CiteSpace visual analysis presents research authors, institutions, and keywords in a discipline through knowledge maps ( Chen et al., 2015 ). This study uses CNKI as the data source, with keywords such as “ESG information disclosure” and “social responsibility disclosure”. Data retrieval spans from January 1, 2006, to March 10, 2025, with source categories including PKU Core Journals, CSSCI (including expanded versions), and SCI journals. A total of 361 valid documents are retained after screening.

This study follows standardized bibliometric procedures. Raw data is exported in RefWorks format during data collection and standardized using CiteSpace 6.2.6 software. Non-research literature such as book reviews and conference notices are manually excluded, and titles, abstracts, and keywords are screened for thematic relevance. Synonymous keywords are merged, and non-substantive keywords are removed. The time span is set from 2006 to 2025, with a time slice of 1. Network nodes are set as authors, institutions, and keywords for visual and collaboration network analysis, generating visual maps to intuitively display productive authors, institutions, research hotspots, and frontier trends.

The number of publications based on the time distribution can reflect the development of the research field from a macro perspective. From 2006 to 2025, the research on ESG information disclosure in China has gone through a cycle from germination and accumulation to rapid expansion, and the publication situation is shown in Figure 1 . From 2006 to 2011, the research in this field was in the exploratory period, with a low annual average number of publications, and was still in the stage of conceptual introduction and theoretical foundation. From 2012 to 2021, with the deepening of green finance policy and capital market reform, the annual publication volume climbed to 10 - 20 papers, with a total of 177 papers published, with obvious fluctuations, but the overall growth is not large, and is in a steady development stage. After 2022, under the dual driving force of the “dual carbon” target and the new mandatory disclosure rules of exchanges, the research will enter an explosive growth stage. It is worth noting that the data drop in 2025 is mainly due to the publication cycle of academic journals, delays in the inclusion of databases and time constraints in downloading data, but the actual research fever has not decreased.

Due to CiteSpace’s limitation in analyzing CNKI co-citations, CNKI’s bibliometric tool is used to count co-citation frequencies of the 361 documents. The top 5 co-cited papers are listed in Table 1 . The study published by Tian Lihui’s team in 2017 in Management World tops the list with 679 citations, which creatively puts forward the paper “mask effect hypothesis”, which confirms that the mandatory social responsibility disclosure policy may exacerbate the risk of stock price crash triggered by intensified information asymmetry through double difference model, providing a new perspective for the assessment of policy effect. In the same period, Tan Xue in Industrial Economics Research reveals the positive correlation between the intensity of industry competition and the quality of corporate social responsibility disclosure through signaling theory, which provides micro evidence for the motivational differences in non-financial information disclosure in the process of marketization. In 2017, Li Zhibin et al. published in Accounting Research based on the institutional context in China, revealing the quality of social responsibility disclosure of internal control as an institutional governance tool to enhance path. Compared with the earlier studies focusing on institutional pressure, the ESG value creation study by Xu Guanghua’s team and Li Zhibin’s exploration of media monitoring published in Accounting Letters in 2022 marks the transition of the research paradigm to “economic consequences-governance synergy”, echoing the interactive influence of multiple governance subjects in the digital era. The Interactive Impact of Multiple Governance Subjects in the Digital Era.

The research scholar collaboration network map can visually present the cooperative relationships among scholars. This study reveals the research scholars in the field of ESG information disclosure through the scholar collaboration network map, as shown in Figure 2 . Table 2 lists the top 10 productive authors. Based on the visual analysis of the 284 scholars’ cooperative relationship, the study found that Yin Hong, Wu Danhong and Lai Yan constitute a small research core, whose node size and connectivity density are higher than that of other scholars, and although they have a certain degree of academic influence, they have not yet formed a large-scale research, and they mainly rely on the individual strength of the core authors. The research scholars present the characteristic of “small core-large periphery”, compared with the gathering of research scholars in the core circle, the peripheral small-scale research scholars are more dispersed, which reflects that the research in the field of ESG information disclosure is still more dispersed and there is the problem of collaboration barriers.

The institutional research collaboration map can visually present the cooperative relationships among research institutions. The collaboration status of institutions publishing in the field of ESG information disclosure in this study is shown in Figure 3 . Universities dominate the research in this field, with the School of Accounting at Zhongnan University of Economics and Law, the School of Economics and Management at Beijing Jiaotong University, the College of Engineering and Technology at Chengdu University of Technology, the School of Accounting at Lanzhou University of Finance and Economics, and the School of Accounting at Nanjing University of Finance and Economics being the top five institutions with high publication volumes. The institutional collaboration network exhibits a “high-yield island” characteristic. Apart from universities, only a few enterprises such as Shanghai Lixin Ruisi and China Everbright Bank have engaged in joint research with academic institutions. Industry-academia collaboration remains particularly weak, and the transformation of theoretical achievements into industrial practice has yet to be fully realized. The top ten institutions in terms of the number of publications in the field of ESG information disclosure are shown in Table 3 .

The frequency of keywords and the density of links between nodes show the research heat and relevance of the field. This study uses the keywords as the knowledge unit to construct the co-occurrence relationship network, as shown in Figure 4 . A total of 283 effective keywords are included in the study, forming 384 co-occurrence lines, and some high-frequency keywords are shown in Table 4 . “Information disclosure” as a key node has a frequency of up to 127 times, with a mediational centrality of 0.69, which shows a large node size and high mediational centrality in the co-occurrence network, i.e., corporate information disclosure behavior is the core research topic in this field. Disclosure behavior is the core research topic in this field, indicating that it is the pivotal position in the whole knowledge network architecture and the key to connecting other research points. The frequency of “social responsibility” is 43 times, and the mediational centrality is 0.43, which reflects that academics attach great importance to the topic of corporate social responsibility disclosure.

Analyzing from the perspective of co-occurrence of keywords, there is an association between “financing constraints”, “corporate governance”, “nature of property rights” and the core theme “information disclosure”. “Existing research focuses on the interactions between information disclosure and corporate financing mechanisms, governance structure models, property rights attributes, and other aspects. For example, how the ESG information disclosure behavior of corporations reacts to their financing constraints and corporate governance effectiveness. Keywords such as “media attention” and “influencing factors” further expand the scope of the study to include the power of external monitoring and the analysis of multiple factors affecting disclosure. Keywords such as “cohort effect,” “private enterprises,” and “state-owned enterprises” appear on the periphery of the core keywords but are connected to other nodes, and are emerging or relatively minor research directions at present. Directions, such as whether there is heterogeneity in the way different ownership firms approach ESG information disclosure. The low frequency and weak centrality of keywords such as “green innovation” and “cost of capital” indicate that they have received limited attention and relevance in the knowledge network. As the concept of sustainable development continues to gain popularity around the world, it is expected that more studies will explore the potential link between “green innovation” and disclosure in the future.

Keyword clustering visually represents the structural characteristics of key nodes and their significant connections, summarizing the core themes of the research ( Huang et al., 2013 ). The study obtained a total of 9 clusters by Log-Likelihood Ratio (LLR), as shown in Figure 5 . The clustering modularity value Q = 0.8402 > 0.3 indicates that the differentiation and independence between clusters are significant; the average profile value S = 0.9762 > 0.7 indicates that the keywords match the clusters where they are located extremely well, and the clustering results are significant and credible ( Chen et al., 2015 ).

In Figure 5 , the larger the network nodes are, the higher the frequency of keywords, and the thicker the lines between nodes are, the stronger the degree of association between different keywords. Based on the co-occurrence and frequency of keywords, nine cluster labels are obtained, and the top hotspots are “social responsibility”, “corporate value”, “green innovation”, “mandatory disclosure”, “listed company”, and so on. The top hotspots are “social responsibility”, “corporate value”, “green innovation”, “mandatory disclosure” and “listed company”. According to the keyword clustering map, the research hotspots in this field are mainly reflected in the following aspects:

Firstly, #1 social responsibility, the topic contains a total of 32 keywords such as sustainable development, responsible investment and collaborative governance. The study found that CSR disclosure needs to be reconstructed in terms of mandatory norms, disclosure indicators and procedural rules ( Zheng, 2021 ). The impact of digital intelligence technology on the quality of ESG information disclosure presents a double-edged sword effect that may be used by companies to implement ESG rinsing behaviors as well as empowering stakeholder monitoring ( Yin et al., 2025 ). The study emphasizes that ESG system construction needs to be integrated with the Chinese context and localized through the integration of party leadership and stakeholder governance mechanisms ( Zhang et al., 2023 ). In addition, external factors such as digital transformation and co-institutionalization have a significant impact on CSR performance, with moderate digital transformation enhancing ESG performance through green technology innovation and information disclosure quality ( Kuang et al., 2025 ), while co-institutionalization may weaken the incentive for ESG investment due to market monopoly ( Lei et al., 2023 ).

Secondly, #2 corporate value, the topic contains a total of 25 keywords such as financing constraints, information transparency and market attention. It is found that ESG performance and information disclosure have a positive promotion effect on corporate value ( Xu et al., 2022 ) and are more significant in non-state-owned enterprises, small-scale enterprises and non-polluting industries ( Zhang et al., 2019; Zhang & Bao, 2023 ), and the specific mechanism of action includes paths such as alleviating financing constraints, enhancing market attention, and enhancing information transparency ( Wang et al., 2022 ). The relationship between ESG and corporate value presents dynamic changes, and there may be a cost effect in the short term and value appreciation in the long term ( Wang, 2022 ). External factors such as institutional investor preferences and economic policy uncertainty can moderate the value effect of ESG ( Bai et al., 2022 ).

Thirdly, #3 green innovation, the topic contains a total of 22 keywords such as ESG ratings, information transparency and green finance. It is found that ESG ratings and information disclosure can promote corporate green innovation ( Ba et al., 2025 ; Shi et al., 2024 ; Li et al., 2022 ), and its mechanism of action is mainly reflected in alleviating financing constraints, reinforcing external supervision, and enhancing information transparency ( Huang et al., 2024 ). In particular, the innovation-driven role of ESG is more prominent when the industry is highly competitive, the firm is in the growth stage, or located in a low-carbon pilot city ( Xiang et al., 2022 ; Da & He, 2024 ). Meanwhile, ESG rating divergence pushes firms to improve green innovation by influencing access to financing resources, market competition pressure and management incentives ( Dong et al., 2024 ; Yao et al., 2024 ).

Fourthly, #4 mandatory disclosure, the topic contains a total of 20 keywords such as evaluation criteria, audit mechanism and voluntary disclosure. Scholars propose to construct a localized ESG information disclosure framework in China ( Hou & Wei, 2023 ) and design three-level disclosure rules for financial institutions ( Lu et al., 2023 ). Mandatory disclosure of climate information has become the focus, and research suggests establishing disclosure standards centered on materiality ( Zhu, 2023 ), and safeguarding disclosure quality through legal regulation and forensic mechanisms ( Kang et al., 2025 ). From the perspective of legal economics, mandatory disclosure needs to balance costs and benefits, limit the breadth of disclosure, and innovate implementation methods ( Yuan, 2024 ). In addition, the field of environmental assessment emphasizes the principle of unified standards and mandatory disclosure ( Yan, 2019 ).

In order to better examine the distribution of research topics on ESG information disclosure, this study uses CiteSpace software to conduct keyword timeline mapping analysis of the literature to visualize the research development history of the relevant literature and obtain a clearer timeline mapping. The keyword timeline mapping analysis can outline the relationship between clusters and the evolution path, revealing the historical span of literature and the evolution of research hotspots in a certain cluster, as shown in Figure 6 .

Based on the time evolution mapping of keyword clustering in the field of ESG information disclosure, this study breaks through the traditional unidimensional analysis paradigm and constructs a three-dimensional framework of “theory-driven-practice-feedback-policy iteration” to deconstruct the knowledge network in the field, as shown in Table 5 , and divides the evolution into three progressive cycles following the stage characteristics of the construction of the ESG system in China. Following the stage characteristics of China’s ESG institutional construction, the evolution is divided into three progressive cycles.

1) In the embryonic exploration period from 2006 to 2015, the policy level was oriented to guide enterprises to “voluntary disclosure”, and mandatory constraints had not yet been formed, and corporate ESG practices were characterized by fragmentation, with only sporadic disclosure of social responsibility and other contents. At this time, the theoretical research focuses on the “legitimacy” of corporate disclosure behavior and “motivation” analysis, trying to answer the basic proposition of “why enterprises choose to disclose”.

2) During the period of standardization from 2016 to 2020, the focus of the policy has shifted to “mandatory pilot”, which pushes some listed companies to disclose ESG reports, and the disclosure rate of enterprises has increased significantly and the content tends to be systematized. Theoretical research has also deepened, with scholars introducing “institutional theory” to explain the impact of policy on corporate behavior, and exploring the internal path of ESG information disclosure on corporate value through the “mediation effect” model.

3) During the period of deepening and perfecting from 2021 to 2024, the policy will enter the stage of “standardization”, clarifying the content, format and forensic requirements of ESG information disclosure, and focusing on the management of the phenomenon of “greenwash” in order to guarantee the authenticity of disclosure in practice. Authenticity. The theoretical level is characterized by interdisciplinary integration. On the one hand, the theory of “signaling” is used to deconstruct the strategic logic of enterprises releasing quality signals to the capital market through ESG information disclosure. On the other hand, scholars from the perspective of “game theory” have begun to draw the behavioral equilibrium among enterprises, regulators and investors in the disclosure game, revealing how the checks and balances of multiple interests drive the dynamic evolution of the ESG ecosystem.

The three phases of China’s ESG information disclosure research show a cycle of “policy traction-practice trial and error-theory iteration” rather than the traditional linear extension. Policy tools have shifted from flexible guidance to rigid regulation, forcing corporate disclosure behavior to evolve from “passive compliance” to “active optimization”. Practical issues such as greenwashing and data distortion in corporate practices have spurred academic theoretical research, while academic achievements have in turn optimized government policies through channels like government projects, forming a dynamic closed loop of “policy-driven practice trials—practice issues driving theoretical innovations—academic consensus feeding policy iterations”. This reflects the evolutionary path of emerging systems from “framework transplantation” to “local reconstruction”.

Aiming at the limitations of CiteSpace in digging out the theoretical roots and analyzing the inner mechanisms and driving factors, this study adopts the supplementary path of qualitative research. On the one hand, we systematically sort out the main research theories of ESG disclosure, structurally present the core theories such as institutional theory and signaling theory, and clarify the explanatory logic of the theories on the practice of ESG disclosure, as shown in Table 6 ; on the other hand, we construct an analytical framework of drivers, internal mechanisms, and economic consequences to demonstrate the role of the internal and external drivers on the behavior of ESG disclosure by enterprises, and at the same time, analyze the role of mechanisms such as reducing information asymmetry to the reduction of corporate risk and the improvement of long-term competitiveness. At the same time, it analyzes the transmission path of mechanisms such as reducing information asymmetry to economic consequences such as corporate risk reduction and long-term competitiveness enhancement, so as to make up for the inadequacy of quantitative tools in the analysis of complex mechanisms, as shown in Figure 7 .

In this study, 361 representative literatures related to ESG information disclosure in China Knowledge Network from 2006 to 2025 were used as sample data, and CiteSpace software was applied to visualize and analyze the knowledge graph, mainly obtaining the following research conclusions:

First, from 2006 to 2025, the number of ESG information disclosure-related articles has gone through the stepwise growth of “exploration period-development period-outbreak period”, reflecting the urgent concern of academia for ESG practice. Secondly, the cooperation network of research scholars is characterized by a “small core-large periphery” structure, with Yin Hong, Wu Danhong, and Lai Yan as the core research scholars in this field with a certain degree of academic influence, but the field has not yet carried out large-scale research, mainly relying on the individual strength of the core author team. Third, the research institutions are dominated by universities such as the School of Accounting of Zhongnan University of Economics and Law and the School of Economics and Management of Beijing Jiaotong University, and the density of inter-institutional cooperation is relatively low, there is a potential space for the integration of academic resources, and the transformation of theoretical achievements into industrial practice has not yet been opened. Fourth, China’s ESG information disclosure research hotspot focuses on the core pivot of “disclosure”, connecting “social responsibility”, “financing constraints” and “green innovation”. “Green Innovation” and other topics, form a dynamic closed loop of “policy first pulling practice trial and error-practice problems forcing theoretical innovation-academic consensus feeding policy iteration”. Fifth, the research focus of the research frontiers in this field has evolved to “green innovation and digital governance driven by dual-carbon goals”, which is mainly embodied in the synergy between green technology R&D and disclosure quality, the reconstruction of enterprise behavior by digital monitoring mechanism, and the study of cohort effect under the perspective of industrial ecology, which not only responds to the dual needs of policy regulation and technological innovation but also highlights the research perspectives. This not only responds to the dual demands of policy regulation and technological innovation but also highlights the dynamic shift of the research perspective from individual enterprises to the institutional environment and industrial ecology.

This study only focuses on China’s ESG disclosure situation and analyzes the CNKI database, which is a single source of data. Although it fits the development characteristics of China’s policy-driven ESG disclosure, there are inevitable limitations in the data coverage, and it fails to incorporate the relevant literature on ESG disclosure in international databases such as Web of Science. In the future, we can combine multiple databases to expand the cross-country comparative perspective. This study utilizes the CiteSpace software system to reveal the evolution of knowledge in the field of ESG information disclosure. Although the theoretical foundations are systematically sorted out through qualitative analysis, and a framework for analyzing the drivers, internal mechanisms, and economic consequences is constructed, the analysis of the differences in the application of the theory and the paths of the mechanisms still needs to be further deepened.