Let us see what this article said then choose your own "poison"
http://seekingalpha.com/article/2000661-...-dividends
///////////////////////////////////////
To summarize:
BRK with Dividend Portfolio
1) Due to the high price of BRK stock, the portfolio was started with $10 million to get a sufficient number of shares.
2) 1% of the portfolio value was removed each quarter (4% a year) as a dividend payment.
3) The stock price was lowered by the amount of the dividend paid, and this price adjustment was cumulative for entire 24-year study period.
4) Since no shares were sold, and there was no dividend reinvestment, the number of shares of BKH held throughout the study period was the same as the number of shares bought at the beginning.
5) The final portfolio value was determined by the final adjusted stock price multiplied by the number of shares owned.
BRK without Dividend Portfolio:
1) The portfolio was started with the same number of shares and the same stock price as the "BRK with Dividend" portfolio.
2) An amount of cash equivalent to the dividend payment from the "BRK with Dividend" portfolio was removed each quarter by selling shares.
3) No dividend adjustment was made to the historical stock price, and the final value of the portfolio was the final stock price multiplied by the final number of shares.
Results
The study ran from Jan 12, 1990 through Jan 27, 2014. During that time, BRK's stock price went from $8200 to $169,511. The adjusted stock price for the "BRK without Dividend" portfolio at the end of the study was $118,616.48 due to the payout of $50,894.52 of dividends over the 24 year period.
The "BRK with Dividend" portfolio both started and finished with 1,219 shares. The "BRK without Dividend" portfolio had shrunk down to 285.36 shares by the end of the study.
The portfolios created a total of $61,719,656.20 for the portfolio owner. (Again, this is with a starting portfolio value of $10 million).
The final portfolio value of the "BRK with Dividend" portfolio was $144,593,491.01, for a return of 11.75%, not including the dividend payouts. The final value of the "BRK without Dividend" portfolio was $48,372,089.38, for a return of 6.77%.
Conclusion
First of all, let me say that I don't agree that the stock price of BRK should be permanently adjusted downward by the amount of the dividend each time a dividend is paid, as I did in this scenario. Although the markets do this automatically on the ex-div date, there is no evidence that this is a lasting effect. Usually within a few days or weeks that drop in stock price is erased. At $169,511, BRK has a PE of a little over 14. If we really did adjust the price downward for each dividend, and BRK's price was only $118,616, then its PE would be about 10.25. I don't think anybody believes that BRK should be trading at such a low PE. No matter how many dividends were paid, the market would bring the price right back up to where it should be based on its earnings. But I did it in this study to cut off that criticism before it ever gets made.
The final value of the "BRK with Dividend" portfolio would have been even higher had I not adjusted the share price. And still, even with lower adjusted the price, the "BRK with Dividend" portfolio created $96 million more value than the "BRK without Dividend" portfolio. Although they both were able to give their owner a 4% yearly payment, while still growing the portfolio's value, the "BRK with Dividend" Portfolio outperformed the "BRK without Dividend" portfolio by over 4 percentage points a year, which translates into almost a 200% larger portfolio after 24 years.
To me, this shows that selling shares to create income is a dangerous strategy. BRK's stock price had an annual return of over 21% from 1990 through 2014, so both strategies were able to be profitable. But how many of us will be able to build a retirement portfolio that will return 21% a year? I'm guessing none of us. With the more reasonable 8-10% annual return we can expect during our retirement, a non-dividend paying portfolio most likely would have run out of money, or, to keep from running out of money, the owner would not have been able to take out as much income as the dividend portfolio owner was able to.
So, although I hesitate to criticize anything Warren Buffett would do or say, in this case I would have to say he is wrong. Even for someone like himself, a master of capital allocation, I feel that for the sake of his own portfolio (not necessarily as a business decision for BRK) he would have been better off had BRK paid a dividend. He would have been able to donate those dividends to charity, and still maintained ownership of all his shares of BRK.
For those of us who cannot expect to produce 21% annual returns like Buffett, using a dividend strategy is not only a more conservative strategy, one that makes it unlikely that you will run out of money, but it is also one that is likely to allow you to continue to grow your portfolio while in retirement. You get the trifecta of security, good income, and continued capital growth.
Thank you for reading my article. I welcome your comments and criticisms.