Safe initial withdrawal rates higher than preciously thought

Research on Safe Withdrawal Rates

Moderator: hocus2004

Post Reply
User avatar
kathyet
Admin Board Member
Posts: 34
Joined: Thu Nov 28, 2002 4:00 am

Safe initial withdrawal rates higher than preciously thought

Post by kathyet »

http://www.fpanet.org/member/press/rele ... alrate.cfm

here is another page on it somehow the first one got lost
hocus2004
Moderator
Posts: 752
Joined: Thu Jun 10, 2004 7:33 am

Post by hocus2004 »

Thanks very much for posting that link, Kathyet.

(The link is to the press release for the study discussed in the Jonathan Clements column referenced in the "Getting Going or Going Bust?" thread.

http://nofeeboards.com/boards/viewtopic.php?t=3107

If you follow the press release to the bottom, there is a link to the study itself.)
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

kathyet's link was better for me. I could not use the alternative because I am not a subscriber to the Wall Street Journal.

[It is a great publication, but it is too big. I would keep on trying to read everything, everyday, which would be a disaster.]

The key flaw was visible right off the bat.

The study looked at a worst case historical sequence. I do not fault the study for doing this. That sequence was with valuations and dividend yields within the historical range. It was at the high end, to be sure, but it was still within the range.

The study produces what I call Calculated Withdrawal Rates. It does not put any confidence limits about them. A Safe Withdrawal Rate corresponds to a lower confidence limit. [A High Risk Withdrawal Rate corresponds to the upper confidence limit.]

The study presumes that the Calculated Withdrawal Rate of a worst case historical sequence is a Safe Withdrawal Rate looking forward.

The error is the same that shows up with any Historical Sequence analysis that fails to take valuations into account. The presumption of safety requires that the stock markets of the past be representative of today's market.

That is false. Today's market, with its P/E10 value of 27 to 28, corresponds only to the very worst cases of the past: 1929 and the mid-1960s. Results that the study identifies as worst case must be treated as typical in light of today's valuations.

I will be studying the article in depth in the future. Because of our investigations, I was able to spot its critical and fatal flaw immediately.

Have fun.

John R.
Post Reply