Answers are grounded in this paper and cite their sources.
Summary The paper analyses longitudinal Compustat data on dividend payments and share repurchases of over 2000 US industrial firms by using 2-yearly averages for the period 1992–2007. First, dynamic auto-regressive models for dividends and share repurchases were estimated by maximum likelihood to yield consistent and efficient estimates. Second, because firms pay dividends at different rates, comprehensive dynamic and static random-effects models for dividends were estimated. Third, models were estimated for share repurchases and the interrelationships between dividends and repurchases were investigated, tackling endogeneity issues. The results showed that dividend decisions are likely to precede those regarding share repurchases, and higher dividend payments decreased the magnitudes of repurchases; the effects of repurchases on dividends were generally insignificant.
No takes yet. Share an insight, caveat, or question.
Alok Bhargava (2010) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: