Four Methods and Three Estimates

Research on Safe Withdrawal Rates

Moderator: hocus2004

Post Reply
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Four Methods and Three Estimates

Post by JWR1945 »

Four Methods and Three Estimates

We are now at the point that we have four methods for calculating a Safe Withdrawal Rates. They are, as follows:

1) Historical Sequence Approach.
2) Monte Carlo Simulation.
3) Mathematical Analysis.
4) Specification and Design.

The first two are familiar to all. The third is based upon gummy's safe withdrawal rate formula and related theorems and calculations.

The third has surfaced just recently. By understanding the mechanisms behind portfolio failure, we can proceed to design an approach that prevents its occurrence. Peteyperson has introduced an innovative approach and our investigations into it are producing valuable insights.

Because of the research into peteyperson's cash buffer management issue, we can add a third method for determining Safe Withdrawal Rates in the current era. Investigations into the 1921-1980 era (for starting a retirement) revealed that 1959-1973 were the worst years and they corresponded to high valuations. They were especially hazardous because they were strung together. Portfolios are most likely to fail because of problems early on. Otherwise, they will grow above the danger zone.

I was able to identify a key mechanism associated with high levels of safety. The dividends in the era (and more than a decade that followed) started with a 3% yield and the real dollar amount stayed very close to that value. Eventually, it even grew a little bit. The safe withdrawal rate calculated by FIRECalc was 4.0%. That means that very few shares of stock had to be sold. After an extended period of decline, there were enough shares of stock remaining to be able to recover and extend the final duration of the portfolio. The dividend shielded the portfolio from price changes...except for one percent.

To a very rough approximation, based on this mechanism, the safe withdrawal rate for January 2000 was 2.2%. The dividend yield at that date was 1.17%.

The three estimates for 2000 that take valuations into account are:
1) William Bernstein 2.0% Monte Carlo method (poorly defined).
2) My original adjustment, using P/E10 levels to adjust for extreme valuations, was 2.3%.
3) This new method, based upon a theoretical understanding of failure mechanisms, is 2.2%.

I draw attention to the fact that we can now reliably improve our withdrawal strategies while maintaining high levels of safety. We can introduce additional asset classes on a sound theoretical basis. Even though we are just at the very beginning, we can specify our requirements and design our strategies to match our goals.

Have fun.

John R.
hocus
Moderator
Posts: 435
Joined: Mon Dec 02, 2002 12:56 am

Post by hocus »

The three estimates for 2000 that take valuations into account are:
1) William Bernstein 2.0% Monte Carlo method (poorly defined).
2) My original adjustment, using P/E10 levels to adjust for extreme valuations, was 2.3%.
3) This new method, based upon a theoretical understanding of failure mechanisms, is 2.2%.


I find it very interesting that you came to a third number in the same neighborhood when using this third approach.

I draw attention to the fact that we can now reliably improve our withdrawal strategies while maintaining high levels of safety. We can introduce additional asset classes on a sound theoretical basis. Even though we are just at the very beginning, we can specify our requirements and design our strategies to match our goals.

My attention is completely drawn!

Thanks for all the effort you are putting into this, JWR1945.

By the way, I will be asking you to look over my "The Truth About Safe Withdrawal Rates" report to check for "bloopers." I sometimes think of you as my personal blooper-detection device. Not that I always take the hint, mind you!

The way I see it is, if you supply the data, I will supply the enthusiastic verbiage, and we'll see whether we can convince anyone out in the big bad world to pick up the pretty blue package with the words "New and Improved SWRs Inside--Get Yours Now!" stamped on the front.
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

hocus wrote:By the way, I will be asking you to look over my "The Truth About Safe Withdrawal Rates" report to check for "bloopers." I sometimes think of you as my personal blooper-detection device. Not that I always take the hint, mind you!
Use them for humor.

Opportunity knocks! If you really got enough things bad enough, you might be able to make them into a separate book.

Have fun.

John R.
peteyperson
**** Heavy Hitter
Posts: 525
Joined: Tue Nov 26, 2002 6:46 am

Re: Four Methods and Three Estimates

Post by peteyperson »

John,

Let me know when the lack of board discussion on your posts becomes an issue. Such work shouldn't go unrewarded.

Petey
JWR1945 wrote:We are now at the point that we have four methods for calculating a Safe Withdrawal Rates. They are, as follows:

1) Historical Sequence Approach.
2) Monte Carlo Simulation.
3) Mathematical Analysis.
4) Specification and Design.

The first two are familiar to all. The third is based upon gummy's safe withdrawal rate formula and related theorems and calculations.

The third has surfaced just recently. By understanding the mechanisms behind portfolio failure, we can proceed to design an approach that prevents its occurrence. Peteyperson has introduced an innovative approach and our investigations into it are producing valuable insights.

Because of the research into peteyperson's cash buffer management issue, we can add a third method for determining Safe Withdrawal Rates in the current era. Investigations into the 1921-1980 era (for starting a retirement) revealed that 1959-1973 were the worst years and they corresponded to high valuations. They were especially hazardous because they were strung together. Portfolios are most likely to fail because of problems early on. Otherwise, they will grow above the danger zone.
Post Reply