The conventional safe withdrawal rate studies claim that the safe withdrawal rate is always the same number and that that number is four. This cannot be. Robert Shiller’s Nobel-prize-winning research shows that valuations affect long-term returns. So the safe withdrawal rate is a number that changes with shifts in valuation levels.
Do the math and you learn that the safe withdrawal rate can drop to as low as 1.6 percent when stock valuations are as high as they were in early 2000 and can rise to as high as 9.0 percent when stock valuations are as low as they were in 1982.
I’m glad that that’s settled.
Q1 2023 hedge fund letters, conferences and more
The 4 Percent Rule
Actually, it’s not entirely settled. Intellectually, it is. But emotionally, lots of us have been reluctant to acknowledge the far-reaching how-to implications of Shiller’s amazing research. There are people in this field who to this day cite the “4 percent rule” as if it were a real thing. And many investors go along.
Acknowledging the effect of valuations would mean coming to terms with the reality that our stock portfolios do not possess nearly the real and lasting value that we have been led to believe they possess. So we pretend.
Still – over time we have been getting more nervous. I have been arguing without much success for 20 years that we need to get those old safe withdrawal rate studies corrected.
Something that I have noticed of late thought is that it is becoming an ever more common practice for advisers in this field to observe that, in the event that retirees using the 4 percent rule notice their portfolio dropping in value at a pace that makes them anxious, they can always take smaller withdrawals for a time.
Safe Withdrawal Rate Analysis
It makes me a little crazy when I hear that suggestion put forward. Here are four reasons why.
One, it counters the most important benefit of safe withdrawal rate analysis for a retiree making use of it to enter retirement with the thought that he is going to be making changes in his plan on the fly.
Retirees can of course make any annual withdrawal that they care to make. They can take withdrawals of three percent, five percent, eight percent, whatever. Theoretically, we could all decide on our withdrawal rate by going with hunches.
The purpose of safe withdrawal rate analysis was to develop a more analytically valid means of making the choice. Done properly, safe withdrawal rate analysis is not guesswork. It makes use of the historical return data to form an assessment of what withdrawal rate offers a high level of confidence. Retirement planning is a serious thing. When analysts in this field identify a withdrawal rate as “safe, that should mean something.
It doesn’t mean much when the analysis does not consider the effect of the valuation level that applies on the day the retirement begins, which is the most important factor. So there is a sense in which it is a good thing that we are now letting retirees know that they may well need to be changing their withdrawal rate as their retirement proceeds.
But wouldn’t it be better to change the methodologies of the studies to make them analytically valid? Properly designed studies would offer retirees a lot more assurance than studies that get the numbers wildly wrong but that come with an acknowledgment that the retiree may need to change his withdrawal rate as his retirement proceeds.
Two, the obvious question that should occur to people who are assurring retirees that they can make adjustments on the fly is – What sorts of adjustments?
If stocks continue to perform in the future anything at all as they have always performed in the past (that is, if the ignored valuation factor continues to matter as much as it always has), the changes that are going to be needed are going to be much larger than what most retirees using the discredited studies are anticipating. For a retiree with a $1 million portfolio, the difference between a 4 percent withdrawal and a 2 percent withdrawal is $20,000 per year.
A retiree who elects to cut his withdrawal rate in half is electing to cut back his living expenses by half. Wouldn’t it be better to make any adjustments needed to make the retirement workable before the retirement begins rather than to craft a plan that stands a good chance of not working out and then make adjustments on the fly?
Three, it is not clear how far into the retirement the adjustments should be made.
The beauty of a properly calculated safe withdrawal rate is that it provides the retiree with assurance that his plan will work even during time-periods when that might not be evident to those who haven’t studied the historical data. As defined in the literature, the safe withdrawal rate is a highly conservative number.
It is the number that works in the worst-case return sequence that we have seen in the historical record. So a retiree who sees five or ten years of bad results should not feel overly concerned about the long-term viability of his plan,
But of course this is not the case for numbers identified without taking the effect of valuations into consideration. Those sorts of withdrawal rates may well bring on retirement catastrophes. Retirees using those sorts of numbers in their plans might well be advised to start making adjustments at the first sign of trouble.
The idea of using bad numbers and then adjusting them on the fly runs counter to the entire idea of a genuine safe withdrawal rate analysis, in my assessment.
Four, using bad adjustable numbers does not force the aspiring retiree to face the strategic questions that would offer him more appealing means of constructing an adequate plan. There are almost always things that an aspiring retiree can do to make his plan safer if only he is told up front that he needs to do so.
He can delay the retirement for a bit. He can plan to take on part-time work during retirement. If it is high stock prices that are pulling his safe withdrawal rate lower than what he requires, he could shift a portion of his assets to non-stock asset classes. It’s generally better to make these choices prior to the beginning of the retirement rather than to wait until poor investing results cause a sense of panic.
Rob’s bio is here.
