"The Dividend Puzzle Revisited, Again: Does Home Country Matter?"

Abstract

Abundant research notwithstanding, dividend policy remains one of the greatest puzzles in corporate finance. In this study, we revisit a central tenet of dividend theory, namely that home country matters to dividend decisions. We pit this view against the notion that increasing globalization results in financial integration and cross-country trends which, while exerting an influence on all firms embedded in a particular country, are not country-specific. We posit that the home country effect should diminish with time, and that institutions may combine in a complementary or substituting manner to affect dividend policies across countries in an equifinal way. We employ hierarchical linear modeling to estimate the home country effect on dividend policies of 10,794 firms, operating within 811 industries embedded within 49 countries, over a period of 11 years. The country effect was statistically significant in some models, but its relative importance was limited and declined over time. These results support the notion of mildly segmented markets and imply that while the home country effect may be declining in relative importance, to date it still affects dividend policies around the world. We discuss implications of these findings for the study of dividend policy, national institutions, and international business.