A Re-Examination of Dividend Policy: A Dynamic Panel Data Analysis

Abstract

Through a cross sectional analysis of 48 firms, over a seven-year period, from 1996 to 2002, we try to shed some light into two sides. First of all, we research whether managers smooth their dividend policies or not. Besides, we outline the main determinants that may influence the dividend policy pattern. Specifically, we attempt to find answers to the following questions: Do Tunisian firms follow stable dividend policies or not? Do dividend yields differ across industries? What are the main factors that determine the dividend policy making? In the first section, Lintner's model is applied. Our results show that Tunisian firms rely on both current earnings and past dividends. However, it seems that dividends tend to be more sensitive to current earnings than prior dividends. Any variability in the earnings of the corporation is directly reflected in the level of dividends. However, dividend policy does vary across financial and non-financial industries. Our results uncover a low target payout ratio of respectively 16% and 27% for non-financial and financial firms while the adjustment speed respectively ranges from 0.68 to 1.56. Non-financial institutions adopt therefore a more smoothing dividend policy. Secondly, through a fully developed model, we highlight some factors that may influence the dividend policy pattern. First, riskier firms with high financial leverage pay out fewer dividends and have lower dividend yields. Furthermore, high-profitability firms with more stable earnings can also afford more dividends. However, larger investment opportunities deprive firms from higher dividends. Similarly, growing firms distribute fewer dividends. Additionally, dividends serve to reduce agency costs between the shareholders themselves only whereas the conflicts between insiders and outsiders seem to be not resolved with dividends. This matter holds true only in non-financial firms. In this vein, it should be noted that our analysis does show significant differences throughout financial versus non-financial industries. Finally, the size of Tunisian corporations has a systematic negative effect on dividend policy.

Keywords: dividend policy, dynamic panel data and Tunisia