Gao et al., 2021) and redefine the ESG variable by converting the Huazheng ESG rating from detailed levels into broader categories. Specifically, a new proxy variable, ESG1, is constructed: Firms rated AAA to A are assigned a score of 3, those rated BBB to B are assigned 2, and those rated CCC to C are assigned 1. The regression results show that the coefficient of ESG1 is −0.0361 and remains significant at 1%, confirming that the main conclusion—that financial mismatch significantly suppresses corporate ESG performance—is robust to alternative variable definitions.

4.3.2. Exclusion of Special Years

To account for potential exogenous shocks to the financing environment and ESG behavior caused by the 2015 stock market crash and the COVID-19 pandemic beginning in 2020, the corresponding years are excluded. The results still reveal a significant negative effect of financial mismatch on ESG performance, providing further support for the robustness of the findings.

4.3.3. Propensity Score Matching (PSM)

We use PSM to mitigate potential sample-selection bias. The procedure involves randomly sorting the sample and setting a seed for reproducibility, which ensures that the matching results can be replicated. Then, firms with high levels of financial mismatch are designated as the treatment group while those with low mismatch are assigned to the control group. Based on the control variables, 1:1 nearest-neighbor matching is conducted, and regression analysis is performed on the matched sample. The results (

Table 4. Robustness Test Results (1).

Variable	(1)	(2)	(3)	(4)
	Alternative Measurement of Dependent Variable	Excl. Stock Crash	Excl. COVID-19	PSM
Finmis	−0.0361 *** (−7.9617)	−0.0657 *** (−7.0547)	−0.0307 *** (−2.8846)	−0.0639 *** (−6.5062)
N	41,311	39,124	24,281	35,960
R2	0.042	0.067	0.048	0.067
Control	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes
Company	Yes	Yes	Yes	Yes

Note: *** p < 0.01; t-statistics in parentheses.

) indicate that the coefficient on financial mismatch remains significantly negative at the 1% level, thereby validating the robustness of the estimation.

4.3.4. Lagged Explanatory Variables

To address potential reverse causality between ESG performance and a firm’s financing environment, we follow Wang Shuhua et al. (S. Wang et al., 2022) and use a one-period lag for both the core explanatory variable and the control variables. The results (

Table 5. Robustness Test Results (2).

Variable	(1)	Variable	(2)
	ESG		ESG
Finmis	−0.0637 *** (−7.0463)	Finmis	−0.0590 *** (−5.9965)
Size	0.2495 *** (16.7025)	L. Size	0.2149 *** (13.3854)
Lev	−0.8728 *** (−16.1315)	L. Lev	−0.7233 *** (−12.2686)
Cash	−0.2355 *** (−3.1671)	L. Cash	−0.1294 * (−1.6752)
Age	−0.1422 (−1.3402)	L. Age	0.0688 (0.5890)
Growth	−0.0997 *** (−8.8133)	L. Growth	0.0506 *** (4.0593)
Roa	0.4203 *** (4.4577)	L. Roa	1.3806 *** (13.6740)
Iot	−0.1154 * (−1.9045)	L. Iot	−0.0349 (−0.5263)
Dual	−0.0056 (−0.3179)	L. Dual	0.0042 (0.2263)
Indirect	0.0087 *** (5.9441)	L. Indirect	0.0047 *** (3.1447)
Top1	0.0027 *** (2.6881)	L. Top1	0.0012 (1.1320)
_cons	−0.5943 (−1.3925)	_cons	−0.6124 (−1.3195)
N	41,311	N	35,859
R2	0.065	R2	0.072
Control	Yes	Control	Yes
Year	Yes	Year	Yes
Industry	Yes	Industry	Yes
Company	Yes	Company	Yes

Note: *** p < 0.01, * p < 0.10; t-statistics in parentheses.

) demonstrate that the coefficient of financial mismatch remains significantly negative regardless of whether variables are lagged, indicating that there is no significant endogeneity bias and the core conclusion holds.

5. Discussion

5.1. Transmission Mechanism Test

The traditional stepwise approach to testing mediating effects has limitations (T. Jiang, 2022). Therefore, following H. Liu and Li (2012), J. Jiang et al. (2018), and H. Song and Lu (2020), we introduced the mediating variable M, which captures financing constraints, internal control quality, and innovation capability. Each mechanism variable is coded as a binary indicator (M = 1 for the high group, M = 0 for the low group) based on yearly or industry-level rankings.Subgroup regressions under M = 0 and M = 1 allow us to assess how financial mismatch differentially affects ESG performance through these channels.

5.1.1. Intensification of Financing Constraints

Financing constraints are a key factor determining whether a firm has sufficient external capital to support ESG investment. We use the WW index proposed by Whited and Wu (2006) to measure the degree of financing constraints faced by firms. The coefficients in Equation (3) are fixed parameters derived from their structural model, rather than estimated in this study. The index incorporates both firm-level financial characteristics and industry sales performance and is calculated as

WW = −0.091cf + 0.062div + 0.021lev − 0.044size + 0.102isg − 0.035sg,

(3)

where cf denotes the ratio of cash flow to total assets, div is a dummy variable indicating whether the firm distributed dividends in a given year, lev represents the leverage ratio, size is the natural logarithm of the firm’s total assets, and sg and isg refer to the sales growth rates of the firm and its corresponding industry, respectively (Qiao & Zhang, 2023).

Based on the year–industry dimension, we divide the full sample into terciles according to the WW index. Firms in the highest tercile are classified as the high-financing-constraint group (WW = 1) while those in the lowest tercile constitute the low-constraint group (WW = 0). The results (

Table 6. Financing Constraint Mechanism.

Variable	(1)	(2)
	ESG	ESG
Group	WW = 0	WW = 1
Finmis	−0.0204 (−1.0809)	−0.0874 *** (−6.1170)
N	11,725	11,937
R2	0.085	0.067
Control	Yes	Yes
Year	Yes	Yes
Industry	Yes	Yes
Company	Yes	Yes
Chow test p	0.000 ***

Note: *** p < 0.01; t-statistics in parentheses.

) show that in the high-financing-constraint group, the coefficient of financial mismatch is −0.0874 and statistically significant at 1%, whereas the coefficient is not significant in the low-constraint group. The Chow test confirms that the difference between the two groups is statistically significant, thereby supporting H2.

5.1.2. Weakening of Internal Control Quality

Internal control is a critical safeguard for ensuring sound operations, mitigating risks, and enhancing organizational efficiency (Fan & Xiao, 2014). Following Zhai et al. (2022), we measure the quality of internal control using the Internal Control Index from the DIB database, where higher index values indicate better internal control quality.

Based on the baseline regression model, the sample is divided annually into terciles according to the Internal Control Index (Ic). Firms in the highest tercile are classified as the high-internal-control group (Ic = 1), and those in the lowest tercile constitute the low-internal-control group (Ic = 0). Separate regressions are conducted for each group. The results (

Table 7. Mechanism of Weakened Internal Control Quality.

Variable	(1)	(2)
	ESG	ESG
Group	Ic = 0	Ic = 1
Finmis	−0.1062 *** (−6.9274)	−0.0008 (−0.0439)
N	12,219	12,438
R2	0.073	0.081
Control	Yes	Yes
Year	Yes	Yes
Industry	Yes	Yes
Company	Yes	Yes
Chow test p	0.000 ***

Note: *** p < 0.01; t-statistics in parentheses.

) show that in the low-internal-control group, the coefficient of financial mismatch is −0.1062 and statistically significant at 1%, while the coefficient in the high-internal-control group is insignificant. The Chow test further confirms that the difference in coefficients between the two groups is statistically significant thus supporting H3.

5.1.3. Hindrances to Innovation Capability Enhancement

Following Quan and Yin (2017), to measure a firm’s innovation capability, patent types are weighted according to their relative contribution to corporate development. Invention patents, utility model patents, and design patents are assigned weights of 3, 2, and 1, respectively. We then construct an innovation capability index (Innovate) as the natural logarithm of the weighted total number of patents plus one.

Using the same subgroup regression method, firms are divided annually and by industry into three groups based on the Innovate index. Firms in the highest tercile are classified as the high-innovation group (Innovate = 1) while those in the lowest tercile are designated as the low-innovation group (Innovate = 0). The regression results (

Table 8. Mechanism of Reduced Innovation Capability—Patent Measure.

Variable	(1)	(2)
	ESG	ESG
Group	Innovate = 0	Innovate = 1
Finmis	−0.0819 *** (−5.4967)	−0.0237 (−1.3315)
N	13,388	13,766
R2	0.080	0.073
Control	Yes	Yes
Year	Yes	Yes
Industry	Yes	Yes
Company	Yes	Yes
Chow test p	0.000 ***

Notes: Robust standard errors are reported in parentheses. *** p < 0.01.

) show that in the low-innovation-capability group, the coefficient of financial mismatch is −0.0819 and statistically significant at 1%, whereas the coefficient is not significant in the high-innovation group. The Chow test confirms that the difference between the two groups is statistically significant, thus supporting H4.

Since relying solely on patent-based measures may not fully capture firms’ innovation activities, we further employ R&D intensity as an alternative indicator to re-examine the mechanism. Following Hu and Jefferson (2009), we measure firms’ innovation input by R&D intensity, defined as the ratio of R&D expenditure to total assets. Using the same subgroup regression method, firms are divided annually and by industry into three groups based on this innovation intensity measure. Firms in the highest tercile are classified as the high-innovation group (Innov = 1), while those in the lowest tercile are designated as the low-innovation group (Innov = 0). The regression results (

Table 9. Mechanism of Reduced Innovation Capability—R&D Measure.

Variable	(1)	(2)
	ESG	ESG
Group	Innovate = 0	Innovate = 1
Finmis	−0.0245 *** (0.0075)	−0.0055 *** (0.0019)
N	13,173	17,394
R2	0.5426	0.5333
Control	Yes	Yes
Year	Yes	Yes
Industry	Yes	Yes
Company	Yes	Yes
Chow test p	0.000 ***

Notes: Robust standard errors are reported in parentheses. *** p < 0.01.

) show that in the low-innovation group, the coefficient of financial mismatch is −0.0245 and statistically significant at the 1% level, whereas in the high-innovation group, the coefficient is also significant but much smaller in magnitude. The Chow test confirms that the difference between the two groups is statistically significant, thereby supporting H4.

5.2. Heterogeneity Analysis

5.2.1. Heterogeneity Based on Corporate Lifecycle

Firms at different stages of the corporate lifecycle exhibit substantial variations in their resource endowments, strategic priorities, and risk tolerance, which in turn influence the effect of financial mismatch on ESG performance. According to agency theory, firms in the growth stage focus mainly on survival and expansion, typically facing lower levels of internal information asymmetry and displaying less opportunistic managerial behavior (W. Li et al., 2013). By contrast, mature firms prioritize financial stability and are more likely to reduce long-term investments (such as ESG) when experiencing financial volatility (L. Zhang & Zhang, 2024). Firms in the declining stage tend to shift their focus inward and often rely on earnings management to embellish financial statements, thereby weakening their commitment to CSR (Sun et al., 2016). Moreover, financial mismatch may further impair credit ratings, prompting declining firms to substitute real CSR activities with earnings manipulation (Guo et al., 2021).

We use the cash-flow-pattern method to classify firms into different lifecycle stages and conduct subgroup regression analysis (S. Liu et al., 2020). The results (

Table 10. Heterogeneity Test Results Based on Corporate Lifecycle and Ownership Type.

	(1)	(2)	(3)	(4)	(5)
	Heterogeneity by Lifecycle Stage			Heterogeneity by Ownership Type
	Growth Stage	Maturity Stage	Decline Stage	Non-SOE	SOE
Variable	ESG	ESG	ESG	ESG	ESG
Finmis	−0.0356 ***	−0.0709 ***	−0.0938 ***	−0.0563 ***	−0.0523 ***
	(−3.2048)	(−2.6613)	(−4.7455)	(−5.0110)	(−3.1743)
N	28,484	4763	7819	23,166	13,163
R2	0.065	0.085	0.064	0.069	0.047
Control	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes
Industry	Yes	Yes	Yes	Yes	Yes
Company	Yes	Yes	Yes	Yes	Yes
Chow test p	0.000 ***			0.000 ***

Note: *** p < 0.01; t-statistics in parentheses.

) indicate that financial mismatch has a negative effect on ESG performance across all stages. However, the effect is weakest for growing firms, stronger for mature firms, and strongest for firms that are declining. These findings align with market preferences for sustainable assets under China’s dual-carbon goals (Q. Li & Zhang, 2021): Growth-stage firms tend to improve ESG performance to attract investors and policy support (X. Liu et al., 2023), whereas such incentives are less prominent for mature firms and least prevalent for declining firms, owing to limited resources and weaker strategic motivation.

5.2.2. Heterogeneity Based on Ownership Type

Misaligned financial resource allocation mechanisms and policy frameworks in China can often result in a leakage effect (F. Lu & Yao, 2004), where SOEs, because of their public service obligations, are more likely to receive policy and financial support. Following the classification approach of Li Jinglin et al. (J. Li et al., 2021), we divided firms into SOEs and non-SOEs using data from the CSMAR database and conducted separate regressions (Table 10). The results show that the absolute value of the financial mismatch coefficient is greater for non-SOEs (−0.0563) than for SOEs (−0.0523), indicating that the suppressing effect of financial mismatch on ESG performance is more pronounced for non-SOEs.

As shown in Table 10, the underlying reasons can be summarized as follows: (1) Differences in financing costs reinforce the crowding-out effect. SOEs, benefiting from soft budget constraints, are more likely to obtain long-term, low-cost funding, which enhances their capacity to maintain ESG investments (L. Shen & Chen, 2020). Non-SOEs, by contrast, face more limited access to financing and bear higher financing costs, making them more likely to reduce long-term investments, such as ESG, under conditions of financial mismatch. (2) Internal incentives and external constraints differ across ownership types. Performance evaluation systems for SOEs often incorporate policy-oriented goals, thereby imposing rigid requirements for ESG practices (Z. Zhang & Deng, 2022). Non-SOEs, meanwhile, tend to focus more on short-term financial performance and are more inclined to reduce ESG investment when faced with financial pressure. Nevertheless, the difference in coefficients between the two groups is relatively small, and both are significantly negative. This may be attributed to China’s ongoing mixed-ownership reform, which has gradually narrowed governance and financing disparities between firms with different ownership (Wei et al., 2023; Wen & Liu, 2019).

5.2.3. Heterogeneity Analysis Based on Industry Pollution Intensity

Ecological governance has increasingly become a vital component of China’s national governance system (H. Shen & Zhou, 2017). The pollution intensity of industries might significantly moderate the relationship between financial mismatch and corporate ESG performance. Based on the classification criteria outlined in the ‘Notice on the Environmental Inspection Industry Classification Directory for Listed Companies,’ issued by the Ministry of Ecology and Environment, as well as the categorization method proposed by Pan et al. (2019), we classify firms into heavy-polluting and non-heavy-polluting industries. A dummy variable (Pol) indicates whether a firm belongs to a heavy-polluting industry, and an interaction term (Pol × Finmis) is introduced to examine the moderating effect.

The results in

Table 11. Heterogeneity Test Results Based on Industry Pollution Characteristics.

	Heavy-Polluting Industry
Variable	ESG
Pol × Finmis	0.0093 ***
	(2.01)
N	41,311
R2	0.300
Control	Yes
Year	Yes
Company	Yes

Notes: Robust standard errors are reported in parentheses. *** p < 0.01.

show that the coefficient of the interaction term Pol × Finmis is significantly positive, indicating that the negative effect of financial mismatch on ESG performance is more pronounced among firms in non-heavy-polluting industries. This finding can be explained as follows: (1) Firms in heavy-polluting industries are subject to stricter environmental regulations and emission-reduction obligations, requiring them to maintain environmental investments to avoid penalties, even under financial pressure (A. Jiang et al., 2023; Porter & van der Linde, 1995). (2) In the ecological civilization context, firms in polluting industries are more likely to receive policy and financial support—such as green credit and government procurement—which helps sustain their ESG commitments (Cui et al., 2023). (3) Owing to their considerable environmental externalities, heavy-polluting industries are subject to heightened scrutiny from the government, the public, and the media. These formal and informal regulatory pressures incentivize firms to strengthen their environmental governance and CSR practices (J. Wang et al., 2024), thereby mitigating the adverse effects of financial mismatch.

6. Conclusions and Recommendations

6.1. Conclusions

We provide empirical evidence that financial mismatch undermines corporate ESG performance. By analyzing a large panel of listed Chinese firms, we find a consistent negative relationship between financial mismatch and ESG ratings, confirming the main hypothesis that distortions in financial resource allocation pose a substantial barrier to a firm’s sustainable development. This finding remains robust across various model specifications and alternative measurements, underscoring the reliability of the result.

Mechanism analyses offer further insights into why financial mismatch impedes ESG performance. We find that higher financial mismatch intensifies a firm’s external financing constraints, thereby limiting the capital available for investments related to long-term sustainability. Moreover, financial mismatch is associated with deteriorated internal control quality, which weakens corporate governance and oversight of ESG initiatives. In addition, firms facing severe mismatch exhibit reduced innovation capability, which hinders the development of green technologies and sustainable practices. These channels—financing constraints, governance deterioration, and innovation stagnation—collectively explain how financial mismatch translates into weaker environmental and social outcomes.

Furthermore, our heterogeneity analysis reveals that the financial mismatch effect is not uniform across all firms. The negative effect is more pronounced for companies in the later stages of the corporate lifecycle, especially those in the declining stage, as these firms have fewer slack resources and more difficulty adapting to financial pressures. Non-SOEs also face stronger adverse effects than SOEs, likely because of their higher financing costs and lack of policy support. By contrast, the effect of financial mismatch effect on ESG is less severe for firms in heavy-polluting industries. This can be attributed to the stringent environmental regulations and external pressures that compel continuous ESG investment for these firms, even under financial strain. These differences underscore the importance of considering firm-specific characteristics and external environments when evaluating the relationship between financial factors and sustainability outcomes.

• Our findings have several policy implications to enhance corporate ESG performance amid financial mismatches.
• Financial regulators and institutions should strive to alleviate financial mismatches by expanding access to long-term financing and patient capital, ensuring that firms—especially private and innovation-oriented ones—can obtain the necessary funding for sustainable projects.
• Policies that strengthen internal corporate governance and risk-management systems can help firms maintain their ESG commitments, even when faced with financial constraints.
• Targeted support should be provided to vulnerable firm groups (e.g., mature or declining firms and non-SOEs) through tailored financial products or incentives to sustain their ESG-related investments.
• For industries that face relatively lower environmental requirements, incentive mechanisms or voluntary guidelines could be introduced to encourage firms to continue investing in ESG practices despite financial pressures. By addressing the root causes of financial mismatch and tailoring strategies based on firm heterogeneity, policymakers can better promote corporate sustainability and align financial resource allocation with the goals of green development and carbon neutrality.

These findings are obtained in the context of China’s bank-dominated financial system and policy-driven environment. Therefore, they mainly reflect the Chinese institutional setting.

6.2. Recommendations

An important limitation of this study is that our measurement of corporate ESG performance relies solely on the Huazheng ESG ratings. Future research could incorporate text-based ESG indicators derived from corporate sustainability reports, CSR disclosures, or other narrative sources, which would enrich the assessment of firms’ ESG practices. Such text-mining approaches have been shown to provide complementary insights beyond quantitative ratings. Nevertheless, because ESG disclosure by some Chinese firms is insufficient—preventing us from obtaining a sample large enough to implement text-based measures—we continue to use this rating-based measure as the dependent variable.

In addition, the results are context-specific, as they are obtained under China’s financial and institutional environment. Future studies could conduct comparative or cross-country investigations to examine whether the effect of financial mismatch on ESG performance holds under different settings.

Another limitation concerns the measurement of innovation capability. Future research could incorporate additional indicators, such as the number of digital transformation projects, or collaborative innovation measures. These dimensions would provide a more comprehensive view of firms’ innovative capacity and help validate the robustness of the findings. In this study, we measure innovation primarily through patent output, which reflects the quantity of patents but not necessarily their quality or innovation processes. Broader indicators—such as collaborative projects, or digital transformation initiatives—would enrich the measurement of firms’ innovation capacity. Finally, a rigorous and coherent metrics system for evaluating the outcomes of China’s policy-oriented innovation in firms remains lacking; building such a framework will be a key focus of our future work.