My blog entry for May 23, 2006, set forth the text of an e-mail sent to me by Wall Street Journal Columnist Jonathan Clements. In the e-mail, Clements defended Bill Sholar’s FIRECalc retirement planning tool from my criticisms of it, while acknowledging that FIRECalc “may not be the last word in safe withdrawal rates.” Set forth below is my response to Clements’ e-mail:

Jonathan:

Thanks for your response.

I agree that the conventional methodology studies have served to lower expectations and that that is a good thing. Bernstein described the Trinity study (the grandfather of conventional methodology studies) as “breakthrough research.” I agree with that assessment.

The fact that there has been one breakthrough in the safe withdrawal rate field does not rule out further breakthroughs. I hope you will take at least a quick look at Russell’s site. He has done some amazing work.

If it would help you to look at summaries of major findings, please let me know the sorts of questions in which you have the greatest interest. I can’t do the statistical work that Russell does. But I am writing a book on his findings. So I very familiar with his work and its implications.
You hint at one of the most important findings with your reference to TIPS. Given the valuations that apply to stocks today, a strategy of shifting a good bit of one’s portfolio to TIPS until such time as valuations return to moderate levels is highly appealing, given what the historical data says about the long-term returns likely to be provided by stocks from today’s valuation levels.

Please let me know if it is okay with you to quote your response to my e-mail in my blog. The question of the analytical validity of FIRECalc is a matter of considerable controversy at the Financial Freedom Community boards. So I set forth the text of my e-mail to you in my blog entry for today. I would like to set forth the text of your response as well (in part in fairness to Sholar and his supporters since to some extent you are defending the FIRECalc tool from my criticisms of it), but I of course will not do so unless you give me permission to do so.

Russell and I are in the process of constructing a calculator that uses the historical stock-return data to generate “predictions” of the range of stock returns we will likely see over 10-year, 20-year and 30-year time-periods and to assign rough probabilities to various points on the spectrum of possibilities. I have hopes that the calculator will be up at my site within three or four weeks. I will send you an e-mail letting you know when it is available. It is the work we did together on safe withdrawal rates that led to our study of the effect of valuations on long-term stock returns, which is of course an issue of even more far-reaching import.

My blog entry for May 22, 2006, set forth the text of an e-mail that I wrote to Wall Street Journal Columnist Jonathan Clements urging him not to put forward further endorsements of Bill Sholar’s dangerous FIRECalc retirement planning tool. Set forth below is the response that I received from Clements (he has given me permission to post his response here):

Rob:

Thanks for the email.

Given its length, my short reply will hardly do it justice. But I think there are a few points worth considering:

1) Firecalc may not be the last word on safe withdrawal rates, but it does what it states — it tells you how your strategy would have fared historically;

2) Surveys suggest that most retirees have a wildly inflated idea of what a safe withdrawal rate is, so Firecalc is a good reality check, even if it isn’t as much of a reality check as you would like;

3) With 10-year TIPS yielding 2.4%, the safe withdrawal rate must be higher than that, assuming a willingness to eat into principal.

Best,

Jonathan

I sent an e-mail this morning to Wall Street Journal Columnist Jonathan Clements urging him not to include further endorsements of Bill Sholar’s dangerous FIRECalc retirement calculator in his “Getting Going” investment column. Here is the text:

Mr. Clements:

My name is Rob Bennett. I am the founder of the Financial Freedom Community (a group of internet discussion boards) and the owner of the www.PassionSaving.com web site, where I report on findings of the research that we have been doing in our community for over six years now.

You included a link to the FIRECalc retirement calculator in your Wall Street Journal column of May 21, 2006, entitled “Make Sure Your Money Lasts as Long as You.” There have been extensive discussions of the FIRECalc calculator in our community over the past four years. We have determined beyond any reasonable doubt that the methodology used in FIRECalc is analytically invalid for purposes of determining safe withdrawal rates. I urge you not to give further publicity to this dangerous retirement planning tool unless the flaws in it are corrected by its creator, Bill Sholar. (I of course have made Sholar aware of the flaws).

The problem is that FIRECalc does not make adjustments to the safe withdrawal rate to reflect the effect of changes in valuation levels. The historical stock-return data shows that the valuation level that applies on the starting date of a retirement is the single most important factor determining what withdrawal rate is safe.

For retirements beginning in January 2000, FIRECalc reports a safe withdrawal rate (for a high S&P portfolio) of 4 percent. Analytically valid studies put the number at 1.6 percent. For a retiree with a $1 million portfolio, that’s the difference between living on annual spending of $40,000 for the last 30 years of his life and living on annual spending of $16,000 for the last 30 years of his life. That’s not a rounding error. That’s a deadly serious mistake.

Valuations are not as high today and so the safe withdrawal rate for a high S&P portfolio is today higher. But it is still nowhere even remotely in the neighborhood reported in FIRECalc. The SWR for a high S&P portfolio today is about 2.5 percent. The odds of an 80 percent S&P/20 percent commercial paper portfolio for which a 4 percent withdrawal applies surviving 30 years are at today’s valuations about 50-50.

Placing one’s entire life savings down on a coin flip is not a safe thing to do. It is an extremely risky thing to do. The words “safe” and “risky” are not synonyms; they are antonyms. FIRECalc is not science. It is science fiction.

William Bernstein, author of The Four Pillars of Investing, has confirmed the analytical flaws of FIRECalc. He has said that any investor giving thought to using a conventional-methodology safe withdrawal rate study (FIRECalc uses the conventional methodology) to plan a retirement would be well-advised to “FuhGeddaBoudit!” Bernstein calculated the safe withdrawal rate for a high S&P portfolio near the top of the bubble to be about 2 percent. That is in accord with what other analytically valid studies have shown. The four years of research done in the Financial Freedom Community backs up Bernstein all down the line.

I have had extensive communication with Dallas Morning News Columnist Scott Burns on the flaws of FIRECalc and a similar study (the conventional-methodology study published by John Greaney at his RetireEarlyHomePage.com site). Burns has told me that he agrees with the conclusions reached by the Financial Freedom Community in its investigations of the flaws of these two retirement planning tools. Burns has won for himself a reputation as an expert on the topic of safe withdrawal rates as the result of a column he wrote some years ago correcting an error Peter Lynch made in overstating the safe withdrawal rate for stocks.

Burns published a column on June 3, 2005, saying: “The established safe-withdrawal-rate rules of thumb are based on long periods of time in which yields were higher than today and stock valuations were lower. A growing school of thought believes future withdrawal rates should be reduced to reflect lower expected future returns. This would knock another 1.5 to 2 percentage points off the safe withdrawal rate.” Burns added that the reason why there has been only a smattering of reports in the general press about these critically important findings is that “it is information most people don’t want to hear.”

Greaney is a highly abusive poster in a number of discussion-board communities in which he participates. He has threatened physical violence on any poster who posts honestly on what the historical stock-return data says regarding safe withdrawal rates. Sholar is not himself abusive, but he permits Greaney defenders to have the run of the board that he owns (the Early Retirement Forum). At one time, Sholar permitted honest discussions of the safe withdrawal rate topic. But after Greaney defenders threatened to burn his site to the ground if he continued to do so, Sholar imposed a ban on honest posting. The ban remains in effect today.

The fact that Sholar has banned honest posting at his board shows that he knows from our community discussions that there is no possible reasoned defense that can be put forward in support of his safe withdrawal rate claims. In the event that stocks perform in the future somewhat as they always have in the past, these demonstrably false claims are likely to cause hundreds of thousands of busted retirements. You should not be putting the good name of the Wall Street Journal behind an endorsement of such an individual or behind an endorsement of safe withdrawal rate claims likely to cause such devastating life setbacks to so many middle-class investors.

I think you would be doing a great service to your readers to report in your column on the safe withdrawal rate findings of the Financial Freedom Community. Much of our best research has been published by John Walter Russell at the www.Early-Retirement-Planning-Insights.com site.

In contrast to Sholar and Greaney, Russell exemplifies the best of the new internet communications medium. He is a retired government engineer who became interested in safe withdrawal rates as the result of our community discussions and has now devoted four years of full-time work to studying in depth what the historical data really says. Giving some publicity to Russell’s findings could well save millions of middle-class retirements from going bust in days to come.

Please let me know if you have any questions about any of the points put forward in this e-mail. Please also know that the fine work you do in your column is appreciated by many, many middle-class workers trying to make sense of the question of how to invest successfully for the long run.

Rob