Payout Policy. Are Dividends Disappearing?

Are Dividends disappearing?
Dividend concentration and the consolidation of earnings
Harry DeAngelo, Linda DeAngelo, Douglas Skinner
Journal of Financial Economics, 2004

• Fama and French (2001) document that the number of industrial (nonutility and nonfinancial) firms that pay dividends decreased by over 50% during 1978 to 1998. This paper has been cited in the financial press as evidence that dividends are disappearing or have become less relevant in the last two decades.
• However, in this paper DeAngelo et al. find that dividends paid by industrial firms actually increased over 1978 – 2000, both in nominal and real terms (by 224.6% and 22.7% respectively). This evidence raises the main research question in this paper: Why have aggregated dividends increased in the face of a radical decline in the number of payers?

Sampling procedure and aggregate dividends, 1978 – 2000

• In this section the authors describe the sampling procedure and the details the aggregate dividend increase from 1978 to 2000 for industrial firms.
• Following Fama and French (2001), the authors sample CRSP industrial firms with SIC codes outside the ranges 4900 – 4949 and 6000 – 6999 (financial and utility firms). They restrict attention to NYSE, AMEX and NASDAQ firms with share codes 10 or 11 (common shares) for at least one month of each year in question, and nonmissing December share price and quantity data. They consider only firms CRSP firms with dividends and earnings before extraordinary items on Compustat.
• Table 1 in the paper shows that the large decline in the number of dividend payers from 1978 to 2000 is confined to industrial firms. While the number of industrial firms that pay dividends fell from 2,250 in 1978 to 926 in 2000, the number of payers among financial and utility firms increased from 852 to 933. This implies that the decline in the number of payers does not reflect a general increase in the managers’ reluctance to pay dividends.
• Table 2 in the paper documents that the aggregate nominal dividends increased by 224.6% for CRSP/Compustat firms between 1978 and 2000, while the aggregate real dividends (in 1978 dollars) increased by 22.7% in the same period.

Dividend concentration and the increase therein over the last two decades

• Table 3 in the paper ranks firms that pay dividends by cash dividends paid in 1978 and 2000, in groups of 1000 firms. Overall, a relatively small number of firms pay the overwhelming majority of aggregate industrial dividends, and this concentration increased over the last two decades. For example, in 1978 the top 100 dividend payers distributed 67.3% of the total dividends in 1978. This percentage increased to 81.8% in 2000.The $ 31.5 billion in real dividends paid by the top 100 dividend payers in 2000 exceeds the $ 31.3 billion paid in 1978 by all dividends payers.
• Table 4 summarizes the cross – sectional distribution of dividends in 1978 and in 2000 categorized by real dollar dividends paid. We can see that the number of firms paying dividends of $ 100 million-plus increased by 34 over 1978-2000, for a dividends increase of $12.6 billion. Over the same period, the number of firms paying less than $100 million decreased by 1,280 with a dollar magnitude reduction of $1.1 billion. Therefore, the top categories of dividend payers now contain more firms and these firms pay substantially higher total dividends, while the bottom categories now contain many fewer firms and these firms pay modestly lower total dividends.

Earnings concentration and payout ratios

• According with Linter’s (1956) dividend supply decision primarily depends on earnings. It suggests that the high/increasing dividend concentration may be the result of high/increasing earnings concentration. The evidence reported in this paper supports this hypothesis.
• Table 5 ranks dividend-paying firms by their earnings in 1978 and 2000. Like dividends, earnings were highly concentrated in 1978, and substantially additional concentration has occurred among dividend payers over the last two decades. In 1978, the top 100 dividend payers generated 57.5% of the earnings of all payers, while cumulatively the top 200 payers generated 71%. These percentages increased for 2000 to 74% and 86%.
• Table 7 reports the distribution of real earnings for payers and nonpayers in 1978 and 2000. It documents a strong positive relation between the level of earnings and the proportion of firms paying dividends. For example, only 2.3% of the firms with earnings of $100 million-plus failed to pay dividends in 1978, compared with 28.1% in 2000. Despite this reduced propensity to pay, aggregated real dividends increased by $ 7.1 billion from 1978 to 2000.
• Table 8 shows that payout ratios exhibit a little change over the last two decades.
• Table 9 in the paper identifies the 25 industrial firms that paid the largest dividends in 2000. These primary dividend suppliers are well established firms such as Exxon Mobil and General Electric. The top 25 payers distributed 54.9% of aggregate industrial dividends paid in 2000.
• Table 10 ranks the 25 nonpayers with the highest 2000 earnings. A handful of technology firms dominate the earnings of nonpayers. These firms’ decisions to forego dividends more likely reflect their continued high growth prospects than a reduced propensity to pay dividends.

Implications and summary
• Why have aggregated dividends increased in the face of a radical decline in the number of payers? Because i) the reduction in payers occurs almost entirely among firms that pay very small dividends, and ii) increased in real dividends from the top payers swamp the modest dividend reduction from the loss of many small payers.
• These findings on dividend concentration cast doubt o the empirical importance of the dividend clientele and signature hypothesis.
• If the demand to satisfy heterogeneous clienteles were the truly first order determinant of dividends policies, we would expect to observe substantial dividend heterogeneity among the prominent firms whose securities are important of well-diversified portfolios. This is not the case, we observe few firms with very high earnings that fail to pay dividends, mainly from the technology sector.
• If managers use dividends to communicate with stockholders, dividend signaling should occur primarily in small, relatively unknown firms with limited access to the financial press. But the vast majority of dividends of dividends are paid by prominent corporations like General Electric, which enjoy major coverage by analysts and press.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License