Does Patient Capital Crowd out the Stabilizing Benefits of ESG? Evidence from Corporate Investment Volatility

Abstract

The market imperfection hypothesis posits that market frictions undermine economic efficiency and amplify economic fluctuations. As an emerging corporate evaluation framework and behavioral norm, ESG (environmental, social, and governance) performance helps mitigate such market imperfections. This study empirically examines the impact of corporate ESG performance on investment volatility and its underlying mechanisms. Using panel data from Chinese listed companies, we find that higher ESG ratings significantly reduce corporate investment volatility. Mechanism tests reveal that ESG practices curb investment fluctuations through two key channels: alleviating information asymmetry and reducing agency costs, thereby addressing fundamental market frictions. Moderating effect tests indicate that patient capital suppresses the smoothing effect of ESG on corporate investment volatility. Heterogeneity analysis further demonstrates that this stabilizing effect is more pronounced in non-state-owned enterprises, larger firms, and financially constrained firms. These findings highlight the economic value of ESG practices in promoting corporate investment stability and provide relevant insights for policy design and market participants.

1. Introduction

Economic volatility remains a central concern in macroeconomics, as maintaining stable fluctuations is essential for fostering positive expectations, expanding domestic demand, promoting employment, and enhancing social welfare [1]. The global business environment has undergone a profound transformation, demarcated by the COVID-19 pandemic [2,3]. In the pre-pandemic era of relative stability and expansion, corporate investment cycles were largely predictable. However, the pandemic triggered unprecedented supply chain disruptions and intense liquidity pressures, exposing the vulnerabilities of conventional business models [4]. As the world transitions into a post-pandemic phase, this crisis has acted as a catalyst, accelerating digitalization processes, shifting investment priorities towards resilience-building, and elevating stakeholder expectations for robust corporate governance [5]. This evolution, while opening avenues for supply chain transformation [6], has also starkly revealed the significant resilience barriers, particularly for micro and small enterprises [7,8,9]. Within this new paradigm of heightened uncertainty, corporate investment behavior—a fundamental driver of macroeconomic dynamics—demands renewed scholarly attention. Investment volatility can trigger short-term economic fluctuations, making its stabilization crucial for broader macroeconomic stability [10].

In this context, Environmental, Social, and Governance (ESG) principles have moved from a peripheral concern to a central focus for firms navigating the post-crisis landscape. A growing body of literature examines how ESG influences corporate investment decisions. Strong ESG performance is associated with lower capital costs [11,12], improved financing access, and a greater propensity for long-term investments in areas such as green innovation and R&D [13]. Firms with high ESG ratings are also more likely to engage in and expand outward investment [14]. Moreover, ESG attributes can redirect capital flows by shaping investor preferences and market valuations, thereby promoting funding for sustainable projects [15]. Evidence further suggests that ESG performance enhances investment efficiency [16], partly through improved disclosure quality that mitigates information asymmetry [17].

Despite these insights, the empirical evidence regarding the relationship between ESG and investment volatility remains mixed and inconclusive. Some studies indicate that high ESG performance dampens investment fluctuations by building resilience. For instance, Eccles et al. [18] observed that firms with robust ESG practices exhibited lower capital expenditure volatility during the 2008 financial crisis. Similarly, Arouri and Pijourlet [19] found that ESG-leading firms adjusted investment less drastically in response to macroeconomic shocks, suggesting a risk-mitigation mechanism. In contrast, other studies suggest that ESG commitments may introduce new sources of rigidity or instability. Landier and Lovo [20] theorized that managerial hesitancy in complex ESG-related decisions could amplify investment volatility. Battiston et al. [21] reported a U-shaped volatility pattern among European energy firms undergoing green transitions, while Pedersen et al. [15] noted that environmental (E) factors might increase volatility in certain sectors, even as governance (G) factors promote stability.

This divergence underscores a critical gap: a lack of understanding of whether and how ESG performance translates into investment stability in the post-pandemic era, particularly in emerging economies like China. While ESG is often discussed as a disclosure framework or governance infrastructure, its role as a direct stabilizer of corporate investment cycles remains underexplored. This study aims to fill that gap by systematically investigating the impact of ESG performance on firms’ investment volatility and positioning it as a key metric for corporate stability in the new normal.

The contributions of this paper are threefold. First, this study introduces and empirically validates a crucial moderating role of patient capital in the ESG-investment stability nexus. While existing literature predominantly examines the direct benefits of ESG, we reveal that the stabilizing effect of ESG on corporate investment is significantly attenuated in the presence of patient capital. This finding challenges the conventional view of ESG as a universally stabilizing mechanism and suggests that long-term institutional ownership and relational financing can functionally substitute for the risk-mitigation role of ESG, thereby offering a more nuanced understanding of how different stability-inducing factors interact within firms.

Second, we provide robust evidence that ESG performance mitigates investment volatility by simultaneously addressing two core market frictions: information asymmetry and agency costs. By formally testing and validating these parallel mediating channels, our research moves beyond establishing a direct correlation to delineating the concrete theoretical pathways—grounded in imperfect market theory—through which ESG translates into real economic stability. This mechanistic clarity significantly advances the scholarly conversation from “whether” ESG matters to “how” it operates.

Finally, our analysis documents substantial variation in the effectiveness of ESG across ownership structures, firm sizes, and financial constraints. Specifically, we find that the investment-smoothing effect of ESG is more pronounced in state-owned enterprises, small firms, and financially constrained firms. These results highlight the conditional nature of ESG benefits and suggest that firms facing higher baseline levels of market frictions derive greater marginal gains from ESG adoption—an insight critical for targeted policy design and firm-level ESG strategy.

2. Literature Review

Classical financial theory posits that under perfect market conditions, firm value is independent of financing decisions [22]. However, real-world markets are characterized by pervasive information asymmetry and agency problems, which exacerbate financing frictions [23]. Specifically, information asymmetry impedes external investors’ ability to accurately assess firm quality, thereby increasing the cost of capital, while agency issues encourage myopic managerial behavior that undermines long-term value. These market imperfections provide a theoretical foundation for understanding how ESG (environmental, social, and governance) performance influences corporate financial and governance outcomes. Recent studies further suggest that ESG practices can mitigate such frictions through signaling and stakeholder trust mechanisms [18].

Empirical evidence indicates that strong ESG performance helps reduce firms’ financing costs and enhances both debt and equity financing capacity [24]. For instance, by disclosing ESG information, firms signal their long-term sustainability and risk management capabilities to the market, thereby alleviating information asymmetry [25]. At the governance level, ESG practices strengthen stakeholder engagement, curb managerial opportunism, and improve decision-making quality. Recent research further highlights the resilience value of ESG during crises. For example, Durst et al. [26] argue that the resilience mechanisms of startups rely on their ESG foundations—maintaining operational stability through social responsibility and governance structures amid resource constraints. Similarly, Iqbal et al. [27] find that micro and small fast-fashion enterprises strengthened their demand resilience and achieved post-pandemic recovery by reinforcing social and environmental responsibilities.

Ownership structure and the nature of capital play key moderating roles in the relationship between ESG and corporate performance. Institutional investors, particularly long-term oriented “patient capital” such as pension and sovereign wealth funds, are more inclined to support corporate ESG investments [28]. Such investors are willing to tolerate short-term earnings volatility in exchange for long-term risk-adjusted returns. Moreover, ownership concentration and shareholder type also influence the effectiveness of ESG implementation. For example, family-owned firms or those with significant state ownership often exhibit greater continuity in ESG practices [29]. Using panel data analysis, Mandal et al. [30] demonstrate that the financing cost reduction effect of ESG is more pronounced in firms with dispersed ownership, underscoring the contingent role of governance structure in the ESG transmission mechanism.

The COVID-19 pandemic, as an exogenous shock, profoundly reshaped corporate operations and governance environments. Studies show that firms with robust ESG foundations exhibited greater resilience and recovery capacity during the pandemic [31]. For instance, El Khoury et al. [32] analyzed the buffering effect of green supply chain practices amid the pandemic and found that firms implementing environmental management were better equipped to cope with supply chain disruption risks. Meanwhile, Guo et al. [33], using data from Chinese listed companies, revealed a positive relationship between government environmental pressure and corporate green innovation during the pandemic, highlighting how macro-level governance environments shape corporate ESG behavior. Collectively, these studies illustrate that, in times of crisis, ESG serves not only as a safeguard for stable operations but also as a strategic tool for responding to macro-level policies and market shifts.

In summary, the existing literature has established a solid theoretical and empirical foundation for understanding the economic consequences of ESG. Research consistently confirms that ESG practices can effectively reduce corporate financing frictions and improve governance outcomes by mitigating information asymmetry and agency problems. Nevertheless, several research gaps remain worthy of further exploration. Future studies should aim to: (1) extend the scope of research to include more diverse types of enterprises, thereby constructing a more generalizable ESG theoretical framework; (2) employ longitudinal tracking and in-depth case studies to dynamically uncover the process mechanisms through which moderating variables influence ESG effectiveness; and (3) develop integrated models that connect macro-level governance, meso-level industries, and micro-level corporate behaviors to more comprehensively assess the value and challenges of ESG in a complex and evolving context. This study seeks to address some of these gaps by examining the role of imperfect market theory and patient capital in shaping ESG outcomes.

3. Research Hypothesis

Within the framework of perfect market assumptions, firms can invest in all projects with positive net present values, and frictionless real business cycle (RBC) models suggest that external financing has no material impact on investment. However, the Imperfect Markets Hypothesis contends that real-world financial markets are rife with information asymmetry, transaction costs, behavioral biases, and institutional constraints, all of which diminish capital allocation efficiency and exacerbate corporate investment volatility [34]. Severe information asymmetry between a firm and external stakeholders can result in systematic mispricing of assets and cash flows, inducing adverse selection and moral hazard problems that distort investment behavior—leading to over- or under-investment, higher financing costs, and reduced investment efficiency [35]. These frictions make it difficult for firms to sustain stable investment levels. Principal-agent conflicts further impede arbitrage, causing stock prices to deviate from fundamentals and amplifying distortions in corporate investment [36]. For instance, overconfident CEOs are more likely to pursue high-risk acquisitions, thereby increasing volatility in firm value [37].

We argue that ESG (Environmental, Social, and Governance) practices can alleviate these market imperfections by enhancing information transparency, mitigating agency costs, and fostering investor confidence. In China, where capital markets are still developing and bank credit remains the dominant source of external finance, information asymmetry between lenders and firms is particularly acute, leading to widespread financing constraints [38]. High-quality ESG performance helps mitigate these constraints through two primary channels.

First, from a signaling perspective, banks perceive firms with strong ESG performance as more transparent and less risky [39]. ESG-oriented firms are more likely to engage in high-quality voluntary disclosures, cultivate responsible corporate images, and accumulate reputational capital, which collectively alleviate information asymmetry and reduce financing costs. Second, ESG adherence aligns with public policy goals, thereby reducing regulatory risks and curbing managerial opportunism in financial reporting [40]. Such firms often benefit from preferential treatment such as tax incentives and fiscal subsidies [41], which further stabilizes financing access and improves capital allocation efficiency. By easing financing constraints and reducing information friction, ESG performance helps stabilize corporate investment [42].

Additionally, ESG practices help reconcile the conflict between managerial risk aversion and shareholder value maximization. Stakeholder theory [43] suggests that ESG integration fosters alignment between managers and a broader set of stakeholders, improves corporate governance, constrains self-interested managerial behavior, and reduces agency costs [44]. This alignment enhances mutual trust and facilitates more stable investment strategies.

Accordingly, we propose the following:

Hypothesis 1.

Corporate ESG performance is negatively associated with investment volatility.

Subsequently, this study examines the mechanisms through which corporate ESG performance influences investment volatility, drawing upon the theoretical foundation of the Imperfect Markets Hypothesis. Within China’s institutional context, where capital market mechanisms remain underdeveloped, bank credit constitutes the primary source of external financing for most firms. However, pervasive information asymmetry between banks and enterprises often leads to significant financing constraints [38]. We argue that superior ESG performance helps mitigate such constraints through two distinct yet complementary channels.

First, based on signaling theory, high-quality ESG performance conveys positive signals to financial intermediaries regarding a firm’s operational transparency and managerial credibility [39]. Firms with strong ESG profiles tend to engage in more proactive and reliable information disclosure, which not only enhances their reputational capital but also reduces perceived lending risks. As a result, banks are more inclined to offer favorable credit terms, thereby alleviating financing frictions and contributing to more stable corporate investment.

Moreover, strong ESG engagement aligns with China’s policy priorities, enabling firms to gain institutional legitimacy and governmental support. Such alignment not only reduces regulatory risks but also discourages managerial opportunism in financial reporting and earnings management [40]. In addition, ESG-committed firms often benefit from policy-driven incentives such as tax benefits and fiscal subsidies, which further stabilize cash flows and lower external financing dependence. Through these combined effects, ESG performance enhances the efficiency of credit allocation in the capital market, ensures financing stability, and ultimately attenuates investment volatility [45]. Accordingly, we propose the following hypothesis:

Hypothesis 2.

Corporate ESG performance reduces investment volatility by mitigating information asymmetry.

A second mechanism operates through the lens of stakeholder theory and agency costs. In traditional corporate finance frameworks, a fundamental conflict exists between managerial risk aversion and shareholder value maximization, often leading to suboptimal investment behaviors. However, firms that embed ESG principles into their governance and operational practices are better positioned to align the interests of managers with those of a broader set of stakeholders, including shareholders, creditors, and employees.

Specifically, robust ESG frameworks strengthen internal monitoring mechanisms and enhance managerial accountability, thereby curbing self-serving behaviors such as excessive risk-taking or short-term earnings manipulation. By improving transparency and fostering trust between key stakeholders, ESG practices help mitigate principal-agent problems within the firm-bank relationship. This, in turn, reduces agency costs and encourages longer-term, more stable investment planning. Based on this reasoning, we propose the following:

Hypothesis 3.

Corporate ESG performance reduces investment volatility by lowering agency costs.

The prevailing literature posits that superior ESG performance acts as a “stabilizer” for corporate investment by mitigating information asymmetry, managing tail risks, and bolstering stakeholder trust [46]. However, this view often implicitly assumes homogeneity in capital’s response to ESG information. This study introduces patient capital as a key moderating variable, framing its interaction with ESG through the lens of governance economics, which involves potential Substitution and Complementary Effects.

The Substitution Effect: Patient capital, by virtue of its long-term orientation and tolerance for short-term volatility, constitutes a potent direct governance mechanism in itself. It can directly ensure the stability of long-term investment projects by actively monitoring and engaging with management to curb myopic behavior [47]. In this context, ESG practices—promoting transparent disclosure and stakeholder management—serve as an indirect governance tool aimed at stabilizing expectations by shaping a favorable reputation and operational environment. If the direct monitoring from patient capital is sufficiently effective, the marginal benefit of ESG as an ancillary stabilizing mechanism diminishes. In essence, patient capital may substitute for the stabilizing function of ESG in mitigating investment volatility.

The Complementary Effect: Conversely, when institutional investors explicitly and systematically integrate ESG principles into their investment decisions and post-investment stewardship, a powerful synergy can emerge. Patient capital provides the necessary temporal runway and risk tolerance for ESG-oriented long-term strategies, while the ESG framework offers structured metrics and engagement levers for patient investors’ oversight. In this scenario, patient capital does not weaken but rather amplifies the monitoring and disclosure benefits of ESG, collectively forging a more profound level of corporate resilience.

The core theoretical contribution of this study is to demonstrate that the net impact of patient capital is determined by the tension between these Substitution and Complementary Effects. The following theoretical model formalizes this intuition.

To elucidate the underlying mechanisms, we construct a parsimonious two-period model.

Model Setup:

Agents: A risk-neutral firm manager; a representative institutional investor.

Timeline: When t = 0, the firm holds an initial investment project valued at ${\mathrm{V}}_0$.

The manager chooses a level of ESG investment $\mathrm{e}\ge 0$, with a cost function $\mathrm{C}(\mathrm{e})={\displaystyle \frac{1}{2}}\mathrm{c}{\mathrm{e}}^2$. When t = 1, the firm suffers an exogenous negative operational shock $\mathsf{ϵ}$, distributed as $\mathsf{ϵ}~(0,{\mathsf{\delta }}^2(\mathrm{e})).$ The post-shock investment value is

$${\mathrm{V}}_1\left(\mathrm{e},\mathsf{ϵ}\right)={\mathrm{V}}_0+\mathrm{g}\left(\mathrm{e}\right)-\mathsf{ϵ}$$

Here, $\mathrm{g}(\mathrm{e})$ represents the value-enhancing benefit of ESG, with ${\mathrm{g}}^{\prime }(\mathrm{e})$ > 0 and ${\mathrm{g}}^{″}(\mathrm{e})$ < 0.

After observing $\mathsf{ϵ}$, the manager decides whether to undertake a follow-on long-term investment with a net present value L > 0. The decision is denoted by $\mathrm{d}\in \left\{\mathrm{0,\; 1}\right\}$.

Investment Volatility: This study focuses on the volatility of investment expenditure between t = 0 and t = 1. High volatility is characterized by the manager forgoing valuable long-term investments due to short-term pressures.

Investor Types: An investor’s “patience” is characterized by their investment horizon and sensitivity to short-term performance.

Myopic Investor: Cares only about the t = 1 valuation ${\mathrm{V}}_1$. If ${\mathrm{V}}_1$ falls below an exogenous threshold $\underset{\_}{\mathrm{V}}$, they exert pressure, forcing the manager to abandon the long-term investment (setting $\mathrm{d}=0$).

Patient Investor: Cares about long-term total value ${\mathrm{V}}_1+\mathrm{d}·\mathrm{L}$. They understand the transitory nature of the shock $\mathsf{ϵ}$ and do not pressure the manager based on a temporary dip in ${\mathrm{V}}_1$, allowing the manager to always choose $\mathrm{d}=1$ if $\mathrm{L}>0$.

The Dual Role of ESG:

Value Enhancement: Directly improves fundamentals through $\mathrm{g}\left(\mathrm{e}\right)$.

Resilience Building: High levels of ESG practices can attenuate the intensity of operational shocks. We formalize this by assuming the shock’s variance ${\mathsf{\delta }}^2(\mathrm{e})$ is a decreasing function of $\mathrm{e}$, i.e., ${\displaystyle \frac{{\mathrm{d}\mathsf{\delta }}^2(\mathrm{e})}{\mathrm{d}\mathrm{e}}}<0$. This captures the resilience-building effect of institutionalized ESG disclosure highlighted by Khurana et al. [4], and the risk-profile transformation of supply chains noted by Min [6].

Equilibrium Analysis:

Baseline Scenario (Myopic Investor):

At t = 1, if the realized shock $\mathsf{ϵ}$ is sufficiently large such that ${\mathrm{V}}_1<\underset{\_}{\mathrm{V}}$, the manager is pressured to forgo the long-term investment ($\mathrm{d}=0$). The probability of this event is $\mathrm{P}\mathrm{r}\left({\mathrm{V}}_0+\mathrm{g}\left(\mathrm{e}\right)-\mathsf{ϵ}<\underset{\_}{\mathrm{V}}\right)$.

The manager chooses e at t = 0 to maximize expected value Increasing e raises

${\mathrm{V}}_1$ via $\mathrm{g}\left(\mathrm{e}\right)$ and reduces the probability of a severe shock by lowering ${\mathsf{\delta }}^2(\mathrm{e})$. Both effects decrease the risk of investment disruption from short-term pressure. Thus, in the presence of myopic capital, ESG exerts a significant negative influence on investment volatility.

Moderating Scenario (Patient Investor):

Patient capital eliminates short-term pressure; the manager always undertakes the optimal long-term investment ($\mathrm{d}=1$). Here, the primary motivation for the manager to choose ee stems from its long-term value creation $\mathrm{g}\left(\mathrm{e}\right)$, not from its utility in fending off short-term market pressure.

Theoretical Core: In an environment with patient capital, the function of ESG in stabilizing investment via the resilience channel (reducing ${\mathsf{\delta }}^2(\mathrm{e})$) is partially weakened. This is because patient capital itself provides “insurance” against volatility, even without ESG absorbing the shock. This embodies the Substitution Effect.

However, the model also reveals the condition for the Complementary Effect: If the patient investor actively practices ESG integration, incorporating ESG performance ee into their stewardship and evaluation framework, the manager has an additional incentive to increase ee. In this case, patient capital and ESG deeply integrate, working in concert to render the investment plan exceptionally stable even in the face of massive external shocks, such as a pandemic.

Based on the theoretical reasoning above, we propose the following set of testable hypotheses:

Hypothesis 4a.

Patient capital weakens the negative relationship between ESG performance and investment volatility.

Hypothesis 4b.

If patient capital systematically integrates ESG factors, it will strengthen the negative relationship between ESG performance and investment volatility.

4. Data, Variables, and Research Methodology

4.1. Data

This study utilizes financial data from A-share listed companies on the Shanghai and Shenzhen stock exchanges between 2012 and 2022. The core explanatory variable, ESG data, is sourced from the Wind database, while corporate investment volatility and other financial data are obtained from the CSMAR database. Macroeconomic data are collected from the National Bureau of Statistics website.

The data processing procedure includes the following steps: (1) elimination of samples from companies with abnormal listing status (ST and PT stocks); (2) exclusion of samples from industries with distinctive financial characteristics, such as financial and real estate sectors; and (3) application of a 1% winsorization to all continuous variables on an annual basis to mitigate the influence of extreme values. The final dataset comprises 11,971 observations.

4.2. Dependent Variable

The variable inv_fluc represents corporate investment volatility. Following the approach of Wang et al. [42], we employ the Hodrick-Prescott (HP) filter to estimate investment fluctuations among Chinese listed firms. Corporate investment is measured as the ratio of cash outlays for fixed assets, intangible assets, and other long-term assets to total assets.

Since the HP filter requires balanced panel data, we exclude samples with missing values to ensure data balance. Investment volatility is quantified as the cyclical component of investment relative to actual fixed asset investment. A positive value indicates overinvestment, while a negative value suggests underinvestment. To maintain consistency in interpretation, we take the absolute value for all observations with negative investment volatility.

4.3. Independent Variable

The Huazheng ESG Index, which has been assessing the ESG performance of A-share issuers and bond-issuing entities since 2009, is widely recognized in both academic and professional circles. This study employs Huazheng ESG ratings as a proxy for corporate ESG performance. The index classifies companies into nine tiers (C, CC, CCC, B, BB, BBB, A, AA, AAA), which we numerically encode from 1 to 9 to reflect ascending ESG quality.

4.4. Control Variables

To mitigate omitted variable bias, this study selects a series of control variables at both the macro and micro levels. At the macro level, gross domestic product (GDP) and GDP per capita are included. At the micro level, the selected variables include the largest shareholder’s ownership ratio (TOP1), Board size (ND), market-to-book ratio (Tobin Q), current ratio (CR), asset- leverage ratio (ALR), return on assets (ROA), firm size (size), cash flow (cashflow), firm age (age), equity-to-asset ratio (EA) and Financial Constraint KZ Index (KZ).

4.5. Model Construction

$${\mathrm{i}\mathrm{n}\mathrm{v}\_\mathrm{f}\mathrm{l}\mathrm{u}\mathrm{c}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\alpha }}_2{\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{E}\mathrm{A}\mathrm{R}\mathrm{F}\mathrm{E}+\mathrm{I}\mathrm{N}\mathrm{D}\mathrm{U}\mathrm{S}\mathrm{T}\mathrm{R}\mathrm{Y}\mathrm{F}\mathrm{E}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(1)

$${{\mathrm{i}\mathrm{n}\mathrm{v}}_{\mathrm{f}\mathrm{l}\mathrm{u}\mathrm{c}}}_{\mathrm{i},\mathrm{t}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\beta }}_2{{\mathrm{E}\mathrm{S}\mathrm{G}}_{\mathrm{i},\mathrm{t}-1}\ast {\mathrm{M}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\beta }}_3{\mathrm{M}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\beta }}_4\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{Y}\mathrm{E}\mathrm{A}\mathrm{R}\mathrm{F}\mathrm{E}+\mathrm{I}\mathrm{N}\mathrm{D}\mathrm{U}\mathrm{S}\mathrm{T}\mathrm{R}\mathrm{Y}\mathrm{F}\mathrm{E}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(2)

5. Empirical Study

5.1. Descriptive Statistics

Table 1. Descriptive statistics.

	(1)	(2)	(3)	(4)	(5)
VARIABLES	N	Mean	sd	Min	Max
esg	11,971	4.147	0.991	1	8
inv_fluc	11,971	0.563	1.157	0.005	8.888
gdp	11,971	10.46	0.765	7.673	11.73
gdpper	11,971	1.923	0.468	0.631	2.931
TOP1	11,971	34.92	15.37	8.804	75.00
ND	11,971	8.917	1.784	5	15
Tobin Q	11,971	1.884	1.241	0.821	8.144
CR	11,971	1.854	1.591	0.260	11.51
ALR	11,971	0.480	0.195	0.0676	0.917
ROA	11,971	0.0359	0.0557	−0.196	0.200
size	11,971	22.85	1.371	19.79	26.58
cashflow	11,971	0.0537	0.0670	−0.155	0.256
age	11,971	15.54	6.100	2	28
EA	11,971	0.235	0.174	0.002	0.734
KZ	11,971	0.564	1.563	0.005	8.888

presents the descriptive statistics for all variables used in the empirical analysis. The mean value of ESG performance is 4.147 with a standard deviation of 0.991, indicating substantial variation across firms. The scores range from 1 to 8, confirming that no firm in the sample achieved a top-tier AAA rating during the sample period. The dependent variable, investment volatility (inv_fluc), has a mean of 0.563 and a standard deviation of 1.157, with values ranging from 0.005 to 8.888. This wide dispersion reflects considerable heterogeneity in investment behavior among Chinese listed firms. The distributions of all control variables are consistent with those reported in prior studies, supporting the representativeness of our sample.

5.2. Baseline Regression Results

Table 2. The relationship of ESG and Corporate Investment Volatility.

	(1)	(2)	(3)	(4)
VARIABLES	inv_fluc	inv_fluc	inv_fluc (Over)	inv_fluc (Under)
esg	−0.1190 ***	−0.0644 ***	−0.0138 **	−0.0951 ***
	(0.0112)	(0.0117)	(0.00631)	(0.0190)
Tobin Q		0.0154	−0.0120 *	0.0144
		(0.0146)	(0.00659)	(0.0228)
CR		0.0633 ***	0.00989 *	0.0955 ***
		(0.0150)	(0.00592)	(0.0236)
ALR		0.594 ***	0.0585	0.832 ***
		(0.142)	(0.0575)	(0.225)
ROA		−1.380 ***	0.120	−1.840***
		(0.306)	(0.256)	(0.468)
size		−0.146 ***	−0.0439 ***	−0.217 ***
		(0.0122)	(0.00745)	(0.0198)
cashflow		−0.477	0.0761	−0.621
		(0.324)	(0.188)	(0.504)
age		0.00148	−0.00134	0.00369
		(0.00210)	(0.00130)	(0.00339)
gdp		−0.0483 ***	−0.0401 ***	−0.0622 **
		(0.0184)	(0.00984)	(0.0300)
gdpper		−0.00255	−0.00257	−0.000897
		(0.0292)	(0.0125)	(0.0485)
EA		−0.136	−0.205 ***	−0.279 *
		(0.0955)	(0.0615)	(0.151)
ND		−0.0153 ***	−0.00148	−0.0247 ***
		(0.00565)	(0.00205)	(0.00945)
TOP1		0.00139 *	−0.000228	0.00245**
		(0.000755)	(0.000380)	(0.00124)
KZ		−0.0366 **	0.00548	−0.0426 *
		(0.0145)	(0.00786)	(0.0222)
Constant	1.098 ***	4.301 ***	1.593 ***	6.399 ***
	(0.208)	(0.369)	(0.219)	(0.609)
YEAR FE	YES	YES	YES	YES
INDUSTRY FE	YES	YES	YES	YES
Observations	11,971	11,971	5176	6795
R-squared	0.074	0.102	0.134	0.139

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the estimation results from the baseline regression models examining the effect of ESG performance on investment volatility. Column (1) presents results from a parsimonious specification controlling only for industry and year fixed effects. The coefficient on ESG is negative and statistically significant at the 1% level (β = −0.1190). In Column (2), we incorporate a full set of firm-level and macroeconomic control variables while retaining fixed effects. The estimated coefficient on ESG remains negative and significant at the 1% level (β = −0.0644), Our calculations indicate that a one-standard-deviation increase in a firm’s ESG rating is associated with a decrease in investment volatility of 0.060451 absolute units, after controlling for other factors. Relative to the sample mean of investment volatility (0.563), this translates to a reduction of approximately 10.74% (−0.060451/0.563 × 100% = −10.74%). The magnitude of this effect demonstrates that the stabilizing impact of ESG performance on corporate investment is not only statistically significant but also economically substantial supporting Hypothesis 1.

We further examine whether the effect holds for both over-investing and under-investing firms. Results from subsample analyses confirm that ESG performance significantly reduces investment volatility in both contexts, reinforcing the baseline conclusion.

5.3. Robustness Checks

To assess the sensitivity of our results, we conduct robustness tests using an alternative measure of corporate investment. Specifically, we redefine investment volatility using the sum of cash paid for fixed and intangible assets and net cash outflow for acquisitions, scaled by beginning-of-year total assets. The results, presented in

Table 3. Robustness test.

	(1)	(2)	(3)	(4)
VARIABLES	inv_fluc3	inv_fluc3	inv_fluc	inv_fluc
esg	−0.336 **	−0.169 *	−0.0568 ***	−0.0482 ***
	(0.135)	(0.0939)	(0.0131)	(0.0131)
Tobin Q		−0.0511		0.0066
		(0.0472)		(0.0178)
CR		0.0354		0.0578 ***
		(0.0603)		(0.0173)
ALR		−0.156		0. 7258 ***
		(0.932)		(0. 1972)
ROA		−8.067 *		−0.9172 **
		(4.663)		(0.3307)
size		−0.558 **		−0.2748 ***
		(0.218)		(0. 0393)
cashflow		11.95		−0.1347
		(9.975)		(0. 3304)
age		0.0156		0.0491
		(0.0135)		(0.0650)
gdp		−0.0735		0.2825 ***
		(0.106)		(0.1368)
gdpper		0.375 *		−0.1709
		(0.202)		(0.1689)
EA		−2.822 **		−0.2972
		(1.378)		(0.1914)
ND		0.0364		0.0043
		(0.0562)		(0.0123)
TOP1		0.00178		−0.0005
		(0.00782)		(0.0019)
KZ		0.0573		−0.0342 ***
		(0.138)		(0.0161)
Constant	1.989 ***	13.00 ***	0.0084	2.9465
	(0.561)	(3.326)	(0. 6694)	(2.0743)
YEAR FE	YES	YES	YES	YES
INDUSTRY FE	YES	YES	YES	YES
ID FE	YES	YES	YES	YES
Observations	9967	9967	11,971	11,971
R-squared	0.020	0.024	0.318	0.327

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

, are consistent with those in the baseline model both in terms of significance and direction, affirming the reliability of our main findings.

Furthermore, to assess the robustness of our findings, we re-evaluated our models using alternative λ values (λ = 50 and λ = 100) for reconstructing the investment volatility measure. The choice of λ = 100 follows the conventional practice for annual data, while the inclusion of an intermediate value (λ = 50) enables us to examine the continuity of the core explanatory variable’s coefficient as trend smoothness varies, thereby enhancing the credibility of our inference. As presented in

Table 4. Robustness test (λ).

	(1)	(2)	(3)	(4)
VARIABLES	inv_fluc (λ = 100)	inv_fluc (λ = 100)	inv_fluc (λ = 50)	inv_fluc (λ = 50)
esg	−0.660 ***	−0.289 ***	−0.628 ***	−0.275 ***
	(0.221)	(0.109)	(0.231)	(0.105)
TobinQ		1.189		1.182
		(0.908)		(0.949)
CR		0.144		0.137
		(0.229)		(0.225)
ALR		1.494		1.840
		(3.376)		(3.577)
ROA		−14.66 **		−12.78 **
		(6.357)		(5.617)
size		−0.423 *		−0.425 *
		(0.216)		(0.226)
cashflow		−7.042		−8.104
		(10.48)		(11.19)
age		0.0943		0.100
		(0.0707)		(0.0759)
gdp		0.0817		0.0896
		(0.196)		(0.183)
gdpper		−0.369		−0.300
		(0.326)		(0.295)
EA		−0.560		−0.412
		(0.849)		(0.831)
ND		0.0335		0.0493
		(0.0820)		(0.0855)
TOP1		−0.000582		−0.00240
		(0.00720)		(0.00643)
KZ		−0.201		−0.228
		(0.395)		(0.418)
Constant	3.605 ***	6.706 **	3.342 ***	6.211 **
	(1.133)	(2.951)	(1.033)	(2.695)
YEAR FE	YES	YES	YES	YES
INDUSTRY FE	YES	YES	YES	YES
Observations	11,971	11,971	11,971	11,971
R-squared	0.018	0.024	0.017	0.023

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

, the negative relationship between ESG performance and investment volatility remains statistically significant at the 1% level across all re-estimated specifications, with coefficient directions consistent with our baseline results. This evidence confirms that our primary conclusion is robust to parameter selection in the HP filter, providing strong support for the empirical findings.

Given the short-panel structure of our data, the baseline model employs time and industry fixed effects. To address this and to test the robustness of our findings, we re-estimate the model by incorporating firm fixed effects. As reported in columns (3) and (4) of Table 3, the coefficient on our key variable of interest, ESG, remains negative and statistically significant at the 1% level, which aligns with the main results.

5.4. Addressing Endogeneity

First, to address potential reverse causality, we re-estimated the model using the one-period lagged value of the core explanatory variable. The results, reported in Columns (1) and (2) of

Table 5. Endogeneity test.

	(1)	(2)	(3)	(4)
VARIABLES	inv_fluc	inv_fluc	inv_fluc(IV)	inv_fluc(IV2)
IV			0.9818 ***
			(0.0339)
IV2				0.5015 ***
				(0.0443)
L.esg	−0.0981 ***	−0.0477 ***
	(0.0115)	(0.0119)
Tobin Q		0.0129	0.0176	−0.0196 ***
		(0.0129)	(0.0114)	(0.0072)
CR		0.0436 ***	0.0677 ***	0.0119 **
		(0.0147)	(0.0116)	(0.0061)
ALR		0.151	0.649 ***	−0.3965 ***
		(0.110)	(0.108)	(0.0687)
ROA		−1.545 ***	−1.624 ***	1.9104 ***
		(0.331)	(0.272)	(0.1792)
size		−0.126 ***	−0.0765 ***	0.2089 ***
		(0.0121)	(0.0140)	(0.0077)
cashflow		−0.149	−0.572 **	−0.6276 ***
		(0.246)	(0.242)	(0.1575)
age		0.00176	0.00607 ***	−0.0177 ***
		(0.00220)	(0.00177)	(0.0013)
gdp		−0.0449 **	−0.0333 **	−0.0008 ***
		(0.0196)	(0.0154)	(0.0114)
gdpper		−0.000364	0.0454 **	0.0367 ***
		(0.0305)	(0.0224)	(0.0182)
EA		0.0552	−0.323 ***	−0.3963 ***
		(0.0988)	(0.0719)	(0.0504)
ND		−0.0105 *	−0.0195 ***	0.0041
		(0.00573)	(0.00451)	(0.0042)
TOP1		0.00140 *	0.00243 ***	−0.0014 **
		(0.000809)	(0.000623)	(0.0005)
KZ			−0.0321 ***	−0.0241 ***
			(0.00961)	(0.0061)
F			195.96 ***	153.79 ***
Constant	1.015 ***	3.841 ***	2.960 ***	−0.1028
	(0.205)	(0.374)	(0.245)	(0.1905)
Observations	10,148	10,148	11,971	11,971
R-squared	0.073	0.097	0.129	0.117

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

, confirm that its negative impact remains statistically significant.

Second, we confront potential reverse causality by implementing an instrumental variable (IV) strategy. Our instrument is the annual average ESG rating of all other firms headquartered in the same province. This variable is theoretically relevant as firms are influenced by their local regulatory and social environment. We argue that it is plausibly exogenous because, while it affects a firm’s own ESG through regional spillover effects, it is unlikely to be directly correlated with the firm-specific unobservables driving its investment volatility. The IV estimates in Column (3) of Table 5 affirm a statistically significant negative effect, corroborating the causal inference from our baseline model.

To further mitigate endogeneity concerns, we employ a difference-in-differences (DID) design as an instrumental variable (IV2) strategy, exploiting the mandatory ESG disclosure policy introduced in Hong Kong as a quasi-natural experiment. Our instrument is the interaction term H × t, where the group dummy (H) indicates the treatment group (1 for A + H cross-listed firms, 0 for A-share-only listed firms), and the time dummy (t) distinguishes the post-policy period (1 for 2016–2022, 0 for 2012–2015). The exclusion restriction is plausibly satisfied as the policy shock, while affecting cross-listed firms’ ESG practices, is unlikely to be directly correlated with idiosyncratic factors driving their investment volatility. As reported in Column (4) of Table 5, the second-stage results of the 2SLS estimation confirm a statistically significant effect of our key explanatory variable, lending strong support to the causal inference from our baseline analysis.

5.5. Moderation Analysis

5.5.1. Tests of the Agency Cost Mechanism

Table 6. Mechanism Test Regression.

	(1)	(2)
VARIABLES	Cost	Information
esg	−0.0191 ***	−0.00177 *
	(0.00333)	(0.00104)
TobinQ	−0.188 ***	0.0130 ***
	(0.00635)	(0.00160)
CR	0.0231 ***	0.00354 **
	(0.00265)	(0.00150)
ALR	0.275 ***	−0.0704 ***
	(0.0309)	(0.0107)
ROA	−1.045 ***	−0.438 ***
	(0.0723)	(0.0350)
size	−0.328 ***	−0.00904 ***
	(0.00616)	(0.00110)
cashflow	0.344 ***	0.0133
	(0.0765)	(0.0270)
age	0.00299 ***	−0.000891 ***
	(0.000667)	(0.000187)
gdp	0.00426	−0.0102 ***
	(0.00523)	(0.00164)
gdpper	0.0195 **	0.00594 *
	(0.00972)	(0.00318)
EA	0.0556 **	−0.0507 ***
	(0.0265)	(0.00808)
ND	0.00334	0.000443
	(0.00221)	(0.000633)
TOP1	0.00612 ***	−0.000333 ***
	(0.000231)	(5.96e-05)
KZ	0.0239 ***	−0.000886
	(0.00368)	(0.00108)
Constant	6.966 ***	0.531 ***
	(0.140)	(0.0352)
YEAR FE	YES	YES
INDUSTRY FE	YES	YES
Observations	11,971	11,971
R-squared	0.660	0.471

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the results of mechanism analysis examining whether agency costs represent a significant channel through which ESG performance affects corporate investment volatility. We measure agency costs using the administrative expense ratio, where a higher ratio indicates greater managerial perquisite consumption or organizational redundancy, reflecting increased consumption of shareholder resources—that is, higher agency costs. The estimates in Column (1) reveal that firms with stronger ESG performance exhibit significantly lower agency costs, which in turn leads to a discernible reduction in investment volatility. This finding supports the view that ESG commitments function as an internal governance mechanism, effectively curbing managerial self-interest and short-term risk-shifting behaviors. By enhancing transparency and accountability, high-ESG firms alleviate incentive misalignment between owners and managers, thereby attenuating over- or under-investment distortions and promoting smoother investment patterns.

5.5.2. Tests of the Information Asymmetry Mechanism

Column (2) presents the mediating role of information asymmetry, which is proxied by a Comprehensive Index of Information Asymmetry reflecting the degree of corporate disclosure quality and transparency. The results confirm that elevated ESG performance significantly reduces information asymmetry, resulting in more stable investment behavior. This outcome is consistent with the signaling view: high-ESG firms convey credible, non-financial signals to external stakeholders, which mitigates adverse selection and moral hazard problems in capital allocation. As a consequence, such firms benefit from lower financing frictions, improved credit terms, and sustained investor confidence—all contributing to a more predictable investment environment.

These results jointly validate Hypotheses 2 and 3, demonstrating that both the agency cost and information asymmetry channels underpin the stabilizing function of corporate ESG practices.

5.5.3. Moderating Role of Patient Capital

Table 7. Moderating effect regression.

	(1)	(2)
VARIABLES	inv_fluc	inv_fluc
esg	−0.141 ***	−0.0575 ***
	(0.0338)	(0.0121)
deptesg	0.879 **
	(0.389)
dept	−3.066 *
	(2.028)
investesg		0.0004 ***
		(0.0003)
invest		−0.0018 ***
		(0.0001)
Tobin Q	0.0304 **	0.000500
	(0.0134)	(0.0127)
CR	0.0528 ***	0.0594 ***
	(0.0120)	(0.0155)
ALR	0.389 ***	0.333 **
	(0.116)	(0.136)
ROA	−1.558 ***	−1.423 ***
	(0.275)	(0.329)
size	−0.130 ***	−0.130 ***
	(0.0103)	(0.0123)
cashflow	−0.299	−0.227
	(0.261)	(0.326)
age	0.00595 ***	0.000335
	(0.00188)	(0.00215)
gdp	−0.0379 **	−0.0465 **
	(0.0156)	(0.0188)
gdpper	0.00963	−0.00662
	(0.0241)	(0.0301)
EA	−0.155 **	−0.156
	(0.0784)	(0.0952)
ND	−0.0133 ***	−0.0131 **
	(0.00471)	(0.00572)
TOP1	0.00133 **	0.00202 ***
	(0.000638)	(0.000778)
KZ	−0.0249 **	−0.0211
	(0.0117)	(0.0131)
Constant	3.914 ***	3.859 ***
	(0.311)	(0.357)
YEAR FE	YES	YES
INDUSTRY FE	YES	YES
Observations	11,971	9977
R-squared	0.094	0.102

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

presents the empirical results testing the moderating role of patient capital in the relationship between corporate ESG performance and investment volatility. We employ two dimensions—relational debt and stable equity—to measure corporate patient capital. For relational debt (dept), it is measured as the ratio of long-term liabilities to total liabilities. For stable equity (invest), the overall shareholding ratio of institutional investors is used to gauge their aggregate investment level.

In Column (1), the analysis focuses on relational patient capital, measured as the strength of debt relationships (dept). The coefficient for the interaction term between dept and ESG is positive and statistically significant at the 5% level (β = 0.0004). This positive interaction coefficient indicates that the negative effect of ESG on volatility becomes less pronounced as relational patient capital increases. In other words, the investment-smoothing benefit of a high ESG rating is diminished for firms embedded in strong, relational debt networks. Concurrently, the direct effect of dept is itself negative and significant (β = −3.066), suggesting that relational debt alone acts as a buffer against investment volatility. This combination of results—a negative direct effect and a positive interaction effect—clearly illustrates a substitution dynamic: the stability provided by long-term, trust-based debt relationships partially supplants the need for the stability derived from a strong ESG profile, thereby suppressing ESG’s marginal utility.

The results in Column (2) for stable patient capital, proxied by stable institutional ownership (invest), are even more pronounced. The interaction term between invest and ESG is positive and highly significant (β = 0.0004). This confirms that the mitigating effect of ESG on investment volatility weakens substantially in the presence of committed, long-term institutional investors. The main effect of invest is also negative and significant (β = –0.0018), reinforcing the notion that stable investors directly instill investment discipline and provide a shield against short-term market pressures. The highly significant interaction term underscores a powerful moderating force. The underlying reason is that stable owners provide an exogenous, governance-based source of stability. When such a powerful stabilizing force is already in place, the earned stability from managing stakeholder relationships through ESG becomes less critical for smoothing investment cycles. Consequently, the efficacy of ESG as a risk-mitigation and stability-enhancing tool is significantly inhibited.

To quantify the moderating effect of patient capital, we computed the marginal effect of ESG on investment volatility at varying levels. The calculations reveal that patient capital attenuates the stabilizing role of ESG. Specifically: For relational debt, the marginal effect of ESG on investment volatility is −12.58% when it is at the sample mean but decreases to −8.97% when it increases to one standard deviation above the mean. For stable equity, the marginal effect is −10.09% at the sample mean and slightly declines to −10.04% at one standard deviation above the mean. These findings consistently demonstrate that as the level of either form of patient capital increases, the mitigating effect of ESG on investment volatility diminishes. This supports the conclusion of a substitution effect between patient capital and ESG, implying that they can, to some extent, act as substitutes in stabilizing corporate investment.

Collectively, the evidence from both columns strongly supports Hypothesis 4a. The consistent pattern—whereby different forms of patient capital exhibit a negative direct effect on volatility but a positive interaction effect with ESG—validates the theory of functional substitution. It demonstrates that the calming influence of patient capital on corporate investment comes, in part, at the expense of the distinct risk-moderating benefits typically associated with robust ESG performance.

5.6. Heterogeneity Analysis

We conduct grouped regressions by dividing the sample into state-owned enterprises (SOEs) and non-state-owned enterprises (non-SOEs). The results of the grouped regressions for the listed company sample in columns (1) and (2) of

Table 8. Heterogeneous regression.

	(1)	(2)	(3)	(4)	(5)	(6)
VARIABLES	inv_fluc	inv_fluc	inv_fluc	inv_fluc	inv_fluc	inv_fluc
esg	−0.0536 ***	−0.0789 ***	−0.0915 ***	−0.0362 **	−0.0675 ***	−0.0575 ***
	(0.0154)	(0.0185)	(0.0159)	(0.0175)	(0.0188)	(0.0144)
TOBIN Q	0.0290	−0.00667	−0.0178	0.0736 ***	−0.0505	0.000678
	(0.0193)	(0.0217)	(0.0165)	(0.0267)	(0.0341)	(0.0148)
CR	0.0716 ***	0.0603 ***	0.0974 ***	0.0213	0.130 ***	0.0433 ***
	(0.0267)	(0.0182)	(0.0223)	(0.0184)	(0.0483)	(0.0146)
ALR	0.559 ***	0.671 ***	0.501 ***	0.805 ***	0.174	0.423 **
	(0.185)	(0.228)	(0.182)	(0.239)	(0.280)	(0.196)
ROA	−1.239 ***	−1.431 ***	−1.367 ***	−1.312 ***	−0.973 **	−1.488 ***
	(0.448)	(0.431)	(0.402)	(0.470)	(0.419)	(0.427)
size	−0.130 ***	−0.195 ***	−0.128 ***	−0.165 ***	−0.164 ***	−0.103 ***
	(0.0145)	(0.0264)	(0.0201)	(0.0194)	(0.0191)	(0.0158)
cashflow	−0.850 **	0.0528	−0.397	−0.853 *	0.467	−0.115
	(0.427)	(0.477)	(0.397)	(0.514)	(0.585)	(0.414)
age	−0.00201	0.0103 ***	−0.00459	0.00861 **	0.00133	−0.00118
	(0.00313)	(0.00393)	(0.00324)	(0.00422)	(0.00335)	(0.00266)
gdp	−0.0583 **	−0.0358	−0.0248	−0.0754 **	−0.0600 **	−0.0319
	(0.0246)	(0.0302)	(0.0240)	(0.0298)	(0.0288)	(0.0238)
gdpper	−0.00377	−0.0271	−0.0438	0.0320	0.0566	−0.0692 *
	(0.0348)	(0.0535)	(0.0405)	(0.0441)	(0.0438)	(0.0401)
EA	0.0521	−0.511 ***	−0.271 **	0.0230	−0.116	−0.175
	(0.136)	(0.152)	(0.129)	(0.148)	(0.159)	(0.124)
ND	−0.0147 **	−0.0158	−0.0143 *	−0.0117	−0.0172 *	−0.0135 *
	(0.00668)	(0.0109)	(0.00858)	(0.00811)	(0.00929)	(0.00696)
TOP1	0.00325 ***	−0.000837	0.000715	0.00196	0.00226 *	0.000271
	(0.000966)	(0.00131)	(0.000986)	(0.00133)	(0.00128)	(0.000938)
KZ	−0.0357 **	−0.0337	−0.0232	−0.0721 ***	0.0948 **	−0.0651 ***
	(0.0177)	(0.0216)	(0.0179)	(0.0238)	(0.0412)	(0.0167)
Constant	4.010 ***	5.300 ***	3.606 ***	5.304 ***	4.508 ***	3.373 ***
	(0.436)	(0.691)	(0.462)	(0.763)	(0.552)	(0.485)
YEAR FE	YES	YES	YES	YES	YES	YES
INDUSTRY FE	YES	YES	YES	YES	YES	YES
Observations	6821	5150	7123	4848	5633	6338
R-squared	0.119	0.116	0.123	0.104	0.122	0.114

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

show that the coefficients for both SOEs and non-SOEs are significantly negative, but the absolute value of the coefficient for SOEs is smaller than that for non-SOEs (0.0536 < 0.0789). This may be because, compared to non-SOEs, SOEs hold a special strategic position in economic development. On one hand, the government provides implicit subsidies to SOEs; on the other hand, SOEs have easier access to stable financing from the banking system. As a result, SOEs exhibit lower degrees of information asymmetry and weaker principal-agent problems. From the perspective of mechanism testing, corporate ESG strengths can mitigate information asymmetry, reduce principal-agent costs, and alleviate financing constraints. Therefore, the smoothing effect of corporate ESG investment on investment volatility is more pronounced for non-SOEs.

This study conducts subsample regressions by dividing the sample into large firms and small firms based on asset size. The subsample regression results for listed firms in columns (3) and (4) of Table 8 show that the coefficients for both large and small firms are significantly negative, but the absolute value of the coefficient for small firms is smaller than that for large firms (0.0362 < 0.0915). This may be because, compared to small firms, large firms exhibit more complex management structures and suffer from more severe principal-agent problems. ESG development can help mitigate principal-agent issues in large firms, thereby reducing agency costs and ultimately exerting a stronger smoothing effect on corporate investment.

To further explore the heterogeneity of this effect, we partition the sample into firms facing strong and weak financing constraints. The results indicate that the stabilizing effect of ESG performance on investment volatility is more pronounced in firms with strong financing constraints compared to those with weak constraints. We attribute this differential impact to two primary factors. First, firms with heightened financing constraints typically face greater information asymmetry and higher perceived risk, which amplify investment instability. Strong ESG performance helps alleviate these issues by signaling superior risk management and long-term operational sustainability, thereby enhancing external stakeholder confidence and easing access to capital. Second, high-ESG firms are often subject to heightened monitoring by investors and other stakeholders, which encourages more disciplined investment behavior and reduces over-investment or under-investment tendencies. This monitoring effect is particularly valuable for financially constrained firms that are more vulnerable to volatile investment cycles. Thus, ESG practices serve as a critical governance mechanism that stabilizes investment decisions precisely where financial constraints are most binding.

6. Discussion

This study provides robust empirical evidence that corporate ESG performance significantly mitigates investment volatility among Chinese listed firms, based on comprehensive analyses of panel data from 2012 to 2022. The findings advance our understanding of how non-financial performance metrics influence corporate investment decisions in emerging markets, while also delineating important boundary conditions through moderation and heterogeneity analyses.

The results demonstrate that the negative association between ESG performance and investment volatility operates through two theoretically grounded mechanisms. First, the information channel functions through ESG’s role in reducing information asymmetry. High-quality ESG performance serves as a credible signaling mechanism to external capital providers, particularly in contexts where traditional financial disclosures may be insufficient. This signaling effect alleviates financing constraints by mitigating adverse selection problems, thereby enabling more stable investment patterns. Second, the governance channel operates through ESG’s capacity to reduce agency costs. By aligning managerial interests with those of a broader stakeholder base and strengthening internal monitoring mechanisms, ESG practices discourage myopic investment behaviors and promote longer-term strategic planning. These parallel mechanisms substantiate the theoretical proposition that ESG practices address the core market imperfections—information asymmetry and agency conflicts—that fundamentally drive investment volatility in emerging capital markets.

A pivotal contribution of this study lies in identifying patient capital as a significant moderator that attenuates ESG’s stabilizing effect. The empirical evidence reveals a “functional substitution” pattern: both relational debt (measured by long-term liabilities ratio) and stable institutional ownership diminish the marginal benefit of ESG in reducing investment volatility. This suggests that when firms already possess substantial long-term, committed capital sources, the additional stability derived from ESG practices becomes less critical. The consistently positive interaction terms between patient capital measures and ESG performance, coupled with the direct negative effects of patient capital on volatility, clearly demonstrate that these alternative stability mechanisms can partially crowd out ESG’s risk-mitigation role. This nuanced finding challenges the conventional wisdom of ESG as a universally beneficial stabilizing mechanism and highlights the importance of considering a firm’s pre-existing financial structure when evaluating the expected returns on ESG investments.

The heterogeneity analysis further enriches our understanding by revealing systematic variations in ESG effectiveness across different firm types. The stabilizing effect proves significantly stronger in non-state-owned enterprises compared to their state-owned counterparts, likely reflecting SOEs’ inherent advantages through implicit government guarantees and preferential financing access. Similarly, the effect is more pronounced in large firms rather than small firms, possibly because larger organizations face more complex agency problems that ESG mechanisms can effectively address. Most notably, ESG’s stabilizing role is particularly evident in financially constrained firms, where information asymmetries are most acute and the certification value of ESG is consequently highest. These conditional patterns align well with the Imperfect Markets Theory, demonstrating that ESG delivers the greatest marginal benefits precisely where market frictions are most severe.

In conclusion, this study establishes ESG performance as a significant determinant of corporate investment stability in emerging markets, operating through complementary information and governance channels. However, this relationship is importantly moderated by the presence of patient capital and varies systematically across firm types. These findings collectively enhance our theoretical understanding of how non-financial performance metrics interact with traditional financial structures to shape corporate investment behavior in imperfect market settings, while also providing practical insights for managers and policymakers seeking to promote sustainable and stable economic development.

7. Conclusions

Based on micro-level panel data from Chinese listed companies between 2012 and 2022, this study examines the impact of corporate ESG performance on investment volatility through the lens of imperfect market theory. The results demonstrate that superior ESG performance significantly reduces corporate investment volatility, suggesting that ESG practices contribute to real economic stability. This finding remains robust after addressing endogeneity issues and conducting multiple robustness checks. Meanwhile, moderation effect tests indicate that patient capital suppresses the smoothing effect of ESG on firms’ investment volatility. Heterogeneity analyses reveal that the negative relationship between ESG performance and investment volatility is more pronounced in non-state-owned enterprises, larger firms, and firms facing strong financing constraints.

Based on the empirical findings of this study, the following policy recommendations are proposed to enhance the role of corporate ESG practices in promoting sustainable investment and economic stability:

Firstly, Policymakers should recognize that the stabilizing function of ESG varies across firm types. For non-state-owned enterprises (non-SOEs) and small- and medium-sized enterprises (SMEs), targeted ESG incentive mechanisms—such as preferential green credit policies, streamlined ESG disclosure procedures, and tax incentives for ESG-certified firms—should be established. These measures can amplify the role of ESG in alleviating financing constraints and enhancing investment stability for structurally disadvantaged firms.

Secondly, given the significant dampening effect of ESG on corporate investment volatility, financial regulators should consider incorporating ESG performance as a soft indicator in systemic risk assessment and macroprudential frameworks. Encouraging banks and institutional investors to include ESG ratings in credit evaluation and portfolio management would not only improve capital allocation but also strengthen the resilience of the real economy to external shocks.

Thirdly, while patient capital provides stability, its moderating effect suggests it may crowd out the risk-mitigation function of ESG. Therefore, policy should guide long-term investors—such as pension funds and insurance companies—to adopt ESG-integrated stewardship principles. This would ensure that patient capital not only provides stable funding but also reinforces, rather than supplants, the corporate governance and transparency benefits associated with strong ESG performance.

Finally, to fully realize ESG’s role in reducing information asymmetry, regulators should move toward mandatory, assured, and standardized ESG reporting for listed firms. A unified corporate sustainability disclosure system, aligned with global benchmarks such as the ISSB, would enhance comparability, curb greenwashing, and enable capital markets to more accurately price ESG performance.