This paper aims to explore how board gender diversity impacts on circular economy (CE) disclosure by a sample of European listed companies. In addition, the paper investigates the moderating role of environmental, social and governance (ESG) controversies in previous relationship.
The study conducted a regression analysis on a sample of 485 companies and 3,761 firm-year observations of European listed companies operating in 19 countries between 2004 and 2021.
The results reveal that the presence of female directors on the board favors the release of higher levels of CE disclosures. Moreover, female directors operating in companies characterized by higher levels of ESG controversies tend to release higher CE disclosures.
First, the paper does not investigate the qualitative dimension of disclosures. Moreover, the research does not examine other elements of differentiation within the boards, such as cultural or religious diversities.
The analysis shows that diversity has an impact on the dissemination of CE information. This should lead companies and policymakers to orient their actions toward both greater diversity in board composition and the higher CE disclosure in fostering sustainable development.
This paper offers novel contributions to existing literature suggesting an objective way to measure CE-related disclosure and investigating the moderating role of ESG controversies in the relationship between gender diversity and CE disclosure.
1. Introduction
In recent years, companies are increasingly requesting to disclose environmental information in response to new regulations (i.e. Directive 2014/95/EU which was recently replaced by the Corporate Sustainability Reporting Directive – CSRD 2022/2464/EU) and stakeholder pressures (Roberts et al., 2022; Sinclair-Desgagne and Gozlan, 2003). Environmental reporting is considered a useful tool for addressing environmental issues (Dagiliene et al., 2020; Zivin and Neidell, 2009), such as climate change, global warming, droughts and deforestation (Yokessa and Marette, 2019; Renstrom et al., 2019; Alrazi et al., 2015). The circular economy (CE) seeks to overcome previous environmental issues by favoring the efficient use of resources through the transition from their traditional linear transformation in production and consumption to a circular transformation (Ellen MacArthur Foundation – EMF, 2012). The circular transformation of resources can be achieved through actions aimed at reducing, reusing, recycling and recovering, better known as the four CE dimensions (Kirchherr et al., 2017). The European Commission (EC), identifying the importance of shifting toward a more CE approach, issued the CE Action Plan in 2015 aiming at reducing waste and reusing resources more efficiently. In a similar vein, the EC recognized the relevance of providing comparable environmental disclosure, requiring companies to perform this disclosure according to specific sustainability standards (Chapter 6a Sustainability Reporting Standards of CSRD) and asked the European Financial Reporting Advisory Group (EFRAG) to prepare a set of sustainability standards. EFRAG issued the first draft of these standards in July 2023, and among the sustainability standards issued, there is a standard specifically devoted to CE-related disclosure (ESRS E5 “Resource use and circular economy”).
Although the CE paradigm is relevant in promoting sustainable development, CE-related disclosure is a poorly investigated topic (Opferkuch et al., 2021; Dagiliene et al., 2020), as is the identification of the determinants of such disclosures (Vitolla et al., 2023; Kuo and Chang, 2021). Although there are different studies investigating the relationship between board composition and environmental disclosure, there are no research on the effects of board of directors on the CE-related disclosure. This paper seeks to fill the existing research gap by investigating the influence of board gender diversity (BGD) on CE-related disclosure, using both the agency and the stakeholder theories.
According to these theoretical frameworks, the presence of women sitting on the board results as a useful control mechanism in reducing the information asymmetries. Women directors, enhancing board effectiveness (Jizi, 2017), increase the disclosure quality (Khemakhem et al., 2022; Ayman et al., 2019). Moreover, female directors are more stakeholder-oriented than male members (Injeni et al., 2022), as they pay more attention to the information released to external parties. In addition, they are more oriented toward sustainability themes, mainly because of their characteristics, such as being supportive, respectful, kind and empathetic (Katmon et al., 2019). Both the agency and the stakeholder theories support a positive relationship between BGD and CE-related disclosure. In line with these theoretical argumentations, most previous studies found a positive association between these variables (e.g. Injeni et al., 2022; Baalouch et al., 2019; Elmagrhi et al., 2019). However, there are also studies showing a nonsignificant relationship (Gallego‐Álvarez and Pucheta‐Martínez, 2020; Garanina and Aray, 2021; Hussain et al., 2018), revealing the need to further investigate the moderating role of other variables. This paper specifically focuses on the moderating role of environmental, social and governance (ESG) controversies in the relationship between BGD and CE-related disclosure. The moderating role of ESG controversies in the relationship between BGD and CE-related disclosure is not present in the literature. To date, few studies have investigated the influence of ESG controversies on environmental disclosure (e.g. Schiemann and Tietmeyer, 2022). Most studies conducted on ESG controversies have analyzed their influence on company financial performance (e.g. Brinette et al., 2023; Shakil et al., 2021).
Companies sometimes do not assign priority to sustainability topics and involve themselves in ESG controversies, consisting of bad news on environmental and/or social scandals or lawsuits (Aouadi and Marsat, 2018). Companies presenting high levels of ESG disputes tend to release more disclosures to mitigate the negative effects of ESG controversies (Schiemann and Tietmeyer, 2022), obtaining legitimation for their actions and achieving stakeholders’ trust (Shakil et al., 2021). In this context, female directors, paying more attention to sustainability issues than male members (Arayssi et al., 2020) and seriously considering ESG controversies (Shakil et al., 2021), should favor higher levels of CE-related disclosure.
The results of the research reveal that the presence of female directors on the board favors the release of higher levels of CE-related disclosures. Moreover, in companies characterized by higher levels of ESG controversies, female directors tend to release higher CE-related disclosures.
This paper contributes to the literature by investigating CE-related disclosure using data collected from the Refinitiv database, suggesting an objective way to measure this phenomenon that could be used by scholars in future research. Almost all previous studies on CE-related disclosure developed a disclosure index based on the CE dimensions (i.e. Dagiliene et al., 2020) and/or some standards that contained suggestions on CE, such as the GRI 306 on waste-related impacts (i.e. Roberts et al., 2022). However, as stated by Opferkuch et al. (2021), there are only a few standards containing suggestions on CE-related disclosures, meaning that it is very difficult for scholars to develop a disclosure index that captures all the CE aspects disclosed by companies. Moreover, most of the previous studies explored the CE-related information provided by companies using data from 2016 or even more recently (i.e. Opferkuch et al., 2021; Dagiliene et al., 2020). This paper adds to previous research by investigating the CE-related disclosure released in the period 2004–2021, that is, both before and after the issue of regulation on CE. Moreover, the presented results show the companies’ CE disclosure behavior in a context of voluntary CE disclosure, that is before the implementation of CSRD. The paper contributes to the research on the determinants of disclosure by investigating the influence of BGD on CE-related disclosure and the moderating role played by ESG controversies. To date, in fact, only the research of Kuo and Chang (2021) investigated the determinants of CE-related disclosure by focusing on the effect of firm-specific variables on CE-related disclosure.
To achieve the pursued objectives, the paper is structured as follows. The next section describes the background of the study. Section 3 illustrates the theoretical frameworks and Section 4 reviews the previous studies that are addressed as useful to develop the proposed hypotheses. Section 5 illustrates the methodology, while Section 6 reports the results of the study. Section 7 discusses the results and concludes the paper by also illustrating the limitations and future research.
2. Background
The paper focuses on the European context. The choice of this setting was guided by various factors. Firstly, the increasing EU attention toward sustainability and, specifically, CE issues. In fact, the CE’s ability to achieve the sustainable development goals (Schroeder et al., 2019; Ghisellini et al., 2016; Rashid et al., 2013) brought the EU in issuing several related regulations (see Table 1).
Regulations, frameworks and standards on CE
| EU regulations on CE | Frameworks and standards on CE |
|---|---|
| 2014. Toward a circular economy: a zero waste programme for Europe 2015. CE action plan 2019. European Green Deal 2019. Report on the implementation of the Circular Economy Action Plan 2020. A New Circular Economy Action Plan for a Cleaner and More Competitive Europe 2020. European Taxonomy Regulation 2020 / 852 2022. Corporate Sustainability Reporting Directive 2464/2022 | 2017. Eco-Management and Audit Scheme (EMAS) 2017. British Standards Institute (BSI) BS 8001 2018. Underwriters Laboratories (UL) 3600 2020. Global Reporting Initiative (GRI) 306 2020. World Economic Forum (WEF) “Measuring Stakeholder Capitalism: Towards Common Metrics and Consistent Reporting of Sustainable Value Creation” 2022. European Sustainability Reporting Standard (ESRS) E5 “Resource Use and Circular Economy” |
Source(s): Developed by authors
The EC introduced the CE concept in 2014 by issuing the communication “Towards a Circular Economy: A Zero Waste Programme for Europe”. In 2015, the EC issued the EU Action Plan aimed at fostering the transition toward CE, by reducing waste and boosting a more efficient use of the resources. The first CE Action Plan marked the beginning of the EU’s formal strategy on the CE, outlining a roadmap for integrating CE principles across the European value chain and stimulating regulatory and industrial changes. In 2019, the EC issued the European Green Deal, which is considered as a key pillar to reduce carbon emissions and promote sustainable growth. In the same year, the EC illustrated the main achievements and potential challenges in fostering the transition toward CE by issuing the “Report on the implementation of the Circular Economy Action Plan”. Based on the analysis conducted in this report, in 2020 the EC published “A New Circular Economy Action Plan for a Cleaner and More Competitive Europe”, which indicates new measures to be started by March 2022. The transition to CE has also been included in the European Taxonomy Regulation 2020 / 852 as one of the six key environmental targets to pursue sustainable finance.
In addition to fostering the transition toward CE, the EU also recognized the relevance for companies to release CE-related disclosure to external stakeholders. A first attempt to introduce the need to release environmental information was conducted with the issuing of Directive 2003/51/EC, followed by Directive 2014/95/EU. Previous directives required companies to provide environmental information, although they did not report detailed suggestions on the CE information to disclose. The explicit requirement to release CE disclosure was contained in the CSRD, according to which companies must release CE information in their sustainability reports starting from fiscal year 2024. In disclosing CE information, companies must refer to the European sustainability standards, in particular to the ESRS E5.
Therefore, before the implementation of CSRD, companies released CE disclosure in a voluntary manner. Moreover, before the issuing of the ESRS E5, there were no frameworks or standards specifically dedicated to CE disclosure (GRI, 2019). Except for the GRI and WEF, the five documents issued on the CE, illustrated in Table 1, were largely focused on the activities for introducing the CE principles within a company’s processes, indicating only indirectly the CE disclosure to release.
The decision to voluntarily release CE disclosure is influenced, among other factors, by the composition of the board of directors (Elmghaamez et al., 2024). This aspect represents the second reason that guided the choice of the European setting. More specifically, this paper focuses on the board gender diversity, that is a topic of great interest for scholars, policymakers and government, due to the frequent underrepresentation of women within the board. To improve the boards heterogeneity, different EU countries introduced reforms aiming at enhancing, on a mandatory or voluntary manner, the percentage of female directors sitting on the board. Due to these reforms, the number of women sitting on the board has gradually risen, mainly in countries that introduced mandatory requirements (Nguyen et al., 2020). Therefore, the European context represents and interesting setting in which investigates the influence of board gender diversity on CE disclosure.
3. Theoretical literature review
The release of CE-related disclosure allows companies to satisfy stakeholders’ needs and increase their legitimacy (Lepore et al., 2022; Stewart and Niero, 2018; Lock and Seele, 2016; de Colle et al., 2014). In addition, the provision of CE-related disclosure favors investors’ understanding of both companies’ operations and their level of risk, making their investment decisions more effective (Kuo and Chang, 2021).
The decision to release CE disclosure or, more generally, environmental disclosure, as well as the analysis of its determinants, has been widely investigated with references to agency, stakeholder, legitimacy or institutional theories. This paper referred to both agency theory (Jensen and Meckling, 1976) and stakeholder theory (Freeman, 1984) to explain the relation between BGD and CE-related disclosure (Bannò et al., 2023; Nicolò et al., 2022), considering the provision of information a way to reduce the conflicts between the company’s management and stakeholders.
According to the agency theory, the agency problems arise when there is a divergence of interests between the principal and the agent, that operates on the principal’s behalf. To reduce the agent’s discretion, the principal can define specific control mechanisms (named agency costs), which influence the behaviours of the agent. In this context, the presence of women sitting on the board is considered a useful control mechanism in reducing the information asymmetries between the agent and the principal. According to the previous literature, a larger presence of women, enhancing board effectiveness (Jizi, 2017), brings to higher levels of disclosure quality (Khemakhem et al., 2022; Ayman et al., 2019).
According to the stakeholder theory, the success of a company depends on its capability to fulfill all stakeholders’ needs (Alatawi et al., 2023; Alipour et al., 2019). Each company, in fact, has numerous counterparts, apart from its shareholders, that can both influence and be affected by the company’s actions in different ways. In this context, a strong and more diversified board structure promotes the reduction of the information asymmetries between the company and all its stakeholders, favoring a better dialog with stakeholders by monitoring the corporate reporting processes (Michelon and Parbonetti, 2012). With specific concern to BGD, the literature suggests that the presence of women on the board favors the release of higher levels of environmental disclosure, as women are more stakeholder-oriented (Injeni et al., 2022) and present greater attention to sustainability questions (Hussain et al., 2018; Matsa and Miller, 2013).
4. Empirical literature review and hypotheses development
To date, few studies have been conducted on CE-related disclosures released by companies. Almost all the studies mainly performed a content analysis of the sustainability report (i.e. Opferkuch et al., 2021; Camilleri, 2015) or the integrated report (i.e. Barnabè and Nazir, 2021) and developed a disclosure index to collect data on CE-related disclosure releases. These disclosure indices were based on the CE dimensions (i.e. Dagiliene et al., 2020) or the few disclosure standards that contained some suggestions on CE-related information (i.e. Tiscini et al., 2022; Roberts et al., 2022). The results revealed that companies released little information (Roberts et al., 2022; Kuo and Chang, 2021) and, when provided, the CE content is superficial, inconsistent (Opferkuch et al., 2021) and largely disclosed in a qualitative form (Dagiliene et al., 2020; Stewart and Niero, 2018). The information released mainly concerns the “reduce” dimension, describing the results and activities carried out to achieve both energy savings and waste reduction (Opferkuch et al., 2021; Kuo and Chang, 2021; Dagiliene et al., 2020). The main conclusion of almost all previous research is that the low level of CE-related disclosure is the consequence of the absence of rules and standards specifically devoted to this topic (Roberts et al., 2022; Opferkuch et al., 2021; GRI, 2019) and, therefore, that a more consistent approach to CE-related disclosure is necessary (Opferkuch et al., 2021). More likely, this target could be achieved in the coming years, thanks to the adoption of both CSRD and disclosure standards specifically dedicated to CE issues, such as the ESRS E5.
4.1 Board gender diversity and circular economy disclosure
With respect to the relation between BGD and CE disclosure, to date no studies have investigated this association. Most previous studies have analyzed the association between BGD and the wider concept of environmental disclosure (Lepore and Pisano, 2023), mainly revealing that the presence of female directors on the board increases the level of disclosed information (Buallay and Alhalwachi, 2022; Effah et al., 2022; García‐Sánchez et al., 2022; Injeni et al., 2022).
According to the agency theory, female directors enhance board effectiveness (Jizi, 2017), by better monitoring the management activities and increasing the commitment toward sustainability issues, and, therefore, improve disclosure quality (Khemakhem et al., 2022; Ayman et al., 2019). According to stakeholder theory, the presence of women directors promotes the release of higher levels of environmental disclosure, as women are more stakeholder-oriented (Injeni et al., 2022) and present greater attention to sustainability questions (Hussain et al., 2018; Matsa and Miller, 2013). This greater commitment toward sustainability issues could be due to some female characteristics, such as being supportive, respectful, kind and empathetic (Katmon et al., 2019; Kend, 2015).
In line with theoretical arguments, most studies have hypothesized and found a positive relationship between BGD and environmental disclosure (e.g. Injeni et al., 2022; Baalouch et al., 2019; Elmagrhi et al., 2019; Fuente et al., 2017; Tamimi and Sebastianelli, 2017; Javaid Lone et al., 2016; Liao et al., 2015; Garcia-Sanchez et al., 2014; Ntim and Soobaroyen, 2013; Rao and Tilt, 2016). However, there are also studies that found a non-significant association between BGD and environmental disclosure (Gallego‐Álvarez and Pucheta‐Martínez, 2020; Garanina and Aray, 2021; Hussain et al., 2018).
Based on the theoretical argumentations, we hypothesize the following:
There is a positive association between BGD- and CE-related disclosures.
4.2 The influence of environmental, social and governance controversies
ESG controversies consist of bad news on environmental and/or social scandals or lawsuits (Aouadi and Marsat, 2018) that could damage the reputation of a company (Shakil et al., 2021). The occurrence of ESG controversies puts the company under stakeholder scrutiny, as ESG controversies enhance uncertainty about the company’s prospects and strategies (Schiemann and Tietmeyer, 2022). As a result, the manifestation of ESG controversies can negatively influence stakeholders’ opinions and damage companies’ legitimacy (Brinette et al., 2023).
The literature suggests that companies characterized by high levels of ESG controversies tend to release more information to mitigate uncertainty due to ESG controversies (Schiemann and Tietmeyer, 2022). In fact, disclosure is considered an instrument to identify and assess the impact of ESG controversies (Amel-Zadeh and Serafeim, 2018). In this sense, companies tend to increase ESG disclosure to protect themselves from negative stakeholder reactions that could appear after the manifestation of an ESG controversy (Hummel and Rötzel, 2019). Releasing more information, companies can legitimate their activities and achieve stakeholder trust (Shakil et al., 2021; Patten and Shin, 2019), especially in the presence of ESG controversies (Brinette et al., 2023). In line with these considerations, the study of Schiemann and Tietmeyer (2022) empirically shows that the existence of high environmental controversies enhances the ESG disclosure released because of the increased pressure from external stakeholders. Therefore, companies presenting high levels of ESG controversies should also report greater CE-related disclosures.
With respect to the relation with CE-related disclosure, women directors could increase the level of information in the presence of higher levels of ESG controversies to contain reputational damage. Female directors pay more attention to sustainability issues than male members (Arayssi et al., 2020) and seriously consider ESG controversies (Shakil et al., 2021). Thanks to their female characteristics, women lead companies to adopt lower levels of harmful and illegal practices (Passas et al., 2022) and, in the occurrence of ESG controversies, tend to limit their negative effects. Therefore, stakeholders value in a positive way the presence of women on the board in the occurrence of ESG controversies because female directors can limit the negative effects of ESG controversies (Brinette et al., 2023).
On this basis, the study hypothesizes the following:
ESG controversies positively moderate the relationship between BGD and CE-related disclosure.
5. Research design
5.1 Sample selection and data source
The study is based on a population of 574 European listed companies (8,748 firm-year observations). Specifically, on the Refinitiv database we selected all the listed companies whose both headquarters and businesses are located in a European member state, which prepared a sustainability report in the period 2004–2021, and that presented a market capitalization equal or higher than 500 million. Due to their characteristics, financial and insurance companies were not included in the sample.
We decided to start the analysis in 2004 to investigate what happened before the introduction of some form of regulation on CE (the CE Action Plan was issued in, 2015). The year 2004 was selected as the starting year as it was the first year after the issuing of Directive 2003/51/EC, with which the EU legislature introduced a first requirement for companies to report environmental information in their management commentary and analysis. The last year analyzed is 2021, as is the year with the most recent available data when the study was conducted. We eliminated 89 companies due to the unavailability of data on Refinitiv. The final unbalanced sample was composed of 485 companies (3,761 firm-year observations) belonging to 18 industries and 19 different countries.
Data on CE-related disclosure, board characteristics and accounting and financial data were collected from the Refinitiv Eikon database.
Table 2 shows the sample selection process (Panel A) and its composition by Country (Panel B) and industry (Panel C).
The sample
| Panel A. Sample selection process | |||||||||
| N. of firms | |||||||||
| Listed companies: - Having the country of headquarters in one of 27 European countries - Having the country of Exchange in one of 27 European countries - Preparing a sustainability report in the period 2004–2021 - Having a market capitalization equal or higher than 500 million | 574 | ||||||||
| Companies with no accounting and financial data | (89) | ||||||||
| Final sample | 485 | ||||||||
| Panel B. sample by country | Panel C. sample by industry | ||||||||
| Country of Headquarter | No. of firms | No. of obs | % of obs | Sector | No. of firms | No. of obs | % of Obs | ||
| 1 | Austria | 17 | 121 | 3.22 | 1 | Accommodation and Food Services | 7 | 54 | 1.44 |
| 2 | Belgium | 16 | 106 | 2.82 | 2 | Admin. and Supp., Waste Manag. and Remed. Services | 11 | 86 | 2.29 |
| 3 | Czech Republic | 1 | 17 | 0.45 | 3 | Agriculture, Forestry, Fishing and Hunting | 1 | 6 | 0.16 |
| 4 | Denmark | 18 | 131 | 3.48 | 4 | Arts, Entertainment, and Recreation | 5 | 29 | 0.77 |
| 5 | Finland | 22 | 196 | 5.21 | 5 | Construction | 24 | 184 | 4.89 |
| 6 | France | 73 | 611 | 16.25 | 6 | Educational Services | 1 | 5 | 0.13 |
| 7 | Germany | 85 | 624 | 16.59 | 7 | Finance and Insurance | 55 | 417 | 11.09 |
| 8 | Greece | 11 | 96 | 2.55 | 8 | Health Care and Social Assistance | 31 | 213 | 5.66 |
| 9 | Hungary | 4 | 48 | 1.28 | 9 | Information | 191 | 1643 | 43.69 |
| 10 | Ireland | 2 | 8 | 0.21 | 10 | Manufacturing | 3 | 27 | 0.72 |
| 11 | Italy | 54 | 367 | 9.76 | 11 | Mining, Quarrying, Oil and Gas Extraction | 2 | 13 | 0.35 |
| 12 | Luxembourg | 9 | 55 | 1.46 | 12 | Other Services (except Public Administration) | 29 | 175 | 4.65 |
| 13 | Netherlands | 23 | 238 | 6.33 | 13 | Professional, Scientific, and Technical Services | 37 | 185 | 4.92 |
| 14 | Poland | 8 | 72 | 1.91 | 14 | Real Estate, Rental and Leasing | 20 | 155 | 4.12 |
| 15 | Portugal | 8 | 70 | 1.86 | 15 | Retail Trade | 16 | 165 | 4.39 |
| 16 | Romania | 5 | 17 | 0.45 | 16 | Transportation and Warehousing | 16 | 73 | 4.39 |
| 17 | Slovenia | 2 | 9 | 0.24 | 17 | Utilities | 28 | 281 | 7.47 |
| 18 | Spain | 36 | 358 | 9.52 | 18 | Wholesale Trade | 8 | 50 | 1.33 |
| 19 | Sweden | 91 | 617 | 16.41 | Tot. | 485 | 3,761 | 100.00 | |
| Tot. | 485 | 3,761 | 100.00 | ||||||
Source(s): Developed by author
5.2 Dependent, independent, moderating and control variables
Table 3 shows all the variables included in the analysis, illustrating how they have been measured, the source of data for each one and previous studies that used the same variable.
Description of variables and measurement
| Variable | Description | Measurement | Source | Previous studies |
|---|---|---|---|---|
| Dependent variable | ||||
| CEDisc | CE-related disclosure | Unweighted sum of 12 environmental items disclosed by companies, assigning value 1 when the company does present information on that environmental aspect and 0 otherwise | Eikon | Gallego‐Alvarez et al., 2017; Du et al., 2010; Kolk and Pinkse, 2010; Chen and Bouvain, 2009 |
| Independent variable | ||||
| BoGenDiv | Board gender diversity | Percentage of female on the board | Eikon | Pozzoli et al., 2023; Lepore and Pisano, 2023; Raimo et al., 2022; Vitolla et al., 2023; Liao et al., 2015 |
| Moderating variable | ||||
| ESGContr | ESG controversies | 100 minus the ESG Controversy Score, that is calculated based on 23 ESG controversy topics | Eikon | D’Amore et al., 2024; Passas et al., 2022; Schiemann and Tietmeyer, 2022 |
| Interaction term | ||||
| BoGenDiv * ESGContr | Interaction term | Two ways interaction term obtained by multiplying the BoGenDiv and ESGContr | Eikon | |
| Control variables | ||||
| Profit | Profit | Return on Assets | Eikon | Pozzoli et al., 2023; Raimo et al., 2022; Ayman et al., 2019 |
| Lev | Leverage | Long-term debt divided by total assets | Eikon | Pozzoli et al., 2023; Lepore and Pisano, 2023; Chan et al., 2013 |
| FirmSize | Firm size | Natural logarithm of total assets | Eikon | Lepore and Pisano, 2023; Vitolla et al., 2023; Michelon and Parbonetti, 2012 |
| BoTen | Board tenure | Average number of years each board member has been on the board | Eikon | Katmon et al., 2019; Shiah-Hou, 2021 |
| BoMeet | Board meetings | Number of board meetings during the year | Eikon | D’Amore et al., 2024; Lepore and Pisano, 2023 |
| BoSize | Board size | Total number of board members at the end of the fiscal year | Eikon | Khan, 2021; Nicolò et al., 2022; Raimo et al., 2022; Aladwey, 2021; Vairavan and Zhang, 2020; Assenga et al., 2018 |
| BoInd | Board independence | Percentage of independent directors on the board | Eikon | Lepore et al., 2022; Garcia-Sanchez et al., 2014; Vitolla et al., 2023; Assenga et al., 2018; Michelon and Parbonetti, 2012 |
Source(s): Developed by authors
The dependent variable, CE-related disclosure (CEDisc), was measured as an unweighted sum of 12 environmental items collected by the Refinitiv Eikon database that measure a company’s level of environmental disclosure on CE topics. Refinitiv measured each item by assigning a score of 1 when the company discloses the specific environmental information and 0 otherwise (Gallego‐Alvarez et al., 2017; Du et al., 2010; Kolk and Pinkse, 2010; Chen and Bouvain, 2009).
The 12 environmental items considered to define the value of CEDisc belong to three different categories: 6 items for emissions, 4 for innovation and 2 for resource use (see Table 4). The emission category describes the company’s commitment to and success in reducing environmental emissions in production processes. The innovation category describes companies’ willingness to reduce costs and environmental obligations by envisioning new economic prospects through environmentally sound technologies and circular practices. Finally, the resource use category describes companies’ attitudes toward adopting the 4R paradigm by avoiding resource waste and improving supply chain management.
Description of environmental disclosure items
| CE-related disclosure | ||
|---|---|---|
| Emissions category | Innovation category | Resource use category |
| Title | Title | Title |
| Description | Description | Description |
| VOC or particulate matter emissions reduction | Eco-Design products | Environmental materials sourcing |
| Does the company report on initiatives to reduce, substitute or phase out volatile organic compounds (VOC) or particulate matter less than ten microns in diameter (PM10)? | Does the company report on specific products which are designed for reuse, recycling or the reduction of environmental impacts? | Does the company claim to use environmental criteria (e.g. life cycle assessment) to source or eliminate materials? |
| NOx and SOx emissions reduction | Take-back and recycling initiatives | Toxic chemicals reduction |
| Does the company report on initiatives to reduce, reuse, recycle, substitute or phase out SOx (sulfur oxides) or NOx (nitrogen oxides) emissions? | Does the company report about take-back procedures and recycling programs to reduce the potential risks of products entering the environment? | Does the company report on initiatives to reduce, reuse, substitute or phase out toxic chemicals or substances? |
| VOC emissions reduction | Product impact minimization | |
| Does the company report on initiatives to reduce, substitute or phase out volatile organic compounds (VOC)? | Does the company report about take-back procedures and recycling programmes to reduce the potential risks of products entering the environment or does the company report about product features or services that will promote responsible and environmentally preferable use? | |
| Particulate matter emissions reduction | GRI sustainability certifications | |
| Does the company report on initiatives to reduce, substitute or phase out particulate matter less than 10 microns in diameter (PM10)? | GRI standards represent guidelines for creating sustainable performance reports. They consist of a modular and interdependent structure to best create reports in the economic, social and environmental areas | |
| Waste reduction initiatives | ||
| Does the company report on initiatives to recycle, reduce, reuse, substitute, treat or phase out total waste? | ||
| E-waste reduction | ||
| Does the company report on initiatives to recycle, reduce, reuse, substitute, treat or phase out e-waste? | ||
Source(s): Developed by authors
The independent variable, i.e. BGD (BoGenDiv) was measured as the percentage of women on the board.
To analyze the moderating effect of ESG controversies (ESGContr) on the relationship between BGD and CE-related disclosure, we included in our regression model an interaction term obtained by multiplying BoGenDiv with ESGContr.
The moderating variable ESGContr was measured using the ESG Controversy Category Score computed by Refinitiv Eikon based on 23 ESG controversy topics. This score measures the company’s exposure to ESG controversies, and the negative events exhibited in the media. The score ranges from 0 to 100. Companies with no controversies obtain a score of 100. Considering that we focused on the level of exposure to ESG controversies, we computed the moderating variable using the reverse version, that is, 100 minus ESG controversies, so that higher values consistently reflect higher ESG controversies. This inversion enhances interpretability and theoretical consistency, as it aligns the direction of the variable with the expectation that higher ESG controversies acts as a moderating factor in the relationship between BGD- and CE-related disclosures.
Finally, to avoid biased results and resolve endogeneity issues related to omitted variables, we included the following control variables:
Profit (Profit): return on total assets;
Leverage (Lev): long-term debt divided by total assets;
Firm Size (FirmSize): natural logarithm of total assets;
Board Tenure (BoTen): average number of years each board member has been on the board;
Board Meetings (BoMeet): number of board meetings during the year;
Board Size (BoSize): total number of board members at the end of the fiscal year; and
Board Independence (BoInd): Percentage of independent directors on the board.
We used one lag period for all the explanatory variables in regression to avoid endogeneity problems.
5.3 Regression analysis
To test the above hypotheses, we developed the following fixed-effect linear regression models for panel data (time-fixed effects) to control heterogeneity problems:
6. Empirical results and discussion
6.1 Descriptive statistics
Table 5 shows the descriptive statistics of all the variables included in the analysis.
Descriptive statistics
| Variables | No. obs | Mean | Sd | Min | Max |
|---|---|---|---|---|---|
| CEDisc | 3,761 | 10.97527 | 0.8691981 | 2 | 12 |
| BoGenDiv | 3,761 | 29.12345 | 14.61531 | 0 | 75 |
| ESGContr | 3,761 | 9.830374 | 23.20216 | 0 | 99.32433 |
| BoGenDiv×ESGContr | 3,761 | 272.8161 | 730.4898 | 0 | 5948.617 |
| Profit | 3,761 | 0.0469432 | 0.06708 | −0.5353444 | 0.5572245 |
| Lev | 3,761 | 22.56063 | 1.848947 | 0 | 28.42058 |
| FirmSize | 3,761 | 0.2068527 | 0.1421267 | 0 | 0.8245783 |
| BoTen | 3,761 | 6.13407 | 3.229729 | 0 | 23.16667 |
| BoMeet | 3,761 | 9.403084 | 6.783995 | 0 | 92 |
| BoSize | 3,761 | 11.35974 | 4.448352 | 0 | 34 |
| BoInd | 3,761 | 52.15801 | 27.3296 | 0 | 100 |
Note(s): Dependent variable. CEDisc: Unweighted sum of 12 environmental items disclosed by companies, assigning value 1 when the company does present information on that environmental aspect and 0 otherwise; Independent variable. BoGenDiv: Percentage of female on the board; Moderating variable. ESGContr: 100 minus the ESG Controversy Score, that is calculated based on 23 ESG controversy topics; Interaction term. BoGenDiv * ESGContr: Two ways interaction term obtained by multiplying the BoGenDiv and ESGContr; Control variables. Profit: Return on assets; Lev: Long-term debt divided by total assets; FirmSize: Natural logarithm of total assets; BoTen: Average number of years each board member has been on the board; BoMeet: Number of board meetings during the year; BoSize: Total number of board members at the end of the fiscal year; BoInd: Percentage of independent directors on the board
Source(s): Developed by authors
On average, the companies in the sample issued almost 11 out of 12 environmental items related to CE issues. Differently from previous studies that conducted a content analysis on sustainability reports (Opferkuch et al., 2021; Roberts et al., 2022; Kuo and Chang, 2021; Dagiliene et al., 2020; Stewart and Niero, 2018), this finding shows high levels of attentions toward CE issues, revealing the relevance to study the CE disclosure using different methods.
The percentage of women sitting on the board is 29.12%, highlighting a limited presence of female directors. The results also reveal that the independent variable varies considerably among the sampled companies, ranging from 0% to 75%.
The ESG controversies are on average equal to 9.83, showing that companies present low levels of exposure to ESG controversies and negative events exhibited in the media. However, this moderating variable varies considerably among the sample companies, ranging from 0 to 99.32.
Passing to the control variables, the sampled companies present an average ROA of 4%, a mean leverage of 22.56 and an average value of the natural logarithm of total assets of 0.20.
The number of years each board member has been on the board is on average equal to 6.13. The average number of board meetings during the year is equal to 9.40, including a limited number of companies that registered 0 as the minimum value, revealing an absence of yearly meetings. The total number of board members at the end of the fiscal year is on average equal to 11.35. On average, 52.15% of board members are independent directors. This control variable varies from 0% to 100%, revealing that there are both companies with no independent directors and companies with all independent board members.
Before performing the regression analysis, we analyzed the correlations between variables (Table 6). BoGenDiv is positively correlated with CEDisc. In addition, almost all the control variables are correlated with CEDisc. More specifically, Profit, Lev, BoTen, BoSize and BoInd are correlated with CEDisc.
Correlation matrix
| CEDisc | BoGenDiv | ESGContr | BoGenDiv × ESGContr | Profit | Lev | FirmSize | BoTen | BoMeet | BoSize | BoInd | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| CEDisc | 1,000 | ||||||||||
| BoGenDiv | 0.0695*** | 1,000 | |||||||||
| ESGContr | 0.0173 | −0.0398** | 1,000 | ||||||||
| BoGenDiv × ESGContr | 0.0188 | 0.1543*** | 0.8777*** | 1,000 | |||||||
| Profit | 0.0574*** | 0.0521*** | −0.0867*** | −0.0610*** | 1,000 | ||||||
| Lev | −0.1142*** | −0.1156*** | 0.3811*** | 0.3224*** | −0.2355*** | 1,000 | |||||
| FirmSize | −0.0125 | 0.0650*** | −0.0257 | −0.0137 | −0.1847*** | −0.0401** | 1,000 | ||||
| BoTen | 0.0423*** | 0.0156 | −0.0093 | −0.0093 | 0.0798*** | −0.0087 | −0.1184*** | 1,000 | |||
| BoMeet | 0.0016 | 0.0231 | 0.0020 | 0.0049 | −0.0848*** | 0.0512** | 0.0603*** | −0.1295*** | 1,000 | ||
| BoSize | −0.0893*** | −0.1783*** | 0.2984*** | 0.2242*** | −0.1459*** | 0.5590*** | −0.0222 | −0.0053 | −0.0535*** | 1,000 | |
| BoInd | 0.0900*** | 0.1788*** | −0.0276** | 0.0221 | 0.0107 | −0.0497*** | 0.0565*** | −0.0131 | 0.1819*** | −0.2956*** | 1,000 |
Note(s): Dependent variable. CEDisc: Unweighted sum of 12 environmental items disclosed by companies, assigning value 1 when the company does present information on that environmental aspect and 0 otherwise; Independent variable. BoGenDiv: Percentage of female on the board; Moderating variable. ESGContr: 100 minus the ESG Controversy Score, that is calculated based on 23 ESG controversy topics; Interaction term. BoGenDiv * ESGContr: Two ways interaction term obtained by multiplying the BoGenDiv and ESGContr; Control variables. Profit: Return on assets; Lev: Long-term debt divided by total assets; FirmSize: Natural logarithm of total assets; BoTen: Average number of years each board member has been on the board; BoMeet: Number of board meetings during the year; BoSize: Total number of board members at the end of the fiscal year; BoInd: Percentage of independent directors on the board. ***p < 0.01, **p < 0.05, *p < 0.1
Source(s): Developed by authors
6.2 Regression results
Table 7 shows the results of the regression analyses conducted using Stata Software. More specifically, the table reports the findings for the tests of H1 and H2. Model 1 shows the findings for the test of the direct effect of BGD and CE-related disclosure. Model 2, instead, shows the findings for the test of the moderating influence of ESG controversies on the relationship between BGD and CE-related disclosure.
Regression results (dependent variable: CEDisc)
| Circular economy disclosure | ||
|---|---|---|
| Variables | Model (1) Direct effect | Model (2) Interaction effect |
| BoGenDiv | 0.0048129 *** (0.0009627) | 0.0036863 *** (0.0010389) |
| ESGContr | −0.0007388 (0.0005663) | −0.0034952 *** (0.0011157) |
| BoGenDiv × ESGContr | 0.0000965 ** (0.0000337) | |
| Profit | −0.9914934 *** (0.236681) | −0.987996 *** (0.2364232) |
| Lev | 0.0888998 (0.0184834) | 0.0885104 *** (0.0184635) |
| FirmSize | 0.0303672 (0.1129386) | 0.0198483 (0.1128738) |
| BoTen | 0.0380269 *** (0.0071799) | 0.0395477 *** (0.0071916) |
| BoMeet | 0.0018116 (0.0024718) | 0.0018628 (0.0024691) |
| BoSize | −0.0026251 (0.0062877) | −0.0034293 (0.0062871) |
| BoInd | 0.0011901 (0.0006732) | 0.0009805 (0.0006765) |
| Year | YES | YES |
| Constant | 8.594448 *** (0.4103221) | 8.649073 *** (0.4103124) |
| Sigma_u | 0.71101916 | 0.71083371 |
| Sigma_e | 0.59241627 | 0.59176306 |
| Rho | 0.59024533 | 0.59065274 |
Note(s): Dependent variable. CEDisc: Unweighted sum of 12 environmental items disclosed by companies, assigning value 1 when the company does present information on that environmental aspect and 0 otherwise; Independent variable. BoGenDiv: Percentage of female on the board; Moderating variable. ESGContr: 100 minus the ESG Controversy Score, that is calculated based on 23 ESG controversy topics; Interaction term. BoGenDiv * ESGContr: Two ways interaction term obtained by multiplying the BoGenDiv and ESGContr; Control variables. Profit: Return on Assets; Lev: Long-term debt divided by total assets; FirmSize: Natural logarithm of total assets; BoTen: Average number of years each board member has been on the board; BoMeet: Number of board meetings during the year; BoSize: Total number of board members at the end of the fiscal year; BoInd: Percentage of independent directors on the board. 485 companies (3,761 firm-year observations). Significance level: ***p < 0.01, **p < 0.05, *p < 0.1. Standard errors in parentheses
Source(s): Developed by authors
We used a panel data fixed effect with a time effects model that provides findings more robust than other models (Eibinger et al., 2024), such as pooled ordinary least squares or random effects models.
The results reported in Model 2 confirm H1, showing a positive and significant relationship between BoGenDiv and CEDisc. More specifically, the coefficient of BoGenDiv is positive and statistically significant at better than the 1% level for explaining variations in CEDisc (β = 0.0036863, p < 0.01). This result is consistent with the theoretical argumentations of the agency and stakeholder theories, as well as with the findings of most previous studies that investigated the relationship between BGD and the wider concept of environmental disclosure, highlighting that female directors within the board enhance the board effectiveness, by better controlling the management activities and favoring the commitment toward sustainability issues, and increase the level of environmental disclosure released by companies to external parties (e.g. Injeni et al., 2022; Baalouch et al., 2019; Fuente et al., 2017; Tamimi and Sebastianelli, 2017; Javaid Lone et al., 2016; Liao et al., 2015; Garcia-Sanchez et al., 2014; Ntim and Soobaroyen, 2013; Rao and Tilt, 2016).
The results reported in Model 2 confirm H1 and are in line with the results reported in Model 1.
The results reported in Model 2 also confirm H2, showing that ESG controversies positively moderate the association between BGD and CEDisc. According to H2, the coefficient of BoGenDiv*ESGContr is positive and statistically significant at less than the 5% level for explaining variations in CEDisc (β = 0.0000965, p < 0.05), meaning that in the occurrence of ESG controversies, female directors tend to increase the CE-related disclosure released. This result confirms the theoretical argumentation, according to which women directors pay great attention to ESG controversies (Shakil et al., 2021), and tend to reveal more information in occurrence of higher levels of ESG controversies to contain negative effects.
The study’s findings overall highlight the need for EU policymakers and governments to pay attention to both the transition toward CE principles by companies and the composition of their board of directors, pursuing reforms that jointly consider these aspects.
With respect to the control variables, the coefficient of Lev is positive and statistically significant in both models, meaning that companies with higher levels of leverage should disclose more CE information to meet the data requirements of their debtholders (Pozzoli et al., 2023; Chan et al., 2013).
In addition, the coefficient of BoTen is positive and statistically significant in both models, meaning that the average number of years each board member has been on the board positively influences the level of CE-related disclosure released. Finally, the coefficient of Profit is negative and statistically significant in both models, meaning that companies with higher levels of profitability tend to disclose less CE information to external parties. The coefficients of the other control variables are not statistically significant.
In addition, in the Table 7, we reported the values of Sigma_u, Sigma_e and Rho to provide useful information for understanding data properties and model behaviour (Stock and Watson, 2005).
Sigma_u is equal to 0.59, represents a measure of variance due to temporal fixed effects. Sigma_e is equal to 0.71, represents the variance of idiosyncratic errors (eit) which capture the variability of observations that is explained neither by time effects nor by the explanatory variables included in the model. Finally, rho represents the proportion of the total variance explained by time fixed effects.
The rho value is 0.59 close to 1, so time effects explain much of the variance in the data.
6.3 Robustness and additional analysis
To demonstrate the robustness of our findings to alternative estimations, we constructed the ordinary least squares regression model, as well as random and fixed effects regression models without time dummies, to test both the direct effects and the interaction effects. The results are substantially similar to those ones obtained with the fixed effects with time effects model, confirming the validity of the findings in the present analysis and demonstrating their robustness. We also estimated the fixed effects with time effects model using no lagged explanatory variables and obtained similar results to those achieved with the main models.
7. Summary and conclusion
The findings of this paper show that the presence of female directors on the board is associated with higher levels of CE-related disclosure. For their characteristics, such as being supportive, respectful, kind and empathetic (Katmon et al., 2019), women pay more attention to sustainability matters (Hussain et al., 2018; Matsa and Miller, 2013), bringing companies to release higher levels of CE-related disclosure. Therefore, although the percentage of female directors within the board is limited in the sampled companies, their presence positively influences the level of CE-related disclosure. This result is consistent with previous studies revealing a positive relationship between BGD and the wider concept of environmental information (e.g. Injeni et al., 2022; Baalouch et al., 2019; Fuente et al., 2017; Tamimi and Sebastianelli, 2017; Javaid Lone et al., 2016; Liao et al., 2015; Garcia-Sanchez et al., 2014; Rao and Tilt, 2016). Considering stakeholder theory, the results of the paper show that female directors are more stakeholder oriented (Injeni et al., 2022), leading companies to provide higher CE-related disclosure to satisfy external stakeholders’ information needs. In addition, considering agency theory, the findings of the paper show that more diversified boards, in terms of gender diversity, favor better dialog with stakeholders, decreasing information asymmetries with the company.
The results of the paper also show that ESG controversies positively moderate the association between BGD and CE-related disclosure. This finding confirms the hypothesis developed, which stated that, in the occurrence of ESG controversies, female directors tend to enhance the level of CE-related disclosure to contain reputational damage. In line with our expectations, the results reveal that, when companies are exposed to higher ESG controversies, female directors favor the release of higher levels of CE-related disclosures. This means that the occurrence of ESG controversies enhances the effectiveness of BGD in stimulating higher levels of disclosures.
The results have important practical implications for regulators and companies. It has been emphasized that gender diversity has an important impact on the level of CE-related disclosure as well. This should lead regulators, and policymakers as well, to orient their acts toward mandatory diversity in the composition of the boards. Diversity can contribute to considering different perspectives, and female directors seem to enhance the level of disclosure in sensitive themes, such as CE, that can be particularly appreciated by stakeholders, supporting the creation of value in the medium and long term. Similar motivations should lead companies to increase their number of female directors on their boards, as this paper highlights that sensitivity to critical issues can be increased when the board is diversified. Therefore, the findings of the study highlight the necessity, for both companies and policymakers, to pay attention to both the disclosure on CE issues and the corporate governance in fostering sustainable development.
There are implications for academia as well. The positive relation between BGD and CE-related disclosure highlight the need to further investigate the association between corporate governance and sustainability disclosure, especially in the light of the effects of the implementation of CSRD and ESRSs. The mandatory adoption of ESRSs in the future could stress the changes produced in this context, as other research could distinguish the level of information in relation to the applied technical standards. Moreover, the use of data collected from Refinitiv to measure CE-related disclosure suggests an objective way for evaluating the information released, which could serve as a useful tool for scholars in future research.
The research presents some limitations as well. The applied model takes into consideration the level of disclosure, without providing an evaluation of the quality of the disclosure. Future research could add a qualitative analysis to the quantitative analysis. In addition, the Refinitiv database addresses only the output related to the presence of CE-related disclosures, without taking into consideration the effort produced to achieve that information; this level of information should also require a qualitative investigation. In this context, a related limitation is due to the fact that Refinitiv cannot identify if a company did not report a CEDisc because the information is not material or because this company did not appropriately address the CE issue. Another limitation is related to the adopted sample. It would be interesting to separately examine countries where there is no legislation and the decision on the composition of the board is free from the adoption of ad hoc requirements. Finally, in the future is interesting to examine the different type of diversity not only gender diversity. It would be interesting to examine interactions with diversity to find new evidence in other fields.
Funding: The research was funded by the following source: research project “Business models, sustainability and circular economy: the role of corporate governance” at the University of Naples Parthenope. Project funded by the Ministry of University and Research with Ministerial Decree 10 August 2021, no. 1061.
Conflicts of interest: The authors declare no conflict of interest.
