The 'Dividend Puzzle': Dividends, Earnings and Income

We like dividend shares at the DIY Income Investor - as described here and here. But how much evidence is there for a relationship between dividend yield and total shareholder return?

Until 2003, the economics 'establishment' believed that there was little relationship between a company's dividend level and its future earnings (and therefore the value of the share). Since then a lot of evidence has emerged to support the existence of such a relationship - but it does not completely answer this 'dividend puzzle'.

The economists Modigliani and Miller in 1961 postulated the 'dividend irrelevance theorem' - in other words, under certain assumptions, a company's dividend policy should not matter, in terms of the future value of the company's shares, as it should make no difference to a company's future earnings if the company used its income to re-invest or to pay dividends. 

However, in one of the most surprising research developments over the last four decades, Robert Arnott and Clifford Asness published an article in the January/February 2003 issue of the Financial Analysts Journal entitled “Surprise! Higher Dividends = Higher Earnings Growth”. In this publication the authors  documented that in every case over rolling 10-year periods from 1946 to 1991, the highest dividend payers had the highest earnings growth. These results were not just on average, but were robust through the strongest and weakest markets as well. The authors concluded that: “The historical evidence strongly suggests that expected earnings growth is fastest when current payout ratios (of dividends) are high and slowest when payout ratios are low.”

While the article focuses on payouts of the S&P500 over time, similar studies were later conducted  in different countries including Canada, Australia, France, Germany, Japan, Netherlands, Switzerland, and the UK. The same relationship was found to hold: higher dividend paying ratios tend to lead to higher earnings growth in the future.

However these studies noted that there was no strong evidence that there was a link with  real growth in dividend earnings. So be clear that the level of dividends is only one of a number of variables that can influence your dividend earnings. It's not simple - there is no clear-cut conclusion on the 'dividend puzzle' but a practical approach based on a number of selection criteria is shown here.



I am not a financial advisor and the information provided does not constitute financial advice. You should always do your own research on top of what you learn here to ensure that it's right for your specific circumstances.