Dividends are like hats. The company can let its stock go up in price. You can sell some, and buy a hat you like, cheap or expensive or none at all. Or the company can pick a single hat that they think will suit every shareholder, and mail the same hat to everybody whether they want it or not, without any consideration of the fit or cost.
...
Well put.
Statements like this will get you stoned as a heretic on many thousands of internet boards & yahoo videos. But you surely know that. ;-)

Well, I did phrase it in a bit of an extreme way, so a pinch of disdain is not unwarranted. I will grudgingly admit that there's nothing wrong with a small dividend here or there, other than messing up my investment bookkeeping. I just have a problem with folks who don't quite get the pros and cons, both among management teams and among shareholders. Whose lifestyle does a small dividend change? A big one as percent of operations is well suited to a cash cow that can't expand. A big one as a percent of stock price is most often a bad sign...if it were a good company, the price would be bid up and the yield would be normal. A super high dividend never lasts. Either the firm is good but oversold and rebounds so the yield falls, or the dividend gets cut.

So the main reasons for a solid dividend are
* The business has no opportunities to deploy that slug of capital at decent rates of return, and management knows it. Nothing wrong with that, sometimes it's the best choice, and management is making an intelligent choice, which is a good sign. This is what See's does within Berkshire.
* They do have good opportunities, but management is dumb, so they think a dividend is a better idea because it's...conventional? Not sure what they're thinking.

The main reasons for a small/token dividend are:
* Management thinks that's what they're supposed to do, not really understanding capital allocation, similar to #2 above.
* Management knows it's not really useful, but they think a dividend will mysteriously make them more popular or taller. Like the Hitchhiker's Guide says, "mostly harmless" I guess.

None of the four situations is super attractive to me. Different strokes, I guess.

There are times that a portfolio with a decent yield makes sense, and I wrote about that recently. mungofitch writes on Dividend investing
(the subtle thing in that post being that a suitable dividend portfolio with turnover will generally maintain its dollar value of income generation right through a bear market, or nearly so, unlike a buy-and-hold portfolio).
But it's a small niche.

The most rational dividend philosophy would, I admit, be very unpopular: if there are opportunities to deploy the capital, or such opportunities are expected to come around soon enough, sit on it. If there is still excess cash beyond foreseeable opportunities, check the stock valuation. If it's high, pay a dividend. If it's low, stop the dividend and do buybacks. Alternate as required. This would be received more favourably if every dividend during high stock prices were called "special".

Jim