Application of the Global Reporting Initiative Standards and Sustainability Disclosure Performance: An Analysis of the Largest Chinese Companies Listed in Hong Kong

2.1 GRI – Background and Worldwide Application

The GRI was established in 1997 in response to the rising call for advancement of corporate social responsibility subsequent to the Exxon Valdez oil spill (GRI, 2022a). Against this historical backdrop, the GRI aims to assist companies and other organizations to bear responsibility for their impacts (GRI, 2022a). As an independent international organization, the GRI was formed by two non-profit organizations, the Coalition for Environmentally Responsible Economies (‘CERES’) and the Tellus Institute, with participation of the United Nations Environment Programme (‘UNEP’) (GRI, 2022a). The CERES is a non-profit organization collaborating with capital market leaders to deal with sustainability challenges, striving for a just and sustainable future (CERES). The Tellus Institute is a leading interdisciplinary, not-for-profit research and policy organization for sustainable development (Tellus Institute, 2022).

The GRI’s original objective was to build up the first accountability system and principles confining corporate environmental behavior (GRI, 2022a). Its scope was later expanded to social, economic and governance aspects (GRI, 2022a). The GRI coordinates with a variety of stakeholders to set up the GRI Standards for sustainability disclosure (GRI, 2022b). The GRI Standards are nowadays widely perceived as one of the main global standards in the field of ESG reporting (Hahn & Kühnen, 2013). The GRI Standards, being referred to by reporting organizations in over 100 countries, are currently the most popular sustainability reporting standards in the world (GRI, 2022b).

2.2 Purpose of the GRI Standards

The GRI Standards cater to the needs of multiple stakeholders, such as investors, policymakers, and civil society (GRI, 2022c). Regarding reporting organizations, the GRI Standards serve the requirements of organizations of various sizes and types, including large companies, small companies, private companies, and public companies (GRI, 2022c). The GRI released the first GRI Guidelines (‘G1’) directing sustainability reporting based on a global framework in 2000 (Fernandez-Feijoo et al., 2014; GRI, 2022a). Given continuous enhancement, the GRI published the second (‘G2’), third (‘G3’), and fourth (‘G4’) sets of standards in 2002, 2006, and 2013 respectively (Fernandez-Feijoo et al., 2014). The GRI put forth the first GRI Standards as international standards for sustainability reporting in 2016 (GRI, 2022a).

The GRI is dedicated to promoting a two-pillar corporate reporting structure as defined below, advocating that financial reporting and sustainability reporting are equally important (GRI, 2022g).

Pillar 1 addresses financial considerations through a strengthened financial report which includes sustainability disclosures, in the context of enterprise value.

Pillar 2 concentrates on sustainability reporting with a focus on all external impacts a company has on society and the environment – and hence their contributions towards the goals of sustainable development.

To achieve this objective, the GRI collaborates with various global and inter-governmental organizations, such as the International Financial Reporting Standards (‘IFRS’) Foundation and the EFRAG, to harmonize sustainability reporting (GRI, 2022g). In 2021, the GRI, together with the EFRAG, participated in the formulation of the European Sustainability Reporting Standards (‘ESRS’) (GRI, 2022g). They assisted the European Commission in raising corporate transparency which facilitates the European Green Deal (GRI, 2022g).

2.3 Information Disclosure under the GRI Standards

According to the GRI Standards, information disclosures are classified into two categories: requirements and recommendations (GRI, 2022d). Regarding requirements, an organization has to report on the information required or comply with required instructions and report in accordance with the GRI Standards (GRI, 2022d). The GRI Standards have incorporated the ‘comply or explain’ disclosure regime for most disclosure items save for a few exceptions (GRI, 2022i). The ‘comply or explain’ disclosure regime was initiated in the 1992 UK Corporate Governance Code (Biondi, 2015; Cheffins & Reddy, 2022). Under the ‘comply or explain’ system, the reporting companies can either disclose the required information or not disclose the required information with explanations for non-disclosure (Biondi, 2015; Cheffins & Reddy, 2022). Applying the ‘comply or explain’ mechanism to the GRI framework, a company is able to report in accordance with the GRI Standards as long as it has explained for the omission of a disclosure or a requirement in a disclosure (GRI, 2022i). There are four acceptable reasons for non-disclosure: (i) not applicable; (ii) legal prohibitions; (iii) confidentiality constraints; and (iv) information unavailable/incomplete. Concerning disclosures falling into recommendations, reporting organizations are urged to disclose the information specified (GRI, 2022d).

The GRI Standards are structured as a modular system broadly covering three sets of standards: Universal Standards, Sector Standards, and Topic Standards (GRI, 2022d). The GRI Standards are posted on the GRI’s publicly accessible website (GRI, 2022g). The most up-to-date GRI Standards have been effective since 30 June 2022 (GRI, 2022h).

2.4 GRI Implementation: ‘With Reference to’ Versus ‘in Accordance with’

The GRI draws a distinction between reporting ‘with reference to’ the GRI Standards and reporting ‘in accordance with’ the GRI Standards, with the differences lying in the level of compliance with the GRI Standards. Companies report ESG issues ‘in accordance with’ the GRI Standards comply with more disclosure items pursuant to the GRI Standards than those report ‘with reference to’ the GRI Standards.

3.1 Theoretical Rationales of Corporate ESG Commitment

From a theoretical perspective, there are four primary theories bolstering corporate ESG commitment, including stakeholder theory, legitimacy theory, institutional theory, and signaling theory. They all suggest that pursuing ESG objectives could boost corporate value in the long term. As stakeholder theory highlights, to operate successfully, companies have a duty to cope with the expectations of diverse stakeholders (Donaldson & Preston, 1995). It is crucial for companies to build positive relationships with the stakeholders who exert a strong influence on them (Gray et al., 1995). Legitimacy theory insists that companies can survive only if they operate in fulfilment of social norms and values (Gray et al., 1996). As legitimacy theorists propose, there is a social contract between companies and members of the society, under which companies have to perform in line with social norms while they operate their business (Johnson & Holub, 2003). Reporting on the companies’ integrating information about ESG performance can strengthen corporate legitimacy and trust, leading to enhanced economic accomplishment (Veldman & Jansson, 2020). Institutional theory upholds that corporate conduct is scrutinized by diversified social structures, such as political establishment and regulations (Campbell, 2007). Due to the variances in institutional features across countries and sectors, different actors may undertake diverse CSR practices (Brammer et al., 2012). Institutional theory advances our understanding of cross-national CSR policies and regulations (Aureli et al., 2020). Signaling theory emphasizes that a party seeks to credibly disseminate information about itself to another party in face of asymmetric information (Spence, 1973). To lessen the information asymmetry regarding a company’s sustainability practices, companies may take the initiative to issue sustainability reports to the stakeholders (Connelly et al., 2011; Yang et al., 2021). By conveying a positive signal of the company’s attitude, management practices, and devotion to ESG issues, the firm will gain sounder reputation (Yang et al., 2021).

3.2 Major Factors Influencing Corporate ESG Reporting

Wide-ranging literature has analyzed how various components on firm level affect corporate ESG reporting practices. It is found that company size, government ownership, using the GRI framework, engagement of independent report assurance and industry are the major elements.

A large body of literature posits that company size has a positive impact on corporate ESG reporting. For example, Meek et al. (1995), Hahn and Kühnen (2013), Yu et al. (2018), and Drempetic et al. (2020) reach the same conclusion with regard to the size effect on ESG disclosure performance.

Some studies point to a positive relationship between government ownership and ESG reporting. For instance, Amran and Haniffa (2011), Gallo and Christensen (2011), Mohamed Adnan et al. (2018), and Özcan (2020) claim that companies controlled by the government achieve superior ESG reporting quality.

A great deal of research has attempted to investigate whether companies applying the GRI Standards and engaging in independent assurance for their sustainability reports attain superior ESG disclosure performance, ending up with contradictory findings. On the positive front, certain studies claim that the application of the GRI Standards and employment of independent assurance raise ESG disclosure quality. As Ballou et al. (2018) assert, third party assurance service refines reporting definitions, scopes and methodologies, leading to higher CSR reporting quality. This positive impact is more significant among companies employing accounting assurance providers rather than non-accounting assurance providers. Applying the GRI Standards facilitates the reporting companies to enhance non-error CSR restatements. Despite this, utilizing the GRI Standards is not conducive to discovering errors and hence its effect is not as significant as that of independent assurance. Independent assurance, if robustly arranged, could serve as an effective tool to upgrade information reliability and relevance (Sonnerfeldt & Aggestam Pontoppidan, 2020). The EU supports the engagement of independent assurance for upgrading sustainability reporting quality. It is suggested that mandatory assurance and verification of non-financial statements would bolster the authority of the NFRD (Tsagas & Villiers, 2020). The EU Corporate Sustainability Reporting Directive (‘CSRD’), which covers around 75 % of the turnover of all limited liability companies with effect from 2023, requires that sustainability reports must be audited by an independent assurance services provider with the participation of the key audit partner(s) (European Commission, 2022). On the negative front, some research points out that the correlation between applying the GRI Standards and ESG disclosure quality appears weak. Michelon, Pilonato, and Ricceri (2015) explore the impact of issuance of stand-alone CSR reports, engagement of third party assurance and applying the GRI Standards on ESG disclosure quality. They indicate that even though these practices give an impression to the stakeholders that the reporting company is dedicated to ESG issues, the ESG reporting quality is not improved. Moreover, the suggestion that companies using the GRI framework submit more complete disclosures is not strongly supported (Michelon et al., 2015). Barkemeyer et al. (2015) analyze 933 corporate GRI reports in 30 countries involving seven industries. As they observe, the GRI succeeded in raising publication and standardization of sustainability reporting. Nevertheless, they find that the report content tends to be uniform without revealing the specific geographical context or stakeholder networks of the companies, and materiality factors are not properly stated.

Besides, some scholars examine how industries are related to companies’ preference of referring to the GRI Standards and engagement of independent assurance service. Fernandez-Feijoo et al. (2014) conjecture that industries have an impact on corporate disclosure and credibility, and companies operating in various industries demonstrate different types of behavior. They classify corporate reporting characteristics into four types of behavior based on two criteria: disclosure and credibility. While disclosure refers to the level of applying the GRI Standards and reporting frequency, credibility refers to third party verification or self-declaration of the level of applying the GRI Standards. The first type of behavior is novice behavior, referring to corporate reporting with a low level of disclosure and also a low level of reporting credibility. Second is cautious behavior, which is defined as corporate reporting with a low level of disclosure but a high level of credibility. Third is chattering behavior, including corporate reporting with a high level of disclosure but a low level of credibility. Fourth is leading behavior, entailing corporate reporting with a high level of disclosure and a high level of credibility. Table 4 summarizes their findings with regard to the association between the type of ESG reporting behavior (level of disclosure and level of credibility) and industries.

3.3 Motivation Behind Applying the GRI Standards

Some scholars have examined why the GRI Standards are referred to by leading global companies, finding multiple reasons. Nikolaeva and Bicho (2010) observe that competitive and media pressures, the company’s CSR media visibility, and CSR publicity efforts play a crucial role in corporate application of the GRI Standards. As a result of institutional pressures and communicators of identity, following the GRI guidelines has become part of the social identity construction process. Companies applying the GRI Standards would gain legitimacy as socially responsible corporations. Through continuous interaction with the company’s stakeholders, the application of the GRI Standards performs some marketing function. Companies paying keener efforts in public relations to spread the information about their ESG commitment and being more visible in the media have a higher inclination to refer to the GRI Standards. Furthermore, peer pressure is a factor pushing companies to opt for the GRI Standards as reflected by the observation that the number of using the GRI framework in an industry notably influences the companies’ choice.

Jain et al. (2022) draw on social contagion theory to illuminate the nature and spread of sustainability disclosure referring to the GRI Standards. As social contagion theory suggests, structural equivalence, via second-hand channels such as mass media, corporate reporting, and industry networks, functions as a network instrument that boosts the diffusion of information and innovation (Jain et al., 2022). The mounting diffusion of applying the GRI Standards is nurtured by extensive media coverage, positive reports released by the early batch of companies using the GRI framework, promotional activities organized by the GRI, and regulators’ recommendations (Jain et al., 2022). Attributable to the institutional diffusion of sustainability reporting practices, application of the GRI Standards has become institutionalised as a reporting norm globally, and companies may feel compelled to imitate the standard practice so as to maintain their public image (Jain et al., 2022).

In the context of China, several studies inspect the impact of using the GRI framework on Chinese firms’ financial performance. As Yang et al. (2021) remark, building on signaling theory, PRC-listed companies using the GRI framework deliver a positive signal of their ESG commitment and therefore achieve improved corporate profitability. Xu et al. (2021) observe that PRC-listed companies referring to the GRI Standards enjoy a lower cost of debt financing relative to those not applying the GRI Standards.

3.4 Global Trend of Applying the GRI Standards

The orientation towards using the GRI framework varies across regions. Halkos and Nomikos (2021) have investigated the trend of applying the GRI Standards in all continents between 1999 and 2019. Among all continents, Europe and Asia have the highest number of companies referring to the GRI Standards. Europe, Oceania and Northern America have already passed the full-grown stage and turned to a downturn stage. Latin America, the Caribbean, and Africa have reached the full-grown stage. Asia is undergoing the spreading out stage with constant growth. A survey carried out by KPMG (2020) shows that around 66 % of the 5200 leading companies in 52 countries applied the GRI Standards for sustainability reporting, and 70 % of the Global Fortune 500 companies applied the GRI Standards. The studies conducted by KPMG (2020) and Halkos and Nomikos (2021) do not differentiate between reporting ‘with reference to’ and ‘in accordance with’ the GRI Standards. In 2019, well over half of the companies included in the S&P 500 Index used the GRI reporting framework (Governance & Accountability Institute, 2020). Among those applying the latest GRI Standards, around 70 % of them reported ‘in accordance with’ the core option, and merely 5 % reported ‘in accordance with’ the comprehensive option.

7.1 ESG Performance and Reporting Quality

The current study decides to employ the Refinitiv ESG company scores as a measure of the sample companies’ ESG performance and reporting quality. Refinitiv, a global leader in providing financial markets data, has calculated ESG company scores for over 11,800 companies all over the world (Refinitiv, 2022a). The Refinitiv ESG company scores take into account the companies’ industry in calculating the environmental and social scores and the companies’ country of incorporation in calculating the governance scores so as to mitigate industry and country bias (Refinitiv, 2022b). Refinitiv has developed the materiality metrics for each industry to adjust the scores assigned to ESG matters corresponding to the materiality level in their respective industries (Refinitiv, 2022b). Refinitiv updates the ESG scores for all covered companies on a weekly basis (Refinitiv, 2022b).

Refinitiv calculates two overall ESG scores for each covered company: ESG scores and ESG combined scores. The Refinitiv ESG scores review a company’s ESG performance, commitment and effectiveness in terms of 10 central themes by reference to publicly reported information (Refinitiv, 2022b). The Refinitiv ESG combined scores, predicated on global media sources, discount the ESG scores for news controversies materially influencing the companies (Refinitiv, 2022b). Suppose a company’s Refinitiv ESG score is 60 based on its reported information. If the company’s ESG disclosure has omitted an ESG matter with a negative impact on its operation but that is later reported by the media, then its Refinitiv ESG combined score will factor in the negative impact of the news report and make necessary deduction from the existing ESG score. As the Refinitiv ESG combined scores are a more comprehensive reflection of the companies’ ESG performance and reporting quality, the current study adopts the Refinitiv ESG combined scores to gauge the sample companies’ ESG performance and reporting quality.

The Refinitiv ESG company scores range from 0 to 100, with a higher score pointing to better ESG performance and a higher level of transparency in disclosing material ESG information to the public (Refinitiv, 2022a). Table 6 presents the interpretations of four Refinitiv ESG score ranges (Refinitiv, 2022b).

The present paper acknowledges that the reliability of the Refinitiv ESG scores is not free from challenges. Berg et al. (2021) postulate that Refinitiv has modified its ESG scores from time to time on grounds of revisions to calculation methodology and unannounced data, resulting in changes in ESG rankings over the past years. Notwithstanding this, this study adopts the Refinitiv ESG combined scores as a proxy for the sample companies’ ESG performance and reporting quality because of two considerations. First, the ESG ratings provided by the other leading vendors may contain similar problems (Berg et al., 2021) and there are no absolutely reliable ESG scores. Second, the Refinitiv ESG scores remain one of the most widely used ESG datasets thus far. As Berg et al. (2021) point out, over 1500 academic articles have utilized the Refinitiv ESG scores as a proxy for assessing corporate ESG performance since 2003. Some of them are published in top journals – for example, Ferrell et al. (2016), Garcia et al. (2017), Dyck et al. (2019), Albuquerque et al. (2020), Rajesh (2020), Uyar et al. (2020), Demers et al. (2021), Khaled et al. (2021), Santamaria et al. (2021), and Thimm and Rasmussen (2022).

7.2 Level of Disclosure

In the present research, the level of disclosure refers to the extent of ESG reporting in compliance with the GRI Standards. It includes three levels from the lowest compliance level to the highest compliance level: no application of the GRI Standards, reporting ‘with reference to’ the GRI Standards, and reporting ‘in accordance with’ the GRI Standards (core option/comprehensive option). This research classifies the sample companies’ level of disclosure based on their statements in ESG reports, either in the form of being incorporated in annual financial reports or standalone ESG reports.

To avoid confusion, this study has pre-determined the measurement for several types of situations under which the level of disclosure cannot be clearly identified counting on the sample companies’ statements. Table 7 outlines the classification of level of disclosure in this research as per the sample companies’ statements.

7.3 Level of Credibility

Prior research studies show that independent report assurance is helpful to enhance ESG reporting quality (Ballou et al., 2018; Sonnerfeldt & Aggestam Pontoppidan, 2020; Tsagas & Villiers, 2020). As inspired by Fernandez-Feijoo et al. (2014), the level of credibility in this study refers to an external and independent verification of the ESG report. This study classifies the sample reports’ level of credibility into three groups: no independent assurance, independent assurance for part of the disclosures, and independent assurance for all disclosures. The classification is grounded on the sample companies’ statements and/or assurance reports presented in their ESG reports.

7.4 Companies Owned by the Government

The Hang Seng Indexes Company Limited has classified all the 100 companies included in the HSCHK100 to H share, Red Chip, and other Hong Kong-listed Mainland Companies (Hang Seng Indexes Company Limited, 2022c). According to the definition provided by the HKEX, all Red Chip companies are controlled by the government (Hong Kong Exchanges and Clearing Limited, 2022d). Other Hong Kong-listed Mainland Companies are controlled by private parties. H shares may be controlled by the government or private parties. This research obtains the information about the sample companies’ ownership from the Wind database.

7.5 Industries/Sectors

The Hang Seng Indexes Company Limited has provided the industry classification for all the 100 companies included in the HSCHK100 (Hang Seng Indexes Company Limited, 2022c). The current research investigates the relationship between the sample companies’ ESG reporting performance and their industries predicated on the Hang Seng Indexes Company Limited’s industry classification. The Hang Seng Industry Classification System is prevalently adopted for researching on Hong Kong-listed companies. For instance, prior studies conducted by the HKEX (2022e), Hong Kong Monetary Authority (Hong Kong’s central bank) (2020), and PwC Mainland China and Hong Kong (2021), adopt the Hang Seng Industry Classification System to inspect ESG reporting in Hong Kong. This study develops the analysis in light of four types of behavior as proposed by Fernandez-Feijoo et al. (2014). These include novice behavior, chattering behavior, cautious behavior and leading behavior (see Table 4).

Table 8 displays the variables involved in this research and the relevant measurement and classification approach.

As the measurement tasks are carried out by one researcher only, there exists no risk of inter-researcher bias. In addition, the researcher has performed self-verification to maintain high consistency and accuracy.

8.1 Rate of Applying the GRI Standards Among the Sample Companies

Considering that the application of the GRI Standards was on a rise in Asia from 1999 to 2019 (Halkos & Nomikos, 2021), the first hypothesis is: The rate of applying the GRI Standards among the sample companies increased from 2017 to 2021.

This study finds that the total number of sample companies reporting ‘with reference to’/‘in accordance with’ the GRI Standards fluctuated in a narrow range in the past five years. The percentage started at 36 % in 2017, rose to 37 % in 2018 and 38 % in 2019–2020, and then dropped slightly to 37 % in 2021. For the sample companies that have reported ‘in accordance with’ the GRI Standards, only one of them opted for the comprehensive option throughout the sampling period. Table 9 and Graph 1 show the trend of using the GRI framework among the sample companies from 2017 to 2021. As such, the first hypothesis is marginally supported.

Graph 1:

Trend of applying the GRI Standards among the sample companies.

In the sampling period, the numbers of sample companies that have mentioned about the GRI Standards are 39 out of 73 sample companies (53 %) in 2017, 45 out of 81 sample companies (56 %) in 2018, 45 out of 84 sample companies (54 %) in 2019, 53 out of 95 sample companies (56 %) in 2020, and 60 out of 100 sample companies (60 %) in 2021. In all the sampling years, a notable number of companies state that their reports have referred to the GRI Standards but do not provide the GRI content index as required. According to the rules set out in Table 7, these companies are counted as companies which do not report ‘with reference to’ or ‘in accordance with’ the GRI Standards in the present research. If all those companies are included, then the percentage of using the GRI framework among the sample companies was on a rising trend (from 53 % in 2017 to 60 % in 2021). The study conducted by Halkos and Nomikos (2021) does not specify whether its analysis has included the companies not providing the GRI content index. The contrasting results generated in the current study may be attributed to different approaches to classifying those borderline cases.

8.2 ESG Reporting Quality of the Sample Companies Applying/Not Applying the GRI Standards

The current research seeks to categorize the sample companies into two groups for conducting a comparative analysis. The first group is composed of companies that utilize the GRI reporting framework, including those reporting ‘with reference to’ and ‘in accordance with’ the GRI Standards (either the core or comprehensive option). The second group covers companies that do not apply the GRI Standards for ESG reporting. Derived from the view that the application of the GRI Standards improves reporting quality (Ballou et al., 2018), this research comes up with the second hypothesis: The ESG reporting quality of the sample companies applying the GRI Standards is higher than the sample companies not applying the GRI Standards, as measured by the Refinitiv ESG combined scores.

As mentioned before, some sample companies were listed on the HKEX in 2018, 2019, 2020 or 2021, hence the Refinitiv ESG combined scores may not be available for those companies in the pre-listing years. In addition, the Refinitiv ESG combined scores assigned to six sample companies for financial year 2021 remain unavailable as at the researching date (4 February 2023), including Kunlun Energy Company Limited, China Unicom (Hong Kong) Limited, Hengan International Group Company Limited, China Feihe Limited, Xpeng Inc., and Baidu, Inc. Two sample companies are not covered by the Refinitiv ESG combined scores, including Jd Logistics, Inc. and Pop Mart International Group Limited.

The current research compares the average Refinitiv ESG combined scores assigned to two categories of sample companies for financial years 2017–2021: (i) those applying the GRI Standards; and (ii) those not applying the GRI Standards. The sample companies which Refinitiv ESG combined scores are not available are excluded from this analysis. In all sampling years, the average Refinitiv ESG combined score of sample companies using the GRI reporting framework is higher than that of their counterparts not applying the GRI Standards. This reflects that the ESG performance and reporting quality of those applying the GRI Standards is of higher standard than those not applying the GRI Standards, therefore the second hypothesis is supported. Table 10 and Graph 2 present the average Refinitiv ESG combined scores of sample companies applying and not applying the GRI Standards in all sampling years.

Graph 2:

Average Refinitiv ESG combined scores assigned to the sample companies applying and not applying the GRI Standards in 2017–2021.

A point to note is that the gap in the Refinitiv ESG combined scores between the sample companies applying and not applying the GRI Standards was diminishing in the sampling period. While the average Refinitiv ESG combined scores assigned to the sample companies applying the GRI Standards climbed mildly from 55 in 2017 to 60 in 2021, the scores assigned to those not following the GRI framework increased more sharply from 47 in 2017 to 56 in 2021. As discussed in Section 4, the HKEX’s ESG Reporting Guide has gradually upgraded corporate ESG reporting obligation from voluntary to ‘comply or explain’ since 2015. It means that all sample companies have to disclose more and more ESG information under the HKEX’s ESG Reporting Guide from 2017 to 2021. The research results may imply that the positive impact of referring to the GRI Standards on corporate ESG reporting performance becomes less significant in wake of tightening ESG reporting regulation.

8.3 Rate of Applying the GRI Standards Between Government-Controlled Companies and Private Companies

Given the widespread finding that government ownership is positively correlated with ESG reporting performance (Amran & Haniffa, 2011; Gallo & Christensen, 2011; Mohamed Adnan et al., 2018; Özcan, 2020), the third hypothesis is: Government-owned companies attain a higher rate of using the GRI framework.

For the period between 2017 and 2021, an average of 35 sample companies studied are owned by the government. In the five-year period under review, the average rate of applying the GRI Standards for government-owned companies stands at 50 %, which is far higher than that of private companies (28 %). This presents strong evidence in support of the third hypothesis.

For the sample companies owned by the government, 16 out of 35 (46 %) in 2017, 19 out of 35 (54 %) in 2018 and 2019, 18 out of 36 (50 %) in 2020, and 17 out of 36 (47 %) in 2021 reported ESG information ‘with reference to’ or ‘in accordance with’ the GRI Standards. As to the sample private companies, 10 out of 38 (26 %) in 2017, 11 out of 46 (24 %) in 2018, 13 out of 49 (27 %) in 2019, 18 out of 59 (31 %) in 2020, and 20 out of 64 (31 %) in 2021 reported ‘with reference to’ or ‘in accordance with’ the GRI Standards. Table 11 and Graph 3 illustrate the rate of applying the GRI Standards by government-owned and private companies in 2017–2021.

Graph 3:

Trend of applying the GRI Standards for sample companies owned by the government/private parties.

These findings are in line with previous studies concerning specific behavior of Chinese SOEs under the socio-economic circumstances in mainland China. While Chinese SOEs are expected to commit more to CSR activities (Chang et al., 2021a), private companies incline to attach greater weight to profits (Chang et al., 2021b; Gong et al., 2021). Using the GRI framework in ESG reporting may be perceived as a way to demonstrate corporate commitment to supporting government policy in favor of enhanced CSR disclosure, and thus Chinese SOEs are keener to take up this initiative.

8.4 Company Industries and Levels of Disclosure and Credibility in ESG Reporting

As Fernandez-Feijoo et al. (2014) posit, corporate ESG reporting performance can be classified into four types of behavior based on company industry, i.e. leading (high disclosure and high credibility), cautious (low disclosure and high credibility), chattering (high disclosure and low credibility), and novice (low disclosure and low credibility). The fourth hypothesis predicts that industries/sectors have an impact on the companies’ levels of disclosure and reporting credibility, and the companies’ ESG reporting practices exhibit four types of behavior according to their industry. As demonstrated in Section 7, the current research adopts the industry classification for all sample companies predicated on the Hang Seng Indexes Company Limited’s industry classification, which fits the nature of Hong Kong-listed companies. This study analyzes the sample companies’ level of disclosure and level of credibility by 12 industries, and then examines the results by reference to the four types of behavior conjectured by Fernandez-Feijoo et al. (2014). Because Fernandez-Feijoo et al. (2014) adopt a different industry classification system, the results produced in this research are not completely comparable to those of Fernandez-Feijoo et al. (2014). As a robustness check, the sample companies’ levels of disclosure and credibility in ESG reporting based on the sector classification system adopted by Fernandez-Feijoo et al. (2014) are presented in Appendix C. Besides, as the sample only covers the 100 largest Hong Kong-listed Chinese companies, it is acknowledged that the results may not be generalizable to other types of companies.

During the sampling years, the average rate of using the GRI framework among all sample companies is 35 %. The average rate of engaging in report assurance in 2017–2021 is 28 %, and most ESG reports were verified by the Big 4 accounting firms with limited assurance. The top three industries with the highest rate of applying the GRI Standards are financials (83 %), energy (65 %), and telecommunications (53 %). The sample companies operating in these three industries also attain the highest rate of engaging in report assurance (financials: 82 %; energy: 65 %; telecommunications: 53 %). The industry with the lowest rate of using the GRI framework is materials (13 %), followed by healthcare (14 %) and properties & construction (17 %). No sample companies operating in healthcare industry engaged in report assurance in all sampling years, and merely an average of 1 % in consumer discretionary industry and 3 % in consumer staples industry involved independent assurance for their ESG reports in 2017–2021.

Table 12 and Graph 4 summarize the findings concerning the sample companies’ rate of disclosure and rate of credibility by 12 industries in 2017–2021. As observed, four industries, including energy, financials, industrials, and telecommunications, attain an average rate of applying the GRI Standards and an average rate of report assurance higher than the average of all industries. These four industries fall within the category of leading behavior. Seven industries record an average rate of applying the GRI Standards and an average rate of engaging in report assurance below the average of all industries. They include conglomerate, consumer discretionary, consumer staples, healthcare, information technology, properties & construction, and utilities. These seven industries’ reporting practice is classified as novice behavior. Only one industry, materials, is categorized into cautious behavior (level of disclosure below the average of all industries and level of credibility above the average of all industries). No industries in the current study demonstrate ESG reporting practice of chattering behavior (level of disclosure above the average of all industries and level of credibility below the average of all industries). Therefore, the fourth hypothesis is partially rejected.

Graph 4:

Sample companies’ rate of applying the GRI Standards and rate of assurance by industries, with corporate behavior types as asserted by Fernandez-Feijoo et al. (2014).

In sharp contrast with Fernandez-Feijoo et al. (2014)’s findings which distinguish four types of ESG reporting behavior by industries, the current study based on the Hang Seng Industry Classification System only identifies three types of behavior (leading, novice, and cautious). In the present research, the rate of report assurance does not exceed that of applying the GRI Standards in all industries except materials. This shows that engaging in report assurance among almost all industries is not as popular as using the GRI framework, explaining why in this study only materials industry falls into cautious behavior. Furthermore, the sample companies operating in the top three industries with leading behavior (i.e. financials, energy, and telecommunications) outperform those operating in the other industries significantly with regard to both application of the GRI Standards and engagement of report assurance, pushing up the average standard. The sample companies operating in the other industries can hardly reach the average rates of applying the GRI Standards and engaging in assurance, and thus almost all fall into novice behavior.

Stakeholder theory, legitimacy theory, institutional theory, signaling theory, and social contagion theory illustrated in Section 3 can be applied to understand the ESG commitment of companies operating in financials, energy, and telecommunications industries. The sample companies operating in these three industries may encounter greater expectation from their key stakeholders, hence they attempt to pursue ESG objectives and show their desire to conform to social norms and values, which can strengthen their legitimacy. Because applying the GRI Standards is a global norm in sustainability reporting (Jain et al., 2022), they tend to report based on the GRI framework to brighten their reputation. Some prior studies have specifically looked into the ESG practices of these three industries. Regarding financial companies, it is suggested that the image of the financial sector worldwide has worsened following the global financial crisis in 2007, attracting scrutiny by a broader scope of stakeholders (del Mar Alonso-Almeida et al., 2014; Jain et al., 2022). Therefore, financial companies were pressured to apply the GRI framework to raise reporting transparency and credibility (del Mar Alonso-Almeida et al., 2014; Jain et al., 2022). In China, the Chinese government has put forth various regulations requiring financial institutions to establish an environmental risk management system and enhance environmental disclosures, thus Chinese financial companies’ sustainability reporting practices have increased gradually and become institutionalized (Dong et al., 2021). With regard to energy companies, since the energy industry is commonly regarded as more polluting, energy companies have a stronger incentive to participate in CSR activities such as releasing social reports and applying the GRI Standards (del Mar Alonso-Almeida et al., 2014). As to the telecommunications industry, Arrive et al. (2019) assert that the success of Chinese telecommunications companies largely hinges on whether they can maintain customer satisfaction and employee loyalty, and pursuing CSR practices is a prevalently pursued strategy to polish their business reputation.

Moreover, the sample companies that are cross-listed on the Shanghai Stock Exchange and operating in the financial industry or industries with a severe impact on the environment are subject to tighter ESG disclosure requirements. 12 out of 13 sample companies (around 92 %) operating in the financial industry are cross-listed on the Shanghai Stock Exchange, which requires financial companies to publish CSR reports on an annual basis (Notice on Doing Well the Work of Disclosing the Reports of Listed Companies, 2008; Self-Regulatory Supervision Guidelines for Listed Companies No. 1 – Standardized Operation, 2022). All the four sample companies operating in the energy industry are cross-listed on the Shanghai Stock Exchange. While three of them are oil and gas producers, one operates in the coal sector. The Shanghai Stock Exchange stipulates that companies operating in industries with a severe impact on the environment, such as thermal power generation, iron and steel smelting, cement production, electrolytic aluminum, and mineral development, have to disclose environmental information in an annual CSR report or make separate disclosure (Notice on Doing Well the Work of Disclosing the Reports of Listed Companies, 2008; Self-Regulatory Supervision Guidelines for Listed Companies No. 1 – Standardized Operation, 2022). Although the rules do not specify whether the reporting requirements apply to companies operating in the oil and gas/coal sectors, the sample companies operating in these sectors may face graver regulatory pressure to disclose more environmental information compared with those operating in the non-environmentally-sensitive industries.

Besides, the percentages of sample companies owned by the government in the financial industry, energy industry, and telecommunications industry are the highest in comparison with those in the other industries (see Table 13). While all the four energy companies included in the sample are government-owned, three out of four (around 79 %) of the telecommunications companies, and 10 out of 13 (around 77 %) of the financial companies are owned by the government. As discussed in Section 8.3, government ownership can be a key driver of applying the GRI Standards in China’s context. Government ownership, in conjunction with the aforementioned industry factors, explains why these three industries attain higher rates of applying the GRI Standards and engaging in independent report assurance.

It bears underlining that the HKEX has noticed the relatively low rate of report assurance among Hong Kong-listed companies and made some recommendations. The HKEX (2022e), selecting a sample that comprises 400 Hong Kong-listed companies through stratified random sampling by industry, observes that merely 6.7 % of the sample companies obtained independent assurance for their ESG reports. Taking account of the companies’ concern about cost or compliance burden caused by independent assurance, the HKEX encourages companies to seek assurance for part(s) of the ESG reports or for particular data.

This study runs two robustness tests examining the sample companies’ level of disclosure and level of credibility by classifying the sample companies into 40 sectors under GRI’s latest classification system approved in 2021 (see Appendix C). In the first test, three sample companies not covered by GRI’s sector classification system are categorized into a sub-group ‘Other’. In the second test, those three sample companies not covered by GRI’s sector classification system are excluded from the analysis. Because no sample companies fall into 16 sectors under GRI’s sector classification system, the tests in Appendix C cover a total of 24 sectors. Although the tests in this Section and Appendix C adopt two different industry classification systems, some similar patterns are observed. Largely consistent with the results in this Section, the tests in Appendix C show that most company sectors fall into novice behavior, almost one third demonstrate leading behavior, and only a tiny minority display cautious behavior (see Table 14). A point of distinction between the results obtained in this Section and Appendix C is that no sectors demonstrate chattering behavior in the current test, yet 1-2 sectors fall into this category in the tests performed in Appendix C. While the results of this Section find that financial, energy, and telecommunications are the top three industries with leading behavior, some comparable sectors in the tests conducted in Appendix C, including banking, insurance, coal, oil and gas, and media and telecommunications, fall into leading behavior as well.