Ownership structure's effect on dividend policy : Evidence from publicly listed Swedish firms

This study examines the effect of ownership structure on dividend policy of 284 firms listed on the OMX Stockholm Exchange in Swedenfrom 2010-2015. Specifically, the purpose of the study is to investigate therelationship betweendifferentinvestor types and dividend policies of firms, measured as dividend yield and dividend payout ratio.Also, the study aims to predict dividend behaviours based on ownership structure which may be useful inthe future since ownership structures of listed Swedish firms havebeen changing over time, with an increased consolidation of ownership and a sharpincrease in institutional ownership. The sample consistsof 1046 observations and was gathered from Thomson Reuters’ Datastream and Eikon databases. This is the first study to examine the relationship between ownership structure and dividend policy in Sweden.The dividend policy is measured using two dependent variables; dividend payout ratio anddividend yield and a multiple regression has been used in orderto test the hypotheses whether any relationships exist between 17 different types of ownership structure used as independent variables, four additional control variables and dividend policy.The findings indicated significant positive relationships between institutional ownership and dividend yield and dividend payout, with one exception being private equity which exhibited a negative relationship with dividend yield. Furthermore, market capitalization, return on assets and price to book value are positively related to dividend payout while debt/equity ratio showed a negative relationship with dividend yield. The results contradict those of the most recent research conducted in Turkey (Al-Najjar & Kilincarsla, 2016) but adds supportin the debateto existing theories of dividends’ relevance to the value of firms developed by Gordon (1963), Lintner (1962) and Walter (1963). Limitations of the study include the geographical delimitation to Sweden which creates some constraints to wider generalization ofthe results to other geographical settings. Furthermore, the datacollected from Thomson Reuters Eikon hadmissing values, showed signs of heteroscedasticity and relevant investor variables such as family ownership were unavailable