>I've heard the story about how earnings per share can
>replace dividends, but it never made sense to me.
>Stock owners never actually get non-dividend earnings.
> They are simply informed of them on the quarterly
>balance sheet before they are reinvested. Stock
>holders are never actually given that money, so the
>earnings only value to investors would be on the
>assumption that the reinvestment will lead to greater
>dividends in the future. The price model would then
>have to be changed to include how the expected return
>on the investments would affect the quantity and
>timing of future dividends that they are expected to
>lead to. This seems highly dubious. In sum, earnings
>certainly can't be simply be interchangeable with
>dividends in any valuation model because investors
>aren't actually paid earnings whereas dividends have
>real value.
You're betraying a very old-fashioned way of thinking! There was a time when some people bought stock for dividends, but now only fuddy-duddies and retirees do. It's all about what the price is going to be tomorrow, or next month, or next year, when you hope to sell it for a capital gain. Since stock prices are at least partly determined by underlying profits, the best guess you can have for the price at which you'll unload the shares is some multiple of future profits. It's one step this side of alchemy, but that's the rationale.
Speaking of stock trading, it seems that on the first day of trading in that Chinese Google, Baidu, the average share turned over ten times.
Doug