Sustainability disclosure and the relationship with board characteristics and assurance

Abstract

This study aims to analyze the influence of internal factors within organizations, such as board characteristics and the practice of external assurance, on the quality of sustainability reports of Brazilian companies. As a gap, to examine the level of compliance of Brazilian companies' sustainability reports with the methodology suggested by IR. The objective is to assess the effects of the external assurance variable on the compliance of sustainability information. The relevance of this study is to assess entities' commitment to socio-environmental responsibility, transparency, and accountability to stakeholders, while fostering sustainable practices and enhancing corporate image nationally. This study provides a discussion on the current state of board compositions, highlighting the high number of directors and the low proportion of women, despite the significant positive influence women have on sustainability disclosure. The results related to the "External Assurance" variable may encourage more organizations to adopt this practice, with the expectation of demonstrating greater alignment with suggested methodologies. An adaptation of the method by Sun et al. (2022) was used, analyzing report quality across nine stakeholder-related topics. An unbalanced panel data analysis was conducted, using random effects estimated by GLS, on a sample of 357 reports from 49 Brazilian companies between 2014 and 2022. The variable "Presence of Women on Boards (PWOB)" was found to be significant in relation to the quality of sustainability disclosures, indicating that diversity contributes to greater prestige and attention to social issues. The variables "Board Committees (BC)" and "External Assurance (EA)" showed a positive relationship, supporting the board's ability to respond to different pressures and reinforcing transparent and reliable approaches in communication with stakeholders.

Keywords:
sustainability; disclosure; boards; assurance; stakeholders

Resumo

O objetivo deste artigo é analisar a influência de fatores internos às entidades, como características dos conselhos de administração e a prática de asseguração externa sobre a qualidade dos relatórios de sustentabilidade das empresas brasileiras. Como lacuna, a observação do nível de conformidade dos relatórios de sustentabilidade das empresas brasileiras em relação com a metodologia sugerida pelo RI. Espera-se verificar os efeitos da variável de asseguração externa sobre a conformidade das informações em sustentabilidade. A relevância do estudo está em avaliar o compromisso das entidades com a responsabilidade socioambiental, transparência e prestação de contas aos stakeholders, além de impulsionar práticas sustentáveis e aprimorar a imagem corporativa no cenário nacional. Este estudo permite discutir o estado atual das composições dos conselhos, relacionando-se ao número elevado de conselheiros e a baixa proporção de mulheres em sua composição, mesmo demonstrando influência positiva significativa sobre a divulgação em sustentabilidade. Os resultados da variável Parecer de Asseguração podem permitir que mais entidades adotem a prática na expectativa de demonstrar maior alinhamento com as metodologias sugeridas. Utilizou-se uma adaptação do método de Sun et al. (2022), que consiste em analisar a qualidade dos relatórios em nove diferentes tópicos relacionados às partes interessadas. Foi realizada uma análise de dados em um painel não balanceado, por meio de efeitos aleatórios estimados por MQG, em uma amostra de 357 relatórios de 49 empresas brasileiras entre 2014 e 2022. A variável “Presença da Mulher nos Conselhos de Administração (CA)” demonstrou-se significativa em relação a qualidade da divulgação em sustentabilidade indicando que a diversidade permite uma maior prestígio e atenção às questões sociais. As variáveis “Número de Comitês Subordinados ao CA” e “Parecer de Asseguração” apresentaram relação positiva e apoiam a capacidade do conselho em responder à diferentes pressões além de reforçar abordagens transparentes e confiáveis na comunicação com as partes interessadas.

Palavras-chave:
sustentabilidade; divulgação; conselhos; asseguração; stakeholders

1. Introduction

The framework of the Integrated Report (IR), published by the International Integrated Reporting Council (IIRC) in 2013, helped organizations identify their material aspects, highlighting the factors that influence the organization's ability to generate value for its stakeholders (Cooray et al., 2020). Stakeholders then issue expectations, and responding to them has become a frequent practice in the process of building reports, with communication capacity being a relevant factor in meeting the needs and demands of target audiences (Tiron-Tudor et al., 2020).

To meet these expectations, corporate governance codes around the world establish the role of the board of directors in making decisions regarding disclosure, aiming to address broader issues such as social and environmental matters (Ramdhony et al., 2023). This premise is based on the perspective that boards, as part of the governance structure, are responsible for protecting the interests of different groups, guiding and leading the organization to prioritize decision-making and improve its perceived value (Qaderi et al., 2022).

The effectiveness of interest protection is often assessed by researchers based on the characteristics of boards, such as board size, director independence, and female participation in the composition (Sun et al., 2022; Qaderi et al., 2022; Tiron-Tudor et al., 2020). These characteristics are considered determining factors in enhancing the disclosure process, as it is influenced by the choices, motives, and values of those involved in the formulation and decision-making of organizations (Cooray et al., 2020; Cucari et al., 2018).

However, board characteristics are not the only internal factors signaling commitment to stakeholders. To ensure balance and the absence of significant errors in the production of report information, organizations seek external assurance, typically through auditing firms, in accordance with the IR framework's guidelines (Sun et al., 2022). The underlying premise is that the level of standard applicability and assurance increases the company's credibility with stakeholders, influencing corporate reputation (Odriozola & Baraibar-Diez, 2017). Additionally, García-Sánchez et al. (2019) found that external assurance of reports, often provided by auditing firms, enables companies to face fewer difficulties in attracting interested investors and presents a complementary determinant, such as reduced capital restrictions for companies with higher quality disclosures.

The level of action disclosure regarding responsibility and sustainability can vary significantly across countries, influenced by factors such as customs, regulations, and internal legislation (Freitas et al., 2019). Hence, there is a gap in assessing the level of compliance of Brazilian companies' sustainability reports in relation to the methodology suggested by the IR framework. This study aims to observe the effects of external assurance on this compliance to verify changes in disclosure outcomes when using a reliability tool.

Using the report evaluation method suggested by Sun et al. (2022), this study aims to analyze the influence of internal factors within organizations, such as the characteristics of the boards of directors and the practice of external assurance, on the quality of Brazilian companies' sustainability reports. The goal is to understand how organizations respond to stakeholder demands. If the top management of an organization fails to meet the needs of its various stakeholder groups, there is a risk of losing their support, which can lead to a reduction in the organization’s legitimacy (Henriques et al., 2022).

The theoretical foundation of this research is based on the principles of Legitimacy theory, an approach that considers society's expectations in order to obtain legitimacy for its operations (Deegan, 2002), and aspects from the Stakeholder theory (Deegan & Blomquist, 2006). Given these perspectives, the guiding question of this study is: What are the relationships between board characteristics, disclosure strategies, and the level of sustainability disclosure quality in Brazilian organizations?

The results provide evidence of the importance of the presence of women on boards of directors, indicating their role in improving companies' responses to socio-environmental issues, as well as the use of external assurance to reinforce the organization's commitment to providing cohesive information to its stakeholders. Additionally, the results related to the variables Board size and Proportion of women diverge from previous research (Qaderi et al., 2022; Tiron-Tudor et al., 2022; Cooray et al., 2020). These findings highlight the complexity of the relationships among these variables and suggest that regulatory changes in the environments where the research is conducted may influence these relationships. The need to consider these regulatory aspects when evaluating companies' sustainability disclosures is emphasized, providing important insights for investors and regulators.

The subsequent sections of the article are divided into a theoretical review, followed by the section developing the research hypotheses. The fourth section outlines the sample and methodology used for data collection. The fifth section presents the descriptive analysis of the data, along with regression analysis and a discussion of the results. Finally, conclusions, limitations, and prospects for future research are presented.

2 Quality of reports, legitimacy and the expectations of stakeholders

The main objective of the legitimacy theory is to explain the behavior of entities about information disclosure, to meet the demands and society expectations. This theory is based on the premise that organizations seek to increase the intersection area among social expectations and their actions, in order to obtain legitimacy and acceptance from the several stakeholder groups with which they relate (Deegan, 2002).

One of the strategies adopted by organizations to maintain legitimacy lies in the disclosure process. They seek to communicate their practices and results transparently, providing relevant information about their economic performance, as well as their social and environmental performance. From the perspective of legitimacy, it can be seen that by meeting the disclosure requirements demanded by society, entities are able to increase their acceptance and reduce the risk of social sanctions, applied by their stakeholder groups when they perceive that the entity is not in line with their prerogatives. Thus, there is an increase in the amount of disclosure and, consequently, new information is accessed by society (Melloni et al., 2017). However, an increase in the quantity of disclosure does not necessarily reflect an increase in the perceived quality of the information. The debate on the definition and empirical separation between quantity and quality of disclosure is complex and there is still no consensus in the literature (Hummel & Schlick, 2016). The lack of clear criteria for assessing the quality of disclosure has led to criticism from investors and financial analysts, who point to the overload of information without a corresponding increase in usefulness for users (Melloni et al., 2017).

Different stakeholders may have divergent perceptions of what is considered “good” or “bad” in terms of disclosure, characterizing the quality of disclosure as a subjective concept and dependent on the context analyzed, and should consider the needs and expectations of different stakeholder groups. This scenario calls for a comprehensive descriptive model to assess disclosure quality, which covers a broad spectrum of data, including the voluntary nature of sustainability information (Helfaya & Whittington, 2019).

Although legitimacy theory is used in this study to support the possible relationships among disclosure and the organization's characteristics, it has limitations in terms of its singular view of society, ignoring disputes and inequalities within the system (Arrington & Puxty, 1991). Knowing that some parts of society may have greater influence in demanding their demands, this study also considers stakeholder theory. Deegan and Blomquist (2006) state that, although implicit in legitimacy theory, stakeholder theory addresses issues of stakeholder power and how a stakeholder's relative power affects their ability to coerce the organization to meet their expectations.

It is essential to obtain the approval and consent of stakeholders, since they are responsible for providing the resources and contributions necessary for business activities, thus contributing to the legitimacy of the entity. Thus, the quality of disclosure is one of the ways to signal that stakeholders' expectations have been met and that the organization is aligned with their demands. However, these expectations can vary, and what is considered relevant by one stakeholder can be seen as irrelevant, or even conflicting, by another (Helfaya & Whittington, 2019). In addition, they can evolve over time, reflecting changes in social, environmental, and economic dynamics.

Due to its complexity, studies have focused on measuring the quality of the information disclosed. While some studies have used aggregate indices for measurement, there has been progress in the discussion about the construction of reports, seeking a more in-depth analysis at the compliance level (Monteiro et al., 2020; Chouabi et al., 2022). Evaluation by aggregate indices can lead to biased responses, since companies can direct their attention unevenly to the different disclosure categories, aiming only to obtain favorable numerical results. Therefore, the qualitative analysis of compliance becomes important in order to more accurately understand the quality of disclosure (Omran et al., 2021).

3 Research hypotheses

Taking care of the relationship with stakeholders becomes a relevant issue in relation to the use of the communication process, represented by the strategy adopted by the entity. Thus, the board of directors plays a notable role in monitoring the activities of an entity, facilitating the management communication process and enabling advances in its disclosures to its stakeholders (Tiron-Tudor et al., 2020). The recognition that the characteristics of boards of directors have an influence on corporate governance is one of the most significant factors in the analysis of these practices (Cucari et al., 2018). This finding suggests the composition of boards, including the multiplicity of directors' experiences and skills, represented by the number of members, is directly related to companies' ability to respond to demands for transparency and disclosure in sustainability (Qaderi et al., 2022).

By having a board made up of individuals with diverse backgrounds, the organization is more likely to have a comprehensive understanding of the challenges and opportunities related to sustainability, as well as being able to offer a wider range of perspectives when making strategic decisions (Cooray et al., 2020). This variety of viewpoints and skills can result in a more dynamic and holistic approach to reporting on sustainable practices.

Although the number of board members alone does not guarantee full coverage of variability, it can be correlated with a multiplicity of interests and skills which, when well-managed, contribute to more effective governance and fuller disclosure of the company's sustainable practices. It is in this context that the first hypothesis of this research arises:

H1: Board size has a positive influence on the level of disclosure quality in sustainability reports.

In general terms, the greater independence of boards of directors is another aspect that is related to the quality of the information disclosed (Holtz & Sarlo Neto, 2014). These directors are concerned about their reputation and tend to be more involved in fulfilling the company's social responsibilities, while at the same time seeking to preserve their status (Ramdohony et al., 2023).

In theory, a board with a high proportion of independent directors will have a greater propensity to better control activities related to board functions, contributing to an increase in the transparency level in company disclosure (Sun et al., 2022). Independent directors have an enhanced ability to manage and analyze a company's operations, offering additional assurance to investors concerned with protecting their interests (Tiron-Tudor et al., 2020). The independence of directors acts in favor of increasing the information disclosed, facilitating the implementation of sustainable strategies, as well as influencing the degree of disclosure quality, promoting reports that meet the interests of different parties (Cucari et al., 2018). The second research hypothesis is raised:

H2: A higher proportion of independent directors on the board of directors has a positive influence on the level of disclosure quality in sustainability reports.

Another observable characteristic of boards of directors is the presence of women on them. Women show a greater degree of sensitivity in building relationships with stakeholders, as well as being more concerned about disclosure and sustainable development (Sun et al., 2022). Therefore, a higher proportion of women on boards of directors tends to result in higher quality voluntary disclosures. Since women encourage discussion, information sharing tends to increase, reducing the level of conflict in the board's decision-making process (Vitolla et al., 2020).

Although the presence of women points to an improvement in sustainability disclosure, it is still possible to observe mixed results in the literature. In their tests, Cooray et al. (2020) found no significant relationship between female participation and the quality of disclosure. The researchers infer that this finding may reflect the low proportion of women among the board members of the companies analyzed. On the other hand, the results of Qaderi et al. (2022) show a positive and significant relationship. Although they also highlight the low proportion of women among directors, the presence of women indicates a significant improvement in the quality of sustainability report disclosure.

Looking at the literature, it is possible to infer that women tend to influence the effectiveness of board monitoring, improving the decision-making process and helping to develop more responsive corporate policies (Ramdohony et al., 2023). With regard to the presence of women on boards of directors, it is possible to raise the following hypothesis:

H3: Greater female proportionality on boards of directors has a positive influence on the level of disclosure quality in sustainability reports.

The promotion of structured and targeted governance inspires companies to be more transparent, addressing different spheres relevant to stakeholders and emphasizing the quality of disclosure directed at them (Cooray et al., 2020). The increase in the number of independent committees allows for this coverage of the several spheres. The presence of an audit committee shows a positive association not only with financial information, but also with the quality of the information disclosed in sustainability reports (Haji & Anifowose, 2016). In addition to the audit committee, Cooray et al. (2020) point to a positive relationship in the existence of a separate risk management committee, indicating that the audit committee alone may not have the necessary expertise to deal with all the risks related to the entities.

Furthermore, Dwekat et al. (2022) point out that the existence of committees focused on Corporate Social Responsibility (CSR) positively influences sustainability disclosure, through the monitoring of policies, CSR performance, and better adoption of sustainability standards. From this perspective, a greater distribution of responsibilities, through a greater volume of committees, can allow them to act more precisely and with greater compliance in the disclosure of information aimed at sustainability (Sun et al., 2022). About the discussion, it is possible to raise the fifth hypothesis of this research:

H4: The number of committees subordinate to the board of directors has a positive influence on the level of quality of disclosure in sustainability reports.

As a corporate initiative, distinct from the characteristics of boards, the assurance of reports seeks to provide stakeholders with assurance on how an organization perceives its sustainability performance and demonstrates it through assured disclosure. However, unlike financial auditing, the objective of the assurance report is not to ensure full reliability of the information provided in the reports (Boiral et al., 2019). Even so, the practice aims to provide stakeholders with assurance on how the organization manages its sustainability performance and how it communicates this factor in its reports (Boiral et al., 2019).

On the other hand, the participation of entities in different aspects of sustainability can influence stakeholders to demand greater reliability of the information disclosed by entities, increasing the number of companies that adopt this strategy (Qaderi et al., 2022). In some cases, the external assurance of reports, demonstrated by assurance opinions, is used as a proxy for the composition of the report evaluation method, strengthening the perception of its positive relationship with the quality of sustainability disclosure (Qaderi et al., 2022; Cooray et al., 2020). Observing the assurance potential to foster the quality of reports, is possible to raise the seventh and final hypothesis of this research:

H5: The external assurance present in the documents has a positive influence on the level of quality of disclosure in sustainability reports.

4 Methodology

This study used an unbalanced panel of data comprising 381 observations, collected from 2014 to 2022, based on the initial composition of 89 companies, which made up the 2022 Ibovespa portfolio, listed on “Brasil, Bolsa, Balcão” (B3).The expectation of the selection is to avoid collecting reports in the early stages of implementation, pointed out as a common practice in the Brazilian context by the work of Maria et al. (2021), in addition to reducing the comparison effect among different company sizes. The selection is also in line with the findings of Hamad et al. (2022), who point out that only large companies are encouraged to use the RI model, the basis for the evaluation method used in this research.

Only companies that published reports without interruption, starting from their first publications, were considered. In addition to it, only companies that published more than two reports during the research period were considered. The sample was reduced to 50 companies, with a total of 381 annual reports. When carrying out the robustness tests using the ESG (Environmental, Social and Governance) Score and ESG Combined Score variables from the Refintiv Eikon® Platform, the sample was reduced to 49 companies, with 357 annual reports, due to the availability of data.

Data on the quality of sustainability reports were collected manually from documents such as Sustainability Reports, Integrated Reports, Annual Reports and ESG Reports. For observation purposes, Table 1 shows the sample distributed across 11 industrial sectors.

Table 1
Distribution of samples by sector

Industry	Sample	Observation	Sample composition (%)
Industrials	5	37	10.36%
Non-Basic Consumption	5	40	11.20%
Consumer Staples	4	28	7.84%
Consumer Discretionary	4	24	6.72%
Miscellaneous	3	24	6.72%
Financials	6	48	13.45%
Mining	6	53	14.85%
Oil / Gas and Biofuels	3	24	6.72%
Health Services	4	15	4.20%
Communication Services	2	11	3.08%
Utilities	7	53	14.85%
Period (2014-2022)	49	357	100%

• Source: Prepared by the authors.

4.1 Econometric Modeling

In order to test the quality of sustainability reports in relation to the characteristics of the board of directors and disclosure strategies, a random effects panel for generalized least squares was used. The definitions of the variables used (Table 4) were taken from previous studies that sought to understand the relationships among the characteristics of boards of directors and disclosure strategies adopted by entities, and the level of quality present in the reports analyzed (Sun et al., 2022; Qaderi et al., 2022; D'Apice et al., 2021; Mion & Adaui, 2019; Boiral et al., 2019; Cooray et al., 2020). To carry out the tests, three control variables were inserted.

The control variables used (1) are Return on Assets (ROA), Natural Logarithm of Total Assets (Assets) and Natural Logarithm of Total Liabilities (Liabilities). The data for the control variables was taken from the Economatica Platform (2023). Due to the presence of collinearity problems among Assets and Liabilities variables, a factor analysis was carried out for both variables, replacing them with one factor (ALf), presented in the regression analyses. The main model of this research is presented below:

$${QSRIR}_{i,t}={\beta }_0+{\beta }_1{TC}_{i,t}+{\beta }_2{IC}_{i,t}+{\beta }_3{PWOB}_{i,t}+{\beta }_4{BC}_{i,t}+{\beta }_5{AP}_{i,t}+{\beta }_6{ROA}_{i,t}+{\beta }_7{ALf}_{i,t}+\sum {SECTOR}_{i,t}+\sum {YEAR}_{i,t}+{\epsilon }_{i,t}$$ (1)

where i denotes the company, where t the year, and E represents the error terms.

4.2 Dependent Variable

There are several important characteristics for studying disclosure, such as the topic covered, the breadth of issues and styles, the nature, scope, coverage, reference period and location (Pistoni et al., 2018). However, in order to assess the quality of disclosure, it is essential to consider the content in its entirety (Unerman, 2000). For this assessment, a methodology is used that involves collecting and coding qualitative and quantitative information contained in non-financial annual reports, establishing specific categories to carry out the analysis (Di Vaio et al., 2022).

Based on this perspective, the measurement of the dependent variable is based on the model adapted and proposed by Sun et al. (2022). The method establishes a disclosure index made up of nine themes related to sustainability: Health and Safety (HS), Public Relations and Social Welfare (PRSW), Environmental and Sustainable Development (ESD), Protection of Shareholders' Rights (PSR), Protection of Customers' Rights (PCR), Protection of Workers' Rights (PWR), Protection of Debtors' Rights (PDR), Protection of Suppliers' Rights (PSR) and Measures to Develop and Improve the Social Responsibility System (MDISRS). These nine themes can be used to understand how organizations are responding to stakeholder expectations.

The construction of the evaluation method is directly related to the Integrated Reporting Conceptual Framework, according to its latest version, published in 2021. The proposal follows the basic principles (Table 3) of IR, set out in item 3 of the Conceptual Framework (IIRC, 2021). So, the method evaluates these principles according to the entities and stakeholders that are considered by the Framework (Table 2).

Table 2
Topics assessed in sustainability reports (TASR)

Topics related to stakeholders	Maximum score per topic: 7
Health and Safety (HS)	Maximum: 7 / Minimum 0
Public Relations and Social Welfare (PRSW)	Maximum: 7 / Minimum 0
Environmental and Sustainable Development (ESD)	Maximum: 7 / Minimum 0
Protection of Shareholders' Rights (PSR)	Maximum: 7 / Minimum 0
Protection of Customers' Rights (PCR)	Maximum: 7 / Minimum 0
Protection of Workers' Rights (PWR)	Maximum: 7 / Minimum 0
Protection of Suppliers' Rights (PSR)	Maximum: 7 / Minimum 0
Protection of Debtors' Rights (PDR)	Maximum: 7 / Minimum 0
Development and Improvement of SR System (DISS)	Maximum: 7 / Minimum 0
Maximum score per report	General Maximum: 63

• Source: Adapted from Sun et al. (2022).

For this assessment, we assign a score of “1” each time it is noted that an item has been described using qualitative and quantitative non-monetary information simultaneously. Secondly, the monetary information (MI) present in the items evaluated will also has a score of “1”. Next, when information describing future expectations and orientations is perceived, as well as past observations (PO), it will receive a score of “1”. In relation to the integration of information, the items articulating with the strategy will have a score of “1” and, next, those with a link to governance will also have a score of “1”. Finally, when the organization demonstrates that there is a dialogue between the stakeholder to whom the information is addressed, a score of “1” will be awarded.”.

Table 3
Evaluation criteria for each topic

Measurement Criteria	Pontuação (Máximo: 7)
Qualitative and Quantitative Non-Monetary Information (QQNMI)	Present: 1; Absent: 0
Monetary Information (MI)	Present: 1; Absent: 0
Future Orientations (FO)	Present: 1; Absent: 0
Past Observations (PO)	Present: 1; Absent: 0
Articulation with Strategy (AS)	Present: 1; Absent: 0
Link to Governance (LG)	Present: 1; Absent: 0
Dialogue with Stakeholders (DWS)	Present: 1; Absent: 0

• Source: Adapted from Sun et al. (2022).

The result for each topic described in table 2 assumes a maximum score of 7, according to each criterion described in Table 3. The level of disclosure quality of the sustainability reports is equal to the sum of all the items assessed, so the highest score for each annual report is 63 (9x7). Therefore, each annual report will receive a score ranging from 0 to 63, as described in the calculation:

$$QSRIR=\frac{{\sum }_{i,t}^{9}TASR}{63}$$ (2)

where QSRIR is the dependent variable of the research, i is the company, t is the year, and TASR is the composition of the 9 topics of the sustainability disclosure index.

It is important to note that the model proposed by Sun et al. (2022) had an evaluation criterion called “Tone of Disclosure”, which when both positive and negative tones were perceived, a score of 1 was added. The total result of the model would be 72 (9x8). However, due to the particularity of measuring this criterion, it was removed from the sample, adapting the proposed model to the maximum score of 63, as described above (2)

4.3 Independent Variables

In order to capture the interaction among board characteristics and the quality level of sustainability reports (QSRIR), the hypotheses raised are expected to have a positive and significant relationship with disclosure. The variables described in Table 4 represent the characteristics of boards of directors and were measured in order to find a correlation with the QSRIR variable.

Table 4
Board characteristics and disclosure strategies

Variable	Description
Board size (BC)	Natural logarithm of the number of board members
Board Independence (BI)	Proportion of independent directors out of the total number of board members
Presence of Women on Boards (PWOB)	Ratio of women to the total number of directors
Board Committees (BC)	Number of committees reporting to the board, weighted from 0 to 1.
External Assurance (EA)	Dummy variable (1 for external audit firms; 0 for others)

• Source: Adapted from Sun et al. (2022), Qaderi et al. (2022), and D'Apice et al. (2021).

The variables BC, BI and PWOB were collected from the reference forms made available by the Brazilian Securities and Exchange Commission. BC and EA variables were collected manually by checking the body of the annual and sustainability reports. 5 Results

5.1 Descriptive Analysis and Correlation Matrix

Table 5 shows the descriptive statistics for all the variables used in the model. The quality level of Sustainability Reports and Integrated Reports showed a minimum value of 0.095 and a maximum value of 0.762. In 357 observations, the variable showed an average of 0.489. For the natural logarithm of board size, the maximum value was 2.773 and the minimum was 1.099, a variation of 1.674.

Table 5
Descriptive statistics of variables

Variable	Number	Average	Median	Standard deviation	Minimum	Maximum
QSRIR	357	0.489	0.492	0.101	0.095	0.762
BS	357	2.220	2.197	0.230	1.099	2.773
BI	357	0.920	1.000	0.133	0.333	1.000
PWOB	357	0.126	0.111	0.114	0.000	0.667
BC	357	0.375	0.417	0.192	0.000	1.000
EA	357	0.610	1.000	0.488	0.000	1.000
ROA	357	3.542	3.962	8.178	-68.598	24.310

• Source: Prepared by the authors.

The independence of directors had an average of 0.92, a maximum value of 1 and a minimum of 0. The average proportion of women on the board of directors was 0.126, with a minimum value of 0.000. The number of committees had an average of 0.367, a maximum of 1 and a minimum of 0. EA variable had a maximum of 1 and a minimum of 0, the presence and absence of the External Assurance in the reports, and an average presence of 0.610. The minimum value of ROA variable is -68.598 and the maximum is 24.310, with an average of 3.542.

Table 6 shows the correlation analysis of the variables. The results show that all the explanatory variables, except for Board Independence, have a significant positive relationship with the dependent variable.

Table 6
Correlation matrix among research variables

	QSRIR	BS	IC	PWOB	BC	EA	ROA	ALf
QSRIR	1
BS	0.126*	1
BI	-0.028	0.067	1
PWOB	0.219***	-0.030	0.119*	1
BC	0.164**	0.235***	-0.041	0.125*	1
EA	0.601***	0.174***	-0.034	0.197***	0.184***	1
ROA	0.024	-0.006	0.029	0.007	-0.141**	-0.015	1
ALf	0.363***	0.157**	-0.275***	0.069	0.135*	0.303***	-0.120*	1

• Note: Pearson's correlation coefficient was used. The indicators ***, ** and * indicate significance at 1%, 5% and 10%, respectively.
• QSRIR = Quality of Sustainability Report and Integrated Reporting; BS = Board Size; BI = Board Independence; PWOB = Presence of Women on Boards; BC = Board Committees; EA = External Assurance; ROA = Return on Assets; ALf = A Factor relating Assets and Liabilities variables.
• Source: Prepared by the authors.

5.2 Regression Analysis

In order to validate the models presented above, tests were carried out for normality of residuals, absence of heteroscedastic residuals, absence of autocorrelation of residuals and tests for multicollinearity of independent variables. Breusch-Pagan and Cook-Weisberg tests were carried out on QSRIR variable to verify the absence of heteroscedastic residues, accepting the null hypothesis that there is variance or dispersion in the data. First-order autocorrelation was detected at a significance level of 1% using the Wooldrige test. The test for the null hypothesis of normal distribution showed a p-value of 0.017.

When the fixed effects of years and sector effects were added, the results showed the presence of multicollinearity. Then, the fixed effects of these variables were removed to avoid biasing the estimators. In order to define the ideal panel, the Breusch-Pagan tests were carried out, which rejected the use of stacked panels and the Hausman tests showed a preference for the random effects model over fixed effects. So, the ideal model for running the research regressions would be the random effects panel estimated by generalized least squares (GLS).

Table 7 shows 4 test models, using dependent, independent and control research variables. Model 1 shows the basic research variables. Model 2 removed the control variables to analyze the impact of independent variables with adverse effects. Two models, 3 and 4, were inserted for robustness tests. Model 3 is in line with the proposal in the work by Degenhart et al. (2024), replacing QSRIR variable with ESG Score variable, which assesses the disclosure of entities through the average disclosure of governance, social information and environmental information, collected through Refinitiv Eikon® (LSEG) platform. On the same premise, Model 4 was added, using a different variable from the platform, ESG Combined Score, which combines ESG Score variable with controversial ESG information captured by different global media sources.

Table 7
Regression models

	QSRIR with Control variables (1)	QSRIR without Control variables (2)	ESG Score Model (3)	ESG Combined Score Model (4)	VIF
BS	0,143	0,251	0,210	0,245	1,118
	(-0,023)	(-0,022)	(-0,039)	(-0,045)	-
BI	0,940	0,948	0,292	0,477	1,126
	(-0,002)	(-0,003)	(-0,075)	(0,062)	-
PWOB	0,006***	0,019**	0,000***	0,021**	1,083
	(0,085)	(0,085)	(0,216)	(0,209)	-
BC	0,010**	0,017**	0,592	0,001***	1,125
	(0,059)	(0,06)	(0,022)	(-0,123)	-
EA	0,000***	0,000***	0,000***	0,000***	1,179
	(0,07)	(0,072)	(0,113)	(0,099)	-
ROA	0,246	-	0,244	0,041**	1,036
	(0,000)	-	(-0,001)	(-0,002)	-
ALf	0,235	-	0,973	0,612	1,239
	(0,000)	-	(0,000)	(0,000)	-
Constant	0,000***	0,000***	0,000***	0,000***	-
	(0,469)	(0,468)	(0,704)	(0,618)	-

• Note: ***, ** and * indicate significance at 1%, 5% and 10% respectively. Breusch-Pagan test: LM = 226.642 with p-value = 0.000** / Hausman test: H = 238.269 with p-value = 0.000**.
• QSRIR = Quality of Sustainability Report and Integrated Reporting; ESG = Environmental, Social and Governance; VIF = Variance Inflation Factor; EBS = Board Size; BI = Board Independence; PWOB = Presence of Women on Boards; BC = Board Committees; EA = External Assurance; ROA = Return on Assets; ALf = A Factor relating Assets and Liabilities variables.
• Source: Prepared by the authors.

5.3 Discussion of Results

In line with the results of Tiron-Tudor et al. (2020), BS variable does not significantly influence the quality of sustainability reports, contradicting hypothesis H1. This finding contradicts those of Qaderi et al. (2022), Sun et al. (2022) and Cooray et al. (2020), who pointed out that the size of the board plays a significant role in the level of disclosure in the reports. There may be more complex interactions among board size and other variables, such as board diversity, the presence of specific sustainability committees, or the level of shareholder involvement. These factors can act as moderators, influencing the strength and direction of the relationship among board size and the quality of sustainability reports, and the failure to incorporate them may have influenced the results found.

Regarding H2, theoretical perspectives indicate that the greater the level of independence present on the board of directors, the less likely it is that the directors will pursue their interests in different spheres of the entity, which possibly reduces the amount of information disclosed by the entity (Helfaya & Whittington, 2019). However, no significant results were observed in the relationship between independence and report quality. The results diverge from the findings of Qaderi et al. (2022) and Ong and Djaiadikerta (2020), which may be related to the first hypothesis of the research, with the size of the boards statistically influencing the behavior of the proportion between independent and dependent directors.

In line with Qaderi et al. (2022), PWOB variable showed a significant positive relationship, allowing Hypothesis 3 of the article to be accepted. This relationship is based on the perspective that the presence of women on boards of directors may indicate greater prestige for the company, offering greater attention to social issues. This finding is relevant due to the low presence of research on female participation on boards of directors in emerging economies, as Al-Nasrallah (2023) notes. He also points out that the UN's 2030 Agenda and internal government policies that benefit female empowerment make it easier to see the impact of women's participation in higher positions.

In both models 3 and 4, PWOB variable showed a positive relationship with the models' dependent variable. It is worth noting that in addition to the results of models 3 and 4, the results of models 1 and 2 for PWOB variable are also in line with the findings of Degenhart et al. (2024). They show a positive relationship with the presence of women on the boards of Brazilian companies and ESG disclosure score of the Refinitiv Eikon® Platform, validating the findings through content analysis.

The results for H4 showed significance at 5% in models 1 and 2, and at 15% in model 4. In line with the research by Qaderi et al. (2022), the findings indicate that a greater number of committees subordinate to the board of directors indicates a greater capacity of the entity to respond and act to different pressures. The result indicates that different committees within the board, such as audit committees, sustainability committees and corporate governance committees, have been associated with higher levels of disclosure in several areas, including disclosures in sustainability reports (Tiron-Tudor et al., 2020).

The result of H5, for EA variable, is in line with the findings of Miles and Righam (2020), Boiral et al. (2019), Farooq et al. (2020) and Sun et al. (2022). The use of External Assurance was the only variable to show a significance level of 1% in all models. This result is linked to the provisions of RI Conceptual Framework, which note that the reliability of information is affected by balance and the absence of significant errors, supporting the practice of assurance (IIRC, 2021). Besides this, the significant results, in all models, reinforce the importance of transparent and credible approaches in communicating with stakeholders. Assurance not only meets stakeholder expectations for transparency and accountability but also contributes to building and maintaining organizational legitimacy (Sun et al., 2022).

The control variables, ROA and ALfFactor, did not show significant results, with the exception of model 4, where ROA was significant at 5%. This finding can be explained by the fact that the sample was limited to companies in the Ibovespa portfolio, aligning the companies in a financially similar group. The results show that the quality of the information published in sustainability reports is independent of the size and profitability of the entities analyzed.

6 Conclusions

The characteristics of the board of directors, including its composition, experience and involvement in sustainability issues, emerge as possible factors in harmonizing corporate governance and disclosure in response to stakeholder expectations. This study aims to examine the influence of board characteristics, together with disclosure strategies, on the quality of Brazilian companies' sustainability reports, seeking to understand how top management reacts to stakeholder demands.

To delimit the sample, the reports published by the companies on Ibovespa 2023 index were analyzed. The methodology adapted from Sun et al. (2022) was used to evaluate 9 different topics aimed at different stakeholders, each evaluated by 7 different criteria. The results did not indicate a relationship between board size and the quality of sustainability reports, raising the discussion about existing dichotomous results for this variable (Qaderi et al., 2022; Tiron-Tudor et al., 2020).

As for board independence, no significant results were found in relation to the quality of reports. The results showed that the presence of women on boards of directors (PWOB) had a significant positive relationship. This underscores the importance of female representation on boards of directors, highlighting their role in improving companies' ability to respond to social and environmental issues. The result is in line with previous research findings that show female participation on boards has a positive association with the quality of sustainability information (Sun et al., 2022; Qaderi et al., 2022; Degenhart et al., 2024). The number of committees reporting to the board of directors has been shown to be related to companies' greater capacity to disclose on a wider range of areas of interest.

The variable External Assurance showed a significant relationship in all the models tested. This value indicates the strong relationship among the quality of the information disclosed in the reports and the entities' choice to ensure the production of this information. In addition to signaling the expectation of IR conceptual framework, it reinforces transparency and responsibility towards stakeholders, influencing the process of building and maintaining legitimacy (Sun et al., 2021). The variable ROA and the ALf Factor were not significant in the main models, indicating a certain alignment among the companies in Ibovespa portfolio.

This research has some limiting factors. Firstly, the independent variable was analyzed manually, without the aid of software. Therefore, it is possible that there are divergences conditioned by the researcher's perception. Secondly, the choice of portfolio was aimed at collecting data from companies with an elevated level of visibility and more uniform results; however, it is not yet possible to generalize the results found for all publicly traded companies in the Brazilian market. Finally, the data showed some specificities, requiring the use of methods to circumvent their negative effects. It is possible that by analyzing a larger volume of reports from different entities, it will be possible to reduce the behavior of these specificities in the sample and increase the relevance of the findings.

We hope to contribute to the literature by demonstrating that the results are not only conditioned by the internal environment of the entities, and that the characteristics of the boards can present different results when analyzed in different contexts. With the results of this research, it is possible to issue a warning to stakeholders, especially investors, about measures such as external assurance of reports in relation to the commonly positive effect on the quality of sustainability reports. The research highlights the importance of considering the implications of regulatory changes and market practices and their effects on the entities' disclosure process. Finally, by analyzing the presence of women on boards, the research contributes to the discussion on the need for gender diversity in corporate governance bodies, demonstrating an increase in the response to socio-environmental aspects related to entities.

For future research, the condition of institutional environments, external to the entity, may represent a significant change factor in the behavior of entities in relation to sustainability disclosure. Researchers can observe market criteria, regulation and development of the legal environment to verify the quality of reports, mainly exploring different contexts for comparative purposes. It is suggested to analyze all companies present on B3, collecting reports in a balanced data panel, with the expectation of analyzing the population of Brazilian companies and reducing errors caused by the data. Finally, the incorporation of the disclosure tone, removed from the calculations of the dependent variable, may present results different from those presented, directly implying possible new contributions to the literature.

References

• Al-Nasrallah, W. (2023). The decade long story of gender equality and female empowerment: a case study of corporate disclosures in Saudi Arabia. Sustainability Accounting, Management and Policy Journal, 14(1), 216-241.
• Arrington, C. E., & Puxty, A. G. (1991). Accounting, interests, and rationality: A communicative relation. Critical Perspectives on Accounting, 2(1), 31-58.
• Boiral, O., Heras-Saizarbitoria, I., & Brotherton, M. C. (2019). Assessing and improving the quality of sustainability reports: The auditors’ perspective. Journal of Business Ethics, 155, 703-721.
• Chouaibi, J., et al. (2022). Do board directors and good corporate governance improve integrated reporting quality? The moderating effect of CSR: An empirical analysis. EuroMed Journal of Business, 17(4), 593-618.
• Cooray, T., Senaratne, S., Gunarathne, A. D. N., Herath, R., & Samudrage, D. (2020). Does integrated reporting enhance the value relevance of information? Evidence from Sri Lanka. Sustainability, 12(19), 8183.
• Cucari, N., Esposito De Falco, S., & Orlando, B. (2018). Diversity of board of directors and environmental social governance: Evidence from Italian listed companies. Corporate Social Responsibility and Environmental Management, 25(3), 250-266.
• D'Apice, V., Ferri, G., & Intonti, M. (2021). Sustainable disclosure versus ESG intensity: Is there a cross effect between holding and SRI funds? Corporate Social Responsibility and Environmental Management, 28(5), 1496-1510.
• Deegan, C., & Blomquist, C. (2006). Stakeholder influence on corporate reporting: An exploration of the interaction between WWF-Australia and the Australian minerals industry. Accounting, Organizations and Society, 31(4-5), 343-372.
• Deegan, Craig. (2002). Introduction: The legitimising effect of social and environmental disclosures-a theoretical foundation. Accounting, Auditing & Accountability Journal, 15(3), 282-311.
• Degenhart, L., da Silva Giordani, M., Piccinin, Y. G., dos Santos Grodt, J. A., & da Silva Zonatto, V. C. (2024). Diversidade de gênero, expertise do conselho de administração e a transparência da divulgação ambiental, social e de governança (esg): evidências do Brasil. Enfoque: Reflexão Contábil, 43(2), 40-56.
• Di Vaio, A., Varriale, L., Di Gregorio, A., & Adomako, S. (2022). Corporate social performance and non‐financial reporting in the cruise industry: Paving the way towards UN Agenda 2030. Corporate Social Responsibility and Environmental Management, 29(6), 1931-1953.
• Dwekat, A., Meqbel, R., Seguí‐Mas, E., & Tormo‐Carbó, G. (2022). The role of the audit committee in enhancing the credibility of CSR disclosure: Evidence from STOXX Europe 600 members. Business Ethics, the Environment & Responsibility, 31(3), 718-740.
• Farooq, Q., et al. (2020). Social and environmental development: Fresh concepts and soft measures towards sustainable development. Sustainable Development, 28(6), 1796-1803.
• Freitas, M. R. O., Santos, S. M., & Crisóstomo, V. L. (2019). Nível de abrangência da informação ambiental divulgada nos relatórios de sustentabilidade de empresas brasileiras com potencial de impacto ao meio ambiente. Revista de Contabilidade e Controladoria, 10(3), 143,161.
• García‐Sánchez, I. M., Martínez‐Ferrero, J., & Garcia‐Benau, M. A. (2019). Integrated reporting: The mediating role of the board of directors and investor protection on managerial discretion in munificent environments. Corporate Social Responsibility and Environmental Management, 26(1), 29-45.
• Haji, A. A., & Anifowose, M. (2016). Audit committee and integrated reporting practice: Does internal assurance matter? Managerial Auditing Journal, 31(8/9), 915-948.
• Hamad, S., Lai, F. W., Shad, M. K., Konečná, Z., Goni, F. A., Chofreh, A. G., & Klemeš, J. J. (2022). Corporate governance code and voluntary disclosure of integrated reporting: Evidence from an emerging economy. Sustainable Development, 30(6), 1497-1510.
• Helfaya, A., & Whittington, M. (2019). Does designing environmental sustainability disclosure quality measures make a difference? Business Strategy and the Environment, 28(4), 525-541.
• Henriques, R., Gaio, C., & Costa, M. (2022). Sustainability Reporting Quality and Stakeholder Engagement Assessment: The Case of the Paper Sector at the Iberian Level. Sustainability, 14(21), 14404.
• Holtz, L., & Sarlo Neto, A. (2014). Efeitos das características do conselho de administração sobre a qualidade da informação contábil no Brasil. Revista Contabilidade & Finanças, 25(66), 255-266.
• Hummel, K., & Schlick, C. (2016). The relationship between sustainability performance and sustainability disclosure quality-hard numbers beat smooth talk. Journal of Accounting and Public Policy, 35(5), 455-476.
• International Integrated Reporting Council. (2021). International <IR> Framework. Disponível em: https://relatointegradobrasil.com.br/home/framework/
  » https://relatointegradobrasil.com.br/home/framework/
• Maria, S. C., d’Angelo, M. J., & Borgerth, V. M. D. C. (2021). Lacunas de engajamento e utilização do Relato Integrado no Brasil. Revista Contabilidade & Finanças, 33(88), 63-80.
• Melloni, G., Caglio, A., & Perego, P. (2017). Saying more with less? Disclosure conciseness, completeness and balance in Integrated Reports. Journal of Accounting and Public Policy, 36(3), 220-238.
• Miles, S., & Ringham, K. (2020). The boundary of sustainability reporting: evidence from the FTSE100. Accounting, Auditing & Accountability Journal, 33(2), 357-390.
• Mion, G., & Loza Adauí, C. R. (2019). Mandatory nonfinancial disclosure and its consequences on the sustainability reporting quality of Italian and German companies. Sustainability, 11(17), 4612.
• Monteiro, A., Dos Santos, T. R., & Dos Santos, G. C. (2020). Índice de sustentabilidade empresarial (ISE) e desempenho econômico-financeiro nas empresas da b3. Ragc, 8(38).
• Odriozola, M. D., & Baraibar‐Diez, E. (2017). Is corporate reputation associated with quality of CSR reporting? Evidence from Spain. Corporate Social Responsibility and Environmental Management, 24(2), 121-132.
• Omran, I. I., et al. (2021). Sustainability assessment of wastewater treatment techniques in urban areas of Iraq using multi-criteria decision analysis (MCDA). Water Practice & Technology, 16(2), 648-660.
• Ong, T., & Djajadikerta, H. G. (2020). Corporate governance and sustainability reporting in the Australian resources industry: An empirical analysis. Social Responsibility Journal, 16(1), 1-14.
• Pistoni, A., Songini, L., & Bavagnoli, F. (2018). Integrated reporting quality: An empirical analysis. Corporate Social Responsibility and Environmental Management, 25(4), 489-507.
• Qaderi, S. A., et al. (2022). Board characteristics and integrated reporting strategy: Does sustainability committee matter? Sustainability, 14(10), 6092.
• Ramdhony, D., Omran, M., & Hussainey, K. (2023). Board attributes and corporate social responsibility disclosure quality based on information typology. Review of Accounting and Finance, 22(4), 508-531.
• Sun, Y., Xu, C., Li, H., & Cao, Y. (2022). What drives the innovation in corporate social responsibility (CSR) disclosures? An integrated reporting perspective from China. Journal of Innovation & Knowledge, 7(4), 100267.
• Tiron-Tudor, A., et al. (2020). The level of European companies’ integrated reports alignment to the framework: The role of boards’ characteristics. Sustainability, 12(21), 8777.
• Unerman, J. (2000). Methodological issues‐Reflections on quantification in corporate social reporting content analysis. Accounting, Auditing & Accountability Journal, 13(5), 667-681.
• Vitolla, F., Raimo, N., Rubino, M., & Garzoni, A. (2020). The determinants of integrated reporting quality in financial institutions. Corporate Governance: The International Journal of Business in Society, 20(3), 429-444.

• 4
  This is a bilingual text. This article was originally written in Portuguese, published under the DOI https://doi.org/10.1590/1808-057x20242148.pt
• 5
  This article stems from a Master's dissertation submitted by the author Nicholas Patrick Garcia Duarte Grodt, in 2024.
• Study presented at the 24th USP International Conference on Accounting, São Paulo, SP, Brazil, July 2024.
• FINANCING
  The authors are grateful to the National Council for Scientific and Technological Development (Conselho Nacional de Desenvolvimento Científico e Tecnológico [CNPq]) for the financial support to this research - Edital Chamada Universal CNPq/MCTI N 10/2023 - Faixa A (Processo 406222/2023-1 / Umbelina Cravo Teixeira Lagioia).

Edited by

• Editor-in-Chief:
  Andson Braga de Aguiar

• Associate Editor:
  Eduardo da Silva Flores

Publication Dates

• Publication in this collection
  16 June 2025
• Date of issue
  2025

History

• Received
  03 May 2024
• Reviewed
  21 May 2024
• Accepted
  21 Oct 2024