Last September I received an email from Rebecca Mitchell, an Investment Analyst in the UK with an irresistible suggestion. Here it is again, with updates to my charts and commentary.

Rebecca: I was looking at your Short History of Stock Dividends. I like the chart, but was wondering if it would be possible to provide an updated one with two different exponential regression trend lines, one for each period (1870-1982, and 1982-present) to match the two averages for the dividend yield?

If one is to think of the focus on dividend yields as having shifted, the focus on price might also have shifted!

Click the first chart to see a larger version incorporating Rebecca's suggestion. The intersection of the two regression slopes does generally coincide with the decline in dividend yield. As I suggest in my monthly dividend update, this change largely resulted from a convergence of demographics and tax policy — the Boomers looking for growth, not income, and the advent of tax-deferred savings plans, which further blurred the distinction between price appreciation and dividend yield.

This chart with the two regressions reminds me of this optimistic chart published a couple of years ago by Jeremy Siegel. Siegel was focused on real operating earnings per share, a different but related topic. He used a regression from the early 1980s to support a bullish market forecast. Along the way he observes that "Finance theory predicts, and historical data confirms, that if a firm pays a lower proportion of its earnings as dividends, then these unpaid earnings must be used to either repurchase shares, lower debt, or invest in capital."

Unfortunately, investors are now beginning to understand that this redirection of earnings from the investor's pocket to the corporate coffer was a slow-motion exercise in wealth destruction. Price growth was partially offset by the decline in dividend yields. But when price growth reversed, dividend yields were far less accommodating than they were during the Great Depression.

In my view, as illustrated in the second chart, a regression trend line through the market from the early eighties to the 2000 peak merely highlights a reversible secular trend comparable to regressions through the 1920s and from 1949 to 1966.

We're now well into the post 2000 cycle, with many investors hopeful, if not confident, that the March 2009 low was the beginning of a new secular bull market. Time will tell.