New SWR Study Counts Valuations —
But Only In One Direction!

Business Week ran an article in its recent Annual Retirement Guide issue arguing that the Old School safe-withdrawal-rate studies get the numbers all wrong — they are too pessimistic! Holy moly!

Juicy Excerpt: One expert now questioning this conventional wisdom is Michael Kitces, 30, director of financial planning for Pinnacle Advisory Group in Columbia, Md. Kitces was frustrated that the 4% rule can result in overly conservative withdrawal rates during certain market conditions and that the market’s mood at the time of the initial withdrawal could greatly affect how much money retirees can drain from their accounts for the rest of their lives.

The reason why I say “Holy moly!” is that the more pressing problem today is the millions of retirements that are likely going to fail because they were constructed in accordance with the overly optimistic claims of the Old School studies. Truth be told, though, much of the argument set forth in Kitces’ paper sounds like it was influenced by him listening in to the discussions that we have been having in the Retire Early and Indexing communities for over six years now (I don’t believe that Kitces was in fact influenced by our discussions, my point is just that he got to the same place that a lot of us did by applying more common sense to his examination of the historical data than did the authors of the Old School studies).

Juicy Excerpt: A growing body of research reveals that, in fact, longer-term returns can be predicted to some extent…. Those extended periods of expanding or contracting P/E ratios (producing long-term real returns above or below historical averages) can often be anticipated in advance — by looking at the valuation of the aggregate market at the beginning of the time-period!

Precisely so. Michael Kitces, take a bow! The testimony of six years of our discussions shows that you have it exactly right.


Unfortunately, after getting it exactly right by accepting that valuations must affect safe withdrawal rates, Kitces drops the ball by jumping to the strange conclusion that valuations pull safe withdrawal rates up but not down. 

Juicy Excerpt: History reveals only three time-periods at which a 5% initial withdrawal rate was not sustainable.

That’s the “Holy moly!” part. Kitces here commits the analytical error core to the Old School studies. We started out trying to determine what withdrawal rate is safe and then made a sudden shift to considering what in the past has been sustainable. As if the two concepts were the same!

This would not pass for “analysis” in any field other than investing. Imagine an alcoholic arguing that it is perfectly safe for him to drive drunk because on three earlier occasions on which he did so he got in crashes but lived. The fact that retirees using a 4 percent withdrawal at times of extreme valuations have always found themselves in big trouble in the past indicates not that a 4 percent withdrawal is safe at times of extreme valuations but that it is risky. The fact that they managed to “sustain” in those three earlier cases does not transform a risky withdrawal rate into a safe one. The words “safe” and “risky” are not synonyms; they are antonyms. 

Is Kitces Old School or New School? Perhaps we should say he’s Nold School. Or Oldew School. Something like that.

How can someone smart enough to see that valuations affect safe withdrawal rates not also be smart enough to see that they affect them in both directions, that the safe withdrawal rate is both sometimes higher than 4 percent and at other times less than 4 percent? And how could the good and smart people at Business Week fail to pick up on this painfully obvious and critically important point? I blame it on the Passive Investing model. Buy into the Passive Investing mindset (and millions did during the out-of-control bull) and you give up your ability to apply human reason to your understanding of valuations-related investing topics. As shocking as it is, that’s the reality we have seen come into play time and time again over the course of our discussions of the past six years.

I view this as a positive development. As more and more of the “experts” (I don’t intend the quote marks as a dig at anyone — my view is that there is no such thing as an investing expert until we all achieve a higher level of understanding of the basics, a level that can only come when we are able to put the Passive Investing model behind us and engage in honest and informed back-and-forth discussion on all sorts of questions once again) become aware that valuations do indeed affect safe withdrawal rates (as we first demonstrated on our boards in May 2002!), it seems inevitable to me that a few will begin to wonder how it could be that this reality would always serve to push safe withdrawal rates up even higher and never to pull them even a teeney, tiny bit below 4 percent. Kitces has advanced the discussions, even if he has not yet been quite bold enough to bring it all back home.

And who knows? Maybe Kitces will devote a bit more thought to all this and make the shift from being a Nold School SWR guy to being a true New School Hero of the First Rank. After posting this blog entry, I will send Michael an e-mail letting him know about our findings and about The Retirement Risk Evaluator.

We’re getting there, people. Slowly. Reluctantly. With feet dragging. With men moaning. With women sobbing. With cats meowing. With babies banging their rattles. But we are over time working ourselves to that place where deep in our hearts we all really want to be.

Or so I pray!

Today’s Passion: William Bernstein is another “expert” who has combined top-notch commentary on the safe-withdrawal-rate matter with some words that suggest that he has been drinking the Passive Investing Kool-Aid a bit too long. Some time back I wrote a blog entry on Bill’s thinking entitled Ball of Confusion.


  1. says

    Juicy Excerpt: History reveals only three time-periods at which a 5% initial withdrawal rate was not sustainable.

    The problem with this is that there are only three major time periods in the historical record. The “5%” number could be 6% or 7%–there would still be only three time periods.

    Have fun.

    John Walter Russell

  2. Rob says

    Stocks are both more risky and less risky than most people today realize. They are more risky at times of high valuations and less risky at times of average or low valuations.

    When you throw all of the historical record into a single big bowl of data soup, you throw off all the numbers. To make sense of the historical record, you have to approach it with an understanding that valuations affect long-term returns. Such an appreciation of the realities is foreign to those who have bought into the Passive Investing concept. The Rational Investing concept begins from a premise directly contrary to the starting-point premise of the Passive Investing model.


  3. says

    Dear Rob,

    I just became aware of your past research in September. Since then, I’ve read archives from many Discussion Boards and websites, and I always find your writing to be very interesting and intriguing.

    I just finished writing up a paper about using earnings valuations to predict withdrawal rates. But I think it is also important to consider dividend yields and bond yields as well, since these are the income sources of returns. With these measures, I find that withdrawal rates continue falling even after 2000. With PE10, you found the lowest withdrawal rates in that year.

    I do cite you and John Walter Russell in my paper as the earliest and strongest advocates of this approach. I understand that John Walter Russell has passed away, sadly.

    If you wish to see my paper, it is at:

    I’m hoping it will be publishable in a finance journal. This is just the first draft, so if you have some comments or anything, I can still make revisions. Thank you very much for your consideration.

    Best wishes, Wade


  1. Is the Safe Withdrawal Rate too safe? Or too aggressive!?…

    As readers of my newsletter know, in May I published research that challenges the safe withdrawal rate as potentially being TOO safe in some environments, where market valuation is not at unfavorable extremes. However, in some feedback I’ve received f…

Leave a Reply

Your email address will not be published.

Comments links could be nofollow free.