In other words, a company should only pay dividends if it is unable to reinvest its cash at a higher rate than the shareholders (owners) of the business would be able to if the money was in their hands. If company ABC is earning 25% on equity with no debt, management should retain all of the earnings because the average investor probably won't find another company or investment that is yielding that kind of return.
Taxes and Dividend PolicyThe tax consequences are tremendously important. If the company is earning 8%, and the individual investor can only hope to earn around 8%, the business should still keep the proceeds because of tax law. Why?
Say, for instance, a small company consistently earning 8% on equity made a profit of $100,000 last year. If the business reinvests that money back into itself, the shareholders could reasonably expect the company to earn $8,000 on the reinvested earnings. If the $100,000 had been paid out in the form of dividends, it would have been subject to the investors individual tax rates. For simplicitys sake, lets say all of the shareholders are in the 30% tax bracket. When the $100,000 was paid out as dividends, $30,000 would have gone to the IRS, leaving only $70,000 available to the investor to reinvest at 8%. At the end of the year, the investors could only expect a return of $5,600 ($70,000 reinvested at 8% = $5,600). Over time, the discrepancy can add up to very significant numbers.
Businesses that operate in mediocre industries (such as steel, railroads, etc.) with low returns on equity would best serve shareholders by paying out profits as dividends. Take ALCOA (the Aluminum Company of America) as an example. According to Yahoo! Finance, the company has a return on equity of 4.89%. Investors can almost certainly earn a higher return, even when adjusting for the adverse tax effects. True to form, the company pays out a lot of its cash flow to shareholders.