Impact of CEO cultural background on internal control quality: Evidence from Cameroon

Abstract

This research examines the relationship which exists between the CEO’s cultural background and firm’s internal control quality. It further investigates whether CEO duality, board size, and CEO tenure moderate this relationship. A sample of 100 unlisted, non-financial firms in Cameroon from 2016 to 2019 was used. A panel probit model was applied to test the hypothesis. The result shows that CEOs with French cultural background have better internal control quality and fewer material weaknesses reported than CEOs with English cultural background. Moreover, the study finds that CEO duality and larger board size enhance the positive relationship which exists between the French cultural background of the CEO and the internal control quality of firms. However, the findings show no significant interaction between the CEO’s tenure and internal control quality of firms.

Keywords:

• CEO cultural background
• internal control quality
• board feature
• internal control material weakness

JEL Classifications:

• M42
• M41
• G34
• M19

1. Introduction

Internal control was defined as the “examination of financial statements by independent public accountants,” which was published by the American Institute of Accountants in 1936, and recently has been a hot topic in the field of accounting research. Prior literature suggests that internal control is helpful to enhance the accounting information quality, reduce enterprise risk, decrease insider training, relieve information asymmetry, and protect investor’s interests (Altamuro & Beatty, 2010). In Cameroon, numerous frauds in accounting and finance, as well as deficiencies in internal controls exist in many organizations. Osifo (2014) documents that a high level of corruption exists in the Cameroon manufacturing sector because of lack of effective governance policies. Similarly, Molungu (2019) highlights that approximately 51.3% of small and medium enterprises lack effective internal control to deter fraud in Cameroon. Since maintaining a good internal control system is becoming increasingly essential for enterprises, Cameroon has gradually begun to improve its internal control laws. COBAC (Banking Commission of Central African States), in which Cameroon is a member, revised in 2016 the “Basic Norms of Internal Control’ and the ‘Supporting Guidelines of Internal Control,” revealing the importance of maintaining a good and effective internal control system in banks and enterprises.

The upper echelon theory believes that the executive characteristics could influence the firm’s choice of strategy, which could further affect other behaviors of the firm (Hambrick & Mason, 1984a). This conclusion has been proven by a large number of empirical studies (Malmendier & Tate, 2005). Additionally, prior research has concluded that senior executives of different ages, genders, and tenures may have significant differences in risk appetite and risk-taking (Bertrand & Mullainathan, 2004), hence affecting corporate behavior. Prior studies examining the relationship between CEO background characteristics and internal control quality, however, nearly work from the firm executives’ features such as age, gender, tenure, and financial expertise (Lin et al., 2014; Oradi et al., 2020). Their results show that CEOs of older ages, longer tenure, and higher financial expertise maintain better internal control quality. Like tenure and age, culture can affect the CEO’s risk appetite and risk-taking, which could further affect corporate behaviors.

On one side, culture is gradually implanted into people’s thinking in the early socialization process and further stimulates and adjusts the actions and choices of people to make them conform to certain social values. On the other side, culture forms the subjective psychological structure of people to explain the problem by influencing the way they deal with information (North, 1990). This means that an individual’s perception of the same information varies from the type of culture in which they live. An empirical study by Sekely and Markham Collins (1988) highlights that culture significantly affects the firm’s decision-making. In the Cameroonian context, since its annexation by France and Britain, there has been a parallel development of both the French and English cultures. Studies regarding these two cultural backgrounds document that English Cameroonians are marginalized, and do not participate on an equal basis in the economic, social, and political affairs of the country (Henry, 2012; Kuate, 2022). Additionally, their results also point out the lack of uniqueness, collectivism and sameness within the English-speaking Cameroonians. A study by Xuanqi (2018) finds that the Chairman’s cultural background characteristics can significantly influence the firm’s audit fees. The results show that a chairman with nomadic cultural background pays lower audit fees compared to a chairman with a cultivation cultural background. Since culture can affect the chairman’s and executive decisions and influence corporate behavior, how does the CEO’s cultural background affect the company’s internal control quality?

The purpose of this study is to empirically examine the relationship which exists between the CEO’s Cultural Background and the internal control quality of firms in the Cameroonian context. This article uses the panel data of small and medium enterprises from 2016 to 2019 as the sample. The native place of the CEO is used to measure his or her cultural background, and a material weakness reported is used to measure the internal control quality of firms. The result shows that CEOs with French cultural background have better internal control quality than CEOs with English cultural background. Moreover, the study finds that CEO duality and larger board size enhance the positive relationship that exists between the French cultural background of the CEO and the internal control quality of firms. However, the findings show no significant interaction between the CEO’s tenure and internal control quality of firms.

The study is essential for policymakers, regulators, and researchers, as it provides new evidence concerning the significance of managerial characteristics, such as the CEO’s Cultural Background in establishing and maintaining an effective internal control system. This article has several academic and practical contributions. Firstly, it adds to the growing literature on the determinants of internal control quality by highlighting the cultural background of the CEO in the Cameroonian setting. Secondly, Lin et al. (2014) suggested future works to examine the relationship between managerial characteristics such as ethnicity, cultural background, personality, and internal control quality. Hence, this study responds to (Lin et al., 2014) by examining the CEO’s cultural background as one of the factors enhancing managerial ability.

The rest of the paper is organized as follows: Section 2 provides the literature review and hypothesis development, section 3 discusses the data and the research design. Section 4 reports and discusses the empirical results. Lastly, conclusions are drawn from the findings in section 5.

2. Literature review and hypothesis development

2.1. CEO cultural background and Internal Control Quality (ICQ)

The Upper Echelon Theory (UET) predicts that the management background characteristics can influence the firm’s outcome, planned choices and the performance levels (Hambrick & Mason, 1984a). The theory further predicts that the more complex a decision is, for instance, a strategic decision, the more important the decision-maker’s personal characteristics such as tenure, age, and financial expertise. The UET predicts that top management person-specific attributes (precisely their preferences and dispositions) to some extent matter to corporate strategy outcomes (Hambrick & Mason, 1984a). Additionally, the principle of the UET identifies that executive’s different traits such as tenure and age can affect their strategic choices which can further affect the company’s choice and performance (Nielsen, 2010). Furthermore, the UET highlights that executives characterized by bounded rationality are more likely to make decisions based on their social, cognitive, and psychological characteristics. Several studies have been carried out on the UET, analyzing the relationship which exists between executive personal traits and organizational traits such as financial leverage decision (Ting et al., 2015) and cash holding (Orens & Reheul, 2013). Consistent with the upper echelons theory, the CEO’s cultural background may influence his/her risk preferences and his/her financial behavior. A study by Xuanqi (2018), based on the UET, finds that the Chairman's cultural background could significantly influence audit fees. However, the empirical significance of these studies remains largely unknown, especially in terms of the relationship between a CEO’s cultural background and the internal control quality of firms. Thus, this paper attempts to bridge the gap by exploring the impact which the CEO’s cultural background has on the internal control quality.

Hofstede (2001) defined culture “as the collective programming of the mind that differentiates members of one group from another.” The definition of culture by Hofstede stresses on norms, beliefs, shared values, and expected behaviors, which are completely embedded, unconscious, and regularly irrational. These shared values often represent the desirable and/or acceptable behavior within a group. As a result, it helps members of a group make and judge the decisions of others. Studies have shown that individuals with different cultures have systematic differences (Chee Yeow Lim Kanagaretnam et al., 2014; Li-Jun et al., 2001). One of the ways culture influences corporate decision-making is by affecting the CEO’s behavior and cognition which further affects corporate decision-making. For instance, Chee Yeow Lim Kanagaretnam et al. (2014) and Ferris et al. (2013), in their respective studies, find that executives under different cultures have different risk preferences and risk-taking, which could make them choose different acquisition strategies and accounting conservatism. Hence, the risk attitude and risk-taking behavior of a CEO is influenced by his cultural background, which in return affects the internal control quality of the firm. Additionally, Cain and McKeon (2016) document that the personal background characteristics of the management team have a significant effect on audit fees. In a more recent study, Shen et al. (2021), using the Chinese urban-rural dual system, examine the impact which the CEO’s rural origin has on the internal control quality of companies. They discover that a CEO’s rural origin could significantly improve the internal control system of firms.

Cheng and Selden (1994) highlight a great difference in the economic treatment of urban and rural residents as far as employment, taxation, educational treatment, medical treatment, labor security, and pension welfare are concerned. Cameroon as an example has great treatment difference between French and English-speaking Cameroonians in several aspects. Since the annexation of Cameroon by France and Britain in 1884, both French and English have been maintained as the two official languages of the country. However, French has always been the domineering language as far as education, politics, laws, and regulations regarding businesses are concerned. A study by Kuate (2022) finds that English-speaking Cameroonians consider themselves marginalized and assimilated in an asymmetric relationship with French-speaking Cameroonians who are the majority of the population. Their results further show that the government has failed to keep the flame and excitement of the reunification alive. Similarly, a study by Ngoh (2004) shows that English-speaking Cameroonians do not participate on an equal basis as the French-speaking Cameroonians in the political, social, and cultural life of the nation. Nfi (2014) further highlights that the English-speaking Cameroonians are in a struggle to rescue their cultural identity threatened by the assimilationist policies of the majority French-speaking Cameroonians. The majority of the laws governing the country are only being published in French (even the present revised internal control regulations are all in French), not forgetting politics, employment, and education. As such, the French-speaking Cameroonians are always put at a greater advantage compared to English-speaking Cameroonians. This structural opportunity inequality makes French-speaking CEOs have higher risk awareness and stronger prevention motivation, thus forming their risk-averse nature. Previous studies in behavioral finance found a strong relationship between the financial behavior of the company and top-management risk preferences. Thus, supporting the finding that enterprises are more likely to maintain poor internal control when their executives are excessive risk-takers (Cain & McKeon, 2016).

In this study’s context, the extent to which a CEO chooses to maintain better ICQ depends on the cultural values that drive individual managers’ values, preferences, and behaviors. In line with prior works (Han et al., 2010; Hope, 2003; Kiridaran Kanagaretnam et al., 2016a), this study will further consider cultural dimensions highlighted by Arrindell (2003) which is individualism versus collectivism. This cultural dimension is used because it has the clearest implications for the choices and behaviors of managers. It is the foremost frequently examined cultural dimensions in corporate governance research (Han et al., 2010; Hope, 2003). Individualism is often defined as the “preference for a loosely knit social framework in a society in which individuals focus on themselves rather than the group to which they belong.” On the other hand, collectivism refers to a “preference for a tightly knit social framework in a society in which individuals focus on the group rather than on themselves.” Decisions based on personal needs tend to prevail in individualistic societies. This sort of behavior can largely be attributed to how individuals see themselves and how they pursue self-esteem. In line with this notion, Han et al. (2010) and Kiridaran et al. (2011) document a higher level of earnings management in individualistic societies than in collectivist societies. Additionally, Shupp and Williams (2008) find that groups were more risk-averse in high-risk situations than individuals.

A study in South Africa by Eaton and Louw (2000) compared English residents (individualist) and Africana residents (collectivist). Their result shows that English residents have the habit of being independent and self-reliant, especially in education, and they mostly focus on their achievement, self-orientation, personal success, and personal uniqueness. Their study further highlights that English residents feel loosely linked with one another which makes them have an individualistic nature. Additionally, the researcher states that Africana is the most domineering language in South Africa compared to English. Hence, the domineering nature of the French people due to the linguistic, political, and economic inequality makes them more collective than the marginalized English people. Moreover, previous studies on Canadian cross-cultural analysis (French and English citizens) characterized English Canadians as having a self-orientation, a relationship of mastery over knowledge, and an individualistic orientation. Likewise, French Canadians were characterized as having a present orientation, a relationship of harmony with nature, and a collectivist orientation (Sussmann & Rashcovsky, 1997). In the Cameroonian context, several authors have pointed out the lack of uniqueness, collectivism, and sameness within the English-speaking Cameroonians (Henry, 2012; Kuate, 2022). Hence, consistent with this, we believe that English Cameroonians who represent Cameroon’s less dominant cultural group are more individualist than the dominant French group. A recent study by Kiridaran Kanagaretnam et al. (2016a) documents a positive association between individualism and the existence of internal control material weakness. Based on the above explanation, hypothesis 1 is proposed.

Hypothesis 1: CEOs with French cultural background have better ICQ and fewer material weaknesses reported than CEOs with English cultural background.

2.2. Impact of CEO duality on the relationship between CEO’s Cultural Background and ICQ

A CEO’s desire to acquire extra power is achievable through the dual role of the CEO and the board chair. Combs et al. (2007) mentioned the existence of two divergent views as far as CEO duality is concerned. A consistent conclusion on the influence CEO duality has on the company’s behavior is yet to be established. On one side, Fama and Jensen’s agency framework (1983) demonstrates that a unified authority structure diminishes the significance of the partition between choice control (chair of the board) and choice administration (CEO). A combined administrative structure might encourage CEO entrenchment, and it is regarded as the essential cause of the misalignment of interest (Kong-Hee et al., 2009). Kong-Hee et al. (2009) conclude that duality could limit the complete transfer of important information between the CEO and the board members. Moreover, when CEO duality exists, the CEO’s control will be amplified, and imperatives will be diminished in this manner, permitting the CEO to create choices agreeing to his individual judgment, thus, affecting the formulation of the enterprise strategy to a large extent, which will further amplify the existence of agency problems. Furthermore, CEO duality will more effortlessly lead to the CEO’s irrational decision behavior, and it could also weaken the supervision of the board of directors on non-rational decision-making of the CEO. In addition, Peng et al. (2007), examining Chinese listed companies, discovered that CEO duality reduced the need for high-quality audit services and additional reduced audit fees. Lastly, Chidambaran et al. (2010) mentioned the existence of a positive relationship between CEO-board connectedness and the likelihood of extortion. From this light, one can deduce that CEO duality can make CEOs easily override or weaken internal control systems and corporate monitoring systems for the benefit of their personal interest.

On the other side, Kiridaran Kanagaretnam et al. (2016a) discovered that a unitary structure could increase operating efficiency and effectiveness, and the management team could intend encourage a strong internal control system in order to better manage internal operations and decision-making. In a recent study, Khlif and Samaha (2019) highlight that board independence was significantly related to internal control quality. While testing for the moderating impact of CEO duality on this relationship, it was significantly positive under combined board leadership structure and negative under separated leadership structure. Moreover, Pathan (2009), while examining the banking industry, finds that CEO power measured with the CEO’s ability to be in command of board decisions was negatively related to the enterprise’s risk-taking. From this point of view, it is also possible that CEOs who serve as chairperson of the board make full use of their authority in order to deal with material weaknesses more carefully and effectively. Furthermore, while examining US firms, Lin et al. (2014) document that CEOs who serve as chairperson of the board had fewer material weaknesses reported as compared to less powerful CEOs. Based on the opinions inferred from previous studies, hypothesis 2 is proposed.

Hypothesis 2: CEO duality will strengthen the negative relationship between CEOs with French cultural background and material weaknesses reported.

2.3. Impact of Board size on the relationship between CEO’s Cultural Background and ICQ

A board is nothing more than a group of persons trying to work together in order to create the best results for the company they are charged to direct and protect. Board effectiveness directly determines the extent of the board function, and the board size remains one of the essential factors which determines board effectiveness. Human beings work best in groups of certain sizes, and over the years, several studies have been conducted on the effectiveness of group decision-making. Agency theory suggests that “the board size affects the extent of a company’s monitoring, controlling and decision making.” Empirical research reports different results, and some researchers discovered that bigger board sizes took more effort to reach an agreement. Thus, smaller board sizes need less extreme and more compromises to make final decisions. Guest (2009) discovered that larger board sizes operate less effectively and experience more communication and coordination difficulties. Additionally, Menozzi et al. (2012) find that bigger board sizes in state-owned enterprises could suffer from coordination problems. Moreover, in a study by Martin and Lorsch (1992), they highlight that it could be more difficult for members to express their ideas and opinions when the board was composed of more than 10 people. Mcnulty et al. (2013) also discover that smaller boards performed better than bigger boards in a risk management function.

In contrast, based on a univariate analysis of public sector companies in South Asia, Ashfaq and Rui (2019) find that board and audit committee characteristics and ownership by government had a significant positive impact on internal control disclosure. Additionally, larger board sizes can be helpful to solve distributional conflicts among different shareholders. Moreover, Samaha et al. (2015) suggest that larger board sizes allow diverse experiences and opinions which can possibly increase a board’s supervisory capacity. Furthermore, in a more recent study, Zubeltzu-Jaka et al. (2020) used a meta-analytic perspective to examine the influence the size of the firm’s board of directors had on corporate social performance. In their results, they highlight that larger and more independent boards can better represent stakeholder’s activities and allow firms to achieve their business objectives. In addition, Oradi et al. (2020) stipulate that a negative relationship exists between the board size and material weaknesses reported. Accordingly, we expect a negative relationship between board size and material weakness reported. Based on the above analysis, hypothesis 3 is proposed:

Hypothesis 3: Larger board size will strengthen the negative relationship between CEOs with French cultural background and material weakness reported.

2.4. Impact of CEO Tenure on the relationship between CEO’s Cultural Background and ICQ

Research has proven that the CEO’s characteristics can essentially influence the enterprise’s decision-making and the CEO’s tenure remains an essential aspect of the CEO’s personal characteristics. Hambrick and Mason (1984b) state that the tenure of the CEO could be predicted to have an important effect on the CEO’s decisions or the decisions of the top management team. Literature review on corporate governance suggests a short CEO tenure since CEO entrenchment could be a function of the period the CEO has been working at the office and an entrenched CEO undermines shareholder’s wealth (Garcia-Mejido et al., 2017a; Goodwin et al., 2009; Azlan et al., 2014). Ali and Zhang (2015), in their study, went further to show that long-tenured CEOs practice timely loss recognition and lower discretionary accruals. Prior studies also suggest that CEO tenure is negatively related to the firm’s performance (Bizjak et al., 2008; Jensen, 1993). In addition, Lin et al. (2014) show that CEO tenure is positively related to a material weakness reported under SOX 404. From this light, it can be deduced that the internal control system and the corporate monitoring system can easily be weakened by longer length of CEO tenure.

On the other hand, the significance of the length of the tenure of the CEO is also well documented in accounting research. Milbourn (2003) reports that CEO tenure is positively related to the market perception of a CEO’s ability. Additionally, Francis et al. (2008), in their study, find the existence of a positive relationship between financial reporting quality and CEO tenure. They further state that tenure builds and increases the CEO’s reputation, which drives the CEO to have better internal control quality. Hence, the CEO’s knowledge and experience on accounting methods and most misreported areas increase when he or she has a longer length of tenure. As a result, the CEO’s ability to prevent and discover any irregular behavior increases. Furthermore, Yazawa (2015) stipulates that the tenure of the CEO is negatively related to the disclosure of a material weakness. Hence, CEOs with a longer length of tenure are expected to be more entrenched, and as a result, positively related with better internal control quality and fewer material weaknesses reported. Based on the standpoints inferred from prior studies, the fourth hypothesis is developed:

Hypothesis 4: Longer CEO tenure will strengthen the negative relationship between CEOs with French cultural background, and material weaknesses reported

3. Data and research design

3.1. Sample selection and data sources

The sample of this empirical research comprises 100 medium-scaled, unlisted firms in Cameroon for the period 2016 to 2019. It gave a total number of 400 observations. These firms are selected because of data availability in different platforms in Cameroon, such as Cameroon Chambers of Commerce and GICAM (Groupement International du Cameroon). The study focuses on non-financial firms and does not consider financial firms because they are more regulated than the others due to the particular role they play within the Cameroon economy and the higher external monitoring they are subject to from shareholders. The years 2016 to 2019 are selected because the basic norms on internal control were revised by COBAC in 2016. Initially, there were 120 firms and 480 observations, but 05 firms were deleted because they had missing CEO characteristic data, 05 firms were created after 2016 and 10 firms had CEOs who were of foreign nationality. Table 1 describes the sample selection procedure.

Table 1. Sample selection

Element	Observation
Initial observation for the period 2016 to 2019 (120*4)	480
Less: firms having missing CEO characteristics data (10*4)	20
Less: firms created after 2016 (5*4)	20
Less: firms with CEO of foreign nationality (5*4)	40
Final sample	400
Final sample of French CEOs (58/100*400)	232
Final sample of English CEOs (42/100*400)	168

Both primary and secondary sources were used to collect the data. The secondary source, which is the main data source, includes both archival and financial reports obtained from the Cameroon Chambers of Commerce and major financial websites. Various authors such as Behr (2015) and Smith (2015) support the idea of collecting data from both archival documents and company reports. Given prior research (Lin et al., 2014; Oradi et al., 2020), the internal control quality of firms is measured by an internal control material weakness disclosed by auditors. Information regarding the material weaknesses of firms, CEO tenure, the board size, and loss were obtained from the financial report. A questionnaire comprising structured questions, which entailed open-ended questions, was administered to get information about the CEO’s cultural background, age, and gender since all could not be obtained from the financial report. A questionnaire was also used because previous studies could not conduct this study due to data unavailability about the cultural background of the CEO in the financial report and thus suggested a questionnaire approach to obtain the data (Lin et al., 2014; Oradi et al., 2020).

3.2. Model and variables

Relying on prior research (Chee Yeow Lim Kanagaretnam et al., 2014; Hollis et al., 2007), the study estimates the following Panel Probit model of the determinants of internal control material weaknesses to investigate the relationship which exists between a CEO’s cultural background and firm’s internal control quality:

(1) $\begin{array}{rl}& \mathit{I}\mathit{C}{\mathit{Q}}_{it}=\hspace{0.17em}{\mathit{\beta }}_{0\hspace{0.17em}+\hspace{0.17em}}\hspace{0.17em}{\mathit{\beta }}_1{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{C}\mathit{U}\mathit{L}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_2{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{D}\mathit{U}\mathit{A}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_3{\left(\mathit{B}\mathit{D}\mathrm{\_}\mathit{S}\mathit{I}\mathit{Z}\mathit{E}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_4{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{T}\mathit{E}\mathit{N}\right)}_{\mathit{i}\mathit{t}}+\\ & {\mathit{\beta }}_6{\left(\mathit{C}\mathit{O}\mathit{N}\mathit{T}\right)}_{\mathit{i}\mathit{t}}+{\epsilon }_{\mathit{i}\mathit{t}}\end{array}$(1)

To investigate the impact of the moderating variables (CEO duality, board size, and CEO Tenure) on the relationship between the CEO’s cultural background and firm’s internal control quality, the following variables were introduced into the baseline probit models:

(2) $\begin{array}{rl}& \mathit{I}\mathit{C}{\mathit{Q}}_{it}=\hspace{0.17em}{\mathit{\beta }}_{0\hspace{0.17em}+\hspace{0.17em}}\hspace{0.17em}{\mathit{\beta }}_1{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{C}\mathit{U}\mathit{L}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_2{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{D}\mathit{U}\mathit{A}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_3{\left(\mathit{B}\mathit{D}\mathrm{\_}\mathit{S}\mathit{I}\mathit{Z}\mathit{E}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_4{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{T}\mathit{E}\mathit{N}\right)}_{\mathit{i}\mathit{t}}+\\ & {\mathit{\beta }}_6{\left(\mathit{C}\mathit{O}\mathit{N}\mathit{T}\right)}_{\mathit{i}\mathit{t}}+{\epsilon }_{\mathit{i}\mathit{t}}+{\mathit{\beta }}_5{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{C}\mathit{U}\mathit{L}\mathbf{\ast }\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{D}\mathit{U}\mathit{A}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_6{\left(\mathit{C}\mathit{E}{\mathit{O}}_{\mathit{C}\mathit{U}\mathit{L}}\mathbf{\ast }\mathit{B}\mathit{D}\mathrm{\_}\mathit{S}\mathit{I}\mathit{Z}\mathit{E}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_7{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{C}\mathit{U}\mathit{L}\mathbf{\ast }\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{T}\mathit{E}\mathit{N}\right)}_{\mathit{i}\mathit{t}}+\\ & \hspace{0.17em}{\mathit{\beta }}_{8\hspace{0.17em}}{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{A}\mathit{G}\mathit{E}\right)}_{\mathit{i}\mathit{t}}+{\mathit{\beta }}_9{\left(\mathit{C}\mathit{O}\mathit{N}\mathit{T}\right)}_{\mathit{i}\mathit{t}}+{\epsilon }_{it}\end{array}$(2)

3.2.1. Dependent variable

The study’s explained variable is internal control quality. On the premise of the existing literature, such as that of Lin et al. (2014) and Oradi et al. (2020), a material weakness is used to measure the internal control quality of firms. Material weakness takes the value 1 when a company’s auditor reports a material weakness and 0, otherwise.

3.2.2. Independent variables

The explanatory variable is culture $\left(CEO\mathrm{\_}CUL\right),$ and the native place of the CEO is used to measure it. It takes the value 1 when a CEO has a French cultural background (west or south region and center or littoral region) and 0 when he/she has an English cultural background (Northwest or Southwest region). It is important to note that all the firms used in the study are located in the French regions of Cameroon.

3.2.3. Moderating variables

This paper has three moderating variables, which are CEO duality $\left(CEO\mathrm{\_}DUA\right)$, board size, and CEO tenure. CEO duality is a dummy variable which takes the value 1 when the CEO and chairman are the same person, and 0, otherwise. The board size is measured by the number of board members, and the CEO tenure is measured by the length of tenure of the CEO.

3.2.4. Control variables

According to prior research, this study controls for 1) Firm characteristics variables, including firm age, sales growth, and firm size, because Doyle et al. (2007) highlight that smaller, younger, and rapidly growing firms tend to report internal control material weaknesses. Thus, the existence of a material weakness will be negatively related to firm size and age, but positively related to sales growth; 2) personal characteristics of the CEO, including CEO’s age, gender, and financial expertise, because Lin et al. (2014) and Oradi et al. (2020) document that a negative relationship exists between the CEO’s age, financial expertise, and the disclosure of a material weakness. Additionally, they as well suggest that female CEOs had better internal control quality than male CEOs, though it was not significant; 3) the company’s financial condition is controlled with the variable Loss, as Doyle et al.(2007) and Hollis et al. (2007) find that financially weaker firms mostly disclose material weaknesses; 4) lastly, the year indicator and firm indicator variables are used to control for the year and firm-fixed effects.

4. Empirical results

4.1. Descriptive statistics and correlation

Table 2 presents the descriptive statistics for the dependent variables and the independent variables. Internal control material weaknesses (MW) is a dummy which takes the value 1 when a firm’s auditor reports a material weakness and 0 otherwise. In the sample firms, 61.22% had poor internal control quality because auditors reported material weaknesses. The mean value of the dependent variable is 0.612%, and the standard deviation is 0.488%. Because of the structural opportunity inequality in Cameroon, 58% of the CEOs of our sample are from the French cultural background, and 42% are from the English cultural background, which gives a mean value of 0.584%. CEO duality represents a dummy variable, which takes the value 1 when CEO duality exists and 0 otherwise. CEO duality has a mean value of 0.765% and a standard deviation of 0.424%. At the same time, board size ranges from a minimum of 2 to a maximum of 10 board members. It has a mean value of 4.339% and a standard deviation of 2.315%. Additionally, CEO tenure has a mean value of 9.423% and a standard deviation of 6.566%. The length of the CEO’s tenure ranges from 1 to 28 years.

Table 2. Descriptive statistics

	(1)	(2)	(3)	(4)	(5)
VARIABLES	N	Mean	SD	Min	Max
Gender	400	0.804	0.398	0	1
CEO_Age	400	45.94	11.23	24	68
Edu_background	400	2.827	1.203	1	5
CEO_cul	400	0.584	0.493	0	1
Long_ten	400	9.423	6.566	1	28
CEO_Duality	400	0.765	0.424	0	1
Firm_age	400	11.43	7.377	1	33
Firm_size	400	24.80	17.38	5	60
Board_size	400	4.339	2.315	2	10
MW	400	0.612	0.488	0	1
Sales_Growth	400	0.078	0.153	−0.53	1.18
Loss	400	0.472	0.500	0	1

With respect to the control variables, 80.36% of the CEOs are of the male gender, and 19.64% are of the female gender. Gender has an average number of 0.804%. The high percentage difference between the genders shows that the majority of firms in Cameroon, like most other African countries, are managed by the male gender. Moreover, the age of the firms as counted from 2016 ranges from 1 to 33 years. Going further, the firm size measured with the number of workers ranges from 5 to 60 people and has a mean value of 24.80%. Furthermore, loss is represented by a dummy variable and takes the value 1 when a firm reports negative earnings and 0, otherwise. In the study sample, 0.472% of the firms reported a loss. In addition, Educational Background has a mean value of 2.827%, out of which 21.17% of the CEOs have GCE Advanced Level, 10.71% have Higher National Diploma, 40.56% have Bachelor’s Degree, 19.39% have Master’s Degree, and 8.16% have PhD. Moreover, the age of the CEOs ranges from 24 to 68 years. Lastly, the average sales growth is 0.078, with the maximum sales growth being 1.18% and the minimum sale growth −0.52%.

Table 3 presents the correlation analysis of the variables of interest. This analysis helps to check for multicollinearity between the dependent and independent variables and between the independent variables. Correlation displays the degree of relationship between variables. Correlation between the independent variables should not be more than 0.7%; otherwise, there is multicollinearity between the independent variables. From Table 3, it is obvious that there is no multicollinearity between the variables, the maximum correlation found is between firm size and firm age with a value of 0.675%.

Table 3. Correlation matrix

	MW	CEOcul	Boardsize	Longten	CEOdual	Gender	CEOage	Edubac	Firmag	Firmsize	Loss	Salegrth
MW	1
CEOcul	−0.369***	1
Boardsize	−0.513***	0.245***	1
longten	0.435**	0.295***	0.625***	1
CEOdual	−0.313***	0.153**	0.205***	0.182***	1
Gender	0.185***	0.0624	0.228***	0.215***	0.0285	1
CEOAge	−0.459***	0.268***	0.608***	0.492***	0.252***	0.244***	1
Edubackgrnd	−0.420***	0.205***	0.244***	0.242***	0.0295	0.205***	0.168***	1
Firmage	−0.453***	0.304***	0.666***	0.449***	0.105*	0.250***	0.495***	0.274***	1
Firmsize	−0.388***	0.304***	0.048***	0.635***	0.106*	0.325***	0.630***	0.201***	0.675***	1
Loss	0.0205	0.0270	−0.0248	0.133**	−0.110*	0.0611	0.0644	0.140**	0.1000*	0.0628	1
SalesGrowth	−0.0732**	0.0266	−0.141**	−0.0819	−0.158**	−0.0609	0.0291	0.0369	−0.0891	−0.143**	0.0312	1

Source: Author construct. Note: these are the level of significance * p < 0.05, ** p < 0.01, *** p < 0.001

4.2. Regression result

4.2.1. Empirical result for hypothesis 1

Table 4, model 1, columns 1 and 2, shows the result of the relationship that exists between the CEO’s cultural background and the internal control quality of firms. From the table, it is evident that being a CEO with a French cultural background as compared to a CEO with an English cultural background reduces the probability of having a material weakness reported by 0.1712%. In other words, CEOs with French cultural background are 0.1712% more likely to maintain better internal control quality as compared to CEOs with English cultural background. Similarly, being a French CEO as against an English CEO reduces the Z-score by 0.8540%, following the coefficient of the estimated probit model in column 1. The results are significant at 1%, Wald chi2 has a value of 55.060% and a probability of 0.000, which signifies that this model has a good fit. The result supports the conjecture of hypothesis 1, which states that CEOs with French cultural background have better ICQ and fewer material weaknesses reported than CEOs with English cultural background. Also, this finding is consistent with the notion that the marginalized English citizens in Cameroon have an individualistic nature, which makes English CEOs more likely to breach internal controls for opportunistic reasons, such as managing earnings to meet performance benchmarks, whereas the collectivism nature of the French CEOs and the structural opportunity inequality with regard to the language, laws, and politics which exist in Cameroon makes French CEOs have higher risk awareness and stronger prevention motivation, thus forming their risk-averse nature. This finding is consistent with studies conducted by Kiridaran Kanagaretnam et al. (2016a) and Han et al. (2010), which document the existence of fewer material weaknesses in societies with high collectivism and higher material weaknesses in individualistic societies.

Table 4. Regression result

Dependent Variable: ICQ measured by Material Weakness reported
	Model 1		Model 2		Model 3			Model 3
	(1)	(2)	(3)	(4)	(5)		(6)	(7)	(8)
VARIABLES	Probit coefficient	Marginal effects	Probit coefficient	Marginal effects	Probit coefficient		Marginal effects	Probit coefficient	Marginal effects
1.CEOcul	−0.8540***	−0.1712***	−0.0394***	−0.1576***	−0.1969***		−0.1855***	−0.8613**	−0.1746***
	[0.309]	[0.062]	[0.742]	[0.053]	[0.738]		[0.054]	[0.541]	[0.063]
Edubackgrnd	−0.7811***	−0.1417***	−0.7000***	−0.1256***	−0.6646***		−0.1212***	−0.7791***	−0.1388***
	[0.187]	[0.024]	[0.159]	[0.022]	[0.174]		[0.025]	[0.189]	[0.024]
CEOAge	−0.0722***	−0.0131***	−0.0563***	−0.0101***	−0.0600***		−0.0109***	−0.0741***	−0.0132***
	[0.020]	[0.004]	[0.017]	[0.003]	[0.018]		[0.003]	[0.019]	[0.004]
Gender	0.1823	0.0331	0.2820	0.0506	0.4036		0.0736	0.1832	0.0326
	[0.413]	[0.075]	[0.336]	[0.060]	[0.341]		[0.062]	[0.431]	[0.077]
Firmage	−0.0633**	−0.0115***	−0.0511**	−0.0092**	−0.0317**		−0.0158**	−0.0891**	−0.0159**
	[0.026]	[0.004]	[0.024]	[0.004]	[0.012]		[0.004]	[0.040]	[0.007]
SalesGrowth	0.7531	0.1366	0.2564	0.0460	0.5216		0.0951	0.7288	0.1298
	[0.781]	[0.142]	[0.647]	[0.116]	[0.711]		[0.130]	[0.785]	[0.142]
Firmsize	−0.0219**	−0.0248**	−0.0168**	−0.0103**	−0.0527***		−0.0451**	−0.0146**	−0.0236**
	[0.007]	[0.005]	[0.008]	[0.004]	[0.016]		[0.018]	[0.006]	[0.008]
Loss	0.0297**	0.0344**	0.0975**	0.0175**	0.0436**		0.0179***	0.0237**	0.0309**
	[0.013]	[0.012]	[0.018]	[0.008]	[0.020]		[0.006]	[0.011]	[0.013]
1.CEOChair			−0.5238***	−0.2337***
			[0.044]	[0.078]
1o.CEOChair#0b.CEOcul			−0.9765***	−0.2887***
			[0.078]	[0.044]
Boardsize					−0.4978***		−0.0838***
					[0.153]		[0.020]
1.CEOcul#c.Boardsize					−0.0521***		−0.0758***
					[0.014]		[0.020]
Longten								0.0322	0.0055
								[0.090]	[0.014]
1.CEOcul#c.longten								0.0022	0.0558
								[0.045]	[0.178]
Constant	7.3254***		6.7708***		7.4840***			7.4296***
	[1.171]		[1.024]		[1.111]			[1.261]
Observations	400	400	400	400	400		400	400	400
Number of Firms	100		100		100			100
Wald chi2:	55.060	.	61.793	.	61.115		.	56.188	.
Prob Wald:	0.000	.	0.000	.	0.000		.	0.000	.

Robust standard errors in brackets

*** p < 0.01, ** p < 0.05, * p < 0.1

With regard to the control variables in model 1, one-unit increase in the CEO’s age variable decreases the probability of having a material weakness reported by 0.0131%, as shown in column 2. Similarly, one-unit increase in the age of the CEO increases the probability of maintaining better internal control quality by 0.0131%. This displays that companies with older CEOs have way better performance with respect to internal control establishment and maintenance than companies with younger CEOs. From the result, it is also evident that the relationship between the gender of the CEO and material weakness is positive but not significant with a coefficient of 0.0331%, as shown in column 2. This signifies that male CEOs are 0.0331% more likely to have a material weakness reported compared to female CEOs. In other words, a female CEO is 0.0331% more likely to maintain better internal control quality than a male CEO, this relationship remains insignificant and maybe as a result of the limited number of female CEOs in the sample. These findings are consistent with the study of Lin et al. (2014), which stipulates that older CEOs have better ICQ than younger CEOs. Additionally, their results indicate that female CEOs have better ICQ than male CEOs, but their findings on the gender of the CEO were also insignificant. Moreover, it is obvious from the results of model 1, column 2, that an increase in the CEO’s educational background decreases the probability of having a material weakness reported by 0.1417%. This finding is similar to the research of Oradi et al. (2020); their results highlight a negative relationship between the financial expertise of the CEO and material weakness.

Additionally, examining the firm's characteristic variables, the study finds that one-unit increase in the age of the firm decreases the probability of having a material weakness reported by 0.0115%, as shown in model 1, column 2. That is, older firms are 0.0115% more likely to maintain better internal control quality than younger firms. Moving to the firm size, the study finds that one-unit increase in the size of the firm decreases the firm’s probability of having a material weakness reported by 0.0248%. These findings are consistent with studies conducted by Doyle et al.(2007), Hollis et al.(2007), and Kiridaran Kanagaretnam et al.(2016a) that discover that firms reporting internal control material weaknesses tend to be smaller and younger. Thus, the existence of material weaknesses is negatively associated with the size and the age of firms. Furthermore, the results also show that an increase in the sales growth unit increases a firm’s probability of having a material weakness reported by 0.1366%, as shown in column 2, model 1. The sale growth result is in line with previous studies (Doyle et al., 2007; Hollis et al., 2007), which stipulate that sale growth is positively associated with the disclosure of a material weakness. Though the sale growth value remains insignificant, it is believed it could probably be due to differences in the business scale and sizes. This study used medium scaled enterprises, whereas previous studies used large scaled businesses. Lastly, loss and material weakness are positively related, as shown in model 1, columns 1 and 2. This signifies that an increase in the loss unit decreases the probability that a firm will maintain good internal control quality by 0.0344%. This finding is consistent with studies conducted by Kiridaran Kanagaretnam et al.(2016a), which stipulate that enterprises disclosing material weaknesses tend to be financially weaker.

4.2.2. Empirical result for hypothesis 2

Table 4, model 2, shows the result for equations 2(2) $\begin{array}{rl}& \mathit{I}\mathit{C}{\mathit{Q}}_{it}=\hspace{0.17em}{\mathit{\beta }}_{0\hspace{0.17em}+\hspace{0.17em}}\hspace{0.17em}{\mathit{\beta }}_1{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{C}\mathit{U}\mathit{L}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_2{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{D}\mathit{U}\mathit{A}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_3{\left(\mathit{B}\mathit{D}\mathrm{\_}\mathit{S}\mathit{I}\mathit{Z}\mathit{E}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_4{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{T}\mathit{E}\mathit{N}\right)}_{\mathit{i}\mathit{t}}+\\ & {\mathit{\beta }}_6{\left(\mathit{C}\mathit{O}\mathit{N}\mathit{T}\right)}_{\mathit{i}\mathit{t}}+{\epsilon }_{\mathit{i}\mathit{t}}+{\mathit{\beta }}_5{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{C}\mathit{U}\mathit{L}\mathbf{\ast }\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{D}\mathit{U}\mathit{A}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_6{\left(\mathit{C}\mathit{E}{\mathit{O}}_{\mathit{C}\mathit{U}\mathit{L}}\mathbf{\ast }\mathit{B}\mathit{D}\mathrm{\_}\mathit{S}\mathit{I}\mathit{Z}\mathit{E}\right)}_{\mathit{i}\mathit{t}}+\hspace{0.17em}{\mathit{\beta }}_7{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{C}\mathit{U}\mathit{L}\mathbf{\ast }\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{T}\mathit{E}\mathit{N}\right)}_{\mathit{i}\mathit{t}}+\\ & \hspace{0.17em}{\mathit{\beta }}_{8\hspace{0.17em}}{\left(\mathit{C}\mathit{E}\mathit{O}\mathrm{\_}\mathit{A}\mathit{G}\mathit{E}\right)}_{\mathit{i}\mathit{t}}+{\mathit{\beta }}_9{\left(\mathit{C}\mathit{O}\mathit{N}\mathit{T}\right)}_{\mathit{i}\mathit{t}}+{\epsilon }_{it}\end{array}$(2) . These results answer the research questions as to whether the negative effect of CEOs with French cultural background and material weakness reported is enhanced by CEO duality, board size, and CEO tenure, respectively, or not. It is evident from the results in columns 3 and 4 that the coefficient of the interactive term between CEO cultural background and CEO duality is statistically significant and negative. The results suggest that the impact of CEOs with French cultural background on internal control quality increases when CEOs with French cultural background double up in the board of management. In other words, the negative association between CEOs with French cultural background and material weakness reported is further enhanced when CEO duality exists. Accordingly, the marginal effect of a change in the CEO’s cultural background on material weakness is no longer only equal to the estimated coefficient but the sum of its coefficient and its interaction term with CEO duality. Thus, all things being equal, the negative effect of the French CEOs with having a material weakness reported increases by 0.2887% due to the significant negative sign of its interaction with CEO duality. The results support the prediction of hypothesis 2, CEO duality will therefore enhance the negative relationship between CEOs with French cultural background and material weakness reported. This finding is consistent with that of Khlif and Samaha (2019), who find that board independence is positively and significantly related to better internal control quality under unitary board leadership structure (CEO duality), and it is negative and significant under dual-leadership structure. The result is also similar to a study by Lin et al.(2014). Their results find fewer material weaknesses reported by CEOs who served as the chairperson compared to CEOs who did not serve as the chairperson.

4.2.3. Empirical result for hypothesis 3

Similarly, columns 5 and 6 under model 2 of Table 4 capture the impact of the moderating of board size on the relationship between the CEO’s cultural background and internal control quality. Here also, the coefficient of the interaction term between CEO cultural background and board size is negatively significant as well as the individual association between material weakness and CEO cultural background. This submits that CEOs with French cultural background and board sizes are more likely to complete each other in boosting internal control quality in the businesses under consideration. Meaning, the impact of the French CEO on internal control is further enhanced when the board size of the business they head increases. In other words, the marginal impact of French CEOs on material weakness will increase when their board size increases. Thus, the marginal effect of French CEOs on internal control is added on by 0.0758% due to the negative coefficient of their interactive term. The result supports the prediction of the third hypothesis: a larger board size strengthens the negative relationship between the CEO with French cultural background and material weaknesses reported. This finding is in line with a study by Oradi et al.(2020) that documents that board size was negatively related to material weaknesses disclosure.

4.2.4. Empirical result for hypothesis 4

In contrast, under model 2, columns 7 and 8 of Table 4, which captures the moderating role of CEO tenure, it is evident that the coefficient of the interaction term between CEO cultural background and CEO tenure is positive but statistically insignificant. This proposes that the negative effect of French CEOs on material weakness reported will reduce when the French CEO’s tenure increases. Thus, all things being equal, the negative effect of the French CEOs on material weakness will reduce by 0.0558% due to the positive sign of its interaction with CEO tenure. However, as observed from the result, the moderation role of a CEO tenure is insignificant on the relationship between CEO cultural background and material weakness. Thus, it can be deduced that CEO tenure has no statistical impact on the association between French CEOs and internal control quality. This finding does not support our prediction of hypothesis 4, hence it is rejected. The study’s result is similar to that of Hollis et al.(2007), which stipulates that CEO tenure is insignificantly related to the internal control quality of companies that are not mandated to comply with SOX 404 in 2003. A more recent study by Lin et al.(2014) also found a similar and more persuasive result based on a larger firm sample following the SOX 404 requirements. They discovered that the tenure of the CEO was positively associated with a material weakness reported under SOX 404.

4.3. Additional analysis

Table 5 offers the comprehensive predictive probability for both the French and English CEOs as well as the other interactive predictor variables at all levels of the sample values on material weakness. Column one shows the predictive probability for CEOs with both French and English cultural background. From the results, it is evident that English CEOs have a greater predicted probability of having a material weakness reported (0.706%) as against the French CEOs (0.562%). Thus, this suggests that French CEOs have better internal control quality than English CEOs. This outcome reflects exactly the expectation of the first hypothesis.

Table 5. Predictive probability of French and English CEOs

	(1)	(2)	(3)	(4)
VARIABLES	Marginal effects	Marginal effects	Marginal effects	Marginal effects
French CEO (French CEO_BD 2)	0.562***	-	0.797***	-
	(0.0379)		(0.0744)
English CEO (French CEO_BD 4)	0.706***	-	0.617***	-
	(0.0390)		(0.0489)
French CEO_BD 6	-	-	0.403***	-
			(0.0564)
French CEO_BD 8	-	-	0.213**	-
			(0.0904)
French CEO_BD 10	-	-	0.0869	-
			(0.0803)
English CEO_BD 2	-	-	0.927***	-
			(0.0375)
English CEO_BD 4	-	-	0.811***	-
			(0.0398)
English CEO_BD 6	-	-	0.616***	-
			(0.0693)
English CEO_BD 8	-	-	0.380***	-
			(0.134)
English CEO_BD 10	-	-	0.180**	-
			(0.146)
French CEO_No CEODUA	-	0.789***	-	-
		(0.0870)
French CEO_ CEODUA	-	0.500***	-	-
		(0.0440)
EnglishCEO_No CEODUA	-	0.781***	-	-
		(0.0809)
English CEO_ CEODUA	-	0.688***	-	-
		(0.0437)
English CEO_ Tenure 1 yr	-	-	-	0.639***
				(0.130)
French CEO_ Tenure 1 yr	-	-	-	0.506***
				(0.109)
English CEO_ Tenure 8 yr	-	-	-	0.695***
				(0.0412)
French CEO_ Tenure 8 yr	-	-	-	0.557***
				(0.0373)
English CEO_ Tenure 16 yr	-	-	-	0.747***
				(0.0878)
French CEO_ Tenure 16 yr	-	-	-	0.607***
				(0.0950)
English CEO_ Tenure 25 yr	-	-	-	0.795***
				(0.159)
French CEO_ Tenure 25 yr	-	-	-	0.656***
				(0.179)
Observations	400	400	400	400

Standard errors in parentheses

*** p < 0.01, ** p < 0.05, * p < 0.1

Column 2 also presents the interactive role of CEO duality on the association between French CEOs and English CEOs, respectively, on the disclosure of material weakness. The result shows that French CEOs have a lower predictive probability of 0.500% of having a material weakness reported as compared to the English CEOs’ predictive probability of 0.688%. This means that French CEOs who double up as part of the management of the business maintain better ICQ and fewer material weakness reported than English CEOs who doubles up as part of the management of the business. Hence, CEO duality enhances the internal control quality of CEOs with French backgrounds better than CEOs with the English background. This result is in consonance with the second hypothesis of the study.

Column 3 in Table 5 also captures the moderating role of board size in the association between CEO cultural background (French CEO and English CEO) and material weakness. The predicted probability for the values of board size was generated from board size of 2 to 10 in the increment of 2 on the association of CEO cultural background and material weakness. From the results, it is evident that the predicted probability of having a material weakness reported by a French CEO is 0.797% if the board size of the company is 2 but decreases to 0.0869% as the board size increases to 10. Similarly, the predicted probability for English CEOs is 0.927% when the board size is 2 and also reduces to 0.180% when the board size is 10. However, juxtaposing the two, it is obvious from the results that CEOs with French Background have a lesser predictive probability of reporting material weaknesses than CEOs with English background when both CEOs are moderated by the board size.

Moreover, column 4 depicts the interactions of the CEOs tenure and CEO’s cultural background on material weakness. The predicted probability for the values of CEO tenure was generated from 1 year of tenure to 25 years of tenure in the increment of 8 years on the association of CEO cultural background and material weakness. The result shows that the predicted probability of having a material weakness by a French CEO is 0.506% when the French CEO tenure is just a year, but increases to 0.656% as the tenure of the CEO increases to 25 years. Equally, the predicted probability for an English CEO is 0.639% when the tenure of the CEO is a year but also increases to 0.795% when the tenure of years increases to 25 years. Comparing the effect of the CEO tenure on the relationship between French and English CEOs on material weakness, it can be deduced that English CEOs have a greater probability of reporting material weaknesses than French CEOs as the number tenure increases for both CEOs. Again, the results suggest that long years of CEO tenure turn out to weaken the internal control quality of firms.

4.4. Robust test using panel logit model

A robust test using a Panel Logit Model was employed. Table 6, models 1 and 2 provide the robust test results for hypotheses 1, 2, 3, and 4. From the result in Table 6, the sign and significance of the parameters have not changed. Moreover, the output or results are not statistically different as envisaged in the marginal effects of both models (panel and logit). Hence, the results of the study are robust even to the estimation of a different model.

Table 6. Robust test using a panel logit model

Dependent Variable: Material Weakness
	Model 1		Model 2		Model 2		Model 2
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
VARIABLES	Logit Coefficient	Marginal effects	Logit Coefficient	Marginal effects	Logit Coefficient	Marginal effects	Logit Coefficient	Marginal effects
1.CEOcul	−0.5171***	−0.1739***	−0.1596**	−0.1608***	−0.3143**	−0.1886***	−1.5759**	−0.1770***
	[0.553]	[0.063]	[1.347]	[0.053]	[1.344]	[0.056]	[0.966]	[0.063]
Edubackgrnd	−1.3891***	−0.1421***	−1.2439***	−0.1253***	−1.1993***	−0.1235***	−1.3869***	−0.1388***
	[0.332]	[0.023]	[0.281]	[0.023]	[0.303]	[0.024]	[0.334]	[0.024]
CEOAge	−0.1270***	−0.0130***	−0.1005***	−0.0101***	−0.1072***	−0.0110***	−0.1309***	−0.0131***
	[0.036]	[0.004]	[0.031]	[0.003]	[0.032]	[0.003]	[0.034]	[0.004]
Gender	0.3454	0.0353	0.5285	0.0533	0.7345	0.0756	0.3474	0.0348
	[0.744]	[0.076]	[0.597]	[0.060]	[0.601]	[0.061]	[0.777]	[0.078]
Firmage	−0.1120**	−0.0115***	−0.0890**	−0.0090**	−0.0510	−0.0052	−0.1604	−0.0161
	[0.046]	[0.004]	[0.043]	[0.004]	[0.039]	[0.004]	[0.141]	[0.013]
SalesGrowth	1.2908	0.1320	0.4406	0.0444	0.9361	0.0964	1.2478	0.1249
	[1.363]	[0.140]	[1.138]	[0.115]	[1.250]	[0.129]	[1.372]	[0.140]
Firmsize	−0.0192**	−0.0329**	−0.0413***	−0.0241**	−0.0890***	−0.0092***	−0.0298**	−0.0110***
	[0.011]	[0.008]	[0.014]	[0.005]	[0.027]	[0.002]	[0.012]	[0.002]
LOSS	0.3482***	0.0356**	0.2020**	0.0204**	0.0756***	0.0178**	0.0163**	0.0317***
	[0.014]	[0.042]	[0.031]	[0.039]	[0.020]	[0.021]	[0.007]	[0.013]
1.CEOChair			−0.8817***	−0.2319***
			[0.076]	[0.081]
1o.CEOChair#0b.CEOcul			−0.7215**	−0.2888**
			[0.097]	[0.064]
Boardsize					−0.9084***	−0.0888***
					[0.289]	[0.020]
1.CEOcul#c.Boardsize					−0.8726***	−0.3787***
					[0.042]	[0.044]
Longten							0.0575	0.0058
							[0.158]	[0.013]
1.CEOcul#c.longten							0.0003	0.0799
							[0.080]	[0.163]
Constant	12.9616***		11.9839***		13.3380***		13.2084***
	[2.108]		[1.871]		[2.027]		[2.286]
Observations	400	400	400	400	400	400	400	400
Number of Firms	100		100		100		100
Wald chi2:	52.414	.	56.424	.	58.004	.	53.378	.
Prob Wald:	0.000	.	0.000	.	0.000	.	0.000	.

Robust standard errors in brackets

*** p < 0.01, ** p < 0.05, * p < 0

5. Conclusion

Culture finance has been a hot topic in the academic field, and the importance of culture to corporate behavior has been put forward by western scholars (Kwok and Tadesse, 2006; Chen et al., 2015). Given that effective internal control is important to guarantee the reliability and ethical reporting of companies’ performance and financial position, it is thus very important for firms to establish a good internal control system and know the factors which affect it. Although there exists literature on the impact of the CEO’s characteristics such as age, gender, and entrenchment on internal control quality, few studies discuss how the cultural background characteristics of the CEO affect the internal control quality of firms.

The study analyzed the impact of the CEO’s cultural background characteristics on the company’s internal control quality and used the CEO’s native place to measure his or/her cultural background. The results show that the CEO’s cultural background significantly influence a company’s internal control quality. Compared to a CEO with the English cultural background, a CEO with the French cultural background maintains better internal control quality. Additionally, the empirical results show that CEO duality and larger board sizes enhance the positive relationship which exists between the cultural background of the CEO and the internal control quality of firms. Thus, supporting the prediction of hypotheses 2 and 3. However, hypothesis 4 is rejected because a negative but insignificant relationship exists between CEO tenure and material weakness reported.

The study offers insights to regulators and lawmakers interested in the effects of CEO characteristics, precisely his or her cultural background, on internal control quality. Importantly, it points out that CEO cultural background is likely to affect the strength of internal control mechanisms. In order to ameliorate the internal control system of firms in Cameroon, the study recommends that the “Basic Norms of Internal Control” and the “Supporting Guidelines of Internal Control,” and all other laws governing the country be published in English, so as to help English-speaking CEOs maintain a good and effective internal control system. It is also recommended that the government should make English Cameroonians participate on an equal basis in the political, social, and economic affairs of the country in order for them not to feel marginalized. In the meantime, it is recommended that management of companies, particularly English CEOs, should engage the work of French translators to help translate all the necessary documents in order to read, understand, and comprehensively make informed decisions to enhance their internal controls. Overall, the study recommends that all CEOs irrespective of their cultural background should maintain good internal control quality, as this will help them have a higher business performance. Higher business performance would result in paying more taxes which the government officials can use for social amenities which can be beneficial for the community.

Unavoidably, this paper still has some limitations. Due to great challenges in data acquisition, some control variables such as seniority of the CEO, board independence, management rights, business segments, leverage, restructure, merger, and dividends were not considered. These are crucial variables which could affect the relationship between the CEO’s cultural background and ICQ. They were not added to our control variables due to data unavailability since the firms used were medium scaled enterprises. Therefore, we propose future studies to conduct a similar research in a developed country such as Canada that has both French and English systems as Cameroon but has bigger and listed companies.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors received no direct funding for this research.