Effect of Family Involvement and Corporate Governance on Dividend Policy: A Study of Non-Financial Listed Firms in Morocco

Current research on the determinants of dividend policy has focused on the study of developed countries, while research on emerging countries remains limited. The objective of this research is to study the effect of the involvement of the family firm on dividend policy. This paper uses Ordinary Least Squares regression, we analyzed data for Moroccan nonfinancial firms listed on the Casablanca Stock Exchange over the period 2015-2017. The results of this paper show that family firm pay less dividends than non-family firms. Thus, ownership concentration and firm size influence dividend policy, unlike the composition of the board of directors, which doesn’t influence it. Our paper contributes to broaden our understanding of the effect of the family firm on dividend policy in one of the emerging countries. It is suggested that the characteristics of corporate ownership determine the share of profits to be distributed, not to say dividend policy, since concentration of ownership limits the power of minority shareholders to make their voices heard. Despite the presence of laws protecting minority shareholders, their room for maneuver remains limited. Our paper enriches the existing literature on the family firm dividend policy by studying the Moroccan context. The results show that family ownership and family CEO involvement do not have the same influence on the dividend policy of family firms, but it depends on the context. Our paper also provides relevant information for practitioners regarding dividend policy in Moroccan family firms.


Introduction
At the end of each fiscal year and after the closing of the accounts, the firm's shareholders' council decides on the payment of profits. Depending on each firm and its mode of governance, the management team proposes solutions to the board to decide on the share of profits to be reinvested and the share to be distributed as dividends to shareholders. These decisions are analyzed by investors and provide information on future profits, according to researchers (Gugler & Yurtoglu, 2003). The dividend policy is a major issue for management; it's called upon to ensure the financing of profitable projects at the lowest possible cost and to satisfy the expectations of shareholders in terms of dividends, especially minority shareholders. A survey conducted in 1999 by Baker et al. (2001) among firms listed on Nasdaq on the factors influencing dividend policy. The authors found that there are 22 different factors that influence the manager's decision of listed firms regarding dividend policy. Among these factors, three are considered important, namely stability of earnings, the level of current earnings and expected earnings. The survey also concluded that the executive managers attach great importance to the dividend policy and review it at least once a year in view of its potential impact on the firm value. The study by Baker et al. (2001) revealed that the factors that impact dividend payment decisions remain like those discussed for half a century by Lintner (1956).
The study of the characteristics of dividend policy in the family firm has attracted the interest of researchers. However, research remains limited (Briano-Turrent et al., 2019). The conclusions put forward are far from unanimous, except for one point, which is the low level of dividends paid by family-owned firms compared with those paid by non-family firms. The literature on the influence of family involvement on dividend policy is mainly focused on studying the effects of this policy on Type I conflict and Type II agency problems, and agency theory remains the preferred method for explaining this problem.
The purpose of the corporate governance system is to protect the interests of stakeholders in a partnership approach and to protect the interests of shareholders in a shareholder approach. When the firm management is entrusted to non-shareholder managers, the interests of the owners may be threatened. This situation creates conflicts of interest or what is called Type I and Type II agency problems. The interests of each actor are protected by law and also by codes of good governance practices even if they remain just recommendations. Odeleye (2018) studied the relationship between corporate governance and the dividend policy of 97 Nigerien non-financial firms between 1995 and 2012. He found that there is a positive relationship between the two variables and the mode of sector classification plays an important role in this relationship. However, the author believes that the number of independent directors needs to be increased but also of institutional investors.
The dividend policy is seen as a governance mechanism in the hands of the directors, and it helps to mitigate the problems of Type I and Type II agencies. According to Setia-Atmaja et al. (2009), dividends and debt mitigate the expropriation of minority shareholders wealth (Type II agency problem). Pindado et al. (2012) also reached the same conclusion, family firm in the Euro zone use dividends as a mechanism to reduce the intent of conflicts between family majorities and minority shareholders. The family's influence on agency problems depends on the family's participation (González et al., 2014). Minority shareholders demand the payment of dividends when they perceive that there are favorable conditions for the family's expropriation of their wealth (González et al., 2014).
Following a study of 369 Indonesian firms characterized by the concentration of ownership (91% of firms are under the control of a single shareholder), Duygun et al. (2018) concluded that firms with conflicts between managers and owners (Type I agency problems) distribute low dividends. The control of Chinese family shareholders may intensify Type I agency problems more than Type II agency problems and this has a significant and negative effect on dividend policy (Wei et al., 2011). In another study, majority-minority conflict (Type I agency problems) doesn't influence dividend policy (Duygun et al., 2018) and Gadhoum et al. (2017) could not prove that the majority (families) expropriate the wealth of minority shareholders. Michiels et al. (2014) studied another aspect of conflict and that is that intra-family conflict leads to a greater propensity to pay dividends and family governance moderates this relationship. According to Wei et al. (2011), a favorable regional environment plays a positive role on corporate governance by mitigating Type I agency problems and encourages the family firm to pay cash dividends. In the Moroccan context, agency problems within firms are pervasive (Baker & Jabbouri, 2016). Investors are not protected, transparency is low, and poor disclosure of information facilitates expropriation by shareholders (Farooq & Jabbouri, 2015;Jabbouri, 2016). High-growth firms use dividends to attract investors looking for firms that pay the maximum dividends. Xu'nan (2011) concluded that these firms pay more dividends and those that do build a friendly reputation with their future minority shareholders. In the Canadian context, dividends are used as a mechanism to protect minority shareholders from expropriation of their wealth by the majority; the power to protect dividends is less effective in the family firm (Gadhoum et al., 2017). In recent studies, Miah (2022) analyzed the association between ownership structure and dividend payout policy of Bangladeshi firms and found that family-owned firms pay out more dividends than non-family-owned firms in order to alleviate fears of expropriation by non-controlling shareholders.
What is new in this paper is that we study the effect of family involvement and corporate governance on dividend policy in an emerging country, Morocco (Sahay et al., 2015), and these countries are less studied than developed countries. It's different in terms of its level of economic development, its culture, and the specificities of its family firms.
The purpose of our article is to study and answer the following question: What is the effect of family involvement and corporate governance on dividend policy? This question is important because the answer is related to the ownership structure (the effect of capital concentration, family involvement, etc.) and the functioning of corporate governance practices of Moroccan firms, which have unique characteristics. Moroccan firms listed on the Casablanca Stock Exchange are predominantly family-owned and characterized by a high concentration of capital either in the hands of the same person or concentrated in the hands of members of one or two families. A paragraph is dedicated to the presentation of the Moroccan context, which remains typical. Our article is the first study that explores the effect of family involvement in the management of listed firms on dividend policy in Morocco. Studying of the Moroccan context it's interesting because it's considered one of the emerging countries but also with its unique characteristics with respect to its culture and belonging. The study of emerging countries broadens our understanding of this issue. Previous studies have focused more on the study of developed countries such as the United States (Admati, Pfleiderer, & Zechner, 1994;Hollowell, 2006;Villalonga & Amit, 2006), the United Kingdom (Becht, Franks, Mayer, & Rossi, 2008), Germany (Andres, 2008), and Japan (Yoshikawa & Rasheed, 2010). Oubenal and Zeroual (2017) provide an analysis of how Morocco's economic elites are formed. These economic elites are the result of three elements, the first of which is the alliance between the families of the traditional old bourgeoisie and large foreign groups. The second is the dominance of the monarchy over economic sectors through its holding firm and its influence on Moroccan employers. The last is that the economic elites rely on the state to consolidate their position. Moreover, the largest groups have financed their investment through public or semi-public credits. According to Oubenal and Zeroual (2017), the central government uses financial engineering to strengthen its control and influence over Morocco's economic elites. The central government influences two institutional investors, the Caisse de Dépôt et de Gestion (CDG) and the Caisse interprofessionnelle marocaine de retraite (CIMR), which are shareholders in most listed firms. Following their analysis by interlocking directorates, Oubenal and Zeroual (2017) found that there is a high concentration among private groups in the payment of participation on boards of directors and an absence of independent members. Of the 561 seats, 105 seats are controlled by 8 groups, 20% of the 105 of which belong to the CDG. These institutional investors play the role of a financier but also a source of information. Their analysis of the network to show the existence of a central group and a fragmented peripheral.

Moroccan Context
The formation process of Moroccan firms has gone through periods of major transformations influenced and marked by neoliberal reforms that have given a singularity and a paradigmatic character to Moroccan capitalism (Catusse, 2008). Morocco's transformation to liberalism is less radical than in other Arab countries such as its neighbors Tunisia and Algeria. Rather, Morocco transformed itself into a state capitalism after its independence (Richards & Wterbury, 1990). This policy gave rise to a dynamic local bourgeoisie composed mainly of small family farms (Clément, 1995).
Moroccan family capitalism has its roots in economic history characterized by autarky and the influence of the Muslim religion (Labari, 2016). In addition, entrepreneurs have entered the Moroccan political environment through interest groups such as the Confédération Générale des Entreprises du Maroc (CGEM) and the Groupement Professionnel des Banques du Maroc (GPBM). According to Larabi (2016), today's known family firms were tribal men during the war and protectorate . Some of them chose to combine business and politics or even political opportunism to create a solid network of support (Larabi, 2016). Moroccan businessmen find their strength in their status as local notables (Larabi, 2016). The holding Akwa group which includes a dozen companies (Afriquia, Akwa Holding, Maghreb Gaz, Maghreb Oxygène, Nouvelle Tribune, Médi télécom, etc.) two of which are listed on the Casablanca Stock Exchange (Afriquia Gaz and Maghreb Oxygène) owned by the Akhanouch family. According to Forbes, the descendant of this family, Aziz Akhanouch, was ranked 8th in 2020 and 12th in 2021 among the richest in Africa. Mr. Akhanouch is currently a member of the Moroccan government. Moreover, 55.97% of the members of the House of Representatives were entrepreneurs, liberal professions, and business executives, which shows the interest of economic elites in Moroccan political life.

Dividend Policy
Dividend policy can be defined as the trade-off between profit retention and the payment of liquidity to shareholders (Albouy, 1990). It's a source of conflict between shareholders and managers (Jensen, 1986). The intensity of these conflicts becomes serious when the firm's free cash flows exceed profitable investment opportunities. In this case, firms choose to go into debt as a precondition for paying off excess cash. The opposite is true, as firms with more growth opportunities have less free cash flow and therefore pay less dividends. Smith and Watts (1992) find that there is a significant negative relationship between dividend yield and stock options.
La Porta et al. (2000) analyzes the dividend, according to two models, the first "outcome model" and "substitute model". The first model considers dividends as an "outcome of an effective system of legal protection of shareholders. "Minority shareholders have the right to force the firm to pay dividends because majority shareholders can influence dividend policy by appointing executives or selling shares to hostile parties that don't pay dividends. The right of minority shareholders also manifests itself in the right to vote so that they can defend their interests. The second model stipulates that the firm has an interest in building a reputation that will help it to have funds on advantageous terms and this through the payment of dividends. This model substitutes the first because if there are no laws on shareholder protection and, obviously, dividend payment, the firm has an interest in paying dividends (La Porta et al., 2000). These authors concluded, after studying the dividend policy of 4,000 large firms in 33 countries around the world, that firms in countries where there is protection of minority shareholders' rights pay more dividends, especially those with slow growth rates. The least protected shareholders are forced to accept what is distributed because they have no other opportunities. In this sense, Gaver and Gaver (1993) found that dividend yield is negatively related to growth opportunities and that there is no significant relationship between dividends paid and growth opportunities. However, firm size influence both, the dividend yield and the dividend paid. These authors also found that growing firms have a low debt/equity ratio and pay less dividends compared to non-growing firms.
According to the customer model developed by Miller and Modigliani (1961), if management expectations of future earnings influence their decision to distribute current dividends, changes in dividends will convey information about future earnings. This explanation can be summarized in the notion of the informational content of dividends, which is formalized in two ways: the first consider that dividends are used as an ex-ante signal on future cash flows, while the second states that dividends provide earnings information such as a description of the sources and uses of funds (Allen & Michaely, 1995). This model has been criticized by researchers, especially for its ability to explain the determinants of dividend policy in the first place (DeAngelo et al., 2004;DeAngelo et al., 2006;Denis & Osobov, 2008). The propensity to pay dividends is higher among the most profitable firms in six countries: the United States, Canada, the United Kingdom, Germany, France and Japan, and the retained earnings of these firms represent a large proportion of equity (Denis and Osobov, 2008). Over the period from 1994 to 2002, Denis and his colleagues found that overall dividends did not decline in the six except in the United States and were concentrated in the most profitable large firms, leaving them to explain the results by agency cost theory based on life cycle theory. Noting that the authors did not find a strong positive relationship between the relative prices of paying and non-paying firms and the propensity to pay dividends, therefore, they have doubts about considering signaling and customer considerations as the primary determinant of dividend policy. Grullon et al. (2002) studied the relationship between the level of dividends paid and the level of systematic risk (increasing dividends implies a decrease in systematic risk and vice versa). The market's reaction to an increase in dividends paid depends on a significant decrease in systematic risk. Maintaining this situation over a long period of time causes a price increase in the next three years. To explain this finding, Grullon et al. (2002) put forward the maturity hypothesis (which is later known by the dividend life cycle theory) according to which the increase in dividends is linked to the firm growth phase and the change in the dividend policy of a firm with its life cycle. From the perspective of this theory, when retained earnings represent a large part of equity and total assets, listed industrial firms pay high dividends and become close to zero when capital is contributed by shareholders and not profits (DeAngelo et al., 2006). The payment of dividends or not evolves over time as profits accumulate and investment opportunities diminish.
Following a study of 736 announcements of dividend changes in German firms between 1992 and 1998, Gugler and Yurtoglu (2003) put forward another explanation of the information transmitted to the market through dividend policy. For these authors, dividends signal the degree of conflict between the controlling majority shareholder and external minority shareholders. They find significantly negative wealth effects for firms where the ownership and control structure expropriate the wealth of minority shareholders. Lower dividends may be due to increased rent extraction and expropriation of minority shareholders' wealth.
According to Michiels et al. (2014), the use of family governance leads to a more efficient dividend policy. Indeed, in the context of Belgian firms, the dividends distributed are low when a family CEO runs the firm and also in the case of family dominance over the board of directors. This trend is especially pronounced in the first generations (Vandemaele & Vancauteren, 2013). In the same vein, with a sample of 87 Latin American firms and over a period from 2004 to 2014, Briano-Turrent et al. (2019) showed that family CEOs pay fewer dividends and invest in fixed assets than non-family CEOs. The quality of governance also influences the dividend payment policy and influences the likelihood of a family member being an executive. While the latter authors found influence, González et al. (2014) concluded that family involvement in management doesn't influence dividend payment policy, and this has been verified in the context of Colombian firms. However, according to the same authors, a proportionate participation of the family in the board of directors has a positive effect on the dividend policy. Pindado et al. (2012) studied the impact of type voting rights on dividend payments. The authors concluded that high dividend payments are pronounced in firms where there is no separation between the voting rights and the cash flow rights of the managing family. However, Xu'nan (2011) found that the dividend ratio decreases with the separation of ownership and control.
Dividend smoothing in the family firm is lower than dividend smoothing in non-family firm (Hussain and Shah, 2015). The speed of adjustment of dividends in non-family firm is twice the speed of adjustment in the family firm (4 years and 2 years, respectively). The authors also conclude that the speed of adjustment of dividends paid is faster compared to developed countries, justifying this finding by the speed of adjustment of dividends in U.S. firms, which reaches 11 years (Lau & Wu, 2010). Yousaf et al. (2019) studied the case of Pakistani firms during the period from 2009 to 2016. They concluded that family firms differ from non-family firms in financial characteristics and that family firms pay lower dividends than non-family firms. However, Yousaf et al. (2019) found that firm size inversely affects dividend policy. Indeed, the dividend payout is correlated with firm size, with large firms paying out more dividends than small firms to raise funds in the market. Family control does not moderate the impact of all firm-specific factors on dividend policy. For these authors, family control, size and tangibility are the main determinants of dividend policy in Pakistan. On his part, Miah (2022) analyzed the association between ownership structure and dividend policy of Bangladeshi firms. Based on the analysis of 993 observations covering the period from 2011 to 2019, the author found that family-owned firms distribute more dividends than non-family-owned firms to alleviate the fears of expropriation by non-controlling shareholders.
However, Madyan et al. (2021) investigated the moderating effect of CEO demographic characteristics on the relationship between family ownership and dividend policy. Following the analysis of data of family-owned non-financial firms listed in Indonesia between 2013 and 2017, the authors concluded that family ownership and family CEO positively influence the dividend payout ratio. They also concluded that the age of the CEO further strengthens the positive effect of the family CEO on the dividend payout ratio. However, CEO seniority does not influence the positive effect of family CEOs on dividend payout ratio. The authors also concluded that ROA and firm size have a significant effect on the payout ratio.
Considering the arguments presented above regarding family firms and dividend policy, we estimate the following hypothesis: H1. Family involvement positively influences dividend policy.
H2. Family CEO influences positively dividend policy.
H3. Concentration of ownership positively influences dividend policy.

Dividend, Size, and Family Firms
In terms of the effect of firm size on dividend policy, medium-sized family firm distributes fewer dividends than large family firm to finance investment opportunities (Bresciani et al., 2016). Thus, the purpose of the dividend payment policy in this case is to attract new investors and retain old ones, and this finding is more pronounced in family-owned firms than in large non-family firms. Medium-sized family firms may take more power from the family to ensure continued development and growth even if liquidity is reduced (Bresciani et al., 2016). Noting that their study showed that family-owned firms have a greater appetite for investment than non-family firm and therefore influences the level of dividends paid, a large proportion of the profits generated during the year are reinvested. This trend is especially pronounced in medium-sized family firms.
Fifty-four families and 49 non-family firm listed in Pakistan, Yousaf et al., 2019 found that firm size negatively affects dividend policy while tangibility affects it positively. They also concluded that family control, size, and tangibility are the main determinants of dividend policy in the Pakistani context (Yousaf et al., 2019). Family-controlled firms use higher dividend ratios, higher debt and lower board independence than non-family firm (Setia-Atmaja et al., 2009).
Considering the arguments presented above regarding family firms' size and dividend policy, we estimate the following hypothesis: H4. family firm size influences dividend policy ijbm.ccsenet.org International Journal of Business and Management Vol. 17, No. 11; 4. Methodology

Sample and Data Source
To test our hypotheses, we use a sample composed of firms listed on the Casablanca Stock Exchange excluding banks, finance firms and insurance firms. This exclusion is due to the fact that this type of firm is highly regulated and has its own management mechanisms. This choice is supported by internal studies such as those of Fama and French (2001) and Duygun et al. (2018). The sample is composed of 41 family firm and 14 non-family firms. The distinction between family and non-family firm is made based on the percentage of capital held by a family or several families, which must be at least 20% (Villalonga & Amit, 2006). It should be noted that most of these firms are controlled by the first generation. The study period is from 2015 to 2017 and the data used are collected from the annual reports of listed firms and from the official website of the Casablanca Stock Exchange.

Dependent Variable
Payout: rate of dividend payment and is the result of the fraction of the dividend per share on the net earnings per share. This measure has been used by researchers such as Vandemaele and Vancauteren (2013)

Independent Variables
Family firms: A family firm is defined as one in which a family controls at least 20% of the shares of the business directly or through another cross-or pyramid-shaped ownership structure (Yang, 2010;Kang & Kim, 2016;Briano-Turrent et al., 2019). A family-owned firm is one whose CEO is a founder or member of the family that is the majority shareholder, while a non-family firm is one whose CEO is external and independent of the majority shareholder (Yang, 2010;Kang & Kim, 2016;Briano-Turrent et al., 2019). This variable is dichotomous, taking the value 1 if the firm is family-owned and 0 if the firm is non-family owned.
CEO family: this variable is dichotomous and takes the value 1 if the management position is held by a family member and 0 if it's an external one.
Concentration of ownership: This is the percentage of capital held by the majority shareholder.

Control Variables
In this study the following control variables are used: the age of the firm expressed in years, the assets of the firm and which is the ln of the total assets for the fiscal year 2017, sector of activity: is a dichotomous variable and takes 1 if the firm operates in the industrial sector and 0 for the other sectors of activity.

Regression
Partial least square (PLS) regression is a technique that generalizes and combines the features of principal component analysis and multiple regression. The objective of this technique is to predict or analyze a set of dependent variables from a set of independent variables or predictors (Abdi, 2003).

Ownership and Dividend Policy
The first model studies the effect of family involvement on dividend payment policy, in fact, the dependent variable is represented by the per value of the dividend paid per share, while the independent variable is the type of business (family or non-family). The model uses control variables such as equity, firm age and listing duration. The summary of Model 1 (Table 3) shows that the positive relationship is statistically significant.
To determine whether a family firm pays more or less than non-family firms, the average dividend paid by the two types of firms were compared. The comparison is made at two levels. The first is a comparison of the average value of dividends paid in 2017. The average par value per share per family firm is MAD 3.169 (In 2017, the sum of dividends paid by the 39 family-owned firms is 123.59 MAD, of which 9 firms did not pay dividends, while the 16 non-family firm paid 58.16 MAD) per share while a non-family firm pays MAD 3.635 per share. Non-family firm pays more than MAD 0.47 per share than the dividends paid by family firms. The second ijbm.ccsenet.org International Journal of Business and Management Vol. 17, No. 11; comparison is between the average dividend paid in the three years prior to 2017. The average dividend paid by family firm over the said period is 23.748 MAD (over the last three fiscal years prior to 2017, family firm (39) paid a total of 926.18 MAD and non-family firm (16) paid 582.85 MAD) per share while non-family firm pays an average of 36.428 MAD per share. Over three fiscal years (2015, 2016 and 2017), a non-family firm paid more than MAD 12.7 million per share compared to family firms. These two comparisons clearly show that family-owned firms pay less dividends than non-family firms, therefore the H1 assumption is accepted. Thus, the dividend payment policy in Moroccan firms follows that of other countries such as Taiwanese family-owned firms (manhwa et al., 2019), Indonesian family-owned firms (Duygun et al., 2018), Chinese family-owned firms (Wei et al., 2011), Mexican family-owned firms (San Martín Reyna, 2017) and Latin American family-owned firms (Gonzalez et al., 2017). However, according to Setia-Atmaja (2010), the Australian family firm pay more dividends than non-family firms. The analysis of the model with the average dividend as an independent variable doesn't reveal a significant relationship (model 2). The two tables below present the results of the test under SPSS 25. A general discussion of this result is presented in the following.

Ownership and Dividend Policy
In this model we consider the dividends distributed as a dependent variable, the concentration of ownership as an independent variable and as control variables we find age, listing duration and equity. The first model in Table 4 shows that there is a positive and statistically significant relationship, in fact, the degree of significance is less than 5%, it's 2.5%. This result affirms the H3 hypothesis, i.e., that there is a relationship between ownership concentration and dividend policy. This positive relationship can be explained by the fact that ownership in Moroccan and particularly listed firms remains concentrated in the hands of a family or another firm. In contrast to these results, Gonzalez et al. (2017) found that concentration of ownership negatively influences dividend policy. Over a period from 2007 to 2014 and with a sample of 1,464 firms from six Latin American countries (Argentina, Brazil, Chile, Colombia, Mexico, and Peru), Gonzalez et al. (2017) found that high concentration of ownership in private hands negatively influences dividend payments. Florackis et al. (2015) studied the effect of concentration of ownership among executives on dividend policy. Their results showed that low managerial ownership has a negative effect on dividend policy. This relationship becomes positive when managerial ownership is high. However, the analysis with the second variable (model 2) doesn't show that there is a statistically significant relationship between the variables, whether positive or negative.

Family CEO and Dividend Policy
The analyze of the family-owned CEO effect on the value of dividends paid has shown that there is no significant relationship at the 5% threshold and therefore the H2 hypothesis is rejected. However, the model remains significant at the 10% threshold. The model can be influenced by the other variable, that of the type of ijbm.ccsenet.org International Journal of Business and Management Vol. 17, No. 11; firm since most of the 49-family firm has a family CEO. Vandemaele and Vancauteren (2013) have risen, among other things, the relationship of socio-emotional wealth with dividend policy. They found that when the majority family dominates the board of directors, the family CEO sets the dividend payment according to the family's socio-emotional wealth.
After studying the effect of demographic characteristics of the family CEO in four American firms, Briano-Turrent et al. (2019) concluded that family CEOs pay less dividends than non-family CEOs. They also pay less dividends if they have direct ownership. James et al. (2017) studied another aspect of CEO influence, which is firm ownership. The authors note that CEO owners have an important influence on dividend policy, but the nature of this relationship depends on the payment history and the level of free cash flow in the firm. The analysis with the second dependent variable (model 2) doesn't find a significant relationship. Table 5 shows the results of the test.

Firm Size and Dividend Policy
The last model is reserved for the study of the effect of firm size on dividend policy. The regression results show that there is a statistically significant relationship between firm size and dividend policy, and therefore the H6 hypothesis is accepted (see Model 1 in Table 6). This conclusion is in line with those of other studies on the subject such as (Rafique, 2012;Sener et al., 2019).
According to Redding (1997), firm size and liquidity explain the decision to pay dividends, while existing informational explanations such as control and signaling to explain the level of dividends paid. Other studies have found no significant relationship between firm size and dividend policy (Sakir et al., 2014;Das, 2017).

Discussion and Conclusion
The objective of this study was to examine the effect of family control on the dividend policy of non-financial firms listed on the Casablanca Stock Exchange. It also investigated whether the family CEO influences the dividend policy. The study of the effect of family involvement and corporate governance on the dividend policy of Moroccan firms listed on the Casablanca Stock Exchange has yielded unexpected results that leave us perplexed. To achieve this objective, we tested two models. The first to analyze the effect of family involvement and corporate governance on dividend policy by taking as an independent variable the dividend paid in 2017. This analysis showed that Moroccan family-owned firms listed on the Casablanca Stock Exchange pay less dividend (H1) than nonfamily firms. This result can be explained by the fact that family firm has a long-term vision and tend to pay less dividends to reinvest the profits made in the realization of projects or the maintenance of the firm's life for future generations. This study revealed that family CEOs (H2) don't influence the level of dividends paid to shareholders. However, ownership concentration (H3) and firm size (H4) influence the dividend policy of family firms listed on the Casablanca Stock Exchange.
We tested two models; results don't show any of a relationship whether positive or negative. The analysis of the first block with dividends paid in 2017 as the dependent variable revealed significant and insignificant relationships, whereas the analysis with the second dependent variable, i.e., the average dividend paid over the last three fiscal years, did not reveal any significant relationship.
The results of our model showed that Moroccan family-owned firms pay less dividends than non-family firms. This result was also found by Duygun et al. (2018), who also studied an emerging country, Indonesia. Shareholding is concentrated in both countries, but its composition may be different; indeed, these authors mention very serious conflicts of interest between shareholders and managers on dividend policy, whereas in Morocco there are no studies that have proven the existence of conflicts of interest between these two parties, but the hypothesis of a "harmonization" of interests is more likely given that the central power influences the economic elites (Oubenal & Zeroual, 2017).
In other non-Arab Muslim countries, the results of the study by Yousaf et al. (2019) showed that family firms differ from non-family firms in financial characteristics and that family firms pay lower dividends than non-family firms. They also concluded that firm size inversely affects dividend policy. However, Miah (2022) found that family firms pay out more dividends than non-family firms in order to alleviate fears of expropriation by non-controlling shareholders. Regarding the role of CEO, in our study we found no effect on dividend policy while in the case of Madyan et al. (2021) study concluded from the analysis of data of family-owned non-financial firms listed in Indonesia between 2013 and 2017 that family ownership and family CEO positively influence the ratio of distributed dividends. They also concluded that the age of the CEO further strengthens the positive effect of the family CEO on the dividend payout ratio. This means that even though both countries are Muslim (Morocco and Indonesia), the results are different.
Shareholding in companies listed on the Casablanca Stock Exchange is characterized by the existence of a majority shareholder and the rest of the shares are held by other companies. Moreover, Oubenal and Zeroual (2017) have made the observation that out of the 561 seats (directors), 105 are controlled by 8 groups, 20% of which belong to an institutional group. The concentration of ownership and control questioned the true definition of a dividend policy itself. In this context, it could be said that it is not a dividend policy but a kind of remuneration because of the characteristics of these companies. La Porta et al. (2000) considers that dividends have two roles (outcome model and substitute model), yet according to our study minority shareholders are institutions and not individuals. These institutions have close relationships with the management of the company but also with other companies. As noted by Oubenal and Zeroual (2017), concentration of ownership leaves the vote ineffective and remains a kind of formality. Majority shareholders have absolute power and at the general meeting, the decision on whether or not to distribute dividends is left to them (Ben Sedrine, 2020). Minority shareholders can only join or prove an abuse of the majority before the courts. However, no shareholder will have an interest in bringing a claim against the shareholders of his company.
There is a law that protects the rights of minority shareholders, but in practice, minority shareholders have no power to influence decisions made within the company (Ben Sedrine, 2020).
In another emerging country, San Martín Reyna (2017) concluded from a study of Mexican firms that the concentration of ownership in family hands negatively influences dividend policy, and the presence of institutional investors positively influences it. Jabbouri (2016) studied dividend policy in emerging MENA countries over a period from 2004 to 2013. He concluded that size, profit for the year, and the level of liquidity positively influence dividend policy, while it's negatively influenced by leverage, growth, and the state of the economy. Moreover, Jabbouri and Jabbouri (2020) analyzed the effect of ownership identity on the performance of firms listed on the Casablanca Stock Exchange before and after the 2008 financial crisis. They concluded that family ownership positively influences the performance of these firms before the crisis and negatively after the crisis. They add that the financial crisis has changed the behavior and attitude of family owners, indeed, they have shifted from actors who seek to preserve decent governance before the financial crisis to actors who pose a threat to external or non-family shareholders. The two authors explain this situation by the fact that family owners have increased their profits in a turbulent environment that has reduced the performance of their firms. These results are based on a quantitative study as it leads to an explanation for the change in behavior. Therefore, the findings should be viewed with caution until qualitative research (e.g., case studies) is conducted to determine whether family owners have changed their behavior and whether this is found in family and non-family firms.
Managers use dividends to transmit (report) information to the market when shareholder protection and governance measures are weak. Jabbouri (2016) draws attention to the presence of agency problems. Over a period from 2011 to 2016, Ouattara and Konate (2017) found that the decision to distribute dividends by firms listed on the Casablanca Stock Exchange is influenced by the change in operating income and the amount of dividend paid in the previous year (Ouattara, 2017).
Moroccan investors prefer high dividends (Baker and Jabbouri, 2017). These authors explain this trend because it may reflect a poor governance environment but also because of poor disclosure. Indeed, Jabbouri (2016) has documented that MENA corporate executives increase dividend payments to reassure investors who are averse to the risk of expropriation by majority shareholders. Following a questionnaire study, Baker and Jabbouri (2016) studied the factors that influence the dividend policy of firms listed on the Casablanca Stock Exchange. Their results showed that the level of current profits, the stability of profits and the needs of current shareholders are the most important determinants of the dividend policy of these firms. These determinants are similar (in comparison based on 25 factors) to those of American, Canadian, Indonesian, and Indian firms. Management believes that dividends affect the value firm. The results of Baker and Jabbouri (2016) Lotto (2020) concluded that profitability, liquidity, firm size, leverage, firm growth, previous dividend, and GDP are the main determinants of dividend policy. The explanation of similarities could be due to the fact that these countries were colonized by Western countries. Morocco was colonized by France while Nigeria was colonized by the United Kingdom and therefore the legacy left by the colonizers but also the training of business managers is generally done in Western countries and therefore, they are influenced by the Western management mode.
To try to understand why there is not a significant relationship between the average dividend paid over three years and the independent variables (family involvement and governance). An analysis of the sample composition was carried out: 42% of the firms distributed less than USD 0.5 per share over the three years, of which 20% did not distribute any dividends in the years prior to 2017. All but three of these firms are family-owned.

Scientific Contributions
Our paper contributes to broaden our understanding of the effect of the family firm on dividend policy in one of the emerging countries. Indeed, our study has highlighted the dividend policy of a country with unique characteristics. The formation of Morocco's economic elites remains quite interesting and to give rise to companies with a concentration of ownership and control. It is suggested that the characteristics of corporate ownership determine the share of profits to be distributed, not to say dividend policy, since concentration of ownership limits the power of minority shareholders to make their voices heard. Despite the presence of laws protecting minority shareholders, their room for maneuver remains limited. Under these conditions, on what basis do majority shareholders determine their dividend policy? And how do minority shareholders behave in the face of this situation? These are questions that show that there are still gray areas in dividend policy in emerging countries in general and in Morocco in particular.

Managerial Contributions
Our study provides relevant information for practitioners regarding dividend policy in Moroccan family firms. Indeed, the families who own and actively participate in the control and management firms become aware that their decision influences the perception of the actors of their environment, especially the investors or their future shareholders. The conclusions of our study can also serve as a basis for investors to make the right decisions that will enable them to achieve their goals. To allow the family firm to understand that the dividend policy followed affects the allocation of resources but also a governance mechanism. A dividend policy that takes into consideration the expectations of non-family shareholders will undoubtedly contribute to the strengthening of confidence and cohesion within the firm, which will have a positive impact on the firm. Moreover, Baker and Jabbouri (2017) noted a weak communication between firms listed on the Casablanca Stock Exchange and investors, which limits market development.
The results of our study will also enable the public authorities to take the necessary measures to influence the behavior of the family firm in cases where dividends are appropriated by family members at the expense of minority interests or external shareholders. Strengthening the legal framework in order to reinforce the protection of shareholders' interests will have a positive impact on potential investors, especially when firms are going through a difficult economic situation. To that end, during the financial crisis of 2008, Jabbouri and Jabbouri (2020) found that the family firm expropriated the interests of minority shareholders.

Study Limits
This study has useful implications for investors and policymakers in emerging markets. However, it should be recognized that this study also has limitations. First, this study focuses only on family-owned non-financial firms listed in Morocco instead of studying both family-owned and non-family-owned firms. Second, this study focuses only on the impact of family involvement on dividend policy in Morocco. Also, the study period is just three years when a longer period had to be considered.  Vol. 17, No. 11;