This study examines whether there exists any evidence to support the notion that variations across firms in observed ownership structures cause systematic variations in observed firms’ dividend pay-out behaviour. In this study, secondary data has been used and collected from Prowess and Capitaline databases. Sample used is BSE 100-Index companies for 15 years from 2000–2014, hence the data is panel data in nature. Behaviour of entities is observed across time under panel dataset. Variables used for this study are selected after extensive review of literature. Independent variables i.e., measures of ownership used in the study are promoters holding, non-promoters holding, non-promoters institutional holding and non-promoters non institutional holding. And dependent variable in this study is dividend pay-out ratio. Panel data regression model is used in the study along with descriptive statistics. As the data is panel data in nature, so in order to get unbiased results pooled regression analysis is first applied and then with the help of f-test and hausman test it is decided which model is suitable for the study. With f-test being significant, panel data regression model fits best for the study. Regression models are applied one by one, taking one independent variable and dependent variable into account at a time. Using panel data regression model in between all ownership structure measures and dividend policy, the empirical findings suggest that the probability value of t-statistic in case of non-promoters institutional holding and non-promoters non-institutional holding is found to be significant. In case of promoters holding and non-promoters holding (as a separate variable) no effect is found on DP ratio of the firm.
Ownership structure, dividend policy, BSE
