Do investor preferences drive corporate dividend policy?

This research paper aims at assessing whether managers adapt their dividend policies to the changing preferences of investors, as predicted by the catering theory of dividends. To answer this question, we used an modified approach based on the method proposed by Baker and Wurgler [2004a] in their studies on dividend catering. We noted a systematic decline in percentage of companies that paid out dividends in a sample of American publicly-traded companies, excluding companies of low capitalization and low profitability. Next, we observed a parallel declining tendency in dividend premiums in our sample. The decrease in the readiness to pay out dividends among companies on the American market can be linked to the fact that investors have assigned less weight to dividends over the years, and so in turn they were less willing to reward dividend-paying companies with higher valuations. Periodic fluctuations in investor mood with regard to dividend-paying companies, and the resulting changes in their relative valuation, influence the propensity of managers to pay out dividends. We showed a statistically significant relationship between changes in dividend premiums in one year, and the proportion of companies that paid out dividends in the following year. Additionally, it looks like companies try to compensate shareholders by paying out dividends in years of worse performing market and are less likely to distribute their earnings when shareholders gain on rising stock price. We found a negative correlation between the change in proportion of companies paying out dividends and changes in the S&P500 index. However, this does not seem to reflect investor preferences and taste for dividends. We found no statistically significant correlations between the change of the dividend premium and changes in the S&P500 index and, surprisingly, we observed relatively worse valuation of dividend-paying firms in years of market downturn. In terms of originality, our work contributes to the ongoing dividend puzzle discussion in a number of ways. First, we use a sample of American companies after excluding small capitalization stocks. Second, we assume a time lag between a shift in investor preferences and a change in corporate payout policy. Finally, our studies also account for the impact of general market conditions on dividend decisions.