Invalid

Research on Safe Withdrawal Rates

Moderator: hocus2004

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

Invalid

Post by JWR1945 »

Invalid

For hocus: We need a word other than invalid.

We can show that intercst, in the Retire Early Safe Withdrawal Rate study:
1) Got the number wrong.
2) The number never has been right (as a prediction as of the time that he wrote it).
3) Used his disclaimer unless the future is worse than the past in a deceptive manner, implying an exceedingly unlikely event, but when pressed on specific issues, twisting its meaning to be no more than the portfolio is guaranteed to be safe unless it fails.

Yet, people do not like to call the study invalid. IMHO, they interpret that word as rejecting the historical sequence method in toto. That is not your intent. You do not intend to reject raddr's sensitivity studies, nor my investigations that have used the FIRECalc, nor many others. I think that that is what people are hearing. It is not what is being said.

Consider this analogy: Suppose that your objective is to plant flowers on the moon. If you went to your local lawn and garden store and bought a book and supplies, you could not grow flowers on the moon. You would need to supply an atmosphere, control the temperature and do a host of other things. But many of those supplies, especially the book and the seeds, would come in handy. They would never be directly applicable on the moon, not by themselves. They would be applicable on the earth and they could be helpful elsewhere. People would prefer not to say that those supplies, including the book and the seeds, are invalid. The reason is that they are valid for the earth.

Have fun.

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

Post by hocus »

People do not like to call the study invalid. IMHO, they interpret that word as rejecting the historical sequence method in toto. That is not your intent You do not intend to reject raddr's sensitivity studies, nor my investigations that have used the FIRECalc, nor many others. I think that that is what people are hearing. It is not what is being said.

I would never want anyone to think that I reject the historical sequence method. The historical sequence method has great value. William Bernstein called the Trinity study "breakthorugh" research, and I agree with that claim. It is the historical sequence method that showed the effects of volatility, and that was a great advance in SWR analysis. Intercst performed a service to the FIRE community by publishing his study at his web site, and by publicizing it at the REHP board. The study offers many valuable insights. I have made profitable use of it in my own planning, and recommend that all aspiring early retires study it and make use of it.

That said, the conventional SWR methodology ignores a critical aspect of SWR analysis. It ignores the effect of changes in valuation levels. My body of posts pointing out the importance of changes in valuation levels is as important a breakthrough in SWR analysis as the Trinity study insight re volatility. Changes in valuation levels and volatility are both critically important aspects of SWR analysis, roughly equal in importance. In the days before it was widely understood that changes in valuation levels had a critical effect on the result, it was reasonable for analysts to ignore this factor. Once the data-based proof we have assembled vindicating me in my claim that changes in valuation are a critical factor in the analysis are publicized, there will be no justification for an analyst to ignore this factor. It would be a deliberate deception for an analyst who knew that changes in valuation affect the number to report a number to the public that did not reflect this effect and to claim that he was reporting the safe withdrawal rate.

Here's how William Bernstein says it on page 73 of the Four Pillars book: "Although the discounted dividend model (DDM) informs us well about expected returns, it tells us nothing about future risk. We are dependent on the pattern of past returns to inform us of the potential risks of an asset. And in this regard, I believe that the historical data serve us well." He is saying that you must combine your findings from analysis of the two critical factors to achieve a reaonable assessment of the SWR.

Volatility is critical to SWR analysis. Any methodology ignoring the effects of volatility is invalid. Changes in valuation levels are critical to SWR analysis. Any methodology ignoring the effects of changes in valuation levels is invalid. You can't do an analytically valid analysis by looking at changes in valuation alone, and you can't do one by looking at volatility alone. Both critical factors must be taken into consideration.

I have no problem with the statement "the conventional methodology is invalid for the purpose of determining SWRs." The "for the purpose of" part should always be mentally added in when I make the shorthand version of the statement that "the conventional methodology is invalid." Analysts can of course use the conventional methodolgy to determine the historical database rate (I am ignoring for the time being some non-valuation concerns that I have with the conventional methodology, which need to be discussed separately from the valuation issues).

The problem is with studies that determine the historical database rate and then purport to have determined the safe withdrawal rate, which is something different. Determining the historical database rate gets you only halfway there. Until you factor in the effect of changes in valuation levels, you do not know the SWR (as that term is defined in the SWR studies).
JWR1945
***** Legend
Posts: 1697
Joined: Tue Nov 26, 2002 3:59 am
Location: Crestview, Florida

Post by JWR1945 »

hocus
I have no problem with the statement "the conventional methodology is invalid for the purpose of determining SWRs." The "for the purpose of" part should always be mentally added in when I make the shorthand version of the statement that "the conventional methodology is invalid."
I like this.

Is there any way to shorten it up or to do something else to make sure that people add the "for the purpose of" part? I would like something sufficient to get people to check into the details instead of misreading your comments.

Have fun.

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

Post by hocus »

Is there any way to shorten it up or to do something else to make sure that people add the "for the purpose of" part?

I'll make it a practice from now on to include the "for the purpose of " part each time I make the invalidity claim.

I would like something sufficient to get people to check into the details instead of misreading your comments.

One thing that will help in this regard is writing a FAQ for this board. One question would be "Why do you say that the conventional methodology is invalid?' and we would provide all the background relating to this aspect of the debate. Then, when the issue came up, we could refer people to "Question Six of the FAQ" for a full explanation of why the invalidity question is important, why the claim of invalidity is made, what the various viewpoints on the question are, and so on.
Post Reply