# Introduction ne of the most important decisions taken by firms is their dividend policy. In the finance sector the decision on the ownership structure, dividend policy and its components is viewed as one of the most extensively researched area. Dividend distributions tend to reduce the agency problem occurring between managers and shareholders (Jensen, 1986 andRozeff, 1982). According to them, managers tend to prefer dividend retentions in order to use the firms' resources for their own personal benefits. Paying out dividends to shareholders tends to reduce the managerial control over the firms' resources and control their opportunistic behaviors, while avoiding the over investment problems of corporations and wasting the firms' resources on unprofitable projects. Hence, dividend policies can effectively reduce the conflicts of interests between managers and shareholders. The dividend policy of any firm is considered one of the most important decisions made for the corporate policies, as it is considered a reward to shareholders for their contribution in raising funds for a company and for bearing the relevant risks. "Dividend policy is the set of guidelines a company uses to decide how much of its earnings it will pay out to shareholders. Some evidence suggests that investors are not concerned with a company's dividend policy since they can sell a portion of their portfolio of equities if they want cash." Once the company decides on whether to pay dividends they may establish a somewhat permanent dividend policy, which may in turn impact on investors and perceptions of the company in the financial markets. What they decide depends on the situation of the company now and in the future. It also depends on the preferences of investors and potential investors. This paper is designed to test the impact of ownership structure on dividend policies of Sri Lankan firms. According to previous researches conducted on this area, there are many types of ownership structure that affect dividend policies, such as the individual, managerial, institutional, concentrated, and state ownership type of corporate structure, along with few more such as the foreign and family ownership structures. For the purpose of this research, however, only three ownership types of corporate structure are examined throughout this research, to determine their impact on dividend policies of firms in Sri Lanka. These are the institutional ownership, the concentrated ownership, and the foreign ownership. # II. # Literature Review The agency theory focuses on mitigating conflicts of interests between managers and shareholders due to the separation between ownership and control (Jensen and Meckling, 1976). This paper identifies some of the literature relevant to this research of how the different types of ownership structure affect the dividend policies. It indicates the relationship of this study to the relevant literature. It highlights some of the theories of corporate governance and identifies their relation to different types of ownership of corporate structure, and the importance of finding the possible relationships between ownership structures and dividend policies. Ownership structure is an influential factor on company policies. Decisions regarding to dividend are one of the Fundamental components of corporate policies (Kouki and Guizani, 2009). # a) Ownership Concentration and Dividend Policy Ownership structure is an important internal mechanism of corporate governance. It is defined by the distribution of equity with regard to votes and capital as well as the identity of the equity owners. These structures are of major importance in corporate governance because they determine the incentives of managers and thereby the economic efficiency of the corporations they manage (Jensen and Meckling, 1976). Ownership structure consists of individual ownership structure, institution ownership structure and foreign ownership structure. Carvalhal-da-Silva and Leal (2004) argued that ownership structure is very important and influential factor in determining the efficiency of the market by giving information about two significant things. First, it shows the extent of risk diversification of shareholders. Second, it gives information about the possible agency problems in the management of the corporation. Ownership structure has an effect on the capital structure and firm performance. Chaganti & Damanpour (1991) found that among the firms in their study, those with a large share of family ownership flavored debt financing while a large share of institutional ownership displayed a positive relationship with return on equityratios. This is supported by Margaritis & Psillaki (2010) who found that a positive relationship between concentration of ownership and higher levels of debt in the capital structure. La Porta, Lopez-de-Silanes, & Shleifer (1999) studied the ownership structure of some of the world's biggest economies in order to find the most appropriate controlling shareholder of the companies included in his sample. Those countries with high shareholder protection have a large amount of shareholders and trading shares through IPO (initial public offering) in the primary market as well as in the OTC market or at least one stock exchange. These types of companies are often controlled by families or the state instead of financial institutions. The controlling shareholder is often involved in pyramids and managerial decisions since the ability to control their surplus of cash flow rights. # b) Ownership Identity and Dividend Policy i. Managerial Ownership and Dividend policy According to Ullah, Fida, and Khan (2012) the managerial ownership is measured by the total proportion of managers, directors, and executives divided by the total capital shares of the firm. Many researchers observed significant relationships between the managerial ownership type of corporate structure and the dividend payout plans. Mehrani, Moradi and Eskandar (2011) found the evidence in support of negative association between the managerial ownership and dividend payment policy. Consequently, we found that many prior studies have found negative association between managerial ownership and dividend payout policy in different circumstances. Jensen (1986) argued that managers prefer to retain earning instead of giving it to shareholders as a dividend. Managers want to use the resources the growth of the firm as well as for the personal benefits. # ii. Institutional Ownership and Dividend Policy Institutional ownership is defined as the investors with the main characteristic that they are legal entities without any physical persons involved in the process. This includes hedge funds, investment funds and pension funds etc. Another characteristic of institutional investors are that they operate independently or are included in large companies or conglomerates. (Celik & Isaksson,). Many prior studies showed a relationship between the dividend policy and management ownership (Rozeff, 1982; Jensen and Meckling 1976) but still there was space to consider the institutional ownership into consideration. Waud (1966), Fama and Babiak (1968) and Short, Zhang and Keasey (2002) suggested that there is significant relationship between the dividend policy and institutional ownership. # iii. Concentrated Institutional Ownership and Dividend Policy Kouki and Guizani(2009) argued that Tunisian companies having concentrated ownership distribute more dividends and show positive relationship between the concentrated ownership and dividend payout. Claessens and Djankov (1999) has empirically found in the context of Czech Republic that more concentrated the owners, higher will be the firm profitability and the level of labor productivity. He further argued that concentrated ownership will allow the owners to monitor the managers in a better way by using their powerful seat in the board of director. The greater the concentrated ownership structure, greater will be the need for monitoring. According to Mitton (2005), there exist the positive association between the corporate governance and the dividend Payout in emerging market and it is further argued that the countries having strong investor legal protection are capable to pay more dividend payment. # Global Journal of Management and Business Research Volume XX Issue III Version I Year 2020 ( ) iv. Foreign Share Ownership and Dividend Policy According to Chai (2010), foreign ownership has important impact on the dividend policy of the concerned firm. Baba (2009) observed the influence of foreign investor's ownership on the dividend payout policy of the firms of Japan and the study further showed that higher foreign investor's ownership is related with the higher dividend policy of firm. # III. # Empirical Analysis Despite a great deal of prior research on the subject, few studies investigated the agency and ownership-based explanations of dividend policy. This paper therefore attempts to provide more insights into the literature by providing an empirical analysis on the relationship between corporate payout policy and ownership characteristics. Empirical evidence is a fancy way of describing facts that can be experienced and tested only through the senses. Empirical evidence is evidence from observations. From my understating it can be through naturalistic observation or experimental. Experimental evidence is much more reliable as naturalistic observations are vulnerable to researcher bias. # a) Data Collection The present study used secondary data for the analysis. Secondary data is the data that have been already collected by and readily available from other sources. Secondary data is often used in social and economic analysis, especially when access to primary data is unavailable. The financial statements which are made up of income statements, balance sheet of the sample listed companies and investors' information were the main sources of data for this study. There were obtained from the annual reports of respective companies. Which were published by CSE in Sri Lanka. Further, scholarly articles from academic journals, relevant text books on the subject and the internet search engines were also used. Specifically, the data of the listed companies in the sample were collected for the period of 2011-2015. This study utilized dividend and different types of ownership structure data. The dividend variables were retrieved from DataStream financial database. In addition, data on ownership was collected from sample companies' annual reports. These annual reports are gathered from the website of CSE. This Paper is analyzed the data description by using the statistical method. All of these following methods are used to investigate the relationship and impact of ownership structure and dividend policy, according this study researcher can use these statistical techniques for interpretation of the data. There are # b) Variables' Measure # Analysis and Discussion # a) Descriptive Statistics Descriptive statistics are used to describe the basic features of the data in a study. In Table 2 descriptive statistics shows a summary of the variables that was taken from the financial statements and the annual reports of sampled firms on the Colombo Stock Exchange. The table reports the mean, standard deviation, minimum, maximum, and the number of observations for each of the dependent and independent variables. In the above table 2 observations have been used for analyzing minimum, maximum, mean and standard deviation for each variable. The mean value of institutional ownership ratio, Concentrated ownership ratio, Foreign ownership ratio, Firm Size ratio, Return on equity ratio, Future growth opportunity ratio are 0.677%, 0.698% , 0.193 %, 10.47%, 21.69%, and 0.193% respectively with the standard deviations 0.275, 0.210, 0.541, 0.716, 15.38, and 3.478. The descriptive statistics show that over the period under study, the ownership structure measured by DPS & DIVYIE averaged.5.188% and 4.292%, with standard deviations of 7.025 and 3.543 respectively. The above table shows the average number of shares held by institutional shareholders was found to be 0.6771, which implies that almost 67.7144% percent of shares ownership is concentrated in hands home. The standard deviation was however smaller, being 0.275, in this case. In terms of ownership variables, the range of firm concentrated ownership represented by the total of ownership owned by five largest shareholders among 20 major shareholders is from 0.30 to 1.815, the average being 0.698 which implies that almost 69.8percent of shares ownership is concentrated in hands of five largest shareholders in Sri Lankan firms with a standard deviation of 0.210. Last independent variable the foreign ownership tends to have an average of 0.193 and to range between a minimum of 0 and a maximum of 3.327, with a standard deviation of 0.541. As per the table, average dividend per share (DPS) is 5.188 ranging from 0 to 45 cash dividend with a standard deviation of 7.025. Having taken decision to pay, almost 5.18% from the earning was paid to the shareholder as a dividend per share. It can be seen that the standard deviation for dividend yield is 3.543while the average dividend yield distributed among the corporations in the sample is 4.292 per share ranging from 0 to 20.72. The result shows that approximately 90% companies earning retained for further investment in new projects due to enhance the business in future. The average firm size calculated as the log of total assets available in the company under study was found to be 10.47. The table indicates an average of 1.889% of future growth opportunity in Sri Lankan listed banks finance and insurance firms, with a minimum of 0 and a maximum of 37.90 since it is control variable. The average ROE is found to be 21.69% ranging from-10.25% to a maximum of 93.47%, indicating that majority of the companies in the sample are moderately profitable. This percentage computed the involvement of net income (local currency) which is invested by the firm's shareholders. # b) Correlation and Multi-co linearity analysis Correlation is concern describing the strength of relationship between two variables. According this study correlation co-efficient analysis is under taken to find out the relationship between Ownership structure and Dividend policy. So indicate what relationship exists among variable. Year 2020 ( ) F Table 3 presents the Pearson correlation coefficient between Ownership Structure and Dividend policy separately. Value of correlation between institutional ownership ratio and Dividend per share is -0.2498 which is significant associated at the 0.05 level. Therefore, there is a Negative association between institutional ownership ratio and DPS and also the correlation between other concentrated ownership and Dividend per share DPS is-0.1558 which is significant at 0.05 levels. This explains low negative correlation between other concentrated ownership ratio and DPS. Value of correlation between foreign ownership ratio and Dividend per share is 0.4903 which is significant associated at the 0.05 level. Therefore, there is a positive association between ratio foreign ownership and DPS. The value of correlation between institutional ownership ratio and Dividend yield is 0.1651 which is significant at 0.05 levels. This represents positive association between institutional ownership ratio and Dividend yield. The value of correlation between other Concentrated ownership and Dividend yield is-0.1791 which significant at 0.05 level. This indicates there is negative correlation between concentrated ownership other ratio and DIVYIE. Among the control variables, there is a positive correlation 0.1916 between dividend per share and firm size was surprising. However, negative correlation had been found between dividend per share and future growth opportunity -0.115 and ROE -0.010. The results also revealed that firm size has positive correlation 0.0872 with dividend yield. And negative correlation was observed between dividend yield and Future growth opportunities -0.0797. There is positive correlation 0.2709 found between one of the control variable ROE and dividend yield. # i. Multi-Co linearity Variance inflation factors (VIF) measure how much the variance of the estimated regression coefficients are inflated as compared to when the predictor variables are not linearly related. Use to describe how much multi-co linearity (correlation between predictors) exists in a regression analysis. Multi-co linearity is problematic because it can increase the variance of the regression coefficients, making them unstable and difficult to interpret. Use the following guidelines to interpret the VIF: VIF = 1 -Not correlated 1 < VIF < 5 -Moderately correlated VIF > 5 to 10 -Highly correlated Multi-Co linearity: Two major methods were used in order to determine the presence of multi-co linearity among independent variables in this study. These methodologies involved calculation of a Tolerance test and variance inflation factor (VIF) (Ahsan, Abdullah, Gunfie, & Alam,2009). The results of these analysis are presented in table 4.10 Test of Co linearity. # c) Regression analysis The strengths of the influence that the indicator of independent variable has on each of the dependent variable are determined by the use of single regression coefficients of the predictor variables In this research According to the table 4.6. Test of Co linearity, none of the tolerance level is < or equal to 1; and also VIF values are perfectly below 10. Thus the measures selected for assessing independent variable in this study reach levels indicate of multi-co linearity. It shows the Mean Vale 1.22. . the regression analysis has been carried out to find out the pattern of variation of the dependent variable (Dividend per share and Dividend yield) in relation to the values of independent variable (Institutional ownership, concentrated ownership, and foreign share ownership). F The Column of unstandardized coefficients' gives coefficient value for the regression model. The constant of 9.778 is intercept, (-3.757) institutional ownership ratio, (-4.600) other concentrated ownership ratio, 6.010 foreign ownership ratio are slope, and X is independent variables and Y is the DPS. The slope of the Coefficient provides with the most important The table shows the R value is 31.53 % this is the correlation between the dependent variable and independent variable (predicted variable). Here, Institutional ownership, concentrated ownership, Foreign share ownership are independent variable and DPS is dependent variable. According to the table, correlation is 31.53% that means there is a positive direction of relationship between Ownership structure and DPS.R square is the square of R and is also known as the coefficient of determination. The table shows the R 2 value is 0.3153 and which is computed to identify the impact of Ownership structure (institutional ownership ratio, concentrated ownership ratio, foreign share ownership ratio) on DPS. From the above table it is crystal clear that Ownership structure (institutional ownership ratio, concentrated ownership ratio and other foreign share ownership) are contributed to determine DPS by 31.5 3%. The remaining 68.5% is influenced by factors other than Ownership structure (institutional ownership ratio, concentrated ownership ratio and foreign share ownership ratio). The Adjusted R Square refers to the best estimate of R square is for the population from which sample was drawn. According to the table adjusted R square is 0.2866. information, it shows by how much the Dependent scores changes for a change in the independent score by one unit. When firm ownership structures no effect on DPS, in the DPS value is 9.778. If the institutional ownership ratio is increased by one unit then DPS value will increase by -3.757, likewise the other concentrated ownership ratio has negative b -4.600 value. This reveals that other institutional ownership ratio, concentrated ownership ratio DPS tends to move in opposite direction. But the foreign ownership ratio and DPS tends to move in positive direction. According to the table it is seen that there is a significant impact of institutional ownership ratio on DPS, here significant value is -5.096.and also there is a significant impact of concentrated ownership ratio on DPS, there is a significant impact of ratio foreign ownership ratio on DPS. # Model 2 Impact of ownership structure on Dividend yield Based on findings in the first step of hierarchical multiple regression, six predictors were entered: FGO, ROE, FSIZE, FOROWN, INSOWN and CONOWN. This model was statistically significant with F = 4.57; and (r 2 ) of the total variance in dividend yield which determines the effectiveness and importance of the independent variables on the model. Since the Adjusted R-Square (0.1608) is close to the R-Square (0.1255), then the independent variables are proved to be effective and important in relation to the dependent variables. The Column of unstandardized coefficients' gives coefficient value for the regression model. The constant of 5.234is intercept, 2.256 institutional ownership ratio, (-3.391) other concentrated ownership ratio, (-0.541) foreign ownership ratio are slope, and X is independent variables and Y is the Dividend yield. The slope of the Coefficient provides with the most important information; it shows by how much the Dependent scores changes for a change in the independent score by one unit. When firm ownership structures no effect on DIVYIE, in the DIVYIE value is 5.234. If the institutional ownership ratio is increased by one unit then DIVYIE value will increase by 2.256, likewise the other concentrated ownership ratio has negative b -3.391 value. This reveals that other concentrated ownership ratio and foreign ownership ratio DIVYIE tends to move in opposite direction. But the Institutional ownership and DIVYIE tends to move in positive direction. V. # Conclusion This study was designed to observe the impact of ownership structure on the corporate dividend policies by examining the sample of 30 bank, finance and insurance firms listed in Colombo stock exchange for the period 2011 to 2015. This study has been examined through descriptive statistics such as mean, minimum and maximum value and standard deviation. Rather than correlation and regression analysis also used to find out the impact between dependent and independent variable by using STATA statistical package. On the basis of correlation and regression analysis institutional ownership and DPS have significantly negative impact, DPS shows an insignificant negative relation with concentrated ownership. Foreign ownership has significantly strong positive relation with DPS therefore the companies should care more on foreign owners of the shares because they provide more benefits to these firms. The ROE relates positively and statistically insignificant with the DPS. The firm size in the regression model revealed a positive affiliation with DPS variable, but it is statistically insignificant and the future growth opportunity shows the negative relationship with DPS. At the same time significantly positive relationship has been found between institutional ownership and dividend yield. While significantly negative relationship has been found between concentrated ownership and dividend yield. And foreign ownership has an insignificant negative relationship with DIVYIE. The return on equity relates positively and statistically significant with the dividend yield. The firm size in the regression model revealed a negative affiliation with dividend yield variable, but it is statistically insignificant and here once again the future growth opportunity shows the negative relationship with dividend yield. 1Year 2020ConceptVariableIndicatorsMeasurementInstitutional ownershipINSTOWNT.shares held by non financial institutions T.capital shares( )ConcentratedT.shares held by top 5 shareholders amongOwnership structureInstitutional OwnershipCONCOWNmajor 20 shareholders T.capital sharesForeign OwnershipFSOWNT.shares held by foreign holders T.capital sharesDividend per shareDPSTotal dividend No of ordinary sharesDividend policydividend yieldDIVYIEDPS / Market price per share*100Firm SizeSIZENatural log of total assetsControl variableFuture growth opportunitiesFGOShare Price Beginning of the year /Net Asset Value per ShareReturn on Equity (ROE)ROENet income / Total equity*100F c) MethodologyThe broader objective of this research is to study the impact of ownership structure on dividend policy of Bank Finance and Insurance companies. According to Creswell (2009), the variables need to be specified in quantitative researches so that it is clear to readers what groups are receiving the experimental treatment and what outcomes are being measured. 2VariablesNumberMeanStd. DeviationMinimumMaximumDPS1505.1887.025045DIVYIE1504.2923.543020.72INSOWN1500.6770.27500.983CONOWN1500.698.0.210.0.3001.815FOROWN1500.1930.54103.327SIZE15010.470.7168.68811.90FGO1501.8893.478037.90ROE15021.6915.38-10.2593.47 3DPSDIVYIEINSOWN CONOWN FOROWNSIZEFGOROEDPS1DIVYIE0.008210.9205INSOWN-0.2498*0.1651*10.00210.0435CONOWN-0.1558-0.1791*0.12510.0570.02830.1274FOROWN0.4903*-0.115-0.1851*0.0001100.16120.02330.9991SIZE0.1916*0.08720.2318*-0.4771*0.089810.01880.28840.004300.2746FGO-0.1152-0.07970.0920.0069-0.0673-0.067210.16030.33240.26280.93350.41310.4137ROE-0.01030.2709*0.03370.0816-0.1404-0.0895 -0.005910.90010.00080.68210.32090.08650.27620.9433 4Year 20201: Test of Co linearity16VARIABLEVIF1/VIFVolume XX Issue III Version IINSOWN FOROWN CONOWN Mean VIF1.05 1.04 1.02 1.04 Table 4.2: Test of Co linearity0.95009 0.96518 0.983806)( FVARIABLEVIF1/VIFGlobal Journal of Management and Business ResearchSIZE CONOWN INSOWN FOROWN ROE FGO Mean VIF1.52 1.43 1.23 1.09 1.03 1.02 1.220.655879 0.700653 0.815169 0.918041 0.971116 0.978007© 2020 Global Journals 52: Coefficients(1)(2)VARIABLESModel 1Model 2INSOWN-3.7572.256**(2.693)(0.886)CONOWN-4.600***-3.391***(1.348)(0.914)FOROWN6.010***-0.541***(1.186)(0.177)Constant9.778***5.234***(2.710)(0.903)Observations150150R-squared0.2850.074Robust standard errors in parentheses*** p<0.01, ** p<0.05, * p<0.1DPS= ?0+?1 INSOWN +?2 CONOWN + ?3FOROWN+e=-3.856-4.705INSOWN-2.730CONOWN+5.988FOROWN 51: Model SummaryRR SquareAdjusted R Square31.53%0.31530.2866 53: Model SummaryRR SquareAdjusted R Square16.08%0.16080.1255The table shows the correlation between thedependent variable and independent variable (predictedvariable). Here, Institutional ownership, concentrated18ownership, Table 5.4: Coefficients(1)(2)VARIABLESModel 1Model 2INSOWN-3.7572.256**(2.693)(0.886)CONOWN-4.600***-3.391***(1.348)(0.914) © 2020 Global Journals * reports 2011,2012,2013,2014,2015 * AG CGunathilaka PSGunaratne Ownership structure and dividend policy in SriLanka: Proceedings of International Conference on Business Management 2009 6 * Agency Costs Dividend policies around the world LaPorta RLopez-De-Silanes FShileifer AVishny R Journal of Finance 60 2000 * Dividend Policy Ratios and firm Performance: a case study of Selected Hotels & Restaurants in SriLanka: Research Repository @ university of Jaffna Priya PNimalathasan 2013 * Ownership concentration and firm performance: Evidence from selected Companies in SriLanka: Paper Presented at Students' Research symposium of Faculty of commerce and Management studies University of Kelaniya WCPerera SWeligamage 2013 SriLanka * Dividend Policy and Firm Performance: Evidence from the Manufacturing Companies Listed on the Colombo stock exchange TVelnampy Nimalathasan KKalaiarasi Global Journal of Management and Business Research 6 14 2014 * Board Composition, Ownership Structure and Dividend Policies in an Emerging Market OAbdelsalam AEl-Masry S&elsegini Managerial Finance 34 12 2008 * The Influence of Ownership Structure on the Firms Dividend Policy Based Lintner Model NMAbdullah ZAhmad SRoslan International Review of Business Research Papers 8 6 2012 * The Ownership Structure and Dividend Payout Policy in Pakistan (Evidence from Karachi Stock Exchange 100 Index) HAhmad AJavid International Journal of Business Management and Economics 1 1 2010 * Agency Costs and Ownership Structure JSAng RACole JWLin Journal of Finance 51 2000 * Dividend Policy: A Review NBhattacharyya Managerial Finance 33 1 2007 * A Test of the Agency Theory of Managerial Ownership, Corporate Leverage and Corporate Dividends JCCrutchley RSHansen Financial Management 18 1989 * Two Agency Cost Explanations of Dividends FEasterbrook Journal of Political Economy 88 2 1984 * Dividend Policy and Institutional Ownership: Empirical Evidence using Propensity Score Matching Estimator JAElston RHofler JLee Journal of Accounting and Finance 11 1 2011 * Firm Performance and Dividend-related Factors: The case of Malaysia SSFoong NZakaria HBTan Labuan Bulletin of International Business and Finance 5 2007 * The Effect of Ownership Structure on Dividend Policy in Jordanian Companies MGharaibeh ZZurigat KHarahsheh Interdisciplinary Journal of Contemporary Research in Business 4 9 2013 * Optimum Investing and Financing Policy MJGordon Journal of Finance 18 1963 * Ownership Concentration and Dividend Policy in Japan KHarada PNguyen Managerial Finance 37 4 362 2011