Could determinants found in the financial statements of U.S. companies predict dividend payout?

Abstract

This empirical research will investigate the relationship between dividend payout and company selected factors for U.S. Stocks. This study identified and discussed dividend theories such as (Modigliani & Miller, 1961) Dividend irrelevance theory, (Gordon, 1963) and (Lintner, 1962) Bird in Hand Theory, Tax Preference Theory, Clientele Effects, amongst others. The sample of this study consisted of 50 U.S. listed companies from different industries, using global industry classification standard (GICS). This study was conducted in the year 2018/2019, and the author assessed the financial statements of companies from the year 2013 to the year 2017. Financial data acquired form Thomson Reuters was assessed on a quarterly basis. The sample of companies chosen for this study has not missed a quarterly dividend payment in the above-mentioned dates. A dividend regression model was conducted for each company and results are averaged using Microsoft Excel. This dividend regression model was constructed by reference to past literature. A hypothesis test was also formulated to check if variables are significant or not, and whether an investor could predict dividend payout by analysing company selected factors used in the dividend regression model. The majority of results achieved were in line with our expectations and with previous studies discussed in Section 2. Significance of variables was achieved at a single company level, however not every single variable was considered as having overall significance. The Size and Leverage independent variables of the model were highly effective in predicting any variability in the dependent variables of the model, and were the only two variables to achieve overall significance.