ESG Performance and Customer Purchase Behavior in China: The Role of Information Exposure on Market Share

Abstract

The effect of corporate ESG performance on firm competitiveness has attracted growing attention from both regulators and market participants. Most studies explore and interpret this effect from the perspective of supply-side factors such as technological innovation; however, the role of customer-side factors remains underexplored. This exploratory study aims to theoretically and empirically analyze the mediation role of the customer-side factors in the impact of corporate ESG on market share. Based on a review of the literature, we develop a theoretical model linking corporate ESG performance to customer purchase behavior. The derived hypotheses are empirically checked using panel data of Chinese listed companies from 2009 to 2023 using two-way fixed-effect regression, three-step mediation analysis, and Sobel test. The results show that the effect of ESG performance on market share is significantly positive, and this relationship is mediated by three variables: corporate reputation, firm visibility, and market coverage. Therefore, we suggest that (i) the Chinese government should strengthen mandatory ESG disclosure requirements and enhance supervision of ESG rating agencies; (ii) corporations should substantially improve their ESG performance and enhance ESG communication capabilities; (iii) customers should pay more attention to public interest, allowing individual benefits to align with social welfare, thereby achieving a win-win outcome for both customers and corporations.

1. Introduction

Corporate sustainability has become a central concern in modern business. Corporations now face pressure from multiple stakeholders to improve their environmental, social, and governance (ESG) performance [1]. Although ESG disclosure provides a way for firms to communicate their sustainability efforts, does this information actually influence customer choices? This is a very important concern for both business strategy and policy design. Many studies concentrate on the effect of corporate ESG on corporate competitiveness, and empirically find a significant positive effect of ESG on market share [2]. This significant positive effect is explained through several mediation channels: technological innovation, financing ability, supply chain [3,4,5,6,7,8,9]. Most of these interpretations focus on supply-side factors, while related demand-side studies remain scarce. Limited related studies have examined customer attitudes toward ESG and found that customers have positive intentions toward products from firms with better ESG performance [10,11]. However, intention is not the same as customer purchase [12]. Meanwhile, although customer-side factors such as corporate reputation, social trust, and market attention appear to explain the effect of ESG on market competitiveness in some studies, these factors have not been integrated into a unified framework and thus lack a coherent theoretical foundation [4,8,13,14,15,16,17,18,19].

China provides a valuable context for studying this question. It has rapidly developed its ESG disclosure framework over the past decade. The Shanghai and Shenzhen Stock Exchanges issued disclosure guidelines in 2018 and 2020, respectively [5]. China established the National Carbon Emission Trading Market in 2021. These developments make China an ideal setting for examining how ESG information reaches customers and influences corporate market share.

This exploratory study aims to investigate the effect of corporate ESG performance on market share and its underlying mechanisms through customer-side factors. Drawing on related theories, we propose that ESG information affects customers through three interrelated dimensions: relative advantage, compatibility, and observability. Empirically, these three dimensions are measured by corporate reputation, firm visibility, and market coverage, respectively. Our analysis uses annual panel data of Chinese listed companies from 2009 to 2023. We employ two-way fixed effects regression to estimate the causal effect of ESG performance on market share. We then test the three proposed mediation variables and their interactions by three-step mediation analysis and Sobel test.

The remainder of this paper is organized as follows. Section 2 provides a literature review. The theoretical framework and the derived research hypotheses are given in Section 3. Section 4 is the empirical section, which covers model specification, data description, baseline regressions, robustness checks, subgroup analysis, and mechanism analysis. Section 5 presents the discussion and policy implications. Section 6 is the Conclusion.

2. Literature Review

2.1. Corporate Sustainability and ESG Disclosure

Sustainability has emerged as a central concern for modern corporations. The concept of corporate sustainability originates from the broader sustainable development goals defined by the United Nations [20], which emphasize meeting present needs without compromising the ability of future generations to meet their own needs [21]. At the corporate level, sustainability encompasses three specific dimensions: environmental, social, and governance (ESG) [1]. This framework provides stakeholders and corporate managers with a comprehensive view of firms’ sustainability [2]. Corporate focus has shifted from corporate social responsibility (CSR) to ESG performance, reflecting a fundamental transition from voluntary and qualitative practices to mandatory and quantitative requirements.

Recently, ESG information disclosure has become increasingly important in China. The Shanghai and Shenzhen Stock Exchanges introduced ESG disclosure guidelines in 2018 and 2020, respectively, marking the beginning of regulatory attention to ESG disclosure [5]. Unlike developed markets, where ESG disclosure standards are well-established through frameworks such as GRI, SASB, and TCFD, Chinese regulatory and disclosure frameworks are still developing. A company’s ESG score can vary substantially across different providers due to differing measurement methodologies, making it harder for stakeholders to assess actual corporate ESG performance [22]. The establishment of China’s National Carbon Emission Trading Market in 2021 and subsequent mandatory environmental disclosure requirements signal a shift toward stricter ESG governance. These policy changes are gradually unifying disclosure standards, which enhance the credibility and comparability of ESG information disclosed by Chinese corporations.

Firms use ESG disclosure as a strategic signal. Well-performing firms are willing to disclose, whereas reluctant disclosers often face underlying challenges. Signaling theory, which focuses on information asymmetry [23], explains why firms voluntarily disclose costly ESG information to capital markets and customers. When information gaps exist between firms and stakeholders, high-performing firms have incentives to signal their quality through credible disclosures [24]. However, related studies also find that if firms perceive ESG disclosure as a means to gain market competitiveness, they may disclose inaccurate ESG information [25].

The credibility of ESG disclosure is still in question. This credibility is mainly challenged by greenwashing [26]. Simply put, greenwashing refers to making misleading environmental claims. Related empirical studies report that firms with high ESG scores are more likely to engage in greenwashing [22]. Public doubts about ESG reporting are mainly caused by three issues: inconsistent rating standards, unclear calculation methods, and a lack of auditing [27]. This is why corporations should be required to disclose ESG information and rating agencies should be required to make their scoring details public. Recent studies suggest that greenwashing undermines consumer trust in environmental claims, thereby weakening the positive relationship between corporate ESG performance and customer purchase behavior [28]. When consumers perceive corporate environmental claims as misleading or unsubstantiated, their green trust declines, which in turn reduces their willingness to purchase products even from genuinely sustainable firms [28,29].

2.2. Customer Purchase Behavior and Market Share

Customer purchases in this study encompass four categories: individual consumer purchases, downstream firm purchases, government procurement, and corporate self-purchases [7,30]. In practice, the first two categories account for the majority of customer purchases and serve as the primary channels through which ESG information influences market share.

In general, customer perception of corporate ESG information serves as the foundation for purchase decisions. Although corporations provide information on detailed ESG indicators, the effectiveness of this information depends on the extent to which customers trust and understand it [1]. Research shows that customers exposed to ESG disclosure exhibit higher purchase intentions, particularly when the information is credible and easy to understand [10]. According to the information processing model, ESG information is disseminated to customers through media coverage and social media discussions. In turn, customers develop more accurate mental models of corporate sustainability performance [31]. In addition, customers are more inclined to trust third-party-verified disclosures than corporate self-reporting [26].

The process from ESG perception to purchase intention is not straightforward. Customers who view a company as having higher ESG performance gradually form positive attitudes toward its products; when these attitudes accumulate to a certain level, purchase intention emerges [11]. This is consistent with the Theory of Planned Behavior, according to which attitudes mediate the effect of information on behavioral intentions. Empirical evidence demonstrates heterogeneous effects across ESG dimensions: environmental performance typically shows the strongest influence on purchase intention, followed by social factors, while governance effects are often weaker [32]. Additionally, customers are willing to pay more for products from companies with strong ESG performance. A related study divided firms into two groups and found that consumers paid approximately 10% more for products from high-ESG companies, controlling for substitutes and brand reputation [27]. Some survey studies have found that younger, higher-income, and highly educated consumers exhibit stronger purchase intentions for products from companies with higher ESG scores; however, the research also reveals a significant difference between consumers’ stated purchase intentions and their actual purchases [12]. This difference suggests that studying purchase intention alone is insufficient.

When customers’ ESG awareness translates into actual purchases, product sales increase, customer loyalty strengthens, and market share grows. Market analysis reports show that products marketed as sustainable have experienced faster sales growth than common products in recent years [12]. Brand switching occurs when customers find that the firms producing substitutes for their habitual product have achieved higher ESG ratings. In short, overall customer ESG awareness has a weak impact on market demand but a strong influence on product choice when products are substitutes [11].

Both customers and the market in China exhibit distinctive characteristics. Chinese customers demonstrate high levels of concern for ESG issues, with 78% of consumers in surveys expressing concern about corporate ESG [31]. However, the gap between Chinese consumers’ purchase awareness and actual purchase behavior is significantly larger. This characteristic can be explained by several factors. First, government policy influences customer awareness. In countries with stronger policy enforcement, residents’ awareness tends to be higher. Second, online social platforms such as Weibo and WeChat facilitate ESG information dissemination. Negative corporate news can often be widely disseminated within a couple of days [32]. Third, as China’s ESG disclosure framework is still developing, customers have relatively low trust in corporate ESG scores and performance [27]. Fourth, due to large differences in consumer incomes, ESG-driven consumption behavior also varies greatly.

2.3. Information Dissemination Mechanisms

ESG information reaches customers through various channels, each with different impacts. Corporations release ESG information through their own sources such as annual reports, ESG reports, and official websites. However, research shows that corporate sources receive little attention and are not trusted [33]. Third-party sources, such as ESG rating agency reports and analyst reports, are trusted but receive little attention. When media report ESG information, more people are influenced, and thus the information achieves higher effectiveness—regardless of whether it is good or bad news [32,33].

Customer trust in corporate ESG information is key to whether such information can influence customers’ purchases. In simple terms, when information comes from a single source, customers form trust judgments based on the source and its specific content. In particular, when customers perceive corporate greenwashing, it becomes difficult to establish trust for a considerable period [33]. When information comes from multiple sources and is inconsistent, customers cannot identify the truth. This causes customer confusion and reduces customer trust. In China, when information conflicts, customers mainly trust official government media [34,35].

Peer effects exist in ESG information dissemination. A study on Weibo shows that ESG discussions on the platform significantly influence stock performance and customer trust [36]. When discussions contain conflicting information, attention grows rapidly. Highly popular online topics attract regulatory attention. This creates an auxiliary regulatory mechanism for information conflicts. As a result, with the rapid development of digital technologies in recent years, the impact of corporate ESG on customer behavior in China has been gradually increasing [37].

ESG information dissemination in China is uniquely policy-driven. On one hand, the government uses social media to promote sustainable development. In China, media tends to closely align with the government in communication, including ESG-related information [37]. On the other hand, in addition to relying on third-party organizations, the government also pays close attention to ESG-related discussions on social media platforms. For example, the government encourages such discussions on platforms like Weibo and WeChat [36].

2.4. Research Gap and Contributions

First, existing studies have examined the positive effect of ESG on market share and identified several mediating mechanisms, such as technological innovation, financing capability, and supply chain [3,4,5,6,7,8,9]. However, these interpretations focus on supply-side factors, thus leaving demand-side mechanisms largely underexplored. To fill this gap, this study investigates how customer-side factors contribute to the ESG–market share relationship, complementing the existing supply-side literature from a demand-side perspective.

Second, most of the limited studies on customer responses to ESG focus on attitudes and purchase intentions [10,11]. This is a critical limitation, as the well-documented intention–behavior gap indicates that stated intentions do not always translate into actual purchases [12]. This study addresses this limitation by using market share as a proxy for actual customer purchase behavior, following the approach of prior studies [7].

Third, although some studies have separately explored customer-side factors such as corporate reputation [4,13,15], firm visibility [14,16,17], and market coverage [8,18,19] in the ESG context, these factors lack a unified theoretical framework. Consequently, the interactions among these factors remain unexplored. This study addresses this gap by integrating these three factors into a single theoretical framework and examining their individual and interactive mediating effects.

3. Theory and Research Hypothesis

A significant body of literature has examined the impact of corporate sustainability, particularly ESG performance, on market share. Most of these studies explain this relationship from the supply side; that is, how ESG performance changes the firm itself, drawing on theories such as resource dependence theory [38]. However, these studies largely overlook the customer side. The effects from the customer side primarily come from consumers and downstream firms. Once ESG information is received by customers, it influences their purchase behavior, which in turn affects corporate market share. Therefore, the theoretical model constructed in this study explains the impact of ESG on market share from the customer side, as shown in Figure 1.

According to stakeholder theory [39,40] and signaling theory [23], customer interests are crucial for long-term corporate development. Companies therefore use ESG performance to enhance their reputation and brand image, sending market signals that shape customers’ perceptions and evaluations of the firm and its products, ultimately influencing purchasing behavior [41]. From a microeconomic perspective, market share is essentially the aggregate result of individual purchasing decisions [7]. From the perspective of brand switching, customer purchase behavior drives the flow of market share between brands. When customers switch from other brands to a specific brand, the company’s market share increases; the opposite occurs when customers switch away [30]. Based on these theoretical arguments, we propose the first hypothesis:

Hypothesis 1 (H1).

Corporate ESG performance positively affects market share through customer purchase behavior.

Corporate ESG information is first disclosed, then disseminated, and ultimately influences customers’ purchasing decisions. However, the diffusion of ESG information differs significantly from traditional communication processes. In traditional communication frameworks such as agenda-setting theory [42], information primarily spreads through mass media reports and spontaneous public sharing and discussion. However, the diffusion of ESG information aligns more closely with Rogers’ diffusion of innovations theory [43].

Rogers’ diffusion of innovations theory was first proposed by Everett M. Rogers in his seminal book Diffusion of Innovations (1962). This theory is one of the foundational theories in communication studies. According to Rogers’ theory, the rate and extent to which an innovation is adopted depend on five key attributes: relative advantage, compatibility, complexity, trialability, and observability. Relative advantage refers to the degree to which an innovation is perceived as better than the idea it supersedes. Compatibility is the degree to which an innovation is perceived as consistent with the existing values, past experiences, and needs of potential adopters. Complexity indicates how difficult an innovation is to understand and use. Trialability captures the extent to which an innovation can be experimented with on a limited basis. Finally, observability is the degree to which the results of an innovation are visible to others. However, the diffusion of ESG information differs from traditional innovation diffusion. There is no complexity or trialability issue in ESG information transmission. For ESG information diffusion, relative advantage can be simply viewed as a firm’s ESG rating. A higher ESG rating leads to stronger corporate reputation. Compatibility refers to the alignment between disclosed ESG information and customers’ values and expectations. Information that is more compatible with customers will be more widely discussed. Observability is relatively intuitive. In the ESG context, it refers to the number of customers who receive the ESG information.

After Rogers’ seminal work, researchers have investigated information diffusion in various contexts, particularly in the domain of corporate ESG information diffusion and its subsequent effects. Corporate reputation corresponds to relative advantage in Rogers’ framework. Signaling theory suggests that ESG performance enhances corporate reputation by signaling credibility and trustworthiness to customers. A strong reputation reduces perceived risk and increases customer purchase intentions [4,13,15]. Firm visibility corresponds to compatibility in Rogers’ framework. Customer-based brand equity theory suggests that visible ESG information creates meaningful brand associations through repeated exposure, shaping customer responses [14]. Higher ESG visibility from rating agencies and mandatory disclosure makes ESG information more accessible and detectable, thereby influencing customer purchase behavior [16,17]. Market coverage corresponds to observability in Rogers’ framework. A larger firm scale reaches a broader customer base, increasing the number of potential observers of ESG information and thereby enhancing economic returns [8]. However, firm size alone is an imperfect proxy; internationalization depth and width more precisely capture the geographical scope of customer coverage, determining how widely ESG information can be observed and acted upon [18,19]. Based on the above reasoning, we propose the following hypotheses:

Hypothesis 2 (H2).

Corporate ESG performance positively affects market share through customer purchase behavior, mediated by corporate reputation.

Hypothesis 3 (H3).

Corporate ESG performance positively affects market share through customer purchase behavior, mediated by firm visibility.

Hypothesis 4 (H4).

Corporate ESG performance positively affects market share through customer purchase behavior, mediated by market coverage.

Rogers’ theory not only identifies five key attributes but also suggests that these attributes may reinforce each other. Subsequent empirical research has further validated the interdependencies among these five attributes; such positive reinforcement across different channels is a common phenomenon in information dissemination [44]. These interdependencies, or multiplicative effects, can be captured by examining the mediating role of interaction terms [45,46]. Accordingly, we propose the fifth hypothesis:

Hypothesis 5 (H5).

Corporate ESG performance positively affects market share through customer purchase behavior, mediated by the interactions among corporate reputation, firm visibility, and market coverage.

4. Empirical Analysis

4.1. Empirical Model

Following related practice in panel data studies of Chinese listed companies, this study employs two-way fixed effects regression with year and industry fixed effects [5]. Firm-level fixed effects are not included because they would largely reduce the degrees of freedom. Given the research interest in how corporate sustainability affects customer purchase behavior, the empirical equation is given by Equation (1). The dependent variable $share$ measures customer purchase behavior; the key explanatory variable $esg$ is the ESG performance index; the matrix $\mathit{X}$ includes all the other explanatory variables and $\mathit{\theta }$ is the corresponding coefficients of the explanatory variables; c is the industry fixed effect and $\delta$ is the time fixed effect; $\epsilon$ is the residual term.

$$shar{e}_{it}={\beta }_0+{\beta }_{1}es{g}_{it}+\mathit{X}\mathit{\theta }+c_j+{\delta }_t+{\epsilon }_{it}$$

(1)

This paper employs a combined approach of the three-step mediation method and the Sobel test to identify the mediation mechanism. The three-step mediation method consists of three regressions, all of which are two-way fixed-effects regressions in this paper. We choose to use the Sobel test for supplementary analysis for two reasons. First, while the three-step mediation method is not entirely objective in determining the mechanism, the Sobel test tests the significance of the joint distribution of the relevant estimators. Second, the three-step mediation method does not identify economic significance, whereas the Sobel test provides the proportion of the mediation effect to the total effect.

4.2. Data Sources and Sample Selection

The study sample consists of annual panel data of Chinese listed companies from 2009 to 2023. Companies with Special Treatment, Delisting Risk, and Suspension of Trading (ST, ST*, and PT) were excluded, resulting in a final sample of 5405 listed companies. Variable information for the empirical analysis is presented in

Table 1. Definitions and Sources for Variables in the Empirical Analysis.

Variable	Definition	Source
$share$	Market share ratio (%), =Principal revenue/Total industry principal revenue × 100	CSMAR
$esg$	Annual mean of Huazheng ESG Index	CSMAR
$lev$	Total debt ratio = Total debt/Total asset	CSMAR
$roa$	Return on total asset	CSMAR
$cflow$	Cash flow ratio = Net cash flow/Total asset	CSMAR
$tobin$	Tobin’s Q index	CSMAR
$goodwill$	Goodwill index (Guan and Zhang, 2019) [47]	CSMAR
$trustp$	Social trust (%), =Proportion of trusting population, (Zhang and Ke, 2002) [9]	CGSS
$trusts$	Social trust (%), =Proportion of trust score, (Zhang and Ke, 2002) [9]	CGSS
$attention$	Analyst coverage, =ln(Number of analyst reports + 1) (Guan et al., 2020) [6]	Eastmoney
$mediai$	Online media coverage, =ln(Number of reports + 1) (Liu et al., 2022) [48]	CNRDS
$mediap$	Print media coverage, =ln(Number of reports + 1) (Xiao and Zhou, 2021) [49]	CNRDS
$depth$	Internationalization depth, =Number of overseas subsidiaries (Liu et al., 2024) [3]	CNDD
$width$	Internationalization width, =Number of countries with overseas subsidiaries (Liu et al., 2024) [3]	CNDD

(1) CSMAR = China Stock Market & Accounting Research Database; CGSS = China General Social Survey; CNRDS = Chinese Research Data Services Platform; CNDD = China Deep Data. (2) Variables with citations follow the construction methods in the cited literature.

. Corporate financial data are obtained from the China Stock Market & Accounting Research Database (CSMAR). To further investigate the mechanisms underlying customer purchase behavior, we identified mediation variables related to corporate reputation, firm visibility, and market coverage from the existing literature. These variables were constructed using data from other databases, following the methods in the relevant literature. In addition to the variables listed in Table 1, this paper also uses the major industry classification (primary classification, 19 categories) and the sub-industry classification (secondary classification, 84 categories) issued by the China Securities Regulatory Commission (CSRC) in 2012.

Corporate reputation reflects how customers perceive a firm’s reliability and, consequently, the quality of information about its products. Tadelis (1999) defines corporate goodwill as a tradable intangible asset [50]. Rogerson (1983) states that reputation serves as a “guarantee mechanism” for the quality of corporate products [51]. Therefore, corporate reputation can be measured through two dimensions: corporate goodwill and social trust. For Chinese listed companies, corporate goodwill ($goodwill$) is primarily calculated using principal component analysis based on 12 financial indicators, including asset value, revenue performance, profitability, and stability indicators [47]. Social trust ($trustp$ and $trusts$) is measured using data from the China General Social Survey (CGSS), capturing the trust embedded in peer and stakeholder evaluations [9].

Firm visibility reflects the extent to which information about a company reaches the market and investors. Related research [52,53] typically measures firm visibility using two main indicators: the number of financial analysts covering the firm and the frequency of media coverage. For Chinese listed companies, analyst attention ($attention$) is typically obtained through web scraping to collect the number of analyst reports from the Eastmoney website (www.eastmoney.com (accessed on 11 May 2025); East Money Information Co., Ltd.) [6]. Most related studies on Chinese listed companies use print media coverage ($mediap$), measured by the number of newspaper reports [49]. These data are available from the Chinese Research Data Services Platform (CNRDS). CNRDS also provides data on the number of online media reports, which recent studies have used as a measure of online media coverage ($mediai$) [48,49].

Market coverage captures the observability dimension of ESG information diffusion, as it determines how many customers can potentially receive and observe a firm’s ESG information. Related research [54] shows that, when studying international market size, a company’s degree of internationalization captures market coverage and competitiveness more effectively than traditional indicators such as the foreign sales-to-total-sales ratio (FSTS). Given data availability, Liu et al. (2024) measure corporate internationalization across two dimensions—the depth and width of internationalization ($depth$ and $width$)—by calculating the number of overseas subsidiaries and the number of countries where these subsidiaries are located, using data from China Deep Data (CNDD) [3]. In addition, some studies suggest that the product of the two ($depth\times width$) is a more appropriate measure of corporate internationalization.

All data analyses were performed using StataMP 16 (StataCorp LLC, College Station, TX, USA). The descriptive statistics for the variables used in the empirical analysis are presented in

Table 2. Descriptive Statistics for Variables in the Empirical Analysis.

Variable	Obs	Mean	Std. Dev.	Min	Max
$share$	48,851	2.399	7.689	−0.206	100
$esg$	44,898	4.154	0.955	1	8
$lev$	60,045	0.451	1.135	−0.195	178.345
$roa$	60,044	0.047	0.622	−51.947	108.366
$cflow$	50,021	0.046	0.116	−11.056	2.457
$tobin$	46,178	2.526	69.645	0.611	14,810.306
$goodwill$	39,290	5.570	2.866	1	10
$trustp$	46,836	86.502	64.817	2.7	218.9
$trusts$	46,836	88.052	64.865	4.5	220.9
$attention$	47,120	13.646	23.091	0	298
$mediai$	45,182	4.962	1.020	0	11.108
$mediap$	45,182	2.995	1.424	0	11.172
$depth$	45,435	1.529	4.201	0	39
$width$	45,435	0.975	2.172	0	17

. The number of observations for $lev$ and $roa$ is approximately 60,000, whereas our key variables $share$ and $esg$ each have more than 10,000 missing observations. Further investigation reveals that many firms, especially in the early years after listing, did not disclose their principal revenue or ESG performance. Therefore, although our research sample is slightly unbalanced, the missing data are unlikely to cause significant bias in the estimation results. In addition, the descriptive statistics show that the minimum value of $share$ is negative. Upon checking, we found that only four observations among more than 60,000 have negative operating revenue. These four negative values arise because the firms temporarily suspended certain operations and experienced substantial returns of previously sold goods.

4.3. Baseline Regressions

Based on the research sample, the effect of corporate ESG performance on market share is estimated using the year–industry two-way fixed effects method. The estimation results are presented in

Table 3. Estimated Results of the Baseline Regressions.

Regression:	(1)	(2)	(3)	(4)	(5)	(6)
Dependent:	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$
$esg$	0.547 ***	0.523 ***	0.742 ***	0.713 ***	0.857 ***	0.827 ***
	(0.039)	(0.039)	(0.037)	(0.037)	(0.032)	(0.032)
$lev$		0.300 ***		0.224 ***		0.160 ***
		(0.039)		(0.036)		(0.031)
$roa$		0.473 ***		0.399 ***		0.408 ***
		(0.073)		(0.068)		(0.057)
$cflow$		4.779 ***		4.002 ***		3.578 ***
		(0.418)		(0.396)		(0.338)
Constant	0.189	−0.082	11.974 ***	11.661 ***	4.996 ***	4.759 ***
	(0.165)	(0.167)	(0.385)	(0.386)	(0.447)	(0.447)
Year FE	No	No	Yes	Yes	Yes	Yes
Broad Industry FE	No	No	Yes	Yes	No	No
Industry FE	No	No	No	No	Yes	Yes
Observations	44,143	44,141	44,143	44,141	44,143	44141
Adj. R2	0.004	0.008	0.140	0.143	0.393	0.395

(1) Standard errors are in parentheses; (2) *** is significant under 0.01.

. The control variables are the debt-to-assets ratio ($lev$), return on total assets ($roa$), and cash flow ratio ($cflow$). As shown in Table 3, Regressions (1) and (2) are OLS estimations without year or industry fixed effects. Regressions (3) and (4) include industry fixed effects based on the major industry classification (19 categories), whereas Regressions (5) and (6) use the more detailed sub-industry classification (84 categories).

The results in Table 3 first show that ESG performance has a significant positive effect on market share, consistent with theoretical expectations. The inclusion of control variables does not substantially change the estimated coefficient of ESG, suggesting that there is no significant endogeneity between ESG performance and the control variables. Comparing the adjusted R2 values in Regressions (2), (4), and (6)—which are 0.008, 0.143, and 0.395, respectively—reveals that the large number of intercept terms generated by the fixed effects explain most of the variation in market share. Using a more detailed industry classification effectively improves regression fit without significantly reducing degrees of freedom, unlike firm-level fixed effects. The adjusted R2 values in Regressions (1) and (2) are 0.004 and 0.008, respectively, indicating that the explanatory power of corporate ESG is comparable to that of the three commonly used control variables combined. Thus, corporate ESG is an important determinant of market share, providing a foundation for further mechanism analysis.

4.4. Robustness Checks

Since market shares are ratios and ESG performance has upper and lower bounds, neither variable exhibits a time trend. Therefore, the estimated effect of ESG performance on market share is not driven by spurious regression due to time trends. However, spurious correlation may also arise from confounding bias. For example, if a factor C simultaneously affects ESG performance and market share while ESG performance and market share are not causally related, the coefficient on ESG could still appear significant. To test this possibility, we take the first difference of all variables and regress the one-period lagged ESG difference term. The results are shown in Regression (7) in

Table 4. Robustness Regression Results.

Regression:	(7)	Regression:	(8)	(9)	(10)
Dependent:	$\mathbf{\Delta }\mathit{share}$	Dependent:	$\mathit{tobin}$	$\mathit{share}$	$\mathit{share}$
$\Delta es{g}_{-1}$	0.047 *	$esg$	1.684 ***	0.765 ***	0.870 ***
	(0.028)		(0.301)	(0.032)	(0.045)
$\Delta lev$	0.002	$lev$	29.968 ***	0.150 ***	0.223 ***
	(0.014)		(0.264)	(0.030)	(0.054)
$\Delta roa$	0.026	$roa$	−21.225 ***	0.363 ***	1.561 ***
	(0.024)		(0.519)	(0.057)	(0.309)
$\Delta cflow$	0.383 **	$cflow$	−50.737 ***	3.170 ***	2.759 ***
	(0.178)		(3.204)	(0.334)	(0.515)
Constant	−0.032	Constant	−10.524 **	6.964 ***	4.199 ***
	(0.255)		(4.195)	(1.450)	(0.666)
Year FE	Yes	Year FE	Yes	Yes	Yes
Industry FE	Yes	Industry FE	Yes	Yes	Yes
City FE	No	City FE	No	Yes	No
Observations	34,345	Observations	44,192	44,140	22,070
Adj. R2	0.014	R2	0.360	0.427	0.394

(1) Standard errors are in parentheses; (2) *, ** and *** are significant under 0.10, 0.05 and 0.01.

. Clearly, after differencing, the effect of ESG performance on market share remains positively significant. We also tested non-lagged, second-lag, and third-lag terms for ESG, but the effects on market share were insignificant. These findings provide strong evidence against confounding bias.

In Regression (8), we substitute the dependent variable market share ($share$) with Tobin’s Q index ($tobin$). The results show that ESG performance has a significant positive effect on $tobin$. This aligns with our theoretical inference that companies with good ESG performance tend to attract customers not only as consumers but also as investors. City fixed effects are introduced in Regression (9), while Regression (10) uses a 50% random sample. The coefficients on ESG and the adjusted R2 do not change significantly in either regression, strongly supporting the robustness of the baseline results. Compared with the adjusted R2 in Regression (6), the inclusion of city fixed effects in Regression (9) does not bring much additional goodness-of-fit. There are 5394 cities in our sample. Given the trade-off between degrees of freedom and explanatory power, city fixed effects are not included in the following analysis.

4.5. Sub-Group Analysis

Many studies have revealed clear differences in trust across different types of firms. ESG performance is an important indicator of corporate sustainability, yet people tend to place greater trust in companies such as state-owned enterprises and large corporations. A moderating effect may be at play here. In simple terms, factors such as firm type, size, and age may interact with ESG performance in influencing customer purchase behavior. However, as this is not the main focus of this study, we briefly verify this through sub-group analysis.

Comparing the results of Regressions (11) and (12) in

Table 5. Estimated Results of the Sub-groups.

Regression:	(11)	(12)	(13)	(14)
Dependent:	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$
Sample:	State-Owned	Non-State-Owned	Large-Scale	Small and Medium
$esg$	1.427 ***	0.369 ***	0.935 ***	0.095 ***
	(0.064)	(0.030)	(0.049)	(0.029)
$lev$	4.813 ***	0.001	7.322 ***	−0.051 ***
	(0.275)	(0.023)	(0.282)	(0.020)
$roa$	2.562 ***	0.316 ***	9.029 ***	0.126 ***
	(0.408)	(0.044)	(0.756)	(0.038)
$cflow$	5.921 ***	3.019 ***	4.872 ***	1.222 ***
	(0.708)	(0.304)	(0.677)	(0.262)
Constant	2.733 ***	2.193 ***	5.313 ***	2.634 ***
	(0.725)	(0.512)	(0.684)	(0.400)
Year FE	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes
Observations	16,177	27,065	22,676	21,465
R2	0.443	0.517	0.497	0.575

(1) Standard errors are in parentheses; (2) *** is significant under 0.01.

shows that the effect of ESG on market share is considerably larger for state-owned firms. Similarly, comparing Regressions (13) and (14) reveals that the ESG effect on market share is substantially larger for large firms than for small- and medium-sized ones.

As shown in Figure 2, we also estimated the effect of ESG performance on market share by year. Between 2009 and 2020, the ESG estimators on $share$ show a steady decline, decreasing from 1.7 to 0.55. This decrease is highly likely caused by the lack of uniformity in ESG disclosure standards. During this period, corporate information disclosure systems were imperfect, but many ESG indicators were published by various institutions. The reported ESG performances of the same corporation varied, gradually reducing public trust in ESG indicators. On the contrary, a sharp increase in the ESG estimator appeared from 2020 to 2023. This rapid growth is likely related to a series of policy changes. In 2020, Chinese President Xi Jinping announced China’s carbon peaking and carbon neutrality goals at the 75th UN General Assembly. Subsequently, also in 2020, the Chinese government reviewed and approved the “Reform Plan for the Legal Disclosure System of Environmental Information”, which is a signal that ESG disclosure gradually shifted from voluntary to mandatory. Furthermore, the National Carbon Emission Trading Market was officially established in 2021, making corporate carbon emission costs explicit and strengthening the pricing role of environmental factors in ESG.

The effects of ESG performance on market share across different provinces are also estimated, as shown in Figure 3. The ESG coefficients for some provinces are insignificant, marked in gray; however, most are significant, with three shades of blue indicating the magnitude of the effect. The geographic distribution in Figure 3 shows no strong spatial pattern. It is important to note that China’s economic development, measured by GDP per capita, is significantly higher in the southeast and lower in the northwest. In contrast, environmental quality is significantly lower in the southeast and higher in the northwest. Many economic variables exhibit strong spatial patterns. Therefore, the absence of a clear spatial pattern in the figure at least suggests that the significance of the ESG coefficients is not driven by spatial spillover or panel correlation.

In this subsection, we perform sub-group regressions to further examine the robustness of the main effect. The results show that the positive effect of ESG on market share remains stable across periods. Moreover, we performed similar time-based sub-group regressions for the subsequent mediation analyses (not reported in the main text), which show a consistent pattern and confirm that changes in sample composition (e.g., firm delisting and new listings) do not materially affect our conclusions.

4.6. Mechanism Analysis

In this section, we explore the mechanisms through which ESG performance affects customer purchase behavior. Unlike related studies that focus on customer psychology and intentions, our theoretical framework is built on information transmission theory and innovation diffusion theory. Accordingly, we analyze ESG information across three interconnected dimensions: relative advantage, compatibility, and observability. These dimensions are expected to shape customer purchase decisions at the macro level, thereby determining market share.

4.6.1. Corporate Reputation

The Baron–Kenny three-step mediation approach is used to estimate the three mediation variables ($goodwill$, $trustp$, and $trusts$), yielding nine regression results. Since the first step is the same as Regression (6), it is not presented in this section. The remaining regression results are shown in

Table 6. Mediation Regression Analysis of Corporate Reputation.

Regression:	(15)	(16)	(17)	(18)	(19)	(20)
Dependent:	$\mathit{goodwill}$	$\mathit{share}$	$\mathit{trustp}$	$\mathit{share}$	$\mathit{trusts}$	$\mathit{share}$
$esg$	0.934 ***	0.328 ***	4.814 ***	0.798 ***	4.807 ***	0.798 ***
	(0.015)	(0.037)	(0.322)	(0.032)	(0.322)	(0.032)
$lev$	0.717 ***	0.344 ***	−0.766 ***	0.164 ***	−0.768 ***	0.164 ***
	(0.028)	(0.066)	(0.270)	(0.031)	(0.270)	(0.031)
$roa$	0.747 ***	0.021	0.115	0.407 ***	0.111	0.407 ***
	(0.026)	(0.061)	(0.534)	(0.057)	(0.534)	(0.057)
$cflow$	6.961 ***	−0.037	1.946	3.587 ***	1.899	3.588 ***
	(0.161)	(0.392)	(2.782)	(0.340)	(2.784)	(0.340)
$goodwill$		0.683 ***
		(0.012)
$trustp$				0.006 ***
				(0.001)
$trusts$						0.006 ***
						(0.001)
Constant	1.780 ***	3.096 ***	20.988 ***	4.673 ***	22.533 ***	4.664 ***
	(0.201)	(0.475)	(4.514)	(0.449)	(4.518)	(0.449)
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Observations	37,669	37,042	44,408	43,695	44,408	43,695
R2	0.196	0.433	0.099	0.398	0.099	0.398
Mediator		$goodwill$		$trustp$		$trusts$
Proportion		66.0%		3.3%		3.3%
Sobel test		0.636 ***		0.027 ***		0.027 ***

(1) Standard errors are in parentheses; (2) *** is significant under 0.01; (3) Proportion = Proportion of total effect that is mediated.

. The Sobel test results for the three mediation variables are all significantly positive. The proportion of the total effect mediated by corporate goodwill is 66%. These results indicate that corporate reputation is one of the key transmission mechanisms in the effect of ESG performance on market share. This supports the view that corporate sustainability enhances corporate reputation through ESG disclosures, thereby increasing customer purchases.

4.6.2. Firm Visibility

The three-step mediation approach is used to estimate the three mediation variables ($attention$, $mediai$, and $mediap$), with results presented in

Table 7. Mediation Regression Analysis of Firm Visibility.

Regression:	(21)	(22)	(23)	(24)	(25)	(26)
Dependent:	$\mathit{attention}$	$\mathit{share}$	$\mathit{mediai}$	$\mathit{share}$	$\mathit{mediap}$	$\mathit{share}$
$esg$	6.001 ***	0.479 ***	0.155 ***	0.606 ***	0.279 ***	0.576 ***
	(0.112)	(0.032)	(0.005)	(0.032)	(0.007)	(0.032)
$lev$	0.953 ***	0.101 ***	0.036 ***	0.096 ***	0.054 ***	0.098 ***
	(0.094)	(0.030)	(0.004)	(0.030)	(0.005)	(0.030)
$roa$	3.173 ***	0.139 **	0.060 ***	0.294 ***	0.094 ***	0.301 ***
	(0.186)	(0.056)	(0.008)	(0.056)	(0.011)	(0.057)
$cflow$	28.679 ***	1.001 ***	0.501 ***	2.546 ***	0.629 ***	2.738 ***
	(0.966)	(0.335)	(0.040)	(0.338)	(0.057)	(0.339)
$attention$		0.060 ***
		(0.001)
$mediai$				1.501 ***
				(0.033)
$mediap$						0.942 ***
						(0.023)
Constant	−9.894 ***	5.389 ***	3.580 ***	−0.663	2.596 ***	2.258 ***
	(1.569)	(0.438)	(0.064)	(0.455)	(0.092)	(0.446)
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Observations	44,824	44,111	43,504	42,841	43,504	42,841
R2	0.163	0.423	0.248	0.424	0.235	0.418
Mediator		$attention$		$mediai$		$mediap$
Proportion		42.2%		27.4%		31.1%
Sobel test		0.350 ***		0.229 ***		0.260 ***

(1) Standard errors are in parentheses; (2) ** and *** are significant under 0.05 and 0.01; (3) Proportion = Proportion of total effect that is mediated.

. The Sobel test results are significantly positive for all three variables, indicating that each mediation pathway is statistically significant. Moreover, the estimated proportions of the mediated effects are substantial, suggesting that firm visibility plays an important economic role in this pathway. Overall, the evidence indicates that both analysts and the media are attracted to firms with better ESG performance, and that strong firm visibility increases customer purchases and thus market share.

4.6.3. Market Coverage

To examine the mediation pathways of the three variables ($depth$, $width$, and $depth\times width$) in the effect of ESG performance on market share, we apply the three-step mediation approach and the Sobel test, with results shown in

Table 8. Mediation Regression Analysis of Market Coverage.

Regression:	(27)	(28)	(29)	(30)	(31)	(32)
Dependent:	$\mathit{depth}$	$\mathit{share}$	$\mathit{width}$	$\mathit{share}$	$\mathit{depth}\mathbf{\_}\mathit{width}$	$\mathit{share}$
$esg$	0.470 ***	0.770 ***	0.289 ***	0.753 ***	−0.727 ***	0.743 ***
	(0.023)	(0.033)	(0.012)	(0.033)	(0.123)	(0.033)
$lev$	0.243 ***	0.923 ***	0.138 ***	0.907 ***	−0.495 ***	0.886 ***
	(0.027)	(0.072)	(0.014)	(0.072)	(0.143)	(0.072)
$roa$	0.216 ***	0.351 ***	0.126 ***	0.346 ***	−0.702 ***	0.330 ***
	(0.038)	(0.060)	(0.019)	(0.060)	(0.205)	(0.060)
$cflow$	1.415 ***	3.406 ***	0.829 ***	3.352 ***	−6.879 ***	3.177 ***
	(0.197)	(0.357)	(0.100)	(0.357)	(1.055)	(0.357)
$depth$		0.190 ***			7.678 ***	0.244 ***
		(0.007)			(0.054)	(0.018)
$width$				0.370 ***	5.584 ***	0.274 ***
				(0.014)	(0.106)	(0.030)
$depth\_width$						−0.017 ***
						(0.001)
Constant	−2.002 ***	4.566 ***	−1.207 ***	4.638 ***	3.07 *	4.696 ***
	(0.308)	(0.452)	(0.157)	(0.452)	(1.656)	(0.451)
Year FE	Yes	Yes	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Observations	42,184	41,542	42,184	41,542	42,184	41,542
R2	0.067	0.414	0.090	0.414	0.793	0.417
Mediator		$depth$		$width$		$depth\_width$
Proportion		10.7%		12.6%		1.7%
Sobel test		0.092 ***		0.109 ***		0.013 ***

(1) Standard errors are in parentheses; (2) * and *** are significant under 0.10 and 0.01; (3) Proportion = Proportion of total effect that is mediated; (4) $depth\_width=depth \mathit{\times } width$.

. One point should be mentioned first. Since the interaction term ($depth\times width$) incorporates the partial effects of $depth$ and $width$, these two terms should also be included as controls when the interaction term is regressed. The Sobel test results indicate that the mediation effects of all three variables are significant. The corresponding proportions for $depth$ and $width$ are approximately 10%, which is not negligible. One possible reason is that internationalization may not be a perfect proxy for market coverage, as a large number of Chinese listed companies have no overseas operations. Although the estimated results do not fully meet our expectations, we believe they support the view that market coverage is an important transmission mechanism in the effect of ESG performance on customer purchase behavior, as reflected by market share.

4.6.4. Interaction Effects Among the Three Pathways

As discussed in the theoretical section, the three pathways are not independent. Therefore, in this subsection, we examine the mediation effects of the interactions among corporate reputation, firm visibility, and market coverage. Because there are multiple mediation variables within each of the three mechanisms, for simplicity we select one from each category. Specifically, we select the variables with larger proportions of the total effect: corporate goodwill ($A=goodwill$), analyst attention ($B=attention$), and the width of internationalization ($C=width$). These three variables generate four interaction terms: $A\times B$, $A\times C$, $B\times C$, and $A\times B\times C$. Note that the component terms of each interaction must be included in the regression to avoid estimation bias caused by their partial effects.

In

Table 9. Mediation Regression Analysis of The Interaction Effects among three Pathways.

Regression:	(33)	(34)	(35)	(36)
Dependent:	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$	$\mathit{share}$
$esg$	0.280 ***	0.248 ***	0.254 ***	0.275 ***
	(0.038)	(0.039)	(0.039)	(0.038)
A	0.454 ***	0.565 ***	0.574 ***	0.450 ***
	(0.014)	(0.014)	(0.014)	(0.015)
B	−0.128 ***	0.031 ***	0.030 ***	−0.124 ***
	(0.006)	(0.002)	(0.002)	(0.006)
C	0.003 ***	−0.013 **	0.002 **	0.00013
	(0.001)	(0.003)	(0.001)	(0.004)
$A\times B$	0.018 ***			0.018 ***
	(0.001)			(0.001)
$A\times C$		0.002 ***		0.00051
		(0.0004)		(0.00046)
$B\times C$			0.000047 ***	−0.00047 ***
			(0.000018)	(0.00016)
$A\times B\times C$				0.000045 ***
				(0.000017)
Constant	4.320 ***	3.558 ***	3.503 ***	4.343 ***
	(0.478)	(0.482)	(0.482)	(0.478)
Controls	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
Industry FE	Yes	Yes	Yes	Yes
Observations	35,603	35,603	35,603	35,603
R2	0.456	0.446	0.446	0.456
Mediator	$A\times B$	$A\times C$	$B\times C$	$A\times B\times C$
Proportion	−10.4%	4.6%	0.6%	2.1%
Sobel test	−0.026 ***	0.012 ***	0.002	0.006 ***

(1) Standard errors are in parentheses; (2) ** and *** are significant under 0.05 and 0.01; (3) Proportion = Proportion of total effect that is mediated; (4) In this table, $A=\mathit{goodwill}$, $B=\mathit{attention}$, and $C=\mathit{width}$.

, only the results of the third-step regressions for the four interaction terms are presented to save space. Consistent with expectations, three of the four interaction terms are significantly positive. Surprisingly, the mediation effect of the interaction between corporate reputation and firm visibility ($goodwill\times attention$) is significantly negative, with a relatively large proportion. This implies that ESG performance reduces this interaction term ($goodwill\times attention$), which in turn reduces market share; for instance, if corporate goodwill is linearly related to ESG performance, analyst attention may have a diminishing relationship with ESG. One possible interpretation is exposure fatigue: firms with high ESG performance already receive extensive analyst coverage, so additional reports attract less attention.

5. Discussion and Policy Implication

5.1. Discussion

Many studies have examined the impact of corporate ESG on market share and consistently identified a significant positive effect. Prior research has largely explained this effect from supply-side perspectives, including technology, financing, and upstream industries. To address this research gap, the present study investigates how customer-side factors contribute to this effect, with our findings confirming this positive effect. We then examine three customer-side mediation variables: corporate reputation, firm visibility, and market coverage. All three play significant mediation roles in the effect of corporate ESG on market share. These three factors correspond to three dimensions in information dissemination theory and Rogers’ diffusion of innovations theory: relative advantage, compatibility, and observability [42,43]. According to the theory, these dimensions function in a multiplicative rather than additive manner; thus, we examined the mediation effect of the interaction term among the three mediation variables. The results show that the product of the three mediation variables has a significant positive mediation effect. This finding strongly supports our theoretical inference that customer purchase behavior is affected by ESG, which in turn leads to changes in market share.

Many studies have explained the positive effect of ESG on market share from the supply side, focusing on mechanisms such as technological progress. Some of these studies further found that the increased competitiveness gained through technological advancement can, in turn, promote greater corporate ESG investment. We acknowledge the existence of such mechanisms. However, the frequent exposure of greenwashing behaviors among listed firms suggests that positive mechanisms such as technological progress rely on strict information disclosure, which is undoubtedly costly. The existence of greenwashing behaviors implies that corporations are quite rational. If the cost of information disclosure decreases, or the quality of disclosed information declines, greenwashing behaviors are likely to increase. Consequently, corporate ESG performance would decline; while general technological capability might still improve, green technology development would stagnate [25,26,27].

This study identified and clarified several key processes underlying the positive mechanism of ESG marketing. In principle, promoting ESG marketing represents a promising yet underexplored alternative compared with traditional disclosure frameworks. In the current era of rapid information technology development, the returns from traditional marketing methods such as advertising are gradually diminishing. In contrast, many customers choose to purchase products from companies with higher reputations, even if the products are not the best. This suggests that the returns from ESG marketing deserve greater attention. Promoting ESG-related competition among firms through customer-side factors offers two clear advantages. First, firms invest their own resources in information disclosure, thereby reducing the disclosure costs borne by the government and regulatory authorities. Second, competition in ESG marketing helps improve the quality of disclosed information. The issue of information disclosure thus shifts from a traditional principal–agent framework to a market-based competitive framework [36,37]. Such a transition could significantly enhance both the quality and quantity of corporate information, while reducing greenwashing behaviors.

This study takes an exploratory approach to investigate how ESG performance influences market share from the customer-side perspective. Through extensive analysis, particularly the mediation effects of the interactions, we provide empirical evidence supporting the role of customer purchase behavior in the effect of ESG on market share. Although our research objectives have been achieved, several limitations remain. First, corporate customers consist of two main groups—downstream firms and consumers—which may differ in the mechanisms through which ESG affects market share. Moreover, customer compositions vary significantly across industries, and the mechanisms may also vary accordingly. In future research, we plan to explore the causes of this variation. Second, market share is an equilibrium determined by both demand and supply. Although the prior literature supports using market share as a proxy for customer purchase behavior [7,30], it remains possible that customer purchases are rational due to genuine ESG-driven demands, or irrational due to overestimated product quality driven by ESG perceptions. To address this, future research should integrate micro-level customer data to better understand both rational and non-rational purchase behaviors, which would help clarify the underlying mechanisms. Third, while we employed the depth and width of internationalization as proxies for observability, we acknowledge that these measures may not fully capture this concept [3]. In future research, we plan to use text mining techniques on social media data to more directly measure observability and further validate its role.

5.2. Policy Implication

For regulators, it is necessary to further improve the mandatory ESG disclosure system. The China Securities Regulatory Commission (CSRC) and the three major stock exchanges in Shanghai, Shenzhen, and Beijing need to clarify the scope of firms required to disclose information, thereby improving the comparability and credibility of the information.

For corporations, efforts are needed in two aspects. First, corporations should shift their development strategy from approaches such as traditional advertising toward improving their ESG performance. ESG scores will improve when firms make genuine efforts in environmental protection, social responsibility, and corporate governance. Second, companies should enhance their ESG marketing. Many companies lack a clear understanding of how to conduct ESG marketing, often using it merely as a product advantage for advertising and promotion. Instead, corporations should position ESG as a strategic development direction and a core part of their corporate culture, so as to gain greater recognition and support from customers and investors.

For customers, it is important to foster a sustainable consumption attitude from the perspective of national green development. Customers should not only pursue lower prices and short-term gains but also pay greater attention to corporate ESG performance. When customers’ and corporations’ sense of social responsibility interact positively, both sides benefit, ultimately achieving a balance between economic and social benefits.

6. Conclusions

In this study, the effect of corporate sustainability on customer purchase behavior is analyzed both theoretically and empirically. Drawing on the related literature, we develop a theoretical framework that illustrates the transmission mechanisms through which ESG affects market share via corporate reputation, firm visibility, market coverage, and their interactions. Empirical analysis is conducted using panel data of Chinese listed companies from 2009 to 2023, employing two-way fixed effects regression, the three-step mediation approach, and the Sobel test. The results show that (i) corporate ESG has a significant positive effect on market share; (ii) this effect is correlated with policy intensity, with an upward trend in ESG coefficients after 2020; (iii) the ESG coefficients are relatively larger for state-owned and large-scale firms; and (iv) corporate reputation, firm visibility, market coverage, and their interactions all have significant mediation effects, although the proportions of the interaction effects are quite small.

Accordingly, several policy implications are offered. First, the Chinese government should, on the one hand, continue to strengthen mandatory ESG disclosure requirements for firms while, on the other hand, enhancing supervision over ESG rating agencies. Second, corporations should, on the one hand, recognize future development trends and genuinely improve their ESG performance; on the other hand, they should gradually develop ESG communication and marketing skills. Inaccurate ESG disclosure may bring short-term benefits but is harmful to long-term development. Only by proactively integrating ESG into corporate strategy can firms achieve sustainable development. Third, customers should be encouraged to shift their focus from personal surplus to social surplus. In this way, customers and firms with strong ESG performance can form positive interactions. Within such a more sustainable system, both parties can achieve win-win outcomes.