No. of Recommendations: 12
There's more than "a bit of variation" in income with a constant 4% withdrawal. Withdrawal amounts might be depressed by around 50% for an extended period:
You kind of have to define the strategy a bit more to make sure we're talking about the same thing.
Here are two.
The usual "4% rule" is usually taken to mean "take 4% of the market value of your portfolio on the day you retire, liquidate and withdraw that much the first year, then inflation-adjust that cash withdrawal amount upwards each year so it is a constant dollar amount". This will generally have an unacceptably large chance of going bust even if you start when your portfolio is at a pretty average valuation level, but REALLY risky if started at a moment of high valuations. Being both old and broke at the same time is no fun. In most things I've read, it has been pretty much debunked and abandoned as a suggestion.
My suggestion was: each year to sell 4% of the number of shares you still own that year. The number of shares sold each year falls on a precisely predictable schedule.
(if they are dividend producing assets, of course, you just subtract the dividends received from that security from the dollar value of how much of that security you've calculated you have to sell)
So, the real income amount that you cash out can not fall more than the real price(s) of your asset(s). You can't ever have a wipe-out even if you're Methuselah, provided your securities don't go bust.
That web site starts with Dec 1999. Taking the same start date, FWIW, a person using my suggested strategy monthly on Berkshire stock would have a worst-case rolling two-year real income drop from its real all-time-high of -23.5%. (one reason the drop was even that big was because the all time high was set during the run-up to Sept '08, rolling two year income up 12% from a year earlier). Despite the withdrawals, real income for the last two years had risen inflation + 2.79%/year since the first two years. The number of shares owned is now down to 37.5% of the original pile.
If you'd done the same strategy with the S&P 500 Equal Weight (starting at end '99, the top of the bubble), the biggest drop in real rolling-two-year income from highest-to-date rolling-two-years was -34.8% for the period ending Aug 2010. (Despite the very unlucky start date for valuations, the real income drop from peak was only -16.6% in the tech bust). Real income in the last two years was up inflation + 2.44%/year compared to the real income in the first two years.
I think figures in that range are pretty acceptable for variation in income, as long as you know it can happen and don't (say) set up your expense profile on the assumption that it can't--especially if recent income from the strategy is at all time highs and has been rising fast. Allowing for a bit of up and down is WAY smarter than simply keeping your withdrawals constant and watching the portfolio value drop to zero. Obviously "cash out" will fall over the long run if the securities don't rise in price faster than inflation + 4%/year over the long run, but at least you can't ever go broke as long as the securities don't go bust.
If you're going to use "formulaic periodic withdrawals for life" as your strategy, it's better to have a little flexibility in the income level and have that certainty of completely eliminating longevity risk. Starting the strategy at a time of high valuations leads to lower income for a fair while--that's not ideal, but better than a diet of dog food.
Jim
