Ownership Structure and Dividend Policy of Conglomerate Firms in Nigeria

Dividend policy is considered an integral policy in company decision that can have a big impact on the investors’ reaction. This study examines the effect of ownership structure on the corporate dividend policy in the line with agency context. This study covers the period of ten years (2001-2010), with a sample of eight (8) conglomerate firms, consisting of 80 firm-observations. Data were extracted from the annual reports of the conglomerate firms. The empirical results depict a positive association between dividend pay-out and institutional ownership as well as block-holders ownership, but a negative association with managerial ownership. The results reveal that the higher the institutional and block-holders shareholdings the higher will be the firm dividend pay-out. The result also suggests that dividend policy is used by managers expropriate the shareholders wealth. In line with the result, it is recommended that conglomerate firms in Nigeria should encourage more shareholdings by institutions as well as block-holders to reduce the opportunistic practices through dividend policy.


Introduction
Dividend policy is one of the pivotal mechanisms of firm strategies and has been viewed as interesting problems in a company to replace large ownership as monitoring tools.Furthermore, larger investors could use their influence to pocket corporate resources for their own private utilization.This could limit the dividend payments of companies which turn to be agency conflict.In view of this argument, it is important to examine the association between larger shareholders and dividend policy to expand better understanding of corporate dividend decisions (Easterbrook, 1984;Jensen, 1986;Rozeff, 1982).
Dividend policy is an engine tool to control and reduce the conflicting interests between the investors and managers.Investors have interest on dividend while managers prefer retain earnings.Managers like to retain earnings for reinvestment, future more income and more control of bigger resources.Corporate governance has received a lot of attention as it aims to mitigate agency problems.Jensen (1986) and Rozeff (1982) argue that firms make use of dividend pay-out policy accordingly if dividends are not settle out to the shareholders, managers will use the money for their personal benefits.Jensen (1986) argues that dividend payments to shareholders decrease the power of managers to control big resources.Stouraitis and Wu (2004) suggest that the dividend could be used to compress the excess investment problems of corporations.The dividend policy will assist in reducing the agency costs and also serve as a signal to give information to the investors about the firm's situations.
Several studies have been conducted in the association between ownership and dividend policy in many developed countries as well as developing countries which show mixed results.Short, Zhang and Keasey (2002) shows that institutional ownership has a positive association with dividend pay-out in the UK.Farinha (2002) and Mancinelli and Ozkan (2006) find that managerial ownership has a negative association with the dividend pay-out.In Malaysia, Ramli (2010) shows that shareholders own companies shares around 40 percent and above and influence the dividend policy to expropriate minority interests.But if the subsequent shareholders have large shares the relationship turns to positive effect.In addition, Al-Nawaiseh (2013) shows that institutional ownership and foreign ownership are significantly and positively related to dividend policy while family ownership is not related.Al-Nawaiseh (2013) finds that institutional ownership influences the company's policy and decisions on higher returnable investments in Jordan, while Arshad, Akram, Amjad and Usman (2013), on the other hand, do not support the association between ownership structure and dividend policy pay-out.
Nigeria as one of the emerging economy differs from other countries in terms of economic activities, legal frameworks, investments, and investors.Thus, there is a need to conduct this research in Nigerian environment to see whether the findings are consistent or different from previous studies.
The purpose of this paper is to investigate the relationship between the dividend policy and ownership structure, to examine whether dividend policies are influenced by the nature of ownership in emerging markets such as Nigeria.The ownership structures considered in the current study are institutional ownership (INOS), managerial ownership (MGOS) and the concentrated or block-holders ownership (BLOS).The remainders of this paper are as follows: Section 2 review of the literature on ownership and dividend policy; Section 3 describes the research methods; Section 4 reports the result of the study; while Section 5 summarises the points and concludes the findings.

Theoretical Perspective of Dividend Policy
The dividend is the driving forces that compel shareholders to have interacted in raising capital for running corporations which gear them to take some huge risk in investment.In the line of this, the management of corporations makes up a dividend policy to share dividend among investors for the contribution they made.Dividend policy makes a significant effect on how much worth is the firm as it has to keep a state of balance between growth and pay-out policies of the firm.Jensen and Meckling (1976) observe that the agency relationship is usually established in a situation where the owners involve managers to carry out some of their responsibilities.Agency cost normally comes up due to the contradicting interest between investors and the managers.Short et al. (2002) posit that dividend policy carried out a vital role in lessen the agency cost.Jensen (1986) opined that payment of dividend brings problem between the investors and managers, the conflict is a result of managers' interest to flow back returns into the business while the owners are interested in collecting returns from their investments.
Despite the interest of the shareholders to obtain dividend than to flow back to the business, managers retain the resources and channel it to unprofitable businesses or other areas for their own personal benefit.By extension, this may raise the conflict between shareholders and the managers.However, the dividend pay-out policy can resolve the conflicts.Rozeff (1982) was of the view that dividend payment could serve as a mechanism for reducing agency cost.Many scholars postulated that shareholders would reduce agency cost through controlling dividend policies.Han, Lee and Suk (1999) reported that dividend pay-out and ownership of investment are interwoven.Leal and Carvalhal da Silva (2005) disclosed that conflict of agency between investors and the managers was the product of insincerity in the use of shareholders fund by the managers as it has been observed in Japanese companies.Stouraitis and Wu (2004) claim that most of the problems of over-investment can easily be overcome by the dividend pay policy particularly, in the area of conflicting interest between the principals and agents.

Institutional Ownership and Dividend Policy
Institutional investors are corporate investors who are different corporate entities like financial institutions, banks insurances, pension funds and other corporate institutions (Koh, 2003).Institutional investors' intervention may change the attitude of the managers towards dividend policy.Institutional shareholders influence the decisions of Investee Corporation on the policy of dividend pay-out simply, due to their substantial shareholding.The study is expecting a significant association between institutional ownership and dividend pay policy.Fama and Babiak (1968) and Short et al. (2002) suggested that the association between the institutional ownership and dividend policy is significant.Zeckhauser and Pound (1990) and Short et al. (2002) established a positive association between institutional ownership and dividend pay-out using four dividend models.Ullah, Fida and Khan (2012) using a sample of Pakistan market KSE-100 index, found institutional ownership has a positive association with dividend policy communicating the higher dividend pay-out.Al-Nawaiseh (2013) studies the dividend policy and ownership structure using a sample 62 listed firms for the period of 2006 to 2006.The study revealed that institutional ownership is positively and significantly associated to dividend policy.Al-Gharaibeh (2013) using the sample of 35 quoted firms in Jordan found that institutional ownership has a large stake in companies.It increases the value of the company because of their influence for a better investment.However, Mehrani, Moradi and Eskandar (2011) reported a negative relationship between institutional ownership and dividend policy.Based on the agency theory, the study hypothesis as follows: H1: Institutional ownership has positive significant relationship with dividend policy of the listed conglomerate companies in Nigeria.

Managerial Ownership and Dividend Policy
The relationships between managerial ownership and dividend policy have been existing for long (Wiberg, 2008).Managerial ownership and dividend policies relationship have been strongly discussed and documented in many studies (Jensen & Meckling, 1976).Jensen's (1986) study posits that managers usually choose to retain resources under their own control than to pay-out dividends to the investors.Previous research indicated the association between dividend policy and managerial ownership with different findings (Rozeff, 1982) but still there is a need to put managerial ownership into consideration due to mixed results the relationship.Jensen (1986) reported that managers retain earnings for the purpose of either expansion of the business or for their own personal benefit.These are some of the reasons for retaining the shareholders resources, not to reward back investors for their investment.Espen Eckbo and Verma (1994) reported that where the managerial ownership power is increasing, the dividend pay-out is decreasing.Jensen (1993) shows that the association between managerial ownership and dividend policy has been reported with a negative effect, i.e where the managerial ownership increase dividend policy pay will decrease.Mehrani et al. (2011) find a negative association between managerial ownership and dividend pay-out policy which in support of the Chen, Cheng and Hwang (2005) and Short et al. (2002) whom indicated a negative association between managerial ownership and dividend payment policy of bank holding company.Similarly, Chen, Cheung, Stouraitis and Wong (2005) reported a negative association between managerial ownership and firm performance.
Hsu and Koh ( 2005) expanded koh's ( 2003) study and examine the extent of the association of long term, and short term managerial ownership effect on the dividend policy in Australia.The study established that managerial ownership for all linear specification is statistically significant but insignificant for the non-linear specification models.However, managerial ownership is found positively related with the dividend policy.Bergstresser and Philippon (2006) relate this finding to the CEO compensation.The more closely a CEO's compensation is tied to the value of stock and option, the more likely dividend increases.Srivastava (2011) found that dividend policy is positively significant within the intermediate area of ownership, which suggested that the entrenchment effect is dominant in specific regions.In addition, they found that the association between managerial ownership and dividend policy is significantly negative within low high regions of ownership, indicating that the alignment effect is dominant in these regions.In a study of Al-Gharaibeh (2013) also found a negative relationship.The above studies showed that monitoring activities seem to be weaker at the higher material ownership level and, therefore, a negative association is predicted between managerial ownership and dividend policy.The study hypothesis that: H2: Managerial ownership has negative relationship with dividend policy of listed conglomerate companies in Nigeria.

Block-Holders Ownership and Dividend Policy
Block-holders are the holders of 5% and the above of the equity stock (Mitton, 2004).Shareholders with block ownership will require more monitoring power to safeguard their investment.Mitton (2004) argues that in emerging market, it is found a positive relationship exist between corporate governance and dividend pay-out policy due to concentrated ownership structure.Studies also reported that countries with strong investor's legal protection are capable of paying high dividends.Ramli (2010) reported that Malaysian quoted firms with concentrated ownership pay higher dividends.
When examining the association of block-holders ownership and dividend policy, the opinions discussed from two argumentative views according to Devaki and Kalaveni (2012).The first view which is in line with agency theory, blockholders suffer loss when they want to sell large stock which result to the fall down of the stock price.Thus, block-holders set up a long term plan to monitor managers in order to produce more benefit to their investments.Block-holders owners have the ability to monitor and voice their concerns and objectives because of their large voting rights.This, in turn, leads to block-holders to provide some monitoring over managers, which enable the block-holders to affect the board of directors' decision (Mancinelli & Ozkan, 2006).In second view, block-holder owners can force control on managers to report favourable financial performance which leads to the enhancement of the stock price of their investments, unlike small shareholders whom have no power to control the managers (Barclay, Holderness, & Sheehan, 2003 Vishny, 1997).
In the case of large block-holders' participation may force managers to involve in income increasing in order to report acceptable financial performance.Bethel, Liebeskind and Opler (1998) reported that block-holders have a positive relationship with management turnovers and disclosed that block-holders put pressure on the managers to take two specific actions or face risk of being dismissed whenever the company performs poorly.Klein (2002) investigated the board characteristic and audit committee composition effectiveness and controlling impact of the block-holders' ownership on dividend policy.The study considered firms whom audit committee included representatives of blockholders with more than 5% of the equity.The study posits a negative association between 5% block-holders sitting on audit committees and dividend policy.Based on the above discussion, the study hypothesizes that: H3: Block-holders ownership has significant relationship with dividend policy of listed conglomerate companies in Nigeria.

Control Variable Free Cash Flow (FCF)
Free cash flows of the firm determine its debt employing capacity.Since debt involves periodical cash outflow, firms with higher levels of free cash flows have higher debt capacity and hence may employ higher levels of debt.Kowalewski, Stetsyruk and Talavera (2007) indicated a strong effect of the free cash flow on dividend policy.Therefore, the more available cash is in the firm, the higher the dividend per share.

Research Method
This study is limited to quote conglomerate companies listed in Nigerian Stock Exchange (NSE).The population of companies consists of all eight (8) quoted conglomerates operating in Nigeria as at 31 st December 2010.The period of the study is ten years (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010).Based on Krejcie and Morgan (1970), the sample size is the same as the population since the population is not more than 20.Data of the study were gathered from secondary sources, annual reports.Figure 1 shows the research framework of the study.

Result and Discussions
Descriptive statistics of the study variables are presented in 283 average block-holders ownerships in the listed conglomerate firms in Nigeria is 0.564 (56.4%)The institutional investors own significant portion of the shares.Block-holders owns on average of 0.594 (59.4% of the shares.Managerial ownership owns on average of 0.3077 (30.8%).The mean of the free-cash flow is 0.5870.The skewness ranges between -0.857 and 1.057 whereas the kurtosis is in the range of -0.798 and 0.236.There is variation in the data and the data are considered normal.Table 2 repors the correlation matrix of the variables.The result shows that the independent variables, institutional ownership (INOS), managerial ownership (MGOS), block-holders' ownership (BLOS), and the control variable free-cash flow (FCF) have positive correlation with the dependent variable.INOS is highly correlated at 0.878.This indicates that an increase in institutional ownership (INOS), managerial ownership (MGOS), block holders' ownership (BLOS) and the control variable free-cash flow (FCF) will also increase the dividend policy ratio.The correlation matrix also reveals that all the independent variables are significantly related but are not more than 0.5 which reveals the absence of multicollinearity problem (Hair JR, Black, Babin, & Anderson, 2010).In relation to INOS, the relationship is positive and significant at 1%.The hypothesis is supported, the regression results reveal that, the coefficient with regards to institutional ownership is 1.646, which implies that for every 1% increment of the institutional ownership, the dividend pay-out will increase by 1.65%.The implication is that, the higher the institutional ownership, the better the dividend payment and thereby improving the level of stakeholders' interest in the firm.This result is in line with Fama and French (2000); Han et al. (1999) and Short et al. (2002) who reported that there is significant and positive association between the dividend policy and institutional ownership.While it contravenes the result of Mehrani et al. (2011) who found negative relationship between dividend payment policy and institutional shareholders.
Managerial ownership has negative and significant association with dividend policy and significant at 1% Thus, the hypothesis is supported.The coefficient of the regression results is 0.062, which implies that for every 1% increment of managerial ownership the dividend pay-out will increase by 0.06%.This signifies that the managerial ownership reduce dividend policy.The implication is that, the more managerial ownership in the firm, dividend pay-out will not be attractive.This result supports the findings of Chen, Cheung, Stouraitis and Wong (2005); Mehrani et al. (2011) andShort et al. (2002).However, the finding contradicts the findings of Bergstresser and Philippon (2006), Hsu and Koh (2005) and Koh (2003).The lock-holders ownership has a significant association with dividend policy of conglomerate firms in Nigeria.The coefficient of the regression results of 4.812 signifies positive relationship between block-holders' ownership and dividend policy of conglomerates firm in Nigeria, at 1% significance level, implying that for every 1% increment of blockholders, dividend pay-out ratio will increase by 4.81%.The findings support the previous work of Barclay et al. (2003), Mitton (2004) and Shleifer and Vishny (1997), but contradict the findings of Klein (2002 andRamli (2010).In relation to the control variable, it has been observed that the coefficient of the free cash flow reveals a positive value of 0.045, and statistically significant at 1%.This implies that, for every 1% increment in free cash flow, dividend policy will increase by 0.05%.This signifies that, free cash flow has a positive association with dividend policy.The collective effect of all the exogenous variables put together is able to explain the dependent variable up to 84% as indicated by adjusted R 2 and remaining 16% is controlled by other factors.Similarly, the F-statistic value of 104.53 implies that the model is well fitted and significant at 1%, as indicated by F-significance.This provides evidence that the model fits the data well and the joint effect of the explanatory variables is statistically significant in explaining the dependent variable.The Durbin-Watson of 1.619 indicates a tolerable serial correction within the period of the study.This shows the appropriateness of fitting the model of the study with the four independent variables.

Conclusion
The significance of the paper is to find out the extent to which selected ownership structure affect the dividend policy in conglomerate firms in Nigeria.The results show a significant positive association between institutional ownership and block-holders with dividend policy of conglomerate firms in Nigeria.The negative relationship of managerial ownership is a concern and effective monitoring mechanisms are required to monitor managers' actions to be in line with the outside shareholders.However, for the control variable, free cash flow was significant and positively related to dividend policy.This led to the conclusion that the possibility of a firm to engage in sound dividend policy is the function of its ownership structure.In other words, well blended ownership structure enhances the degree of dividend pay-out to shareholders.
Consequently, upon the findings and conclusions of this study, we recommend that the Nigerian conglomerate firms should consider these corporate ownership structures as benchmark in embarking on sound dividend policy that will alleviate the conflicts that could arise between managers and owners of the firm.As a matter of policy input, managers, regulatory authorities and other stakeholders of interest should ensure that the diverse interests held in a firm by owners are taken into consideration, as this will go a long way in mitigating agency problems exist between managers and owners of the firm.

Figure 1 .
Figure 1.Research Framework Model specification of the study is as follows: DIYit = it + 1INOSit + 2MGOSit + 3BLOSit + 4FCFit + it Where: DIY = dividend pay-out: cash dividend per share/price per share INOS = institutional ownership: is the proportion of shares owned by institutional investors MGO S= managerial ownership: is the proportion of shares own by the executive directors and their families.BLOS = block-holders ownership: is the proportion of shares owned by block-holders.FCF = Free cash flow: capital expenditure less operating cash

Table 1 .
The average payment of the dividend is 13.45.The

Table 1 .
Summary of the statistics

Table 2 .
Summary of Correlation Matrix of Variables Table 3 provides pooled OLS regression result.Based on the table, the model is as follows: