Corporate governance and anti-corruption disclosure

Purpose – In recent years, the role of environmental, social and governance (ESG) disclosure has become crucial. The aim of this paper is to study how corporate governance affects one part of ESG disclosure: anti-corruption disclosure. Design/methodology/approach – This study examined 140 corporate social responsibility (CSR) reports from companies listed on the Italian stock markets and 50 CSR reports from other companies, thenthisstudyanalysed theadoption oftheGlobalReporting Initiative (GRI) standardno.205. Findings – The results show a low level of disclosure, and that corporate governance issues matter. In particular, the analysis found a positive relationship between the presence of female and outside members, thenumberofboardmembersandthe levelofanti-corruptiondisclosure. Research limitations/implications – This study acknowledges some limitations. Firstly, the research is based on a one-year sample. Secondly, the research hypotheses are confirmed only when considered in relation to a single section of the GRI standards. Thirdly, this study has a bias towards relatively large enterprises. Practical implications – It could be worthwhile introducing a soft regulation regarding the composition ofthe boardofdirectorsthatrequires a certainquantitative andqualitative composition. Originality/value – To the best of the authors’ knowledge, this is one of the few studies, the first in Italy, thatshedslightonanti-corruption disclosureandits determinants.


Introduction
In recent years, the role of environmental, social and governance (ESG) disclosure has become crucial (Baldini et al., 2018). ESG disclosure is often based on a voluntary choice and is left to management discretion (Reverte, 2009). And even when it is mandatory, there is a big difference compared to financial disclosure because ESG disclosure does not follow a standardised format (Elzahar et al., 2015). Given these variations and differences, it becomes relevant when analysing this kind of disclosure to identify which factors determine a high or low level of ESG disclosure. Some scholars have identified the country-and industry-levels as being relevant (Ioannou and Serafeim, 2015;Baldini et al., 2018). Other scholars have focused their attention on the company-level, in particular, on the role and characteristics of the board of directors, such as gender, educational level, nationality, age, autonomy and independence (Harjoto et al., 2015). Our study contributes to the literature regarding the relationships between corporate governance issues and ESG/ corporate social responsibility (CSR) disclosure (Agyei-Mensah, 2017a; Khan et al., 2019). We respond to the recommendation by Miras- Rodriguez and Di Pietra (2018, p. 568) that "there is scarce research on how the corporate governance mechanisms are influencing other CSR disclosure decisions such as GRI adoption". More in particular, to the best of our knowledge, only a few scholars have examined one particular area of ESG disclosure and Global Reporting Initiative (GRI) standards, that is anti-corruption disclosure (Barkemeyer et al., 2015;Healy and Serafeim, 2016;Islam et al., 2015;Joseph et al., 2016;Saenz and Brown, 2018), and most of them only observed the effects at the country-and/or industry-levels and did not examine the other determinants that could impact this type of disclosure. The study by Barros et al. (2022) studied the relationship between political connections and anti-corruption disclosure. Sari et al. (2021) studied the extent of anticorruption disclosures in 117 companies, and they found a positive association between the dependence on government tenders and foreign ownership and the level of anti-corruption disclosure. They also found that the United Nation Global Compact (UNGC) did not exert any influence on anti-corruption disclosure, even if other studies (Transparency International, 2009;Branco and Matos, 2016) found a positive relationship between UNGC membership and anti-corruption disclosure practices. In a report with a similar aim to ours, Masud et al. (2019) found a positive association between the presence of external experts on a board and corporate corruption disclosure. Our research aims to fill this research gap. We hypothesise that company-level determinants, and in particular the composition and characteristics of the board of directors, impact anti-corruption disclosure measured through GRI adoption. From this point of view, we also respond to the following recommendation from Katmon et al. (2019, p. 474) "We recommend future studies to focus on the complementary or substitutive impact of each of the board diversity characteristics on CSR, since the board diversity implementation is generally very costly to firms". We tested our hypothesis by analysing 190 Italian companies. As suggested by Baldini et al. (2018, p. 94) "In the on-going debate on the determinants of ESG disclosure practices, researchers should also restrict their analysis to specific countries". In addition, the Italian Stock Exchange and more in general the Italian setting are characterised by large controlling shareholders that considerably influence the board of directors' nominations and decision-making processes. To solve these agency conflicts, voluntary disclosure can play a critical role. So we agree with Allegrini and Greco (2013, p. 189) who state, "These features make the Italian setting appealing for a research about the interplay between governance and disclosure". Finally, we agree with Barkemeyer et al. (2015, p. 349) "the quality of corporate reporting practices has an important role to play in constraining corruption". And this is even more true in a country, like Italy, where corruption is defined by the Italian and European Governments as systemic and pervasive. In fact, Italy has ranked around 50th in the Corruption Perception Index provided by Transparency International for many years. This contextual situation makes Italy a suitable locus for the study. Our results evidence that corporate governance issues matter. In particular, there is a positive relationship between the number of board members, the presence of female and outside members and the level of anti-corruption disclosure. To the best of our knowledge, this is the first study in Italy that sheds light on anti-corruption disclosure and its determinants. Our study contributes to the academic debate about ESG and anti-corruption disclosure and its determinants related to corporate governance issues. At the same time, our study gives a useful insight to policymakers and regulators into how to answer the demand for more transparency and accountability in non-financial reporting. It could be worthwhile introducing a soft regulation that promotes the presence of women and outside members on company boards. The rest of the paper is organised as follows: in second section, we provide the background of the study; in third section, we review the literature and define our hypotheses; in fourth section, we motivate our research and methodology; in fifth section, we test and discuss our hypotheses; and we provide concluding remarks in sixth section.

Background
Corruption is a widespread phenomenon, with several economic and social implications. Shah (2007) distinguished four types of corruption. From the bureaucratic corruption related to public officials that abuse their office by demanding bribes to grand corruption that concerns large amounts of money stolen by public officials. From the corruption that occurs when public officials are trapped by private companies to patronage and clientelism. Peerthum and Luckho (2021) stated that "corruption has now become a very pertinent issue affecting all spheres of our modern society". Several studies found that corruption damages the quality of public institutions Cerchiello, 2018a, 2018b;Chan et al., 2020;Bauhr and Charron, 2020) and reduces productivity, investment and economic growth (Ellili, 2022;Liu et al., 2017;Wu et al., 2017;Cooray et al., 2017;Capasso, 2019;Schomaker, 2020;Akimova, 2020). As regards anti-corruption measures Baniamin and Jamil (2018) stated "The limited success of anticorruption measures points to the need to generate a better understanding of the problem of corruption and to explore effective measures for controlling it". Also Previtali and Cerchiello (2017) found that when "dealing with corruption there are no simple answers". One possible answer for the anti-corruption issue is ESG disclosure (Adeyeye, 2011), as stated by Barkemeyer et al. (2015, p. 349) "the quality of corporate reporting practices has an important role to play in constraining corruption". ESG disclosure implies a series of benefits, both for the regulators and public institutions, and for shareholders and investors. Related to the need for a sustainability report, is the degree of transparency of the reporting systems and legitimisation in the eyes of the community and public opinion (Alareeni and Hamdan, 2020;Fahad and Busru, 2021). For example, several studies link the cost of access to finance and equity capital to ESG disclosure (and in particular social disclosure), by outlining how better disclosure leads to a decrease in the cost of capital (Cheng et al., 2014;Ioannou and Serafeim, 2015;Dhaliwal et al., 2011). At the same time, this disclosure can entail costs. It can lead to proprietary costs and litigation costs. In fact, two large surveys conducted by KPMG (2005) and by Transparency International (2009) on sustainability reports issued by the world's largest companies reported a very low use of the section focused on policies for bribery and anticorruption. While on the one hand, companies disclose a strategic commitment to fight bribery and corruption (Agyei-Mensah, 2017b), on the other hand, they do not disclose information about the procedures, policies and systems that could support this strategic commitment.

Literature review and research hypotheses definition
In the field of ESG disclosures, scholars use different theories (Tagesson et al., 2009;Ioannou and Serafeim, 2012;Baldini et al., 2018;Baalouch et al., 2019). At an industry-level of analysis one of the most applied is institutional theory, especially when scholars want to develop the relationship between country-level characteristics and ESG disclosure (Jackson and Apostolakou, 2010;Muthuri and Gilbert, 2011). In this case, companies are often studied on the basis of their response to pressures from the external and institutional environment, typically rules and regulations, which include coercive, normative and mimetic isomorphic change (DiMaggio and Powell, 1983). Another important theory is the legitimacy theory which focuses on societal acceptance and awareness of a company in its context. In this case, scholars have studied the relationship between social stakeholder expectations and ESG disclosure (Beelitz and Merkl-Davies, 2012;Palazzo and Scherer, 2006). Also by adopting appropriate disclosure practices that enable the company to increase its reputation and its strategic networks (Boyd, 1990). This need to conform to the expectations of external stakeholders and to increase legitimacy was analysed by Branco et al. (2019). They found a positive relationship between companies that are cross-listed, companies that are members of the UNGC and anti-corruption disclosure. Related to this, Blanc et al.
(2019) described how Siemens changed its compliance and corruption disclosure practices to repair its legitimacy after the 2006 corruption scandal. For the aim of this paper which focuses on a company-level of analysis and in particular on the characteristics of the board of directors, agency theory, resource dependence and resource-based theories play a fundamental role. According to the scholars who applied agency theory (Fama and Jensen, 1983), disclosure is important to minimise agency costs and to reduce the misalignment between a company's goals and the stakeholders' interests. This is especially true when referred to the role of outside directors (Bear et al., 2010). According to resource dependence theory the board of directors can be considered as a resource. So the more the board of directors consists of members with different expertise, competencies, backgrounds, etc. the more the company can avoid dependence on the external environment. Similar to resource dependence theory, a resource-based view is another frequently adopted theoretical perspective to explain the relationship between board characteristics and CSR (Barney, 1991;Galbreath, 2016;Yu and Choi, 2016;Katmon, 2019). These theories have led us to focus our attention on board independence and board diversity such as gender, education, skills, competencies, etc. as unique and valuable resources that can contribute to the decision-making process used in the adoption of corporate social disclosure practices. Concerning directors' independence, in a survey on Hong Kong-listed firms, Chen and Jaggi (2000) found a positive relationship between board independence and voluntary disclosure. The same results were found by Lattemann et al. (2009) and specifically by Martinez-Ferrero and Garcia-Sanchez (2017) with regards to CSR disclosure. The study by Jaggi et al. (2021) based on a sample of 234 Europeanlisted companies found that the presence of independent directors and female directors is positively associated with corporate corruption disclosures. All these studies argued that independent boards provide more effective internal controls, also of disclosure policies. In addition, the attention to a company's reputation and long-term goals encourages the disclosure of both financial and non-financial information, such as anti-corruption information. However, in this case too, the literature shows mixed findings. Eng and Mak (2003) and Haniffa and Cooke (2005) found that an increase of outside directors decreases CSR disclosure. Yet other studies found that there is no association between board composition and disclosure (Matolcsy et al., 2012;Cormier et al., 2011;Michelon and Parbonetti;Frias-Aceituno et al., 2013). A further issue concerns chief executive officer (CEO) duality. Samaha et al. (2015) found that CEO duality was associated with poorer disclosure. Gul and Leung (2004), studying a sample of 385 Hong Kong-listed companies, showed empirical evidence that CEO duality is associated with lower levels of voluntary disclosure. Ho and Wong (2001) and Cheng and Courtenay (2006) did not find any significant relationships. These studies lead us to develop the following research hypotheses: H1. There is a relationship between the independence of the board of directors and anticorruption disclosure.
H2. There is a relationship between the presence of executive directors on the board of directors and anti-corruption disclosure.
H3. There is a relationship between the presence of CEO duality and anti-corruption disclosure.
With regards to board diversity Hoang et al. (2018) found a positive relationship between board diversity and corporate social disclosure. Previously, other scholars reached the same conclusion, and found in particular that female directors on the board improve CSR reporting (Rupley et al., 2012;Hafsi and Turgut, 2013;Liao et al., 2015;Al-Shaer and Zaman, 2016). Jaggi et al. (2021) also argued that female directors pay more attention to more comprehensive and transparent disclosure, especially of non-financial information. Female directors are considered to have higher ethical values, social orientation and awareness of stakeholders' demands. All characteristics that lead to a greater orientation towards controlling and monitoring corruption and fraud activities. At the same time, Pucheta-Martinez et al. (2018) posits that there is a non-linear association, concretely quadratic, between independent and institutional female directors on boards and CSR reporting. Katmon (2019) studied 200 listed firms in Malaysia and found a positive relationship between board education level and board tenure diversity on CSR disclosure. Other studies focused on the relationship between nationality diversity and quality of CSR disclosure. A study by Muttakin et al. (2015) found a positive association between the two variables while other studies did not find any significant association (Barako and Brown, 2008). These studies lead us to develop the following research hypotheses: H4. There is a relationship between the diversity of the board of directors and anticorruption disclosure.
H5. There is a relationship between the presence of foreign board members and anticorruption disclosure.
Finally, one last variable that could be worthwhile studying is board size (Ji et al., 2015). Correa-Garcia et al. (2020) found a positive association between larger companies and anti-corruption disclosure. Also Faisal et al. (2022) argued that "The larger a company is, the greater the number of stakeholders involved in the activities of the company and the higher their expectations regarding CSR practices". Rahman and Bukair (2013) found a positive relationship between larger sized boards and disclosure and Rodríguez-Ariza et al.
(2014) discussed the relationship between board size and GRI application in European companies. However, the evidence provided by these two researchers is scarce. These studies lead us to develop the following research hypothesis: H6. There is a relationship between the size of the board of directors and anti-corruption disclosure.

Method
To develop our analysis, we downloaded the non-financial reports related to legislative decree no. 254/2016 from the register managed by the University of Siena called "Observatory of non-financial disclosure of Italian companies". In Italy, legislative decree no. 254 transposes the European Directive 2014/95. It makes non-financial disclosure obligatory for public interest entities. In particular, it is applied to issuers of securities traded on Italian or European regulated markets, banks, insurance companies and to all the companies which exceed some requirements in terms of number of employees and total net asset value or revenues. The other information related to board of directors, size and profitability, were gathered directly from the investor section in the company websites and from the annual corporate governance reports, which are compulsory for all listed companies. Considering all these points, the final data set included the entire set of companies available in the database of the Observatory. It is composed of 140 CSR reports from companies listed on the stock markets and 50 CSR reports from unlisted companies of public interest. The listed companies belong to the following industries: basic materials and oil and gas and chemicals (9 companies), consumer goods (26), consumer services (15), financial services (24), industrial (44), tech and telco (10) and utilities (12). The unlisted companies belong to the following industries: basic materials (1 company), consumer goods (1), consumer services (1), financial services (30), industrial (10), tech and telco (10) and utilities (7). The listed companies are articulated by total assets as follows: 84 companies have less than e1.000m of total assets, 34 companies from e1.000m to e10.000m, 12 companies from e10.000m to e50.000m, 5 companies from e50.000m to e100.000m and 5 companies have more than e100.000m of total assets. The unlisted companies are articulated by total assets as follows: 13 companies have less than e1.000m of total assets, 25 companies from e1.000m to e10.000m, 9 companies from e10.000m to e50.000m, 2 companies from e50.000m to e100.000m and 1 company has more than e100.000m of total assets. Although non-financial disclosure is compulsory for all these companies, the content is not. So companies have a lot of discretion about whether and how to publish information and data regarding anti-corruption. For this reason, we decided to analyse the level of anti-corruption disclosure as a dependent variable. As applied in previous studies (Rodríguez-Ariza et al., 2014), to do this we focused our attention on the different application levels of the GRI. We chose GRI as in the field of sustainability reporting, GRI isto datethe most diffused standard, used by several thousand companies. In fact, all the companies in the sample had adopted it. Because of the aim of our research we considered the GRI no. 205, and in particular the three subsections as reported in Table 1.
Here, similar to other scholars (del Mar Alonso-Almeida et al., 2015;Dong et al., 2014;Lock and Seele, 2016;Pucheta-Martinez et al., 2018;Hoang et al., 2018) we measured each of the abovementioned items as a variable with the following values: Full disclosure of information about anti-corruption using the GRI reports. This item was given a value of 1.
Partial disclosure of information about anti-corruption without using GRI reports and so giving just a part of the information required by the standard recommended by the GRI. This item was given a value of 0.5.
Finally, a value of 0 was given when firms did not report any information about anticorruption.
We considered each subsection as a unique variable appropriately created. More specifically we summed up values for each macro-section -205.1, 205.2, 205.3so as to produce three scores which assess the level of expressed disclosure. Considering subsection 205.1 which is composed of 2 items, the associated calculated score has 0 as minimum value (no disclosure at all) and 2 as maximum value (full disclosure). With regards to subsection 205.2, composed of 5 items, the associated calculated score has 0 as minimum value (no disclosure at all) and 5 as maximum value (full disclosure). Finally, for subsection 205.3, composed of 4 items, the associated calculated score has 0 as minimum value (no disclosure at all) and 4 as maximum value (full disclosure). The period of study is for the fiscal year 2020. The independent variables are the following: board size, measured by the number of board directors; gender ratio, measured by the number of female members on the board/number of board directors; Total number and percentage of operations assessed for risks related to corruption Significant risks related to corruption identified through the risk assessment 205-2 a) Total number and percentage of governance body members that the organisation's anti-corruption policies and procedures have been communicated to b) Total number and percentage of employees that the organisation's anticorruption policies and procedures have been communicated to c) Total number and percentage of business partners that the organisation's anti-corruption policies and procedures have been communicated to d) Total number and percentage of governance body members who have received training on anti-corruption e) Total number and percentage of employees who have received training on anti-corruption 205-3 a) Total number and nature of confirmed incidents of corruption b) Total number of confirmed incidents in which employees were dismissed or disciplined for corruption c) Total number of confirmed incidents when contracts with business partners were terminated or not renewed due to violations related to corruption d) Public legal cases regarding corruption brought against the organisation or its employees during the reporting period and the outcomes of such cases presence of executive directors on the board, measured by the number of managers on the board/number of board directors; independence ratio, measured by the number of independent members on the board/ number of board directors; presence of foreign members on the board, measured by the number of foreign members on the board/number of board directors; and CEO duality, measured by a dummy variable equal to 0 if the CEO does not also serve as the president of the board, and equal to 1 if the CEO serves as the president of the board. Table 2 sums up the most important characteristics of the board of directors.
The hypotheses were tested through the estimation of multivariate linear regressions. This choice was naturally driven by the measurement level of the target variables, namely, the disclosure of anti-corruption measures. Because our main aim is to assess the importance of even partial disclosure of information about anti-corruption, the target variable is treated as a purely numerical one. All the analyses were run accordingly.

Empirical results
Concerning the level of disclosure of anti-corruption measures, as defined by the GRI 205 standards and shown in Table 3 we have the following remarks. The majority of the observed companies adopted a low level of disclosure. In particular, as concerns the standards for 205.2, a very low level of disclosure related to the communication of anticorruption policies and procedures, both related to governance body members and external business partners is reported. The same is true for training programs received by governance body members. A low level of disclosure was reported for "the operations Table 2 Characteristics of the board of directors in the observed companies Average number of directors , whose: 10,3 number of companies with less than 9 board members 93 number of companies from 10 to 14 board members 73 number of companies with 15 board members or more 24 % of outside members out of the total of the board members (independent ratio) 45.5% % of female members out of the total of the board members (gender ratio) 32.8% % of foreign members out of the total of the board members 6.9% Number of companies where the CEO coincides with the president of the board 34 assessed for risks related to corruption" (205.1a). Also for the "confirmed incidents when contracts with business partners were terminated or not renewed due to violations related to corruption" (205.3c), and for "the public legal cases regarding corruption brought against the organisation or its employees during the reporting period and the outcomes of such cases" (205.3d). On the contrary, the only standard for which a good level of full disclosure is reported is the 205.3a which regards the "total number and nature of confirmed incidents of corruption". And to a smaller extent the disclosure of "significant risks related to corruption" (205.1b).
Besides these more general aspects, our research aimed to ascertain whether there are other company factors that influence the disclosure of corruption. Concerning the research hypotheses and in particular board independence, we found that the independence of the board of directors (H1) is pretty significant and positively related to anti-corruption disclosure, when considering both 205.1 and 205.2 as a target variable. Also, we found a significant and negative relationship between the presence of CEO duality and anticorruption disclosure (H3), when considering 205.2 as a target variable. We did not find any significant relationships between the presence of executive directors on the board of directors and anti-corruption disclosure (H2).

Discussion
Considering the academic implications, our results fit with the main research on anticorruption practices disclosure as an integral part of CSR themes or sustainability reporting for firms (Rodriguez et al., 2006;Svensson, 2005;Barkemeyer et al., 2015;Transparency International, 2009;Issa and Alleyne, 2018). The level of disclosure in the observed reports is significantly low. One explanation might be that corruption by its nature is secret, complex, occult and sensitive and not popular with companies (Juliao-Rossi et al., 2022). It does not imply the same emotive weight as other business and social issues. The fact that anti-corruption legislation still suffers from implementation deficits at national and cross-national levels also impacts on disclosure in a negative way. As regards the studies focused on the relationship between corporate governance and disclosure, our results fit with the studies that found a positive relationship between the presence of an outside director and performance, also in terms of disclosure (Guest, 2008;Lattemann et al., 2009;Johansen et al., 2017;Martinez-Ferrero and Garcia-Sanchez, 2017;Jaggi et al., 2021). These results suggest a managerial practice that promotes the presence of outside directors. And this is true especially when the task orientation of the board is towards monitoring and controlling. Moreover, these are tasks typically performed by outside directors, in line with the agency theory tradition, international best practices and with the codes of good governance of many countries. All these practices and codes suggest the presence of outside directors as a critical support to an independent and autonomous decision-making process. Outsider directors are more likely to be independent and objective in these tasks than insider directors. In fact they wish to signal their competence to other potential employers and frequently already have monitoring experience. In contrast, insiders are less independent because their careers within the firm are dependent on the CEO. In addition, internal managers could perceive that following GRI is a way of controlling their activity. Consequently, they do not support carrying out these disclosure decisions. On the contrary, outside managers could argue that these disclosure practices could enhance their credibility with regards to the stakeholder community. Also the result of CEO duality associated with poorer disclosure confirms the results suggested in other studies (Gul and Leung, 2004;Samaha et al., 2015). On this point we can argue that the absence of separation between decision control and decision management reduces the transparency of the board. It may reduce checks and balances, overall reducing the orientation to disclose. The results fit also with the literature that argued that female directors on boards help the company to empathise with stakeholders and to show a higher sensitivity towards their demands and needs. (Liao et al., 2015;Al-Shaer and Zaman, 2016;Jaggi et al., 2021). Female directors have a leadership style open to ethical and social issues and this leads to a greater concern for anti-corruption matters. The resource dependence approach argues that female directors, given their capacity to maintain outside connections with stakeholders and organisations, will positively affect the anti-corruption disclosure because their presence on boards is a relevant tool for firms to gain legitimacy with stakeholders and society. On this point, the managerial implications are the introduction of a soft regulation that provides guidelines for a better balance in terms of gender composition of the board to the company. Finally, our results found a positive relationship between larger boards and better performances, as demonstrated by several studies (Cheng, 2008;Bukair, 2013;Rodríguez-Ariza et al., 2014;Faisal et al., 2022). According to the research, it is preferable to have a larger board composed of people with different experience, to ensure a greater number and a high level of competencies, also in terms of disclosure and compliance. In addition, larger companies have more stakeholders' needs to respond to, and this leads to gaining a higher level of legitimacy. Also in this case, a soft regulation that provides some indications in term of correlation between a company's size and the board size would be worthwhile.

Summary and conclusion
In 2009, the first TRAC report notes that anti-corruption information from the observed companies did not have a lengthy reporting history compared with information on environmental issues. More than 10 years later there is still much work to be done. This article begins from the need to fill the gap present in literature about ESG disclosure and in particular anti-corruption disclosure. The results show that, on average, leading Italian companies still have a long way to go in demonstrating that they are embedding anti-corruption practices into their organisations. The only standard the majority of the analysed companies fully disclosed regards the total number and nature of confirmed incidents of corruption. If we shift our attention from the number and nature of incidents of corruption to other related aspects and outcomes, such as employees dismissed or disciplined for corruption, contracts with business partners that were terminated or not renewed and public legal cases, then we register a low or very partial level of disclosure. The level of disclosure is also low for the risk assessment implementation across the company, regarding both the assessed operations and the identified risk factors. The same level of low disclosure was confirmed for communication and training related to anticorruption policies and procedures, for both governance body members and employees. These results confirm as suggested in several studies (Reverte, 2009;Elzahar et al., 2015;Baldini et al., 2018) that ESG disclosure is based on management discretion and it does not follow a standardised format. These conditions lead to a great variation in terms of the level and completeness of disclosure. Fortunately, not all companies and not all GRI standards are at the same level of ineffective disclosure. Hence, the interest in identifying whether there are some determinants that can affect the level of disclosure. From a theoretical perspective, this research contributes to the literature that extends the use of resourcebased theories to the study of board of directors characteristics. The more the board of directors consists of members with different expertise, competencies, backgrounds, skills, etc. the more the company can avoid dependence on the external environment Katmon, 2019). Our research contributes to existing research on anti-corruption disclosure by demonstrating that company-level determinants, and in particular the composition and characteristics of the board of directors, have an impact on anti-corruption disclosure. Corporate governance matters, especially with reference to the size and composition of the board of directors, in terms of the number of board members, and the presence of female and outside members. Concerning the directors' independence, our results demonstrated a positive relationship between board independence and anticorruption disclosure, as previously suggested by other studies (Lattemann et al., 2009;Martinez-Ferrero and Garcia-Sanchez, 2017;Jaggi et al., 2021). These results argued that independent boards provide more effective internal controls, also of disclosure policies. And at the same time the outside directors are more interested in gaining external legitimacy through several types of stakeholders. With regards to board diversity we found a positive relationship between board diversity and anti-corruption disclosure, in particular as regards the presence of female directors (Al-Shaer and Zaman, 2016;Hoang et al., 2018;Jaggi et al., 2021). These studies confirm the studies that suggest female directors have greater social orientation and awareness of stakeholders' demands, including an orientation towards controlling and monitoring corruption and fraud activities. Finally, we found a positive relation between larger boards and anti-corruption disclosure. Board size impacts the level of disclosure (Correa-Garcia et al., 2020;Faisal et al., 2022) in a positive way. From a practical perspective, our findings can help companies and government enhance the actions they can take in preventing and contrasting bribery and corruption. Regarding companies, as stated by Saenz and Brown (2018, p. 269) "it is necessary to increase the disclosure requirements for companies so that it is known what they are doing to improve their anticorruption management systems". Companies are gradually focusing their strategies on ESG goals; anti-corruption disclosure can have a significant influence on helping progress towards these goals. In addition, as stated by Faisal et al. (2022) "enhancing the level of anti-corruption disclosure demands complete support from top management". So also from this point of view, the composition of the board of directors can play a crucial role by providing support to the company's anti-corruption programme. Regarding government and policymaking, it could be worthwhile introducing a soft regulation regarding the composition of the board of directors that requires a certain quantitative and qualitative composition. As argued by Jaggi (2021) "when regulations/ directives do not contain concrete requirements or guidelines for disclosures, managers' disclosure decisions are influenced by a host of factors, where corporate governance features would have a strong role to play". This study acknowledges some limitations. Firstly, the research is based on a one-year sample. Secondly, the research hypotheses are confirmed only if they are considered in relation to a single section of GRI standards. Thirdly, as with previous studies into sustainability reporting, this study has a bias towards relatively large enterprises as larger companies are more likely to engage in non-financial reporting in the first place. Finally, although the framing of the analysis enabled us to investigate the extent to which companies communicate their anti-corruption engagement, it did not allow us to shed light on their actual engagement in anti-corruption measures. These limitations open up a number of avenues for future research. To start with, future research could examine the link between the communication of anti-corruption initiatives and actual levels of corporate engagement in this area. In addition, it could be very interesting to conduct an inter-temporal study and extend the insights presented here. Another point of interest could be a study across different regions and countries. Lastly, the model of analysis could be extended to other determinants of the relationship between companies and anti-corruption disclosure, such as size, industry and business models.