Yahoo "Finance Quiz"

[wabmester]

It can't be both. The data has to say one of those two things and not the other, right?

Are you serious? If so, I have to point out two fundamental flaws in your logic:

1) You assume there's only one "correct" way to interpret the historical data. This is absurd. Stock market performance has more correlated variables than the weather, and is much harder to predict. To assume that you can take one of those variables, like the P/E ratio, and come up with the one true interpretation is, ehrm, misguided at best.

2) You assume you can extrapolate whatever your misguided historical analysis tells you. Not unless you've discovered some new law of the stock market universe.

IMHO, one should always question the assumptions of any forward looking prediction.

For example, the underlying assumption of intercst's analysis is that the performance of the stock market in the next 50 years (or whatever your timeframe is) will be no worse than it was in the last 120 years. That is equivalent to assuming that the weather in the next 5 days will be no worse than it was in the last 12 days. Personally, the size of the data set doesn't give me a lot of confidence given the length of the withdrawal period.

The underlying assumption of the "Gordon equation" is that the GDP will continue to grow as fast as it has during our amazingly productive recent past.

The underlying assumptions of PE10 are not only that the mean P/E is the "right" P/E (i.e., that the risk premium shouldn't change over time), but that RTM will continue to happen at the same frequency it has in the past.

The more implicit assumptions you make like this, the more likely you are to be wrong.

 

[intercst]

For example, the underlying assumption of intercst's analysis is that the performance of the stock market in the next 50 years (or whatever your timeframe is) will be no worse than it was in the last 120 years.   That is equivalent to assuming that the weather in the next 5 days will be no worse than it was in the last 12 days.   Personally, the size of the data set doesn't give me a lot of confidence given the length of the withdrawal period.

I'd say it's more like the equivalent of assuming that "the weather in the next 5 years will be no worse than it was in the last 12 years.

Sure global warming or some other event might make the weather better or worse, but its unlikely.

I agree with your comment that stock market performance is dependent on so many intercorrelated variables, that to try to predict the future based on just one (e.g. PE ratio) is ludicrous.

intercst

 

[wabmester]

I'd say it's more like the equivalent of assuming that "the weather in the next 5 years will be no worse than it was in the last 12 years.

Sure global warming or some other event might make the weather better or worse, but its unlikely.

Well, I didn't want to put too fine a point on it, but we're really talking about the "net weather" which is harder to visualize over the long term.

So, while we may have some confidence that no single year of the next 50 will be much worse than any single year in the past 120, it's much harder to be confident that the entire 50-year sequence will be no worse than the past based on seeing just a couple of independent sequences.

Specifically, we've historically seen a lot of RTM to the upside after crashes.   If we were to use Japanese stock market data instead of US market data, we wouldn't see such a rosey picture.   And, personally, I don't see anything that makes Japan *fundamentally* different from the US wrt stock market behavior.

 

[hocus]

Re: Yahoo "Finance Quiz"Are you serious?   If so,

You assume there's only one "correct" way to interpret the historical data.

I don't say that there is only one correct way to calculate SWRs. The Bernstein way is not the same as the JWR1945 way. I view both of these approaches as analytically valid. It is my hope that we will be seeing lots of additional approaches to SWR analysis in days to come.

Stock market performance has more correlated variables than the weather, and is much harder to predict.   To assume that you can take one of those variables, like the P/E ratio, and come up with the one true interpretation is, ehrm, misguided at best.

The approach that JWR1945 and I have put forward (and also the approach that Bernstein uses) incorporates every factor incorporated into the conventional methodology, plus one (the effect of changes in valuation levels). If you view the conventional methodology numbers as helpful, then you should view the data-based methodology numbers as super-helpful. We take nothing away from what is included in the conventional methodology studies. We add in a critical factor which those studies ignore.

Not unless you've discovered some new law of the stock market universe.

I understand that there are many who believe that it is a law of the stock market universe that long-term returns are entirely unpredictable. I put up a post earlier in this thread in which I set forth five quotes from Bernstein showing that he does not share that view. Bernstein is not the only one who has come to the conclusions he has come to as a result of his analysis of the historical data.
 
One should always question the assumptions of any forward looking prediction.

This is a very true observation and a very important observation, in my view.

I think it is fair to say that every SWR analysis is rooted in some sort of assumption as to how stock investing works. One simple way of getting to essence of the differences in the numbers is to ask--What long-term return does this analysis assume?

The conventional methodology assumes that changes in valuation levels have zero effect on long-term returns. The Bernstein approach assumes that the Gordon Equation provides a good estimate of the long-term return. The JWR1945 approach uses statistical tools to assign confidence levels to various possible future scenarios, given what has happened in the past.

I think that the important thing is that the reader of the study be put on notice as to what the assumptions are. Say that you as an investor place no confidence in the Gordon Equation. Then you know that the Bernstein approach to SWR analysis is not for you. I think that is fine.

I have two big problems with the conventional methodology: (1) the assumption that changes in valuation levels have zero effect seems far-fetched in the extreme to me; and (2) the conventional studies do not set forth any justifications for the assumptions made. If justifications for the assumptions were included in these studies, it would put investors using them on notice as to the assumptions they were buying into when they relied on them to make investing decisions. That would do a lot to diminish the dangers associated with them, in my view.

I would like to see all SWR analysts make it a practice to be up front about the assumptions that go into development of their methodologies. I think that formal studies done in the future should include discussions of the underlying assumptions and the researcher's justifications for making those assumptions.

The more implicit assumptions you make like this, the more likely you are to be wrong.

What you are talking about a tool that tells people the number that is 100 percent safe presuming that the future is like the past, you need to include all the factors that bear on that question. If you leave any out, you are guaranteed to get the wrong answer. It's better to use reasonable assumptions on all factors than to entirely ignore one known to be of critical importance.

 

[Michael]

...Japan *fundamentally* different...

Since 1871 Japan has moved from despotic rule by warlords and emperor to a relatively free society that respects private initiative.  Japan was also devastated by WWII in a way the United States was not.  Their cities were bombed, which heavily damaged their industry, and then were occupied by foreign troops.

By contrast, the US since 1871 has had more or less the same respect for freedom and private initiative, and has never had any of its cities seriously damaged by foreign bombers.  The troops always fought far from home. No foreign power ever managed to conquer us, occupy our cities, or force a change of Constitution upon us.  This relative stability resulted in a much different situation for our economy.

Unless Russia nukes our cities into oblivion, China drags us into a major land war in Asia, or terrorists get a lucky hit with bio weapons, our economy may stay relatively stable for our retirement years.  Japan's market will probably recover better than it has in the past, as long as they stay out of another major war.

 

[intercst]

Hmmm,

I was hoping for a very simple answer, just for reference. Come on' *****, you're not running for office!

Something Like
____________________________________________
Currently   TIPS  - 30%, IBonds, 35%, and CDs 35%.

and at  todays P/E levels the Dow would have to be in the 4000 - 5000 range for me to be interested in stocks.

_________________________________________                        

I don't want any more theory, I'm just interested in what you're doing with it?  - Can you try to answer my question again in Just a sentence or two?

Excellent point!

Frustrated readers on both the Motley Fool and NoFeeBoards posed the same query repeatedly, only to be met with 20-paragraph tomes that didn't answer the question.

I see that we're well on the road to the same end on this board. <LOL>

intercst

 

[Cut-Throat]

What you are talking about a tool that tells people the number that is 100 percent safe presuming that the future is like the past, you need to include all the factors that bear on that question. If you leave any out, you are guaranteed to get the wrong answer.

Wrong! Wrong! Wrong! - It is 100% safe provided that the future is exactly like the past! we already know the past, because we have all the factors in the return history.

You can never provide a number that will be safe in the future. That is why there is no such thing as a SWR only HSWR!

Could you please go back a couple of posts and answer (very briefly) my question that I posed to you? - Thanks! I just want to see what you are doing.

 

[wabmester]

Michael, the US has had its share of drama too.  But the bottom line is that both the economy and the stock market are unpredictable, *especially* in the long run.   And while looking at our brief market history (or something like the Gordon Equation which is derived from that history) may give you a warm fuzzy feeling, it couldn't hurt to consider alternative outcomes.

Just so I don't get labeled as a nay-sayer (or worse, a cybersaur), a few alternative approaches I like include:

* safe and predictable TIPS
* owning your own business(es)
* real estate, especially in areas in which demand always exceeds supply
* more international exposure than most people have, especially in countries that don't have some of the demographic nasties we often discuss here

 

[hocus]

I'd say it's more like the equivalent of assuming that "the weather in the next 5 years will be no worse than it was in the last 12 years. Sure, global warming or some other event might make the weather better or worse, but its unlikely.

Just how unlikely do you think that it is that the valuation levels we entered in the late 1990s will cause long-term returns for retirements beginning from that time forward to drop below the historical norm, intercst? Do you think that we have a 1 percent chance of lower returns? 10 percent? 50 percent? 90 percent?

I think it is fair to say that raddr is more than a little skeptical of Bernstein's SWR claims. Yet here is the bottom-line conclusion he comes to in an analysis of the question.

Raddr: At current dividend yields, the chance of average or better market performance going forward is pretty small.

http://www.nofeeboards.com/raddr/gordon.htm

Your study assumes that changes in valuation have zero effect on long-term returns. Yet the raddr analysis and the Bernstein analysis and the JWR1945 analysis and the Shiller analysis and the Smithers analysis and the Easterling analysis and the Arnott analysis all say that the chances that changes in valuation will have zero effect are exceedingly small.

Given the testimony of these many other researchers, are you still as confident as you were two years ago today that the 4 percent number will work? Do you think that there is any chance whatsoever that Bernstein is saying something important when he says that the conventional methodology numbers are "misleading" when we are at the sorts of valuation levels that we are at today?

 

[intercst]

Given the testimony of these many other researchers, are you still as confident as you were two years ago today that the 4 percent number will work? Do you think that there is any chance whatsoever that Bernstein is saying something important when he says that the conventional methodology numbers are "misleading" when we are at the sorts of valuation levels that we are at today?

Actually, Bernstein is quite clear on the matter of safe withdrawals in this article

http://www.efficientfrontier.com/ef/901/hell3.htm

Now, let’s return to the above table. The historically naïve investor (or academic) might consider reducing his monthly withdrawals to a very low level to maximize his chances of success. But history teaches us that depriving ourselves to boost our 40-year success probability much beyond 80% is a fool’s errand, since all you are doing is increasing the probability of failure for political, economic, and military reasons relative to the failure of banal financial planning.

Mind you, this is not a call for wild abandon. The above table constrains the retiree desiring a theoretical 97% success rate (of portfolio survival) from spending more than 3% per year of the initial real amount of his nest egg. Taking the accident propensity of the species into account would allow him to spend about 4%. But if you believe that we’re about to encounter a bad returns sequence or simply wish to leave a few baubles to your heirs, you’re right back to 3% again.

So live a little, and enjoy your money, for tomorrow we may be consumed by the ghosts of Hitler, Lenin, and Attila the Hun. And at withdrawals of 3% to 4% of your nest egg, don’t spend it all in one place.


Dozens of very smart people have objected to your continued misquotes of Bernstein's work (I'm not aware of any mention by Bernstein endorsing a 2% retirement withdrawal), yet you persist. Do you have any written confirmation from Bernstein that he concurs with your misunderstandings about his work?

While you're at it, perhaps you can bring us up to date on Dallas Morning News financial columnist Scott Burns' response to your open letter of Feb 20, 2004.

Mr. Burns:

I am a big fan of your column. I strongly believe that you have done a great deal of good for many middle-class investors with your writings.

I am writing to ask that you not offer further endorsements of the safe withdrawal rate (SWR) study published by John Greaney at his RetireEarlyHome Page.com web site. I have been studying the SWR concept since late 1995, and am the founder of the SWR Research Group discussion board (see link below). I have determined by making reference to the historical data that the methodology used in the Greaney study (the so-called conventional SWR methodlogy) is analytically invalid. Studies using this methodology do not accurately report the withdrawal rate that is safe presuming that stocks perform in the future as they have in the past.

The most important flaw of the conventional methodology is that it makes no adjustment in the SWR for changes in valuation levels. William Bernstein says in his book “The Four Pillars of Investing” that changes in valuation levels affect future returns as a matter of “mathematical certainty (page 12),” and that, therefore, the historical return data examined in conventional methodology SWR studies “can, at times, be misleading (page 43).” Since the late 1990s, we have been living in the sorts of times during which the problem is a serious one......

As you know, Scott Burns has a degree from MIT and has been writing about financial topics for about 30 years. Surely if there was anything of substance in your ravings, you would have heard from him by now.

I await a much coveted one or two sentence response, but I'm not hopeful.

intercst

 

[John Galt]

Hello intercst! Even though your post was not
directed at me, just a quickie.................I have run the
numbers backward and forward, listened to the pundits,
potentates and pontificators. I truly believe that
(overall) the 4% SWR is about as good as mere mortals can do, given the available data.

John Galt

 

[Cut-Throat]

Excellent point!

Frustrated readers on both the Motley Fool and NoFeeBoards posed the same query repeatedly, only to be met with 20-paragraph tomes that didn't answer the question.

I see that we're well on the road to the same end on this board. <LOL>

intercst

*****,

Since you have read my very simple question and refuse to answer it or acknowledge it, I for one will not be participating in your silly games.

You came to this board and were basically welcomed. I now understand the 'bad blood' on the other forums. I will no longer read your posts or respond to you on this forum. I am already bored with you and your posts.

I suggest that you get a girlfriend.

 

[Nords]

Re:  Meteorological persistence & problem posters

The prime gold standard of weather forecasting to which all meteorologists aspire is "Persistence". Persistence says that tomorrow's weather will be the same as today's weather (or whatever forecast period you're using). The attractive thing about Persistence is that no one has to argue about units, timeframes, locations, data quality, or any of that crap.

Persistence has a better record than EVERY meteorologist, human or machine, who/that has ever published a forecast. And Persistence is only logging about 50%.

BTW, Bernstein's "Retirement Calculator from Hell" is my favorite on the subject. And I believe that quantifying anything over an 80% success rate over the rest of my life-- or at least the next 30 years-- is an exercise in rising frustration and diminishing validity. I think Persistence will beat all the other forecast techniques there, too.

Intercst, you're a moderator and you know *****. Presumably your experience and background gives you the potential to avoid being dragged into this shouting match. It's hard to believe that you're worried he'll somehow kill off this board (or your board), and it's disappointing to see you indulge in the unprofessional behavior of baiting him. It's exceptionally disappointing to watch you rise to this troll's bait. You're the responsible adult in this childish diatribe, you've made your point, and it's time to move on to other topics instead of validating *****' "jihad".

Dory36, I know this isn't your fault, but I have a software question. Instead of allowing others to give problem posters a podium (or even worse, martyrdom), could YABB add an "Ignore" feature like M*?

 

[hocus]

Scott Burns has a degree from MIT and has been writing about financial topics for about 30 years. Surely if there was anything of substance in your ravings, you would have heard from him by now.

Scott Burns and I have exchanged a number of e-mails on various SWR questions since I sent to him the e-mail you quoted above. I have not yet requested his permission to report on that correspondence to the Retire Early board community. I will take a small chance here and report a small snippet of one of his e-mails to me. I will send him an e-mail tomorrow morning (since today is a holiday) and let him know that I quoted him here on this one point.

Scott Burns: "I agree with you and like the thought process.... I would like to talk with you and write a column on the subject....What's your phone number?"

 

[intercst]

Re:  Meteorological persistence & problem posters

The prime gold standard of weather forecasting to which all meteorologists aspire is "Persistence".  Persistence says that tomorrow's weather will be the same as today's weather (or whatever forecast period you're using).  The attractive thing about Persistence is that no one has to argue about units, timeframes, locations, data quality, or any of that crap.  

Persistence has a better record than EVERY meteorologist, human or machine, who/that has ever published a forecast.  And Persistence is only logging about 50%.

BTW, Bernstein's "Retirement Calculator from Hell" is my favorite on the subject.  And I believe that quantifying anything over an 80% success rate over the rest of my life-- or at least the next 30 years-- is an exercise in rising frustration and diminishing validity.  I think Persistence will beat all the other forecast techniques there, too.

Intercst, you're a moderator and you know *****.  Presumably your experience and background gives you the potential to avoid being dragged into this shouting match.  It's hard to believe that you're worried he'll somehow kill off this board (or your board), and it's disappointing to see you indulge in the unprofessional behavior of baiting him.  It's exceptionally disappointing to watch you rise to this troll's bait.  You're the responsible adult in this childish diatribe, you've made your point, and it's time to move on to other topics instead of validating *****' "jihad".  

Dory36, I know this isn't your fault, but I have a software question.  Instead of allowing others to give problem posters a podium (or even worse, martyrdom), could YABB add an "Ignore" feature like M*?

The "***** problem" on the Motley Fool and NoFeeBoards wasn't solved until about 75% of the participants reached the conclusion that he wasn't posting anything of value and tired of the disruption.

I suspect you'll need the same consensus on this board before any corrective action is taken.

intercst

 

[John Galt]

And everyone tells me I should write a book. Even with
my enormous ego, I assume some of them are just
being polite.

John Galt

 

[intercst]

*****,

Since you have read my very simple question and refuse to answer it or acknowledge it, I for one will not be participating in your silly games.

You came to this board and were basically welcomed. I now understand the 'bad blood' on the other forums.  I will no longer read your posts or respond to you on this forum. I am already bored with you and your posts.

I suggest that you get a girlfriend.

Home wrecker!

He's already married and has kids. <LOL>

intercst

 

[cute fuzzy bunny]

I think there is an ignore feature on this, it just has to be installed or enabled.

Its worth noting though that in my substantial posting career, I've never been "ignored" or ignored anyone. Thought about it a couple of times but then it occurred to me that the "problem" that made me want to ignore someone was usually mine.

However, if we had the feature, I would probably use it. Just not on *****.

He seems long winded, well intentioned, and certainly persistent. Whether I agree with him or not isnt relevant. I dont think any "withdrawal rate" calculation except for historic ones is workable, predictable or worthwhile except as an academic discussion. Since we wont be living in the past, that makes it all pretty irrelevant.

Markets move short term because mass market psychology moves it. In unpredictable manners. Nobody has ever come up with a "system" that consistently beats "the market" and nobody ever will. Most patterns or "correlations" (theres that word again) we "discover" in the historic data is coincidental, unpredictable in future cycles, and may or may not ever repeat.

We'll continue to see historic returns as long as our economic cycle continues to operate like it has historically. When it stops doing that, things will get better or worse depending on which direction our civilizations economic cycle takes. With history as a guide, that would be downward at some point.

Nobody will be able to predict when or where that directional change will take place. Owning the market in indexes and across a broad range of diverse asset classes over long periods of time will be a winning strategy as long as everyone else doesnt figure that out. Being lucky a few times in getting out at extreme market tops and in at bottoms, as I did in early 2000, can make or break your retirement. But theres no calculation, strategy or spreadsheet that can do that for you, and you're as likely to be wrong as right.

All THAT having been said, the idea of avoiding investments that are overpriced and buying asset classes that are beaten down may make sense. Unfortunately every person has their own idea of what is under or overpriced and they'll likely never agree.

In summary, as far as I'm concerned ***** and everyone else can have their say. I can choose whether to read or pass by their dissertations. If I do read a post, I'll say whether I agree or disagree...if I feel like it. What I hope I never devolve into is someone with nothing better to do than follow someone else around with a jar of tar and a bag of feathers. I'd rather go back to work.
:

 

[db]

The more implicit assumptions you make like this, the more likely you are to be wrong.

Wabmester,

How right you are. This is a concept, like confounding of variables, is often missunderstood or forgotten. By the way, beautiful presentation of the assumptions underlying the various approaches: succinct and clear.

I do agree with intercst, though, that the weather analogy might better be in years rather than days.

db

 

[hocus]

There have always been two different sorts of discussions going on when this community and other Retire Early communities have taken up the topic of SWRs. There are the substance-oriented discussions that I refer to as The Great SWR Debate. And there are the process-oriented discussions that I refer to as The Debate About Having a Debate.

There is lots of evidence in the Post Archives that community members absolutely love the substance stuff. There is also lots of evidence that people absolutely hate the process stuff. The substance stuff has pulled all sorts of great posters into the various communities; JWR1945 is the most outstanding example. The process stuff has driven scores of fine posters away. There are a good number who have put up posts telling us that they were leaving. A far larger number just got up and left when they reached a point where it became clear in their minds that it was time to go.

I very much want these communities to thrive. Nothing has done more to cause them to thrive over the course of the past two years than The Great SWR Debate. For that I am grateful to every poster who has participated.

It pains me to see these communities dimished. Nothing has done more to diminish them over the course of the past two years than The Debate About Having a Debate. In regard to that I ask that other community members do what they can do to keep things at least somewhat reasonable when they see the potential of things spinning out of control.

There are not two positions on SWRs. There are probably 22. Wabmaster does not agree with *****, but Wabmaster does not agree with intercst either. Dory36 agrees with intercst more than he does with *****, but by no means does he agree with intercst on all points. JWR1945 shares *****' view that the conventional methodology is analytically invalid, but there are a number of points on which JWR1945 and ***** have exchanged dozens of e-mails trying to work through their differences.

This is wonderful. This is what you want to see on a discussion board--a healthy diversity of opinion on the most important question before the communty. Not only does this mean that you are going to see a lot of posts generated. It means that the posts generated are going to be on-topic and significant.

We can't lose when we engage ourselves in this sort of debate. There are three possible outcomes: (1) We reach a consensus that I was 100 percent wrong in everything I ever said re SWRs; (2) We reach a consensus that I was 100 percent right in everything I ever said re SWRs; or (3) We reach a consensus that I was right about some things I said about SWRs and wrong about others. All three are great outcomes for the comnmunity. Getting to any one of those three places requires that we learn together, and learning together is what this new internet discussion-board communications medium is all about.

My recommendation is that we all do what we can to highlight the wonderful aspects of this debate and to downplay the destructive aspects. We need to normalize discussions of SWRs. When people click on an SWR thread, we want them to have the same feeling they have when they click on any other sort of thread. We don't want people clicking with the half-expectation that they are going to see blood on the floor. It demeans us for us to allow too much blood on the floor and it does great harm to the efforts of those trying to build these boards into valuable information resources.

You could draw a chart showing where all of the various posters fall on the spectrum of SWR opinion possibilities. I am at one end of the chart. Intercst is at the other. Dory36 is somewhere in the middle. That's great. You've got long-time, well-known posters covering the three major spots on the spectrum.

I have learned a lot about SWRs over the course of the past two years. Dory36 has learned a lot too; I know because I can compare his posts from the old days to the ones from the post-May 13, 2002, era. It is my impression that lots and lots of people have learned. I asked intercst yesterday whether he has learned too. He didn't answer the question.

My belief is that he has learned, and my belief is that it was my decision to put my doubts about the conventional methodology on the table that caused him to learn. I am glad that I was able to teach intercst something because it is fair payback for earlier times when he taught me. I hope that intercst will answer my question in days to come. I hope that he will acknowledge that he has learned. I think that a response like that from him would do an awful lot to help normalize these discussions.

 

[John Galt]

SWR? Short Written Replies

John Galt

 

[wabmester]

Wabmaster does not agree with *****, but Wabmaster does not agree with intercst either.

*****, thanks for the mention. Maybe all three of us should write books on the subject. Since I tend to be terse, my book is likely to be very short. In fact, I'll include the entire contents below for review:

Hope for the best, plan for the worst, nimbly adapt to your environment, and if all else fails, withdraw 4% or less.

 

[Workwayless]

Re: Yahoo "Finance Quiz"What do you mean by specif

This post borders on the incredible. I see why you call yourself "WorkWayLess". Could you explain to me why anyone should work way more, to convince someone he doesn't even know to do something that will not be paid for and that will not in any way help the person who is being asked (and not very politely I may add) to do the work involved?

I would say do it yourself, or just continue in whatever course you have decided on. It will work or not, and you will be responsible.

Many things worth knowing require some effort.

Mikey,

Hmmm, you seem to be making assumptions about who I do or do not know, whether or not there is money involved in the payoff and even whether or not such research would benefit me personally.

Looking at the responses, I'm sure that you can see that given the huge amount of interest generated by my suggestion then perhaps the whole idea is a bit credible that you thought, eh Mikey?

 

[wabmester]

Alright, WWL, I can't speak for Mikey, but I'll admit you provided a valuable service by goading JWR1945. However, I also have to say that I find your boy-george avatar somewhat disturbing

 

[hocus]

I'll admit you provided a valuable service by goading JWR1945.  

I have been an active participant in these discussions from the first day. In fact, it was me who put up the post that kicked them off. I have seen zero evidence that JWR1945 has ever been in need of “goading” to do the work that needs to be done to advance our understanding of the realities of SWRs.

In this post,

http://boards.fool.com/Message.asp?mid=17229181

he was the first to respond to my request that the community look at the historical data to determine whether changes in valuation levels might have an effect on the question of which withdrawal rate is truly safe. The post is titled “***** is Really Onto Something." The date of the post is May 16, 2002, Day Four of the Great SWR Debate. This post obtained 44 recommendations, showing that there was a widespread community appreciation for JWR1945’s responsiveness in the early days of this debate.

JWR1945: "Maybe it makes sense to add to that cash cushion every now and then if we can define what “way overvalued” means. We do not need a precise definition. I doubt that that is possible. But we could use an accurate definition even if it is coarse. As long as we do not demand precision, such a definition must exist....I have begun looking at whether “way overvalued” might be meaningfully defined.

“...We should not conclude too much from these early results. But the fact that there were only three bad periods to start a retirement in the twentieth century indicates that you probably can vary allocations successfully as long as your actions are based on decade long variations and not one or two year changes. So far, it seems as if some changes in the percentage of stocks and in the type of cushion (commercial paper, TIPS, 5 year treasuries, long term treasuries) do make sense.”

Here is a JWR1945 post from May 18, 2002, Day Six. It is called “***** Started It All" (the lead sentence is “and I Am Having a Ball”). This one got 59 recommendations.

http://boards.fool.com/Message.asp?mid=17238256

JWR1945: "***** has brought up some really good issues. In a few cases these and related issues can and have been addressed directly. IMHO, however, the best responses are those that have generated new ideas and opened up whole new areas worth looking at. I personally appreciate all of them, even a few abrupt, dismissive responses. They reveal gaps in our knowledge. I regret that a few posters have been abusive. But I am pleased that ***** has stayed around. It has helped a lot.

"I have made a brief sensitivity analysis of the safe withdrawal number. Roughly speaking, a change of 0.1% changes the safe withdrawal period by a decade....***** is very much aware of this sensitivity. But before he brought it up on a recent post, I was not aware of it and I doubt that many you were either....***** has also been looking at the time history of failures in withdrawal rates, thresholds and periods. This is worth knowing.”

There is no statistically skilled poster in the Retire Early community more willing than JWR1945 to take on the work needed to help out those of us who do not possess statistical skills but who have questions that we would like those who possess statistical skills to expore in greater depth. He is a godsend to the community.