"Set It And Forget It" In Retirement Doesn't Work In Bear Markets

In the April 2011 edition of Financial Planning magazine, Temma Erhenfeld writes in an article Outliving Your Money sights a study from T. Rowe Price that shows that those who retired in 1/1/2000 using the standard withdrawal formula of 4% of assets the first year with yearly 3% inflation adjustments would only have a 6% chance of success by 2011! This is with a typical 55% S&P500 and 45% Total US Bond portfolio. The study suggests the best option is to reduce withdrawals by 25% for three years after each bear market bottom (in which there were two in the period studied). Following this option brings your chances of success to 43% in 2011. Frankly that's still not good enough odds in my book. My suggestion would be, in addition to this suggested technique, to diversify the portolio even more (at least add foreign and emerging market stocks and bonds, REITs, and commodities) in addition to using Mebane Faber's timing model: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=962461. Furthermore, their study assumed a 10% stock return and a 6.5% bond return. That's really pushing it in my view. Without any bear markets, the expected chance of success under these conditions was 89%.