Companies in the Standard & Poor’s 500 Index really love their shareholders. Maybe too much. They’re poised to spend $914 billion on share buybacks and dividends this year, or about 95 percent of earnings, data compiled by Bloomberg and S&P Dow Jones Indices show. Money returned to stock owners exceeded profits in the first quarter and may again in the third. The proportion of cash flow used for repurchases has almost doubled over the last decade while it’s slipped for capital investments...This is a familiar theme to readers of this blog.
Here are my own updated numbers. The figure shows dividends and total payouts for the S&P 500 and the nonfinancial corporate sector as a whole, for rolling five-year periods ending in the year shown. Payouts are given as a share of aftertax profits.
|Shareholder payouts as a fraction of aftertax profits, 5-year moving averages|
I haven't broken out the S&P 500 before. (This is based on the current index membership -- it didn't seem worth the trouble to find historical indexes. So for the early years we are talking about a relatively small number of firms.) As you can see, the picture is basically similar. The rise in S&P payouts comes a bit later. And unlike the broader population of firms, there is no rise in dividends relative to profits in the 1980s and 1990s -- the entire increase in payouts comes from buybacks. The other difference -- not immediately evident from the chart -- is that profits, not surprisingly, are more stable in the S&P 500 than in the smaller firms outside the index. You can't tell from the figure, but the big spike in the black lines comes from a collapse in profits in the non-S&P firms, not an increase in payouts. The corporate sector excluding the S&P 500 reported substantial aggregate losses in 2001-2002, meaning a much lower denominator for the ratio in the early 2000s.
Incidentally, this figure was produced in R, which I am finally switching to after years of using SAS and (hangs head in shame) Excel. If you are starting a graduate program in economics -- and I know some readers of this blog are -- I strongly, strongly advise you to learn R and get in the habit of producing all your work in LaTeX with embedded R code, using sweave or knitr. Kieran Healey explains why. You should never cut and paste a graphic from one application to another, or copy statistical results by hand into a table. I think this is the single piece of advice I most wish I'd gotten when I started graduate school.