Ownership structure and financial distress: is the tale from Sub-Saharan Africa different?

Abstract

Purpose

The paper investigates the effect of ownership structure on the financial distress of firms listed in sub-Saharan Africa.

Design/methodology/approach

Using secondary data from 106 non-financial firms listed in 9 selected SSA countries from 2016 to 2021, the research using paired t-tests and conditional logistic regression model analysed a sample of 174 distressed observations matched with 174 non-distressed observations.

Findings

T-tests determined significant differences between distressed and non-distressed groups concerning institutional, foreign, and local ownership. Conditional logistic results established that institutional, foreign, and state ownership significantly reduce distress. However, managerial ownership does not influence financial distress while a significant positive relationship is observed between local ownership and financial distress.

Originality/value

This is the first study to investigate the influence of ownership structure, including local ownership, on financial distress in SSA, employing a unique methodology of matched design and conditional logistic regression analysis. Furthermore, the paper presents cross-country evidence from emerging frontier markets, highlighting the importance of governance frameworks in firms’ stability.

Citation

Abdulkadir, M., Kariuki, S.N. and Kariuki, P.W. (2024), "Ownership structure and financial distress: is the tale from Sub-Saharan Africa different?", African Journal of Economic and Management Studies, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/AJEMS-03-2024-0153

1. Introduction

A reflective analysis of global and regional financial crises highlights the importance of strengthening firms’ financial stability as a critical element to their survival. With regard, firms deploy several tools to create and sustain their value and stability in the long run. Notably, the company’s financial stability has been a concern for both the shareholders and the management. The opening up of foreign markets and the increasing economic liberalisation and uncertainties are creating a riskier and stiffer competitive environment for companies (Md-Rus et al., 2013). Thus, management’s protection of shareholders’ interests through firm value increment and sustaining financial stability has been considered to mirror the level of stewardship actions of the management (Al-Absy, 2020). Further, Alkurdi et al. (2021) affirm that the firm’s level of sustainable returns strengthens its performance and is crucial to its survival.

Remarkably, the ownership structure (OS) is one of the board’s critical tools for improving the financial planning, value creation, and performance of the firm (Alkurdi et al., 2021). The various OS can be employed to address agency problems, thus enhancing the interaction of the management with the firm’s other stakeholders (Kirimi et al., 2022). Equally, the existing procedures, standards, and governance practices fail to deliver the needed checks and balances to the company with which to cultivate sound business practices, especially when the firm is in the most needful situations (Younas et al., 2021). Thus, the ownership structure is an effective control tool for the firm’s management to enhance financial performance, thereby reducing the firm’s distress.

The relationship between OS and financial distress has been a fundamental subject of global discussion, particularly in the aftermath of major economic and financial crises, especially the Asian contagion of 1997 and the global economic crisis of 2007. This issue has gained increasing attention due to its potential impact on decision-making, firm value, and financial stability. Numerous studies perceive OS as a significant aspect influencing effective governance and is critical in preventing the failure of firms (Kirimi, 2024; Mai et al., 2024; Saona and Azad, 2018). These studies show that OS diversity can impact board efficiency significantly, consequently influencing business decisions, and, ultimately, the firm’s stability and distress possibilities. For instance, Kirimi (2024) found evidence of a positive relationship between OS and the performance of firms in Kenya. Similar results were reported by Saona and Azad (2018) in the Asian context. Furthermore, Mai et al. (2024) document strong evidence of OS explaining the performance of Indonesian Islamic banks.

Alternatively, other previous works claim the existence of conflicts of interest associated with ownership structures thereby negatively impacting performance (Alshirah and Alshira’h, 2023; Kirimi et al., 2022; Zahid et al., 2023, among others). In this respect, Kirimi et al. (2022) maintain that ownership structures lead to conflicts between management and shareholders, intensifying agency problems. These problems deepen risky decision-making by the management, which is detrimental to the firms’ value and stability. Furthermore, Zahid et al. (2023) show that OS negatively influences performance in European Frontier Markets (EFM). Meanwhile, Fauzi and Locke (2012) observed a non-linear relationship between OS and firm performance in New Zealand.

Contentious as it is, OS is a fundamental concern in shaping a firm’s CG mechanisms in today’s business environment. Its role in complementing weak and regular regulatory inefficiencies in protecting investors is an issue of debate. Despite the subject’s significance and controversy, little research has been conducted on SSA (Geoffrey et al., 2020) making the SSA an interesting context for the study. Additionally, SSA, an emerging market with diverse economies and growth potential, is often affected by bureaucracy and governance issues related to OS. The issues are related to management transparency, agency problems and inadequate investor protection frameworks (Kirimi et al., 2022). Many firms suffer market deficiencies due to inconsistency in standards and inadequate oversight constraining their capital sources (Akotey et al., 2022).To that end, the study explored the critical question of whether a firm’s OS can influence and help mitigate the financial distress (FD) of firms. Provoked by this question, the study focuses on two critical aspects. Firstly, the study intends to establish the effect of the ownership structure of listed firms on their FD. Contrarily, the second purpose is to fill the related empirical gap concerning OS and FD within a developing market context, particularly in SSA.

Notably, this study is motivated by the inconclusive results from emerging economies (Queiri et al., 2021; Udin et al., 2017; Younas et al., 2021), as well as the limited works that have considered investigating the influence of local shareholding on financial distress, particularly in a developing market context such as the SSA (Boachie, 2021). This study is thus considered necessary, considering local investors’ increasing roles and participation in their respective countries’ financial markets. Lastly, the findings will assist legislators in policy formulation regarding governance frameworks and aid managers in modifying their decisions to mitigate FD. Furthermore, investors and stakeholders can develop an accurate picture of the usefulness of the various OS in improving their decisions around the company’s performance, particularly in light of the ongoing regional stock markets integration.

Going forward, the paper is structured as follows: the subsequent section covers the literature review and hypotheses development while the following sections present the data and methodology, the empirical findings, and the conclusion.

2. Literature review

2.1 Theoretical framework

The separation of ownership and control in the modern business set-up creates a principal-agent relationship leading to conflict of interests (Jensen and Meckling, 1976). This contractual dilemma exists where the agents (managers) pursue their self-interests at the expense of the principals’ (shareholders) interests resulting in agency problems (Kuada, 2021). Agency theory thus anticipates two impediments to effective monitoring of the principal over the agent; moral hazard and adverse selection. A moral hazard problem occurs when managers exploit their access to information to maximise their wealth as shareholders’ monitoring capabilities are limited. Similarly, adverse selection occurs when the managers leveraging the existing information asymmetries make incorrect decisions around the organisation’s policies. Hence, to counter these problems, the theory suggests that managers’ and shareholders’ interests be aligned with the corporate objectives through OS (Alshirah and Alshira’h, 2023). Maintaining an effective OS influences the managers’ behaviours, subsequently reducing asymmetries and agency problems thereby improving the firm performance (Din et al., 2021).

Conversely, the resource-based view (RBV) theory underscores the shareholding’s crucial role in providing firms with external linkages and critical resources such as capital, technology, expertise and knowledge (Pfeffer and Salancik, 1978). RBV asserts that shareholders, including institutional, foreign and state shareholders, have other multiple investments that provide them access to information and resources. Therefore, they form linkages for firms to access external resources and gain sustainable advantage. Although the theory’s view on OS is relatively new (Ozdemir and Kilincarslan, 2021), the approach helps understand the implications of OS in assisting firms acquire best practices, and ultimately scarce and unique resources contributing to their competitive advantage, hence minimising distress.

2.2 Empirical review and hypothesis development

The distribution of the ownership of a company has been argued to be an essential CG tool in enhancing the performance and stability of companies (Kirimi, 2024; Mai et al., 2024; Saona and Azad, 2018; Udin et al., 2017). Accordingly, the consequence of the shareholding structure on the firm’s odds of falling into bankruptcy has been a subject of discussion in contemporary finance literature. For instance, Udin et al. (2017) suggest that a company’s ownership structure heavily influences its success or failure. Consequently, the dispersion of shareholders in today’s business landscapes limits shareholders’ powers to manage the firms giving rise to separation of ownership and control (Oppong et al., 2023). In this manner, managers are delegated with strategic decision-making with an expectation to realise the shareholders’ wealth maximization objective while the shareholders retain control over the managers (Kirimi, 2024).

The managers, however, may pursue self-interests in final decision-making resulting in agency costs detrimental to the firm’s performance and shareholders’ wealth (Boachie, 2021). To address the agency problems, Jensen and Meckling (1976) propose the adoption of various ownership dimensions as they dictate and influence the extent of managers’ decision-making powers. Vital in control and decision-making, Alajmi and Worthington (2023) argued that the different ownership distribution classes hold different strategic decision-making powers and abilities, ultimately impacting the firms’ performance and long-term value. In this regard, the study anticipates the various ownership structures to influence the firms’ resourcing capabilities and adopted governance practices consequently influencing their exposure to financial distress.

2.2.1 Managerial ownership and financial distress

Compliance with most CG codes demands attributing part of the firms’ ownership to the managers and executives. Preceding works by Donker et al. (2009) Luqman et al. (2018), and Manzaneque et al. (2016a, b) established that firms that have the managers own substantive equity shares are less susceptible to encountering FD. Mai et al. (2024), observed managerial ownership (MANOWN) to have a positive relationship with financial performance. Jensen and Meckling (1976), grounding their argument on social psychology, hypothesised that managers desire to make the best use of their convenience. Thus, they may be opportunistic in their behaviour, giving rise to ethical conflicts. In this regard, Calomiris and Carlson (2016) indicate that by attributing ownership to managers, firms can lower the possibility of distress, thus, enhancing their performance. Directors who own firms’ shares are more risk-averse and ensure greater management efficiency. Wei et al. (2016) using a sample of listed firms in Taiwan from 2006 to 2014 observed that concentrating equity to the management charges them with incurring any wealth loss, thus reducing agency problems. Theoretically, Kirimi et al. (2022) posit that MANOWN can drastically decrease agency problems. Agent-principal interests’ alignment, as adduced by the agency theory, could be achieved through insiders’ participation in firm ownership. Additionally, Gerged et al. (2022) concluded that firms with higher MANOWN are always risk-averse, consequently minimising their likelihood of bankruptcy and thus, the following is developed as the first hypothesis:

H1.

Managerial ownership has a negative effect on financial distress.

2.2.2 Institutional ownership and financial distress

Institutional ownership (INSOWN), similarly, play a crucial role in the CG context and significantly influences the firm’s decisions about management and governance practices (Kirimi et al., 2022). Their active role and increasing participation in capital markets pose substantial pressure on the management, thus strengthening firms’ CG. Institutional investors, actively pursue their controlling role in management, which enhances the performance (Farooq et al., 2022) consequently reducing financial constraints. To this end, Gerged et al. (2022) employing a sample of 110 FTSE 350 firms in the UK from 2014 to 2019, concluded that INSOWN negatively influences distress risks. Theoretically, Institutional investors, as large stockholders, are endowed with resources hence better suited for effective monitoring. In support, Li et al. (2021) found that a listed company is less likely to fall into distress when an institution holds a stake in it. Similarly, Younas et al. (2021) established that institutional ownership in Pakistan is positively related to the Z-Score model, implying that the distress of firms could be decreased by increasing institutional ownership. Nevertheless, Kirimi et al. (2022) in the Kenyan context observed an insignificant relationship between INSOWN and commercial banks' performance. Similarly, Jodjana et al. (2021), Udin et al. (2017), and Manzaneque et al. (2016a, b) established no significant impact of institutional investors in forecasting financial distress. Consequently, we hypothesise that:

H2.

Institutional ownership has a negative effect on financial distress.

2.2.3 Ownership by domicile and financial distress

Established evidence regarding the linkage of ownership structure by domicile with financial distress is mixed. For example, a strand of earlier works, including Ali et al. (2022), Kirimi et al. (2022), and Udin et al. (2017), amongst others, have shown that foreign ownership (FOROWN) is a valuable tool for enhancing quality governance practices and mitigating agency problems' impact. For instance, Udin et al. (2017) employed panel data from 146 Pakistani companies listed from 2003 to 2012 to investigate the nexus between OS and distress likelihood. The results suggest that foreign stockholders have better incentives to monitor managers and are more concerned with profit indicating that FOROWN is more valuable in enhancing companies’ performance and reducing distress. Boachie (2021) reported that Ghanaian banks performed better with the increase in foreign ownership. As the agency theory opines, foreign investors help reduce information asymmetries in boards, reducing the prospects of firms falling into distress (Ali et al., 2022). Similarly, Kao et al. (2018) analysed a sample of Taiwanese firms over the 1997 to 2017 period. The findings from the panel estimation established that FOROWN improves firm performance in the Taiwan context. Foreign investors complement the weak monitoring by domestic institutions (Udin et al., 2017); hence, prospects of distress are reduced with additional foreign equity holdings. On the other hand, Alshirah and Alshira’h (2023) analysed a sample of 94 firms listed from 2014 to 2017 in Jordan and observed that FOROWN is insignificant in explaining the banks' performance. Further, an empirical study by Younas et al. (2021) highlighted an insignificant relationship between financial distress and foreign ownership. This confirms the conclusions by Cheng et al. (2016) that foreign ownership insignificantly influences banks’ default risks in China. As it stands out from the literature presented, we develop the following hypotheses:

H3.

Foreign ownership has a negative effect on financial distress.

H4.

Local ownership has a negative effect on financial distress.

2.2.4 State ownership and financial distress

Numerous studies, including those of Donker et al. (2009), Md-Rus et al. (2013), Wang and Deng (2006) and Jin et al. (2018), among others, analysed the influence of government ownership on corporate distress. These studies found that state ownership negatively influences the financial distress of firms. Li et al. (2021) for instance, employed discrete-time survival models on a panel-data structure of over 17 years for 2,824 Chinese firms and found that state ownership negatively impacts distress. RBV asserts that the direct association of SOEs with government institutions ensures institutional support and access to resources lowering their default risk. In agreement, Zouari and Neila (2014) concluded that state investors are beneficial to the corporate performance of Islamic banks in the OIC countries. As suggested by agency theory, by reducing firms’ exposure to information asymmetries and facilitating benefits and risk-sharing mechanisms, the state acts as an agency cost mitigator consequently improving performance. Similarly, Kirimi et al. (2022) confirm a positive relationship between state ownership and banks’ performance in the Kenyan context. Government-owned firms are often argued to be too big to fail. In times of default, the government injects additional capital, provides subsidies, and undertakes tax relaxation measures (Younas et al., 2021). Thus, STATEOWN positively influences performance and is anticipated to lower the firms’ chances of distress. To this end, we hypothesize the following:

H5.

State ownership has a negative effect on financial distress.

3. Methodology

3.1 Data and sample size

The data was drawn from published annual reports of non-financial firms listed in active security exchanges across 9 selected Anglophone SSA; Kenya, Uganda, Tanzania, Rwanda, Botswana, Malawi, Ghana, Namibia, and Zambia. Motivated by their unique regulatory requirements, compliance, and financial reporting standards, the study excludes financial firms from the sample. The focus on Anglophone countries is informed by the need for homogeneity in the sample, based on commonality in legal systems, language, financial reporting frameworks, and CG structures. Furthermore, Anglophone countries have the largest and most active security exchanges that account for over 80% of SSA gross domestic product and stock market capitalisation (Waweru, 2020). Distinctively, the CG standards and business practices in anglophone countries are largely influenced by Anglo-Saxon models enhancing comparability (Ntim and Soobaroyen, 2013). However, South Africa, Nigeria, and Zimbabwe are excluded since Zimbabwe experienced hyperinflation between the study periods while South Africa and Nigeria have high listing and advanced CG practices (see Table A1 in the Appendix for sampled countries and firms).

The study used a matched pair research design drawing on Mangena and Chamisa (2008) and Manzaneque et al. (2016a, b), where distressed firms are paired with observed non-distressed firms for the same period based on the same industry, year, and similar size (log of firms’ total assets). Matching allows for comparison between the two groups while minimizing the effects of confounding variables (Mangena et al., 2020). Additionally, by matching, the variability in the outcomes is reduced and the model’s precision is increased (Ragab and Saleh, 2021). The study identified 175 observations (firm/year) with complete ownership structure data for 2016–2021 as financially distressed and matched them with non-distressed observations. Subsequently, as observed in Table 1, the matching procedure yielded a sample size of 174 distressed and 174 non-distressed observations, giving a total of 348 paired observations, after dropping 1 financially distressed observation, for which its appropriate matching was not found.

We also assess our sample’s appropriateness by estimating the maximum allowable error for a finite population. A small error leads to considering the sample size as a valid population representation (Manzaneque et al., 2016a, b). A resultant error of 4.4% at α = 95% is observed; hence, our sample of 308 observations (firm/year) is considered a valid representative of the population (see Tables 1 and 2). Finally, the study period 2016–2021 was chosen for two apparent reasons: first, a global economic crisis was nearly triggered between 2019 and 2020 by the coronavirus (COVID-19), and companies were exposed to severe financial and economic difficulties during this period. Second, many companies had their annual reports published in those years.

3.2 Measurement of variables

Financial distress, FDA, is primarily constructed from Altman’s Emerging Market model (Altman, 2005) as used in prior studies (Jacoby et al., 2019; Khurshid and Sabir, 2019). This score is presented below:

(1)Z=3.25+6.56v1+3.26v2+6.72v3+1.05v4

Where: v1 is working capital/to total assets, v2 is retained earnings/total assets, v3 is earnings before interest and tax/total assets, and v4 is the book value of equity to total liabilities ratio.

Significantly, we use a dummy variable of 0 for non-distressed firms, whose Z score (in Equation (1) above) is more than 4.15, and 1 for financially distressed firms, whose Z score is less than 4.15.

As for the independent variables, the OWNSHIP includes managerial ownership (MANOWN), institutional ownership (INSOWN), foreign ownership (FOROWN), local ownership (LOCALOWN) and state ownership (STATEOWN).

Lastly, the study controls for potential endogeneity and firm-level differences by employing a set of firm-related controls motivated by previous works. BSZ measures the board size while BDG denotes board-gender diversity. We measure firm size (FSZE) as the log of total assets (log TA) (Udin et al., 2017), liquidity (LIQ) is measured as a current assets ratio (Hamza, 2024); firm age (Fage) is proxied by firm’s operational years since its establishment (Jacoby et al., 2019). In addition, to control for cross-country differences, we utilize the GDP growth rate to measure economic growth (ECG), while the governance estimate indicated by the World Governance Index measures regulatory quality (RQ). See Table 3 for variable definition.

3.3 Econometric model specification

Following the literature, we apply a conditional logistic regression model (clogit) to test the study’s hypotheses. The model uses a conditional likelihood function to fit the regression equation with a conditional mean bounded between zero and one, overcoming linear regression constraints in estimating parameters with binary dependent variables (FDA = 1/0) (Hosmer et al., 2000). The conditional likelihood function preserves the sample’s paired character in estimating the model parameters. Thus, the estimation of the model’s parameters, βi  , is conditional on the number of cases (distressed) and controls (non-distressed) in cluster k for k = 1, 2,…n matched by a vector of X variables. Correspondingly, the econometric model is specified in Eq (1).

(2)FDAit=β0 +βi OWNSHIPit+∑Control variablesit+nt+di+εit

Where: FDA is a dummy variable (coded 0 for non-distressed and 1 for distressed firms proxying financial distress); OWNERSHIP is a block representation of the five types of ownership structure. Control variables include board gender diversity, board size, firm age, liquidity, firm size, economic growth, and regulator quality. The subscripts t and i represent the time period and cross-sectional units, respectively; nt denotes the time effect, di is individual effect, while ε denotes the disturbance term.

4. Empirical results and discussions

4.1 Descriptive summary

The variables’ summary statistics for the entire sample are reported in Table 4 below. In addition, Table 5 provides the key descriptives and the significance test in differences of the mean. MANOWN averages 9.8% of the shares, which signifies a low control power by the directors. Institutional investors with a mean of 63.6% have higher ownership compared to the rest of the ownership classes. INSOWN is higher in SSA than in Spanish-listed companies and Pakistani-listed firms. For instance, Manzaneque et al. (2016b) reported 20.7% for Spanish-listed firms, while in the case of Pakistani-listed firms, 12.8% was reported by Din et al. (2021). Foreign investors have lower mean participation in shareholding compared to their local peers (34.3% vis-a-vis 56.1%). On average, STATEOWN is at 11.4%, with a standard deviation of 19.88%, signifying a high variation in the data of state equity ownership

The results from the mean comparison tests in Table 5 report no significant systematic difference in the level of MANOWN between non-distressed and distressed groups (see Table 5). However, firms under the distressed category tend to have low institutional ownership – mean (median) of 55.897% (59.986%) compared to – mean of 71.235% (median of 73%) of non-distressed firms. Additionally, FOROWN is higher in the non-distressed group than in the distressed category. On average, the non-distressed group has 40.903% of their shares held by foreigners. In comparison, the FOROWN of distressed firms stands at 27.748% on average. Table 5 further reveals that for non-distressed companies, the proportion of LOCALOWN is lower, with an average value (median) of 42.908% (34.729%) compared to 69.239% (88.087%) for distressed companies. No significant difference is observed between non-distressed and distressed firms regarding state ownership.

Concerning the controls, distressed firms are inclined to have a slightly larger board (around 9 members) and where are more likely to have a higher gender diversification (19.8%), than distressed companies (an average of 7 members and 18.3% gender diversified). T-test results, however, indicate a significant difference exists in the board size only between the two categories. Concerning the rest of the controls, no systematic difference between the two groups is noted in firm size, liquidity, and firm age.

4.2 Correlation analysis

Table 6 displays correlation analysis results between our independent variables, controls, and financial distress. We observe the correlation coefficients between all our explanatory variables are lower, except between FOROWN and LOCALOWN with a coefficient of −0.855*. This is, however, expected since a firm whose equity is mainly held by foreign institutions and individuals has relatively low local ownership of its shares. Significantly, we find negative weak correlations of −0.290*, and −0.215* between INSOWN and FOROWN, respectively, with the FDA. We further find a positive correlation between LOCALOWN and FDA. The highest correlation coefficient (apart from that between foreign and local ownership) is observed between the state ownership (STATEOWN) and firm size (FSZE) (r = 0.336, p < 0.1). This implies that state-owned enterprises (SOEs) or firms with a majority of STATEOWN have larger pools of resources at their disposal compared to those owned by institutions (r = −0.140, p < 0.1).

4.3 Conditional logistic regression results

Table 7 reports the results obtained after performing clogit regression. The results from Model 1 of Table 7 depict the coefficient of managerial ownership (MANOWN) to be positive, nonetheless an insignificant one. This suggests that managerial shareholders have an insignificant effect on corporate distress in SSA. This could be explained by the low magnitude of ownership associated with the directors and top-level managers compared to the other classes of shareholders in the region. This is congruent with prior findings of Udin et al. (2017) and Ragab and Saleh (2021). Nevertheless, the results are contrary to those of Mai et al. (2024) who reported a positive relationship between managerial ownership and the Indonesian Islamic banks' performance. In Vietnam, managerial ownership was reported to foster managers’ commitment and effectiveness by Truong (2022) resulting in lower distress levels. The finding thus, implies that the influence of MANOWN on a firm’s financial success or failure is grounded upon contextual and regional dynamics. From the results, our H1 was not confirmed. Hence, the results do not confirm the interest alignment argument as argued by agency theory protagonists.

Institutional ownership exerts a negative influence on financial distress, at a 1% significant level. This confirms our H2. This implies that institutional shareholding is a powerful CG mechanism that reduces FD. From Model 2 of Table 8, the coefficient of −0.02 infers that a percentage increase in INSOWN would reduce the prospect of bankruptcy by 0.02%. This is because institutions in SSA are actively embracing their control and monitoring roles, putting substantial pressure on the management as they focus on long-term performance. Their increasing participation in security exchanges has thus reduced information asymmetries and agency costs lowering the distress risks of listed companies. Our results confirm the findings of Donker et al. (2009), Gerged et al. (2022), Li et al. (2021), Mariano et al. (2021) and Younas et al. (2021) all of whom insisted on the active role of INSOWN in mitigating FD.

The results concerning foreign ownership, as presented in model 3 of Table 7, indicate that foreign ownership (FOROWN) negatively influences financial distress at a p < 0.01. Possibly, foreign investors facilitate the transfer of advanced technology and best governance practices, thus reducing agency problems, which may accelerate the performance rate and diminish the risks of FD. Concurring with Ali et al. (2022) and Udin et al. (2017) in the UK context, we observe that FOROWN can positively stimulate the firms’ performance by motivating the managers through incentives. Therefore, the findings suggest that FOROWN can resolve agency conflicts and consolidate the governance systems. The low coefficient of 0.015, however, implies that foreign capital flow is still limited in SSA due to the lack of an enabling environment favourable for investment. As for our hypothesis, H3 has been statistically confirmed by the results in Model 3.

As for LOCALOWN, a significant positive effect at 1% is observed in model 4 of Table 7. This means that our H5 has not been statistically accepted. We observe that higher ownership of firms by local institutions and individuals increases financial distress. This is because agency problems are prevalent between local investors and the management. Dahlquist and Robertsson (2001) observed that domestic institutional shareholders in developing economies cannot actively monitor the management due to political constraints, lack of adequate regulatory systems, and immature capital markets. Congruent with the findings of Zahid et al. (2023) in the EFM context, we maintain that domestic investors lack the appropriate systems and mechanisms to effectively control managers’ opportunistic behaviours compared to their foreign counterparts. As Boachie (2021) concluded in the Ghanaian context, higher LOCALOWN limits FOROWN decision-making powers, giving rise to a conflict of interest. This threatens the firms’ profitability, increasing the risks of financial distress.

Coinciding with Hu and Zheng (2015), Li et al. (2021), and Wang and Deng (2006), we find STATEOWN to negatively influence the odds of financial distress of SSA-listed firms. This implies that our H5 has statistically been accepted. Notably, we observe that state-controlled firms have lower odds of getting into distress as state intervention through additional acquisition of equity shares (1% SO increase) would lower the chances of financial distress by 2.1%. Governments enhance firm performance through policy interventions and regulations which minimise agency conflicts (Hossain et al., 2023). Generally, firms with the state holding large chunks of ownership rarely suffer distress and are often argued to be too-big-to-fail as governments provide plentiful resources to bail them out of hardship. Government subsidies are much higher in politically connected firms, which in turn enhances firm value (Jin et al., 2018).

Concerning the controls, the study finds that a larger board raises the risks associated with distress. Regardless of the OS, larger boards may experience coordination difficulties (Younas et al., 2021), resulting in a slower decision-making process. Contrarily, FSZE is observed to influence firms’ distress negatively. Large firms attain higher operational efficiency, thereby positively influencing their market performance. However, BDG, liquidity, and Fage are observed to have no significant impact on FD. Regarding country controls, ECG is observed to influence financial distress. This positive influence might stem from rising regulatory and compliance costs often accompanying economic growth. RQ, on the other hand, reduces the chances of distress. Oliveira and Raposo (2020) in explaining this phenomenon opine that, the market disciplines firms in countries with low adherence to regulations and policies with higher probabilities of distress.

4.4 Robustness

The study developed some further analysis to assess the robustness of the results. Literature supports that profitability is a critical element influencing the distress prospect of firms (Khurshid and Sabir, 2019). Further, different years present different economic conditions which are likely to influence the possibilities of distress (Manzaneque et al., 2016a, b). To control that effect, we deploy year dummies. Similarly, firms can respond to adverse corporate performance by intensifying board activities (Brenes et al., 2011).

Equally, a firm’s external auditor is an antecedent of business failure (Al-Bassam et al., 2018). Firms engaging one of the Big 4 audit firms as their external auditor are observed to have better market performance. Higher-quality audit firms have greater independence and can effectively limit managers’ opportunistic activities (Khalil and Ozkan, 2016) due to their vast experience, expertise, financial muscle, information, and knowledge. Of interest to our study, Meah et al. (2021) and Mensah and Onumah (2023), note that audit quality positively influences firms’ performance. We, therefore, re-estimate our model by additionally including ROE as a profitability measure, board meetings, and audit quality; a dummy variable 1 if a firm has engaged the audit service of one of the big 4 audit companies or 0 otherwise. The results from the robustness analysis in Table 8 correspond to those of baseline regression in Table 7, indicating the consistency of our results.

5. Conclusion

The study investigates the influence of ownership structure on the financial distress of non-financial firms listed in ten selected Anglophone countries in SSA using a paired sample of 348 observations. The findings demonstrate that OS significantly influences firms’ financial distress in SSA. Particularly, the results indicate that MANOWN has no significant influence on distress. Directors still have a low level of equity ownership in the region. Further, we established a negative connection between INSOWN and financial distress. This may be due to the improved control, accountability, and transparency that institutional investors provide in SSA. The results further established that FOROWN influences financial distress, while LOCALOWN increases the firm’s default risks. Foreign investors have a better capacity to monitor and control managers, reducing asymmetries and enhancing performance compared to their local peers.

Similarly, the results established that STATEOWN is critical for firms’ stability while BSZE and ECG positively influence the probability of distress across all the ownership classes. Economic growth in developing markets often comes with high regulatory requirements and compliance costs that usurp firms' resources. FSZE and RQ negatively impact financial distress, suggesting that as firms grow in size, their chances of facing financial distress reduce while markets reward firms better when operating in countries that can formulate and implement sound policies and regulations. As for BDG, liquidity, and Fage, no significant associations were established with financial distress.

5.1 Theoretical implications

This work fills an empirical gap regarding the influence of OS on financial distress in developing economies and contributes to the extant literature by providing evidence of the usefulness of ownership theories in comprehending the intricate relationship between OS and corporate distress in SSA. Generally, the findings validate the role of ownership in reducing agency costs and creating firms’ competitive advantage through enhanced resourcing capabilities, as contended by agency and resource-based view theories.

5.2 Managerial and policy implications

The findings pose the following implications: first, the study highlights the importance of OS in mitigating financial distress. Managers should therefore formulate OS policies that enhance the firms’ stability through sound governance practices. Additionally, the study highlights the significant role of institutional shareholders in minimizing agency costs through effective monitoring and control, underscoring the need for improved market infrastructure and strengthened governance in SSA to cater to agency problems influenced by local market forces. Lastly, the findings emphasise the need for SSA governments to devise policies that promote good governance to attract foreign direct investments and enhance efficiency and accountability.

5.3 Ideas for future research

To the best of the authors’ knowledge, this is the first study that investigates the influence of ownership structure on financial distress taking SSA as a case study hence immensely contributing to extant literature. The findings, however, are subject to several limitations. The study employed a single measure of corporate distress, Altman’s emerging market score. Future research may, therefore, use other proxies, such as the Zmijewski score. Second, our focus on CG mechanisms in the study was limited to OS only, suggesting future research may consider a variety of other CG aspects, such as internal governance measures.

Table 1

Population composition and selected sample size

Panel A: composition of matched sample size	Firms	Firm year observations
listed non-financial firms across 10 SSA exchanges	140	840
less
Observations with incomplete corporate governance data or listed than one year	34	204
Observations of final listed non-financial firms’ population	106	636
Identified matched sample		348

Panel B: sample per industry	N	%
1) Agriculture	20	5.75%
2) Mining	38	10.92%
3) Manufacturing	92	26.44%
4) Energy	46	13.22%
5) Construction	16	4.6%
6) Wholesale and Retail Trade	24	6.9%
7) Transportation	56	16.09%
8) Accommodation and Food Service	8	2.3%
9) Information and Communication	48	13.79%
Total sample	348	100%

Source(s): The Authors

Table 2

Yearly sample distribution

	Distribution of samples per year
	2016		2017		2018		2019		2020		2021
	N	%	N	%	N	%	N	%	N	%	N	%
Distressed firms	21	50	23	50	29	50	31	50	38	50	32	50
Non-distressed firms	21	50	23	50	29	50	31	50	38	50	32	50
	42	100	46	100	58	100	62	100	76	100	64	100

Source(s): Authors

Table 3

Variables measurement

	Symbol	Measurement
Independent variable:
Managerial ownership	MANOWN	Percentage of shares held by directors
Institutional ownership	INSOWN	Percentage of shares owned by institutions
Foreign ownership	FOROWN	Percentage of shares owned by foreign investors
Local ownership	LOCALOWN	Percentage of shares held by local investors
State ownership	STATEOWN	Percentage of shares held by state
Dependent variable:
Financial distress	FDA	Altman emerging market score
Firm-level control:
Board size	BSZ	Total number of directors
Board gender diversity	BDG	Female directors to total directors’ ratio
Firm size	FSZE	Log of total assets
Liquidity	LIQ	Current assets to current liabilities
Firm age	Fage	Number of operational years of the firm
Country-level control:
Economic growth	ECG	GDP growth rate
Regulatory quality	RQ	World Bank index on country’s regulatory quality

Source(s): The Authors

Table 4

Sample descriptive summary

Variable	Mean value	S. Deviation	Minimum	Maximum
MANOWN	9.809	18.163	0	70.573
INSOWN	63.579	26.501	0	99.72
FOROWN	34.325	30.703	0	96.361
LOCALOWN	56.073	34.169	0	100
STATEOWN	11.356	19.88	0	73.919
BSZ	7.997	3.031	0	16
BDG	0.191	0.134	0	0.667
FSZE	6.432	1.326	3.033	9.426
LIQ	3.544	15.989	−37.515	197.984
Fage	46.686	26.193	0	119
ECG	3.431	3.976	−8.726	11.37
RQ	−0.237	0.374	−0.786	0.736

Source(s): The authors

Table 5

Distressed and non-distressed mean comparison tests

Variables	Distressed firms					Non-distressed firms					t-tests
	Mean	25%	Median	75%	S. Dev	Mean	25%	Median	75%	S. dev
MANOWN	9.268	0.000	0.004	4.766	18.229	10.351	0.0004	0.05	8.364	18.133	0.55
INSOWN	55.897	32.042	59.986	79.443	30.167	71.235	61.942	73.00	84.492	19.456	5.65*
FOROWN	27.748	0.629	10.528	55.534	30.756	40.903	9.965	51.00	66.236	29.289	4.1*
LOCALOWN	69.239	42.499	88.087	96.732	32.938	42.908	23.00	34.729	71.829	30.137	−7.8*
STATEOWN	11.528	0.000	0.000	25.30	19.351	11.183	0.000	0.000	14.722	20.449	−0.15
BSZ	8.517	6.000	8.000	11.000	2.613	7.477	5.000	7.000	10.000	3.324	−3.25***
BDG	0.198	0.125	0.200	0.273	0.133	0.183	0.091	0.200	0.300	0.134	−1
FSZE	6.36	5.506	6.353	7.825	1.465	6.504	5.675	6.458	7.675	1.17	1
LIQ	3.196	0.291	0.544	0.833	21.26	3.893	1.219	1.555	2.488	7.783	0.4
Fage	46.77	23.00	40.5	60.00	30.89	46.601	33.00	40.00	60.00	20.502	−0.05

Note(s): ***,**,* denote significance at p < 0.01, p < 0.05, and p < 0.1 respectively

Source(s): The Author

Table 6

Pairwise correlation analysis

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)
(1) FDA	1.000
(2) MANOWN	−0.030	1.000
(3) INSOWN	−0.290*	−0.137*	1.000
(4) FOROWN	−0.215*	−0.062	0.127*	1.000
(5) LOCALOWN	0.386*	0.010	−0.124*	−0.855*	1.000
(6) STATEOWN	0.009	−0.042	−0.301*	−0.353*	0.118*	1.000
(7) BSZ	0.172*	−0.291*	−0.120*	−0.126*	0.121*	0.298*	1.000
(8) BDG	0.054	−0.146*	−0.164*	−0.209*	0.154*	0.228*	0.325*	1.000
(9) FSZE	−0.055	−0.001	−0.140*	−0.053	−0.023	0.336*	0.316*	0.131*	1.000
(10) LIQ	−0.022	−0.028	−0.017	0.201*	−0.256*	−0.088	−0.087	−0.147*	−0.200*	1.000
(11) Fage	0.003	−0.214*	0.069	−0.186*	0.238*	−0.021	0.167*	0.175*	0.204*	−0.135*	1.000
(12) ECG	0.098	0.071	−0.087	−0.110*	0.151*	0.097	−0.048	−0.031	0.144*	−0.154*	0.085	1.000
(13) RQ	−0.045	−0.070	−0.044	−0.158*	0.104	0.158*	−0.048	−0.046	−0.326*	0.078	−0.254*	−0.123*	1.000

Note(s): ***,** ,* denote significance at p < 0.01, p < 0.05, and p < 0.1 respectively

Source(s): Author

Table 7

Conditional logistic regression model

Financial distress (FDA)
Variable	Model 1	Model 2	Model 3	Model 4	Model 5
MANOWN	0.012(0.009)	–	–	–	–
INSOWN	–	−0.02(0.006) ***	–	–	–
FOROWN	–	–	−0.015(0.005) ***	–	–
LOCALOWN	–	–	–	0.032(0.006) ***	–
STATEOWN	–	–	–	–	−0.021(0.009) **
BSZ	0.292(0.061) ***	0.231(0.063) ***	0.222(0.062) ***	0.17(0.067) **	0.351(0.072) ***
BDG	0.535(1.01)	0.061(1.043)	0.318(1.046)	0.205	0.608(1.012)
FSZE	−1.147(0.323) ***	−1.104(0.344) ***	−0.94(0.329) ***	−0.861**(0.364)	−1.162(0.34) ***
LIQ	0.002(0.008)	0(0.008)	0.002(0.008)	0.004(0.009)	0.002(0.008)
Fage	0(0.006)	−0.001(0.006)	−0.004(0.006)	−0.009(0.007)	0(0.006)
ECG	0.368(0.092) ***	0.255(0.088) ***	0.354(0.096) ***	0.379(0.113) ***	0.395(0.094) ***
RQ	−0.845(0.459) *	−1.319(0.486) ***	−1.453(0.529) ***	−2.049(0.586) ***	−0.379(0.488)
2 log-likelihood	−93.144	−87.179	−89.287	−71.505	−91.275
Model Chi-square	53.54***	65.47***	61.25***	96.82***	57.28***
McFadden’s R2	0.223	0.273	0.255	0.404	0.239
Cragg and Uhler’s R2	0.355	0.42	0.398	0.571	0.376

Note(s): ***, **,* denotes significance at p < 0.01, p < 0.05, and p < 0.1 respectively

Source(s): The Authors

Table 8

Further conditional logistic regression analysis

Financial distress (FDA)
Variable	Model 1	Model 2	Model 3	Model 4	Model 5
MANOWN	0.008(0.01)	–	–	–	–
INSOWN	–	−0.02 (0.06) ***	–	–	–
FOROWN	–	–	−0.021(0.006) ***	–	–
LOCALOWN	–	–	–	0.051 (0.09) ***	–
STATEOWN	–	–	–	–	−0.025 (0.01) **
BSZ	0.307(0.07) ***	0.256 (0.74) ***	0.254(0.72) ***	0.27(0.092) ***	0.387 (0.082) ***
BDG	1.639 (1.162)	1.23 (1.208)	1.846 (1.23)	3.133 (1.505) **	2.025 (1.2) *
FSZE	−1.01(0.356) ****	−1.015(0.371) ***	−1.066 (0.364) ***	−1.584 (0.466) ***	−1.054 (0.367) ***
LIQ	0.001 (0.008)	−0.001 (0.08)	0.001 (0.09)	−0.001 (0.011)	0.001 (0.008)
Fage	0.002 (0.006)	0.001 (0.06)	−0.003 (0.06)	−0.01 (0.008)	0.003 (0.006)
ECG	0.343 (0.096) ***	0.225 (0.09) **	0.366 (1.06) ***	0.483 (0.143) ***	0.373 (0.099) ***
RQ	−0.915 (0.489) *	−1.357(0.514) ***	−1.936 (0.625) ***	−3.591 (0.874) ***	−0.437 (0.513)
ROE	0.003 (0.089)	−0.004 (0.094)	0.031 (0.101)	0.167 (0.13)	0.038 (0.087)
BM	−0.101 (0.05) **	−0.114 (0.054) **	−0.145 (0.055) ***	−0.276 (0.081) ***	−0.109 (0.051) **
Audit quality	−0.388 (0.383)	−0.268 (0.389)	0.373 (0.474)	1.55 (0.589) ***	−0.458 (0.384)
Year dummy	Yes	Yes	Yes	Yes	Yes
2 log-likelihood	−89.389	−83.287	−83.667	−58.99	−86.333
Model chi-square	54.12***	66.32***	65.56***	114.912***	60.23***
McFadden’s R2	0.232	0.285	0.282	0.347	0.259
Cragg and Uhler’s R2	0.367	0.435	0.431	0.661	0.402

Note(s): ***, **,* denotes significance at p < 0.01, p < 0.05, and p < 0.1 respectively

Source(s): The Authors

Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Declaration of interest statement: The authors declare that there is no conflict of interest.

Data availability statement: Data supporting the findings of this study is available in a data repository through the following link, DOI 10.6084/m9.figshare.25134029

These authors contributed equally to this work

Appendix

Table A1

Country and firm sample

Panel A: country sample
Anglophone countries with active exchange	15
Countries with advanced CG mechanisms	2
Countries with inflation between study periods	1
Data constrained countries	3
Final sampled countries	9

Panel B
List of sampled countries	Initial observations	Pairing sample
1) Botswana	60	40
2) Ghana	81	70
3) Kenya	202	95
4) Malawi	24	9
5) Namibia	30	24
6) Rwanda	21	2
7) Tanzania	65	47
8) Uganda	42	18
9) Zambia	70	43
	607	348

Source(s): The Authors