Dividend persistence and return predictability

Evidence of dividend yield return predictability has been presented so widely and consistently that the result has tended to be generally accepted. This paper shows that return predictability of the dividend yield is a spurious result that is due to dividend persistence and finds that standard dividend behaviour explanatory models are also affected by the spurious regression problem. A simulation procedure is utilized to take account of a spurious correlation that compounds the spurious regression problem when the dependent and independent variables in a time series regression are ratios composed of common component variables. The paper’s results therefore imply that extreme care should be taken when using ratios as predictor or explanatory variables in time series regression. The paper introduces a reformulated Lintner first difference dividend behaviour model that is not subject to spurious regression.