This paper aims to examine the extent of sustainability reporting by state-owned enterprises (SOEs) in Sub-Saharan African (SSA) countries and its relationship with selected board characteristics (size, independence, gender diversity and meetings).
Data was gathered from the annual reports of 78 SSA SOEs during 2012–2022 through a content analysis approach, and analysed using panel data regressions.
SSA SOEs’ sustainability reporting is moderate. Additionally, the selected board characteristics and the extent of sustainability reporting have an insignificant relationship. This conclusion holds even when the sample is split between listed and unlisted SOEs.
The insights can inform authorities and policymakers to strengthen the corporate governance mechanisms of all SOEs, as the findings reveal a low level of reporting and insignificant relationships with the extent of sustainability reporting.
To the best of the authors’ knowledge, this is the first study to examine the relationship between board characteristics and the extent of SOE sustainability reporting in SSA using a multi-country sample of listed and unlisted SOEs. It also extends the literature by showing that within the SSA context, the nexus between corporate governance characteristics and sustainability reporting by SOEs does not vary between listed and unlisted SOEs.
1. Introduction
Effective and functioning corporate governance is key to the success and promotion of sustainability practices in organisations (Post et al., 2015; Tang et al., 2020). Recently, scholars have examined the nexus between corporate governance mechanisms and firms’ sustainability reporting. Despite the growing literature on this topic, state-owned enterprises (SOEs) have not received the same level of attention as private firms (Abang’a and Tauringana, 2024; Manes-Rossi et al., 2020). Due to differences in mandates, purpose, stakeholders, governance and funding sources between private firms and SOEs, the findings for private firms may not necessarily apply to SOEs.
Furthermore, the few studies on SOEs primarily focused on advanced countries like Australia, Italy, Spain and the UK (Garde Sánchez et al., 2017; Manes-Rossi et al., 2020; Nicolò et al., 2019; Ruiz-Lozano et al., 2022). Studies on SOEs in Sub-Saharan African (SSA) countries are scarce (Abang’a and Tauringana, 2024). This is because sustainability reporting in SSA countries is mainly voluntary (Tilt et al., 2021). Moreover, organisations choose what and how to disclose, unlike in advanced countries where sustainability reporting is regulated (Nicolo’ and Andrades‐Peña, 2024). Furthermore, SSA firms suffer from weak corporate governance frameworks that potentially affect their strategic choices (Ogbechie and Arije, 2024; Okeahalam, 2004; Wezel and Carvalho, 2022). Therefore, the findings from advanced countries may not hold in weak institutional settings, like SSA, due to the differences in institutional mechanisms.
We examine the relationship between board characteristics and the extent of SOEs’ sustainability reporting in SSA for the following reasons. First, weaknesses in corporate governance exist within SSA (Ogbechie and Arije, 2024; Okeahalam, 2004). This problem is even worse in SOEs due to political interference from incumbent governments. Specifically, most board appointments to SOEs in the region are made by the incumbent government without much recourse to the appointees’ competencies, skills and abilities (Abang’a and Tauringana, 2024; Okeahalam, 2004). Changes in board membership usually occur in conjunction with government changes. As such, directors may pursue objectives that benefit their political parties rather than sustainability initiatives that could help the entire populace (Abang’a and Tauringana, 2024). Moreover, most SOEs in SSA countries are mostly unlisted, less regulated and highly protected by incumbent governments. These institutional lapses can affect SOEs’ levels of accountability, including sustainability disclosures. Furthermore, in most cases, essential sectors, like mining and energy, which have significant environmental and social impacts, are owned by governments. Hence, examining SOEs in Africa and their sustainability reporting is an interesting research topic.
Some studies have examined the relationship between board characteristics and the extent of sustainability reporting by SOEs in SSA (Abang’a and Tauringana, 2024; Masoud and Vij, 2021). Abang’a and Tauringana (2024) only analysed the annual reports of 45 Kenyan SOEs from 2015 to 2018. They reported negative relationships of board gender diversity, board age and board subcommittees with sustainability reporting, whereas other board variables showed insignificant associations. Considering that the study focused on a single country and used a short sample period, the generalisability of the findings to the remaining SSA countries is difficult and problematic. Additionally, the short data period does not allow for an extended period of trend analysis of the evolution of sustainability reporting among SOEs in the sub-region.
Next, while Masoud and Vij (2021) solely focused on Libyan SOEs, they did not explicitly examine the relationship between board characteristics and sustainability reporting. They reported that firm size, firm age, type of sector and corporate social responsibilities (CSR) are positively related to sustainability reporting. Meanwhile, other variables, like managers’ educational backgrounds and female managers, showed insignificant relationships with sustainability reporting. While managers can play an essential role in corporate governance and the implementation of sustainability initiatives, they are not part of corporate board membership. Therefore, these findings are not as relevant in the context of the relationship between board characteristics and sustainability reporting in SSA.
Similarly, Rahaman et al. (2004) focused on a single SOE in Ghana. They used a qualitative research design involving interviews with employees and government officials, and performed document analysis to reveal the motives for social and environmental reporting by the Volta River Authority. The authors reported that pressure from funding agencies like the World Bank drives sustainability reporting. However, the study failed to focus on board characteristics and their relationship with sustainability reporting. Moreover, its single-country focus does not permit the results to be extended to other SOEs or countries in SSA.
Beyond the African continent, an international study on the nexus between board characteristics and sustainability reporting by Nicolo’ and Andrades‐Peña (2024) only selected SOEs in the Refinitiv database. Consequently, the sample of 253 SOEs was mainly dominated by SOEs from Western and Asian continents, as only one SOE from Africa (Egypt) was included. The authors reported that the board size, gender diversity, board independence and board meetings were significantly associated with SOE sustainability reporting. With only one African country in the sample, generalising the findings to SOEs in Africa or SSA would be deceptive and problematic. Next, Manes-Rossi et al. (2020) relied on cross-sectional data from only 18 European SOEs that published integrated reports for their study. Similarly, Argento et al. (2019) examined Swedish SOEs that published sustainability reports between 2013 and 2017. After excluding SOEs that disclose sustainability information in annual and other reports, several SOEs were excluded from the study. As such, the results of these two studies are likely to be impacted and lack generalisability to SOEs that do not produce sustainability or integrated reports.
Furthermore, the literature suggests that corporate governance variables have endogeneity issues that can affect results and must be addressed (Wintoki et al., 2012). Yet, extant research on the nexus between board characteristics and SOE sustainability reporting does not tackle these endogeneity problems (e.g., Abang’a and Tauringana, 2024; Nicolo’ and Andrades‐Peña, 2024). SOEs are critical tocountries’ economic fortune and sustainable development (Bebbington and Unerman, 2018; Farneti et al., 2019). However, SOE directors’ efficiency and effectiveness are crucial for achieving these goals (Okeahalam, 2004). Therefore, more research is needed to examine the relationship between corporate governance mechanisms and sustainability reporting by SOEs to contribute to both theory and practice. Consequently, a quantitative study that focuses solely on a sample of SOEs in Africa or SSA countries, uses extensive longitudinal data retrieved from various sources and addresses endogeneity issues is highly needed to overcome the limitations identified in extant studies and augment the literature.
Through the theoretical lenses of agency and stakeholder theories, this study fills the aforementioned research gaps by examining the extent of sustainability reporting by SOEs in SSA, and the relationship between selected board characteristics and the extent of sustainability reporting by the SOEs. To achieve these objectives, content analysis is used to analyse data collected from the annual and sustainability-related reports of 78 SOEs from 2012 to 2022. Panel data regression is used to test the proposed hypotheses. Next, the system generalised method of moments (GMM) is used to address endogeneity concerns and improve the robustness of the findings. The findings show that board size, independence, gender diversity and meetings do not have a significant relationship with the extent of sustainability reporting. The results remain the same even after unpacking the sustainability index.
This study contributes to the literature in several ways. First, it contributes to the limited literature on sustainability reporting on SOEs in emerging countries with institutional voids. To the best of our knowledge, this study is among the few to examine listed and unlisted SOEs, and show that, in the SSA context, the relationship between board characteristics and sustainability reporting is the same for both listed and unlisted SOEs.
Second, the study uses 11-year data from 78 SOEs across 16 African countries. This helps generalise the results to several SOEs and for observing the evolution of sustainability reporting compared with prior studies that use a single country, industry, or short sample period (Abang’a and Tauringana, 2024; Argento et al., 2019; Garde Sánchez et al., 2017; Masoud and Vij, 2021; Rahaman et al., 2004). We also extend the research which preforms region-specific analyses (Manes-Rossi et al., 2020; Nicolò et al., 2019). It also enhances trend analysis and helps address the problem of anomalies arising from small sample sizes and periods.
Third, we used system GMM to address potential endogeneity issues not addressed in prior studies (Abang’a and Tauringana, 2024; Argento et al., 2019; Nicolo’ and Andrades‐Peña, 2024; Yustina et al., 2024). Furthermore, our study goes beyond prior studies by widening data sources (annual, CSR, sustainability and integrated reports) to fully capture SOEs’ sustainability reporting. By comparison, extant studies used single data sources like annual, sustainability or integrated reports (Abang’a and Tauringana, 2024; Argento et al., 2019; Manes-Rossi et al., 2020). Thus, we respond to the call of scholars (Abang’a and Tauringana, 2024; Manes-Rossi et al., 2020) to analyse various reports to have more SOEs in the sample, and fully and reliably capture SOE sustainability disclosures. The study extends the literature on stakeholder and agency theories from the context of SOE sustainability reporting in emerging economies, as few such studies exist in the literature, as indicated by Nicolo’ and Andrades‐Peña (2024).
Overall, the study contributes to the global discourse on sustainability reporting within public sector organisations, as it reveals the complexity of relationships within the sector in terms of governance and stakeholder management. The result implies that public sector organisations should prioritise effective corporate governance mechanisms in line with the agency theory rather than mere board restructuring techniques such as expanding board size, increasing board meeting frequency and gender diversity, which seem to have less impact on SOEs’ sustainability disclosures. Furthermore, in line with stakeholder theory, the study highlights the need for active stakeholder participation and dialogue as opposed to mere symbolic representation of stakeholders to enhance the extensiveness of sustainability reporting within the public sector. Finally, the study’s findings provide evidence that sustainability reporting standards, policies and laws should be context-specific and not universal.
The remainder of this article is organised as follows. Section 2 discusses the theoretical framework and Section 3 presents the literature review and hypotheses. Section 4 elaborates on the research design, while Section 5 presents the results. Section 6 discusses the results and Section 7 concludes.
2. Theoretical framework
2.1 Agency theory
Business owners or investors commonly use the services of others to steer a company’s affairs on their behalf. This creates an agent–principal relationship (Jensen and Meckling, 1976). However, potential problems could arise from the different motives of the principal and agent. Additionally, the agents may possess information that the principal may not be privy to. This can also lead to opportunistic behaviour from the agents due to this information asymmetry (Donnelly and Mulcahy, 2008). To overcome these problems, agency theory (Fama and Jensen, 1983) proposes specific tools to be implemented in organisations to monitor and control the agents’ activities and promote the principal’s welfare. These tools include effective corporate governance mechanisms, like the board of directors (Haniffa and Cooke, 2002).
This agency problem is also prevalent in SOEs, as the government, being the principal, uses the services of directors as agents to work on its behalf and that of the citizens (Abang’a and Tauringana, 2024). Additionally, effective corporate governance is a solution to the principal–agent relationship problem (Buniamin et al., 2011; Ienciu et al., 2012). According to agency theory, effective corporate governance can facilitate responsible business practices like reliable and timely disclosures, which have long been a corporate governance problem in SSA, and undertake initiatives that promote public welfare. However, most board and managerial appointments to SOEs are made through the incumbent government (Nana Yaw Simpson, 2014). Hence, in some instances, directors and SOE managers may take actions that benefit their political parties or for their personal gains against public interests (Jensen and Meckling, 1976). Again, within SOEs, a dual-agency problem exists (Wang et al., 2025). This is because, after being empowered by citizens, the government also appoints top executives based on political party affiliation to run the SOEs to seek the party’s interests, contrary to the interests of citizens and minority shareholders (Wang et al., 2025). This creates some complexities regarding accountability, leading to problems like weak disclosures, inefficiencies and weak corporate governance. This also weakens the government’s supervisory authority over these appointees, who capitalise on this weakened supervision for personal gains against that of the citizens and minority shareholders (Fan et al., 2013; Wang et al., 2025). Finally, CEO duality, wherein the same person serves as the chairperson of the board and chief executive officer (CEO), gives one person so much power in an organisation (Jizi et al., 2014). This can prevent other directors from initiating actions that benefit the public and adversely impact their monitoring capabilities (Tuggle et al., 2010).
2.2 Stakeholder theory
Organisations deal with various stakeholders, including employees, customers, regulators and communities (Roberts, 1992; Freeman, 1984). Each of these stakeholders has unique needs and demands that they seek from their organisations (Deegan et al., 2000). Stakeholder theory (Freeman, 1984) posits that organisations should balance the needs of all stakeholders, and not prioritise only the needs of shareholders or investors. This can help the firm maintain good relationships with diverse stakeholders, which can translate into benefits for the firm in the long run (Tang et al., 2015). Stakeholder theory also suggests that balancing these stakeholders’ differing needs lies with the organisation’s directors (Khaireddine et al., 2020). Indeed, firms can use sustainability initiatives and disclosures to manage their diverse stakeholders (Orij, 2010). Effective corporate governance has the potential to help organisations strengthen their relationships with stakeholders through their corporate sustainability initiatives (Michelon and Parbonetti, 2012). This theory also advocates for the representation of other stakeholders external to the organisation. This advocacy can include independent directors as part of an organisation’s corporate governance framework, representing diverse stakeholders’ interests (Hussain et al., 2018).
In the context of SOEs, directors represent the general public’s interest by monitoring managerial actions to enhance social welfare. However, an issue associated with stakeholder theory is its tendency to satisfy the needs of the more powerful stakeholders (Ntim et al., 2017). This is potentially the case with SOEs in SSA, wherein the government usually makes appointments to the board (Abang’a and Tauringana, 2024). This can lead to a situation wherein these appointees prioritise the incumbent government’s interests over those of the general public. Moreover, CEO duality can weaken the influence and power of independent directors, affect their effective functioning and monitoring capabilities and lead to low accountability.
Overall, SOEs can use sustainability reporting to inform the public of the impact of their activities on the environment, society and people. Based on this, the public can query and assess SOE executives’ operations and choices (Ntim et al., 2017). This also helps SOEs effectively manage their diverse stakeholders and maintain a commitment to always be accountable to them with responsiveness, fairness and honesty. These actions can foster trust among the public towards these public organisations.
SOEs in African countries are tasked with providing services that meet public needs. However, they are also supposed to make some financial gains from their operations. These companies must also act responsibly and ethically for the benefit of all stakeholders. Moreover, they must transparently inform the public about their performance. Corporate governance mechanisms are the key to achieving all these objectives.
We use agency and stakeholder theories to provide a comprehensive analysis of sustainability disclosures by SOEs in SSA countries. Specifically, agency theory is used to highlight the numerous corporate governance challenges and agency problems that exist within SOEs in emerging economies. Meanwhile, stakeholder theory is used to explain the disclosure of sustainability information to manage SOE stakeholders. Integrating these two theories helps capture the role of corporate governance mechanisms and stakeholder needs in promoting and shaping sustainability reporting by the SOEs of SSA countries.
3. Literature review and hypothesis development
3.1 Board size
Several studies have examined the role of board characteristics in sustainability reporting performance. For instance, scholars have explored the relationship between board size and sustainability reporting (Abang’a and Tauringana, 2024; Beji et al., 2021; Elmagrhi et al., 2021; Husted and Sousa-Filho, 2019; Nicolo’ and Andrades‐Peña, 2024). This is because for effective board functioning, a certain number of directors are needed to form the board. From the agency theory perspective, a firm with many directors can benefit from diverse expertise, improved monitoring and better control of the managers, which can strengthen its sustainability reporting practices. However, a large board size could also lead to inefficiencies on the part of the board (Prado-Lorenzo and Garcia-Sanchez, 2010) due to delayed decision-making, ineffective communication, poor monitoring and control, thereby affecting the firm’s accountability (Dey, 2008). From the stakeholder theory perspective, large boards are likely to represent diverse stakeholders whose needs and expectations must be managed by the firm. Most corporate governance codes in SSA and elsewhere worldwide have board-size requirements. SOEs are required to adhere to these requirements.
However, prior empirical results on this relationship have been mixed. Abang’a and Tauringana (2024) reported an insignificant association between board size and the extent of sustainability reporting of Kenyan SOEs. Other studies have also reported an insignificant relationship between board size and sustainability reporting (Elmagrhi et al., 2021; Ntim et al., 2017). Conversely, Nicolo’ and Andrades‐Peña (2024) found a negative significant relationship between board size and sustainability reporting by SOEs. Few studies have reported positive findings on board size and sustainability reporting (Bhatia and Makkar, 2019; Ben Fatma and Chouaibi, 2021). Based on agency and stakeholder theories’ arguments on board size and the positive findings of some empirical works, we hypothesise the following:
Board size and the extent of sustainability reporting by SOEs in SSA are positively related.
3.2 Board independence
According to agency theory, one of the most effective monitoring and control mechanisms in an organisation is the presence of independent directors (Hussain et al., 2018). This is because independent directors can effectively and constantly check the organisation’s management to promote transparency and accountability. From a stakeholder theory perspective, having more independent directors reflects diverse stakeholder representations in the organisation (Jizi et al., 2014). Additionally, independent directors have greater autonomy, and face less pressure from investors or owners to make self-servient decisions. The relationship between board independence and sustainability reporting has been explored, with the majority of studies reporting a positive relationship (Abang’a and Tauringana, 2024; Ben Fatma and Chouaibi, 2021; Bhatia and Makkar, 2019; Elmagrhi et al., 2021; Nicolo’ and Andrades‐Peña, 2024). However, Abang’a and Tauringana (2024) found an insignificant association between board independence and SOE sustainability reporting. Nicolo’ and Andrades‐Peña (2024) reported a significant positive relationship using a global SOE sample. Hussain et al. (2018) also reported a significant positive relationship for non-SOEs. Jizi et al. (2014) reported similar positive findings using a US sample in their study. Based on the agency and stakeholder theories’ positions and findings of previous studies, we expect a positive relationship between board independence and sustainability reporting among SOEs in SSA. Accordingly, we hypothesise the following:
Board independence and the extent of sustainability reporting by SOEs in SSA are positively related.
3.3 Board gender diversity
The relationship between board gender diversity and organisational outcomes, particularly sustainability reporting, has been a topic of considerable academic interest. From a stakeholder theory perspective, female directors are concerned about others’ welfare, especially that of children and youth. This makes them more protective of the environment (Desrochers et al., 2019) and seek social initiatives that are beneficial to the public (Orij, 2010). Thus, female directors can induce the board and management to use sustainability disclosures to manage the expectations of various stakeholders (Konadu et al., 2022). From an agency theory perspective, including women on the board enhances its effectiveness, especially with their high rate of meeting attendance (Adams and Ferreira, 2009). This enables them to discuss their social and environmental concerns, leading to better sustainability disclosures.
Indeed, women in Africa for so long have had fewer opportunities to be members of the top hierarchy of organisations, as most African countries are patriarchal (Areneke et al., 2023). Hence, the issue of gender equality is prevalent on the continent (Mori, 2014). However, African women have recently assumed various leadership positions in companies and even national governance (Dawuni and Kang, 2015). This can help tackle gender inequality on the continent (Abang’a and Tauringana, 2024)
However, research on the relationship between board gender diversity and sustainability reporting has yielded mixed results (Abang’a and Tauringana, 2024; Argento et al., 2019; De Masi et al., 2021; Elmagrhi et al., 2021; Nicolo’ and Andrades‐Peña, 2024; Ntim et al., 2017). For instance, Elmagrhi et al. (2021) reported a negative relationship in the UK. Conversely, Ntim et al. (2017) also reported a significant positive relationship in UK higher education institutions. In the context of SOEs, Nicolo’ and Andrades‐Peña (2024) reported a positive influence of board gender diversity on sustainability reporting. However, Abang’a and Tauringana (2024) reported a negative relationship between board gender diversity and sustainability reporting among Kenyan SOEs. Meanwhile, Argento et al. (2019) and Garde Sánchez et al. (2017) found no relationship for Swedish and Spanish SOEs, respectively. Other studies (Manita et al., 2018; De Masi et al., 2021; Masoud and Vij, 2021) also reported an insignificant association, suggesting that a low number of female directors does not significantly influence sustainability disclosures. Still, we expect women on SOE boards to show more social and environmental concern towards the public. Accordingly, we hypothesise the following:
Board gender diversity and the extent of sustainability reporting by SOEs in SSA are positively related.
3.4 Board meetings
Agency theory (Jensen and Meckling, 1976) argues that frequent board meetings are a corporate governance tool for dealing with the problem of information asymmetry in organisations. Additionally, Orazalin and Mahmood (2021) suggested that board meetings contribute to the board’s due diligence practices, allowing them to deliberate on issues concerning the organisation. In effect, this can strengthen the organisation’s transparency and accountability towards its various stakeholders (Hussain et al., 2018).
Several studies have examined the relationship between the frequency of board meetings and sustainability reporting. While some studies reported an insignificant relationship (Abang’a and Tauringana, 2024; Aly et al., 2024; Garcia-Sanchez et al., 2014; Giannarakis, 2014), others reported a positive relationship (Elmagrhi et al., 2021; Jizi et al., 2014; Nicolo’ and Andrades‐Peña, 2024). Meanwhile, a study on Kenyan SOEs showed an insignificant impact (Abang’a and Tauringana, 2024). However, Nicolo’ and Andrades‐Peña (2024) reported a positive relationship using a global sample. We agree with the argument that frequent board meetings help organisations become diligent, transparent and accountable to the public. Accordingly, we hypothesise the following:
The frequency of board meetings and the extent of sustainability reporting by SOEs in SSA are positively related.
4. Research design
4.1 Sample
The sample included 78 SOEs operating in 16 SSA countries. While choosing the sample, we first identified all English-speaking countries in SSA. Given the authors’ lack of proficiency in other languages, this ensured that we only selected SOEs whose reports were issued in English. Second, we limited the country selection to only countries where SOEs issue financial reports following the International Financial Reporting Standards. This was done to improve the comparison of the financial reports of SOEs of different countries. Next, we eliminated all SOEs that had not published reports in our sample period. Table 1 lists the countries and the number of SOEs in our sample.
4.2 Data
We gathered data from the annual, CSR, sustainability and integrated reports of 78 SOEs during 2012–2022. However, not all SOEs had all 11-year yearly reports. Some SOEs were missing certain variables. Consequently, the final sample comprised an unbalanced panel of 698 observations.
4.3 Measurement of variables
4.3.1 Dependent variables.
The main dependent variable was the sustainability reporting index (SRI), which comprised the environmental (ENV) and social subindices (SOC). These variables were generated by analysing the content of SOEs’ annual and sustainability-related reports. The index and subindices were developed following the methodology used in previous studies, in which the presence or absence of an item of disclosure in the checklist was used to measure the extent of reporting (Argento et al., 2019; Greiling et al., 2015; Masoud and Vij, 2021). The authors used 12 environmental and 7 social items to build the SRI. This index was used to measure the extent of reporting by the SOEs and served as the main dependent variable. Appendix provides a list of the items used to build the index.
4.3.2 Independent variables.
The independent variables include board size (BODSZ), board gender diversity (BOGD), board independence (BODIND) and frequency of board meetings (BODMET). These variables were collected from the SOEs’ annual reports.
We examined board characteristics and sustainability reporting in SSA SOEs because boards are key to governance, driving performance and strategic initiatives like sustainability reporting (Post et al., 2015; Tang et al., 2020). Certain board features, like independence, meetings, size and gender diversity, are critical for achieving these goals. As such, many corporate governance frameworks worldwide have mandatory requirements for board size, independence and meetings, while the implementation of gender diversity quotas keeps rising (Abang’a and Tauringana, 2024). Again, unlike other board characteristics, data on these variables are readily available in the annual reports of SOEs. There were instances in which we sent emails to some SOEs for data, but never received a response. We thus focused on these variables in our study as we wanted to avoid excluding more SOEs from our sample due to data unavailability. We have stated this as a limitation of our study and advocated for future research to focus on variables like sustainability committees, audit committees, educational background and board experience.
Additionally, we included four control variables based on previous studies (Abang’a and Tauringana, 2024; Nicolo’ and Andrades‐Peña, 2024): SOE size (SOESIZE), leverage (LEV), SOE years (AGE) and profitability (ROA). The size of an SOE was chosen because large SOEs are more visible to the public and face more considerable scrutiny. Therefore, they make sustainability disclosures to satisfy and manage their many stakeholders. Leverage was included because highly leveraged SOEs are unlikely to engage in extensive sustainability reporting. Age was also included because SOEs that have existed for a long time possess the resources to undertake special initiatives, like sustainability reporting. Finally, SOEs with more profits can engage in sustainability reporting as they can employ the services of sustainability experts, make extensive disclosures and use them to signal their commitment to environmental and social initiatives. Together, these variables helped isolate the predictive powers of the primary dependent variables. The operationalisation of all the variables is reported in Table 2.
Operationalisation of variables
| Variable | Operationalisation | Previous studies |
|---|---|---|
| Dependent variables | ||
| Sustainability score | Items reported/total items to be reported | Greiling et al. (2015) and Manes-Rossi et al. (2020) |
| Environmental score | Items reported/total items to be reported | Greiling et al. (2015) and Manes-Rossi et al. (2020) |
| Social score | Items reported/total items to be reported | Greiling et al. (2015) and Manes-Rossi et al. (2020) |
| Independent variables | ||
| Board size | Total number of directors | Abang’a and Tauringana (2024) and Nicolo’ and Andrades‐Peña (2024) |
| Board gender diversity | Proportion of female directors | Argento et al. (2019) and Nicolo’ and Andrades‐Peña (2024) |
| Board independence | Proportion of independent directors | Abang’a and Tauringana (2024) and Elmagrhi et al. (2021) |
| Board meetings | Number of meetings held annually | Abang’a and Tauringana (2024) and Nicolo’ and Andrades‐Peña (2024) |
| Control variables | ||
| SOE size | Natural log of total assets | Abang’a and Tauringana (2024) and Nicolo’ and Andrades‐Peña (2024) |
| Leverage | Total debts/equity | Abang’a and Tauringana (2024) and Nicolo’ and Andrades‐Peña (2024) |
| Profitability | Return on assets | Abang’a and Tauringana (2024) and Nicolo’ and Andrades‐Peña (2024) |
| Firm age (AGE) | Number of years since incorporation | Abang’a and Tauringana (2024) |
4.4 Model specification
We used panel data regression to test the hypotheses. To choose the best model, we performed various estimations, including ordinary least squares (OLS) and random- and fixed-effects regressions. The Breusch–Pagan test for random effects yielded a value of Prob > Chibar2 = 0.0000. This indicates that the random effects model was the best compared with the pooled OLS model. A Hausman test was performed to choose between random- and fixed-effects models. The resulting Prob > Chi2 = 0.000 indicated that the fixed effects was the best model. Hence, a fixed-effect was used to continue the study. Nonetheless, random effects and system GMM estimations were used to check the robustness of the results.
4.4.1 Empirical model
Notes:
β0 = Constant;
β1-8 = Variable coefficient;
SRI = Sustainability reporting index;
BODSZ = Board size;
BODIND = Board independence;
BOGD = Board gender diversity;
BODM = Board meetings;
ROA = Return on Assets;
LEV = Leverage;
SOESIZE = SOE size;
AGE = SOE age; and
ε = error term.
5. Results
5.1 Univariate analysis
Table 3 reports the summary statistics. The average SRI reporting score during the sample period is 36.06%. This indicates that the extent of reporting by the sampled SOEs during the sample period is low. Among the dimensions, the environmental dimension has the lowest reporting level, with a mean value of 17.62%, followed by the social dimension (54.5%). The high social dimension score supports prior findings that SOEs disclose more social information (Tagesson et al., 2009). The results also indicate that the environmental dimension receives less attention in disclosures, whereas the social dimension is prioritised. The size of the board ranges from 3 to 23, with an average of 10 members. Board gender diversity ranges from 0 to 85.71%, with an average of 30.59% women on the boards of SOEs. On average, 80% of the board members are independent directors, with a minimum of 25%. The average number of board meetings held by SOEs per year is approximately 8. SOEs in SSA, on average, make losses of around 5% (−0.048). The average size of SOEs in Africa is not distinct; the smallest size is 3.16, while the average is 5.83.
Descriptive statistics
| Variable | Obs | Mean | SD | Min. | Max. |
|---|---|---|---|---|---|
| SRI | 698 | 36.06 | 13.40 | 0 | 65.93 |
| ENV | 698 | 17.62 | 8.93 | 0 | 69.23 |
| SOC | 698 | 54.50 | 22.16 | 0 | 100 |
| BODSZ | 698 | 9.99 | 3.32 | 3 | 23 |
| BODIND | 698 | 80.27 | 13.17 | 25 | 100 |
| BOGD | 698 | 30.59 | 17.12 | 0 | 85.71 |
| BODM | 698 | 7.90 | 4.73 | 1 | 42 |
| ROA | 698 | −0.05 | 1.36 | −32.00 | 2.52 |
| LEV | 698 | 0.64 | 0.54 | 0.01 | 8.69 |
| SOESIZE | 698 | 5.83 | 0.96 | 3.16 | 8.11 |
| AGE | 698 | 43.63 | 26.50 | 1 | 119 |
The variables are explained as follows: sustainability reporting index is SRI; environmental reporting index is ENV; social reporting index is SOC; board size is BODSZ; board independence is BODIND; board gender diversity is BOGD; board meeting is BODM; return on assets is ROA; leverage is LEV; SOE size is SOESIZE; and firm age is AGE
5.2 Bivariate analysis
Table 4 reports the bivariate analysis results. The two sub-dependent variables (ENV and SOC) strongly and positively correlate with the primary dependent variable, SRI. In terms of the independent and dependent variables, board size positively correlates with SRI and all sub-independent variables (ENV and SOC). Board independence does not exhibit any significant correlation with the dependent variables. Board gender diversity shows a significant negative association with SRI. Board meetings have a positive relationship with the dependent variables, except for environmental scores. Our data does not suffer from multicollinearity, as the highest correlation between the two independent variables is 0.478. According to Gujarati (2009), multicollinearity is serious when the correlation between two independent variables exceeds 0.8. Moreover, the variance inflation factor test yields a mean value of 1.72, indicating the absence of multicollinearity (Gujarati and Porter, 2004).
Correlation matrix
| Variables | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| (1) ENVSOC | 1.000 | ||||||||||
| (2) ENV | 0.641*** | 1.000 | |||||||||
| (3) SOC | 0.951*** | 0.372*** | 1.000 | ||||||||
| (4) BODSZ | 0.149*** | 0.240*** | 0.084** | 1.000 | |||||||
| (5) BODIND | −0.032 | 0.031 | −0.051 | 0.034 | 1.000 | ||||||
| (6) BOGD | −0.119*** | 0.003 | −0.145*** | −0.047 | 0.151*** | 1.000 | |||||
| (7) BODM | 0.075** | 0.051 | 0.070* | 0.117*** | −0.010 | 0.072* | 1.000 | ||||
| (8) ROA | 0.122*** | 0.081** | 0.115*** | −0.017 | 0.051 | −0.015 | −0.059 | 1.000 | |||
| (9) LEV | 0.023 | 0.022 | 0.019 | 0.002 | −0.068* | −0.126*** | 0.008 | −0.023 | 1.000 | ||
| (10) SOESIZE | 0.273*** | 0.282*** | 0.217*** | 0.262*** | −0.218*** | 0.033 | 0.119*** | 0.053 | 0.086** | 1.000 | |
| (11) AGE | 0.118*** | 0.128*** | 0.091** | 0.342*** | −0.112*** | −0.084** | 0.250*** | −0.102*** | 0.103*** | 0.478*** | 1.000 |
***p < 0.01; **p < 0.05 and *p < 0.1
5.3 Regression analysis
Table 5 reports the regression analysis results. Board size (BODSZ) has a positive insignificant relationship with SRI (β = 0.201, ρ = 0.493), ENV (β = 0.320, ρ = 0.353) and SOC (β = 0.082, ρ = 0.866). Similarly, board independence (BODIND) exhibits a positive insignificant relationship with SRI (β = 13.300, ρ = 0.214), ENV (β = 5.759, ρ = 0.205) and SOC (β = 20.841, ρ = 0.231). Board gender diversity (BOGD) shows a negative insignificant relationship with SRI (β = −8.1050, ρ = 0.382), ENV (β = −1.322, ρ = 0.765) and SOC (β = −14.888, ρ = 0.303). While board meeting (BODM) has a positive insignificant relationship with SRI (β = 0.057, ρ = 0.536) and SOC (β = 0.003, ρ = 0.985), it has a significant relationship with ENV (β = 0.111, ρ = 0.083). Among the control variables, profitability (ROA) has a positive significant relationship with SRI, ENV and SOC (β = 1.076, ρ = 0.000; β = 0.488, ρ = 0.003; β = 1.665, ρ = 0.000). Leverage (LEV) exhibits a positive insignificant relationship with SRI, ENV and SOC (β = 0.091, ρ = 0.894; β = 0.084, ρ = 0.855; β = 0.099, ρ = 0.933). SOESIZE has a positive significant association with all dependent variables (β = 2.974, ρ = 0.046; β = 1.083, ρ = 0.091; β = 4.865, ρ = 0.046). Firm age (AGE) has an insignificant association with SRI (β = −0.062, ρ = 0.299) and ENV (β = 0.073, ρ = 0.309) and a significant association with SOC (β = −0.198, ρ = 0.000).
Fixed and random effects’ regression results
| Variables | Fixed effect | Random effect | ||||
|---|---|---|---|---|---|---|
| M 1 SRI | M2 ENV | M 3 SOC | M 4 SRI | M 5 ENV | M 6 SOC | |
| BODSZ | 0.2008 (0.6892) | 0.3201 (0.9352) | 0.0816 (0.1696) | 0.2282 (0.9221) | 0.3699 (1.2197) | 0.0829 (0.1999) |
| BODIND | 13.2999 (1.2528) | 5.7593 (1.2778) | 20.8405 (1.2079) | 11.2634 (1.2131) | 5.6398 (1.3732) | 17.4755 (1.1360) |
| BOGD | −8.1050 (−0.8795) | −1.3220 (−0.3004) | −14.8881 (−1.0377) | −8.0508 (−1.0608) | −0.6199 (−0.1647) | −15.4546 (−1.2663) |
| BODM | 0.0570 (0.6211) | 0.1111* (1.7547) | 0.0029 (0.0184) | 0.0717 (0.8700) | 0.0971* (1.6742) | 0.0415 (0.2712) |
| Control | ||||||
| ROA | 1.0764*** (4.5935) | 0.4879*** (3.1062) | 1.6650*** (4.8796) | 0.9675*** (4.7426) | 0.3975** (2.4322) | 1.5537*** (5.5071) |
| LEV | 0.0911 (0.1340) | 0.0837 (0.1830) | 0.0985 (0.0850) | 0.2409 (0.4195) | 0.2059 (0.4634) | 0.2626 (0.2485) |
| SIZE | 2.9739** (2.0322) | 1.0833* (1.7140) | 4.8646** (2.0248) | 3.9108*** (3.6438) | 1.8114*** (2.9393) | 5.8751*** (3.3078) |
| AGE | −0.0622 (−1.0455) | 0.0734 (1.0240) | −0.1978*** (−3.6434) | −0.0076 (−0.1861) | 0.0152 (0.4542) | −0.0311 (−0.4705) |
| _cons | 4.0009 (0.3725) | −3.4230 (−0.4956) | 11.4249 (0.6735) | −3.4699 (−0.3937) | −5.9064 (−1.0208) | −0.6216 (−0.0436) |
| N | 698 | 698 | 698 | 698 | 698 | 698 |
| R-sq | 0.1868 | 0.0999 | 0.1857 | 0.1909 | 0.1249 | 0.1843 |
| F-test | 10.19*** | 4.12*** | 17.73*** | |||
| Wald chi2 | 207.61*** | 85.92*** | 282.92*** | |||
***p < 0.01; **p < 0.05 and *p < 0.1. Robust t/z-statistics in parentheses
5.4 Robust analyses
Two additional estimations were executed to confirm the robustness of the results and determine whether they would hold under different regression techniques. The first estimation was performed using random effects regression. Owing to space constraints, we also present the random effect results in Table 5 under the heading random effects. This also facilitated a quick comparison of the findings. The results obtained using both techniques are very similar. Only AGE showed a significant negative relationship with SOC. Thus, the results generally remained consistent for the SRI and its subdimensions. Thus, the findings are robust. To further prove the robustness of our results regarding endogeneity problems, we used the system GMM estimation technique. The results are presented in Table 6. Again, the findings are very similar to the results of both the fixed- and random-effects analyses.
System GMM regression results
| Variables | M1 | M2 | M3 |
|---|---|---|---|
| SRI | ENV | SOC | |
| Dependent variables | |||
| L.SRI | 0.3685*** (2.8541) | ||
| L.ENV | 0.4006*** (4.2479) | ||
| L.SOC | 0.3871** (2.3022) | ||
| Independent variables | |||
| BODSZ | 0.1206 (0.6153) | 0.2292 (1.4286) | 0.7592 (1.1241) |
| BODIND | 4.4263 (0.8169) | 2.8498 (0.7494) | 5.8883 (0.5333) |
| BOGD | −2.5622 (−0.6716) | −0.1668 (−0.0901) | −13.2576 (−1.1208) |
| BODM | 0.0826 (0.7090) | 0.0151 (0.2156) | 0.5755 (1.6079) |
| Control variables | |||
| ROA | 0.9325*** (3.5219) | 0.4439* (1.9116) | 2.6370 (0.6640) |
| LEV | 1.4806 (0.9241) | 0.0776 (0.2088) | 2.8903 (0.8838) |
| SOESIZE | 2.4710** (2.2967) | 1.1192** (2.3876) | 9.7386** (2.0304) |
| AGE | −0.0186 (−0.5813) | −0.0066 (−0.2582) | −0.4213 (−1.4692) |
| _cons | −1.7668 (−0.2212) | −2.5302 (−0.4790) | −26.3407 (−1.1031) |
| N | 613 | 613 | 613 |
| Year effects | YES | YES | YES |
| Arellano–Bond (AR1) | −2.70 (p = 0.007) | −2.95 (p = 0.003) | −2.69 (p = 0.007) |
| Arellano–Bond (AR2) | −0.90 (p = 0.368) | −1.64 (p = 0.101) | −1.40 (p = 0.162) |
| Sargan | 117.54 (p = 0.000) | 164.59 (p = 0.000) | 99.83 (p = 0.000) |
| Hansen | 59.94 (p = 0.239) | 57. 50 (p = 0.312) | 50.09 (p = 0.430) |
***p < 0.01; **p < 0.05 and *p < 0.1. L. SRI, L. ENV and L. SOC represent the lagged values of the dependent variables. t-statistics in parentheses
Next, to further enrich our analysis, we ran a regression based on whether the SOEs were listed and unlisted. This was done to check whether the results would change based on the listing status of SOEs, as the listing status has been suggested to influence sustainability disclosures. The results show that the relationship between board characteristics and sustainability reporting is the same for listed and unlisted SOEs. The results are summarised in Table 7.
Regression of listed and unlisted SOE
| Variables | Fixed effect | Random effect | ||
|---|---|---|---|---|
| Listed | Unlisted | Listed | Unlisted | |
| SRI | SRI | SRI | SRI | |
| BODSZ | 0.0133 (0.0196) | 0.1983 (0.6002) | 0.7597 (1.2848) | 0.1981 (0.7061) |
| BODIND | 12.1843 (0.7956) | 16.3836 (1.2993) | 21.1274 (1.6098) | 11.9462 (1.1007) |
| BOGD | 15.9898 (1.2543) | −9.8599 (–0.9981) | −9.3076 (–0.5883) | −8.1103 (–1.0098) |
| BODM | −0.1474 (–0.7489) | 0.1115 (1.0690) | 0.2631 (0.8816) | 0.0850 (0.8858) |
| Control variables | ||||
| ROA | –9.5519 (–1.4059) | 1.2807*** (8.5975) | −11.8016 (–1.5546) | 1.1148*** (8.8061) |
| LEV | −8.9308 (–1.1678) | −0.1809 (–0.2886) | −17.4099*** (–3.6823) | 0.2388 (0.4449) |
| SOESIZE | 1.5570 (0.1647) | 2.2658 (1.5935) | 3.8374* (1.6739) | 3.7257*** (3.6038) |
| AGE | –0.7205 (–0.1777) | –0.0524 (–0.8455) | 0.0456 (0.6977) | −0.0094 (–0.2312) |
| _cons | 50.7263 (0.2383) | 5.4405 (0.5214) | −8.7641 (–0.5489) | −2.9506 (–0.3142) |
| N | 101 | 597 | 101 | 597 |
| R-sq | 0.4607 | 0.1913 | 0.3157 | 0.1874 |
| F-test | 701.98*** | |||
| Wald chi2 | 22.51** | 66.59*** | 7885.71*** | |
***p < 0.01; **p < 0.05 and *p < 0.1. Robust t/z-statistics in parentheses
6. Discussion
Overall, the panel regression results indicate that board size has a positive insignificant relationship with the extent of sustainability reporting. Moreover, the results are consistent across all the models. Thus, we reject H1, which proposes that board size has a positive significant relationship with the extent of sustainability reporting. These findings align with prior studies that reported an insignificant relationship between board size and the extent of sustainability reporting (Abang’a and Tauringana, 2024; Nicolò et al., 2019; Ntim et al., 2017). In particular, the consistency with Abang’a and Tauringana’s (2024) study on Kenyan SOEs depicts the relevance of contextual factors in sustainability reporting. However, our conclusion contradicts the findings of Nicolo’ and Andrades‐Peña (2024), who reported a significant negative association. This may be because the aforementioned authors only included a single African SOE in their sample, limiting their ability to adequately reflect SSA realities compared with the present study, which has a representative sample of SOES from SSA. The finding also holds true for the environmental (M2) and social (M3) dimensions, indicating that a large board size is not associated with the extent of social and environmental disclosures within the SSA public sector.
This may be because board appointments to SOEs in SSA are mostly political, without considering the expertise, skills and competencies of the person being appointed (Abang’a and Tauringana, 2024; Tsamenyi et al., 2007). Consequently, the government may appoint many directors who lack expertise on sustainability-related issues, worsened by coordination problems and inefficiencies associated with large board sizes, as suggested by agency theory. Furthermore, these appointments, though deemed a monitoring and transparency tool for the government, actually weaken the oversight powers of the government over the appointees. As a result, the directors tend to be ineffective in promoting transparency and sustainability disclosures within the public sector. This contradicts agency theory, which proposes that large boards benefit from diverse skillsets that promote strong monitoring, mitigate information asymmetry and create a balance between the objectives of management and those of the organisation. Additionally, these appointments may not actually reflect broader stakeholder active participation in SOEs’ decision processes or serve as responsive mechanisms to their demands for proper accountability. The appointees mostly represent their political parties and appointers rather than the public, who are the ultimate principals or stakeholders. This leads to low transparency and poor accountability and contradicts the postulations of stakeholder theory. Finally, due to the low awareness of sustainability reporting among people in SSA countries (Tilt et al., 2021), even with a large board size, the absence of pressure will not make the SOE board focus on sustainability disclosures. Accordingly, the theoretical arguments of stakeholder and agency theories that a large board enhances corporate transparency, reflects diverse stakeholder representation and strengthens checks and balances are flawed in the context of SOEs in SSA, as board size shows no significant relationship with the extent of sustainability reporting. This result emphasizes the effectiveness of SOE board members and the active participation of stakeholders as critical factors in enhancing sustainability reporting in public sector organisations, rather than just appointing many directors.
Next, board independence did not show any significant association with SRI or the environmental and social disclosure indices. Accordingly, we reject H2, which proposes a positive relationship between board independence and sustainability reporting. This aligns with the findings of Abang’a and Tauringana (2024) but contradicts those of Nicolo’ and Andrades‐Peña (2024). As previously mentioned, inadequate African representation in the sample of SOEs used by the latter authors and contextual factors are key to this contradiction. Unlike SSA, the highly represented regions in the sample may have strong corporate governance mechanisms that promote sustainability disclosures. Our study provides empirical evidence of SOEs in the unique context of SSA, suggesting that analysis should be conducted at the regional level and not at the global level. This is unsurprising considering that government nepotism and favouritism often influence board appointments to SOEs in SSA (Tsamenyi et al., 2007). Consequently, independent directors’ monitoring capabilities, oversight powers and pressure on SOE managers to embark on valuable initiatives, as suggested by agency theory, may be compromised (Abang’a and Tauringana, 2024). This leads to reduced attention to sustainability initiatives and reporting within the public sector of emerging economies.
Second, CEO duality is prevalent in African SOEs, where the government appoints one person as both CEO and board chairperson. This consolidation of power could subject independent directors to excessive pressure and neutralise their power to monitor and compel SOE managers to embark on initiatives that benefit the citizens (Nicolo’ and Andrades‐Peña, 2024; Voinea et al., 2022). This adversely affects accountability and transparency within public sector organisations. Additionally, independent directors might only seek their interests and those of their political parties, and not the wider public, contrary to the postulations of stakeholder theory. Furthermore, their appointments are likely to be symbolic and fulfilment of regulatory requirements, as most corporate governance frameworks stipulate a certain proportion of independent directors. This can make independent directors more passive in organisational decisions and affect their ability to enhance transparency and reporting practices within the public sector of emerging economies. Overall, our finding fails to support the stakeholder and agency theories’ positions that independent directors represent diverse stakeholders’ interests, serve as effective monitoring mechanisms in SOEs and promote proper accountability in ways that improve the extent of sustainability reporting. This result calls for the promotion of quality and active collaboration between directors of SOEs and the empowerment of independent directors to promote sustainability reporting and accountability within the public sector.
Similarly, board gender diversity has an insignificant negative association with the extent of sustainability reporting. Similar results are observed for the environmental and social dimensions. Thus, we reject H3, which states that board gender diversity has a significant positive relationship with the extent of SOE sustainability reporting in SSA. Overall, this finding contradicts prior studies that reported a significant association (Abang’a and Tauringana, 2024; Nicolo’ and Andrades‐Peña, 2024). Meanwhile, the result supports the findings of Nicolò et al. (2019), who reported an insignificant association.
The contradictory findings could result from the small sample size, the use of only annual reports to capture SOE sustainability disclosure, a focus on a single country, the short data period (Abang’a and Tauringana, 2024) and not addressing endogeneity issues in the data (Abang’a and Tauringana, 2024; Nicolo’ and Andrades‐Peña, 2024). Our study contributes to the literature by addressing all the limitations of the above studies. The alignment of our study with that of Nicolò et al. (2019), who focused on 34 European SOEs, supports our claim that regional analysis is more essential than global studies due to contextual factors. Thus, we extend the literature by being the first to provide a regional analysis of SOE sustainability reporting in the SSA context.
Further, the summary statistics show that African SOEs have very few female directors. Until recently, they were not given the necessary opportunities and support to occupy top positions in organisations. However, this could only be symbolic or politically motivated appointments that do not necessarily promote better stakeholder management or reflect diverse stakeholder representation and advocate for their needs through high-quality disclosure, such as sustainability disclosures. This contradicts what stakeholder theory arguement about the presence of women on the board. Furthermore, their small size might lead to the suppression of their views, prevent them from exerting the required influence or make their voices heard adequately during the board’s decision-making. This reduces their active participation in board processes and ability to enhance the board’s oversight responsibility on SOE managers, contrary to what the agency theory suggests. Again, they may also align themselves easily with the positions of their male counterparts regarding corporate decisions. This may explain their insignificant influence on sustainability reporting. Moreover, their appointment is primarily made by the government; hence, it is possible to focus on financial performance instead of sustainability disclosures. Thus, the stakeholder and agency theories’ arguments that board gender diversity makes the board effective and improves accountability to stakeholders through sustainability disclosures are not supported. This finding suggests that the appointment of women to the board of SOEs should be complemented with an enabling environment for their voices to be heard and valued during corporate decision-making to promote accountability and transparency within public sector organisations.
Finally, board meetings do not show any significant association with the extent of sustainability reporting. This result fails to support H4 but aligns with the findings of prior research (Abang’a and Tauringana, 2024). However, these results contradict the findings of Nicolo’ and Andrades‐Peña (2024). Meanwhile, we observe a positive and significant association for environmental disclosures. This may be because of the low African SOEs representation in the aforementioned study. Our study solves this problem and presents a true reflection of SOEs in SSA.
A potential explanation for the lack of support for H4 could be that the directors are not interested in sustainability issues. Even if they attend more meetings, sustainability issues remain peripheral to other topics and would not significantly impact sustainability reporting. Thus, frequent board meetings, which should indicate effective board oversight, informational flow, enhanced accountability and disclosures as per agency theory, do not hold in the SSA public sector. This is because these meetings are merely undertaken to fulfil compliance requirements and do not enhance board efficacy. Furthermore, the insignificant result raises doubts about whether the issues discussed during meetings are relevant for non-shareholders and creates an opportunity for the directors to actively engage with stakeholder concerns, as it should be according to the stakeholder theory. The boards of directors of SOEs in SSA are likely to receive per diem and other benefits based on the number of meetings attended. This may influence the number of board meetings held in a year; moreover, discussions at such meetings may be of low quality or irrelevant to other stakeholders’ needs. Hence, frequent board meetings do not necessarily lead to proper SOE accountability or disclosure practices in SSA. Therefore, the theoretical arguments of stakeholder and agency theories, which state that frequent board meetings help organisations to be diligent, transparent and accountable to the public, are not supported in this context. This finding highlights the need for public sector organisations to prioritise meetings characterised by quality and valuable discussions on sustainability and accountability instead of frequent compliance-filling non-impactful meetings.
Among the control variables, profitability has a positive relationship with the extent of sustainability reporting in all models. This finding contradicts those of prior studies that reported an insignificant association (Abang’a and Tauringana, 2024; Argento et al., 2019; Nicolo et al., 2021; Nicolo’ and Andrades‐Peña, 2024). Thus, SOEs in SSA only make sustainability disclosures when they make a profit. Leverage shows an insignificant positive association with the extent of sustainability reporting in all models. This may be because SSA SOEs usually seek support from the government during periods of cash constraints, using this funding for their core operations rather than for sustainability ventures. The SOE size exhibited a significant positive relationship with the extent of sustainability reporting. This result supports prior studies with similar outcomes (Argento et al., 2019; Garde Sánchez et al., 2017; Masoud and Vij, 2021) but contradicts others (Abang’a and Tauringana, 2024; Nicolo’ and Andrades‐Peña, 2024).
7. Conclusion
We examined the extent of sustainability reporting by SOEs in SSA and explored the relationship between selected board characteristics and the extent of sustainability reporting. Data was collected from the annual and sustainability-related reports of 78 SOEs in SSA during 2012–2022. Panel data regression was used to test the hypotheses. The findings show that the extent of sustainability reporting by SOEs in SSA is very low. None of the examined board characteristics (board size, independence, gender diversity and meetings) had a significant relationship with the extent of sustainability reporting. The environmental dimension receives lower disclosure attention than the social dimension.
This study makes several contributions to the sustainability reporting literature. First, it is among the few studies to examine the relationship between board characteristics and sustainability reporting by listed and unlisted SOEs in SSA countries. We show that the main findings hold for both listed and unlisted SOEs. We make a crucial contribution by providing empirical evidence from SSA, a region characterised by weak institutions and high political interference in SOEs’ corporate governance. Second, we contribute to the literature on the relationship between board characteristics and the extent of SOEs’ sustainability reporting, particularly in emerging economies. The literature includes a few such studies (Nicolo’ and Andrades‐Peña, 2024). Third, the study performs trend analysis and helps generalise the findings by using 11-year data from 78 SOEs from 16 different African countries, unlike prior studies that focus on a single country and industry or use a short sample period (Abang’a and Tauringana, 2024; Argento et al., 2019; Garde Sánchez et al., 2017; Masoud and Vij, 2021; Rahaman et al., 2004). Thus, we extend the literature of studies that focused on regional-level analysis by using SOEs from a particular geographical region (Manes-Rossi et al., 2020; Nicolò et al., 2019). Fourth, methodologically, we use the system GMM technique to check the robustness of our findings, potentially addressing endogeneity issues not addressed in prior studies and presenting more robust findings (Abang’a and Tauringana, 2024; Nicolo’ and Andrades‐Peña, 2024).
Finally, this study contributes to the global discourse on public sector sustainability reporting by improving our understanding of how sustainability reporting is evolving within the public sectors of economies with institutional voids, high political interference and weak corporate governance frameworks. We show that in such economies, traditional governance mechanisms and stakeholder concerns may not drive sustainability reporting as in developed economies. Accordingly, policies, standards and guidelines concerning sustainability reporting should be more context-specific than universal.
7.1 Theoretical and practical contributions
This study extends the literature on agency and stakeholder theories by showing that in the context of SOEs, traditional board characteristics do not drive sustainability reporting. This is contrary to theoretical suggestions. Specifically, the assumptions of agency theory that strong board mechanisms could help tackle information asymmetry, enhance transparency in organisations and mitigate managerial opportunism are challenged in the context of SSA SOEs. Several reasons could account for this, including the possibility that traditional governance mechanisms in SSA countries are merely symbolic, subject to high political interference and lack the required institutional support. This limits their ability to promote extensive sustainability disclosures. Effective and new sustainability-specific governance mechanisms, like board sustainability committees, sustainability officers and sustainability assurance committees (Agnese et al., 2024), backed by strong institutional mechanisms, could better improve SOEs’ sustainability reporting. From a stakeholder theory perspective, our findings suggest that relying on traditional corporate governance or board attributes and mere stakeholder representation is not enough to promote extensive sustainability reporting within the public sector. Factors like active stakeholder participation, pressure imposed by funding agencies or donors and strong institutional mechanisms seem to be effective means to drive sustainability disclosures in the sector (Rahaman et al., 2004). These mechanisms are likely to compel SOEs to engage in strong disclosure practices.
In general, the findings extend the agency and stakeholder theories as they highlight the significant role of contextual and institutional factors in enhancing corporate governance effectiveness and promoting enhanced disclosure practices within the SOEs. Thus, in the context of SOEs in emerging economies, effective corporate governance mechanisms and active stakeholder participation and dialogue, as proposed by agency and stakeholder theories, are more likely to impact sustainability reporting instead of mere board restructuring and stakeholder representation.
This study has several implications for SSA policymakers, regulators and managers. As shown by the low level of disclosure and tested relationships, making sustainability disclosures mandatory for all SOEs can be valuable to promote sustainability reporting. Additionally, corporate governance frameworks in the sub-region must be comprehensively strengthened to enhance sustainability reporting, as they still seem weak in facilitating enhanced disclosures. Specifically, we suggest that political interference in board appointments should be minimised to allow people with expertise to be appointed to SOE boards and serve the public interest. Furthermore, we call for specific institutions to be tasked with sustainability reporting issues. Moreover, managers must embed sustainability issues in SOE operations. Further, this study informs policymakers and managers to improve environmental disclosures, as they receive less attention than other dimensions. To achieve this, we recommend the formation and implementation of specialised environmental committees with designated environmental sustainability tasks. Among theoretical implications, the empirical results weaken the agency and stakeholder theories’ positions on board characteristics and general stakeholder interests, enhancing SOE sustainability reporting in SSA. This may imply that only regulations and external legitimacy pressures will be helpful for advancing reporting.
7.2 Limitations and future research avenues
First, we omit francophone countries from the sample. Future research should focus on SOEs in French-speaking countries to fill the gap in the literature. Second, future research can focus on specific sectors within the public sector to determine whether the findings hold and vary within these sectors. For instance, empirical evidence from state-owned financial institutions can substantially contribute to the literature, considering that many of these institutions are regulators. Furthermore, sustainability reporting is deemed to vary by sector. As such, a study on specific sub-sectors within the public sector will be highly valuable to the literature. Future studies can also examine other governance variables not examined here, like sustainability committees, board sustainability experience, board tenure, normalisation committees, CEO duality, sustainability assurance committees and the personal characteristics of SOE CEOs. These variables have been found to impact sustainability disclosure in the private sector, but have not yet been adequately tested in the public sector. The sustainability committee, as a new form of sustainability governance mechanism, has yet to receive the required academic attention, especially within the SOEs of emerging economies. Empirical evidence is needed to determine whether it is also symbolic or will significantly help in addressing stakeholders’ sustainability needs. A study of this nature will make a valuable contribution to the literature.
Acknowledgements
The authors wish to express their profound gratitude to the Editor in Chief, Professor Warren Maroun and the anonymous reviewers for their valuable comments and suggestions, which helped to improve the quality of the manuscript.
References
Further reading
Appendix
List of disclosure items
| Environmental index, max 12 | Social index, max 7 |
|---|---|
|
|
