Hur påverkar utdelningar företagsvärde?

: en kvantitativ studie av utdelningsfrekvenser på Nasdaq och New York Stock Exchange

Abstract

Dividend is a relatively well known and researched topic, however, no definite explanation seems to exist for the effect on a firm’s value. Previous studies suggest that researchers have reached a grasp variety of different results throughout time. As a result of an inconclusive and broad range of results this area of studies can be defined as unexplored which leads to the purpose of this essay of quantitatively analyze the impact and effect of dividend frequency on a company’s value. The survey used for analysis is based on two charts: Nasdaq and New York Stock Exchange. For more updated results this survey included 11144 observations in which were registered between the years of 2009 and 2013. The U.S. market uses mainly four different types of periodicity when it comes to appliance of dividends, those distributions consist of the following: annual, half-yearly, quarterly or monthly. There are also companies registered whose management does not apply the dividend strategy as part of their policy. All five types of dividend appliance were considered for the purpose of this study. All the data collection was made through the Datastream database and all the statistical calculations were created through SPSS. Four hypotheses were created based on agency theory, Miller and Modigliani's (1961) article about the irrelevans of dividend, the bird in hand and prospect theory. As a conclusion, the survey determines that the firms with the highest value apply the quarterly dividend which has proven to be the most used method. A summary of the mean values shows that the monthly dividend distribution leads to the lowest enterprise value while the regression analysis clearly determines that the yearly dividend has the least successful result on valuations. Based on several different tests throughout the entire study, it was shown that the quarterly dividend was the significantly best dividend policy.