SWR of 6.21% for 26 years

[John Galt]

Hey TH! Your jester hat has always been obvious to me.
BTW, Anna Nicole Smith is seriously F**ked up.
Future generations are not likely to be much better.

Keep smilin'

John Galt

 

[cute fuzzy bunny]

Oh I know she's a piece of work, but she'd just be an emergency backup wife, as I already have one and she's terrific.

Come to think of it, maybe I should tell her about this plan...

Ok, I just asked, and she said if I become as rich as the guy that married anna nicole smith, it would be ok for me to marry her granddaughter when I'm 90.

Isnt she the best?

 

[ataloss]

2) We are generally in agreement that the number generated by the conventional methodology studies does not provide true safety in the event that a worst-case scenario pops up in one's retirement. There have been a good number of posters who have said that those using the 4 percent number or the 6 percent number need to include a good bit of slack in their plans to cover themselves in the event of a worst-case scenario. By definition, the SWR number is the number that works in the worst case, so not much slack should be necessary if the number were being calculated properly.

I want to know the maximum that I can withdraw so that I don't spend too little and leave a big estate or spend too much and run out of money so I was very happy to see this.... but then I saw this:

The approach that I take with the data-based SWR concept is to include a caveat that the number works if the future is not worse than the past. By the phrase "if the future is not worse than the past," I mean to cover the possibility of economic calamities like hyperinflation or nuclear war or political instability,<b> anything so bad that it could cause the U.S. stock market to perform worse than you would otherwise expect to see it perform given how it has performed in the past.</b> I do not include within the caveat the possibility that we will see lower returns as the result of starting from higher valuation levels because that it is a result you would expect given that higher valuation levels have always caused lower long-term returns in the past.

IOW, the "data based swr" would be like the regular (?non data based swr) except that it would consider valuation. I am not sure that I can imagine let alone list all of the events that might cause lower performance of the market than in the past

I understand that there is no number that will work with 100 percent certainty. I am not sure that it is such a big deal not to cover the possibility of things like nuclear war, however. In the event of nuclear war or hyperinflation, all bets are probably off anyway. There simply is no way to effectively account for that sort of thing in your investment strategies.

here I can think of many other possible calamaties (other than high p/e ratio) and I assume that there are more than I can imagine and I think that it would be a "big deal" to include all of them so I am struggling to see how uncertainty has been reduced....

The more precise way to say things is to say "the conventional methodology is analytically invalid for purposes of determining SWRs."

?maybe this means something to you swr afficianados

I have trouble reconciling these two statements:

It is a matter of different strokes for different folks, not a matter of declaring some folks right and other folks wrong.

That said, the conventional methodology number might work for retirements that do not experience a worst-case scenario. In any event, some retirees may reasonably view the true SWR number as overly conservative and may elect to employ the conventional methodology number in their plans.

if there is a true swr number aren't the people using the "other" number wrong?

I have failed to understand ***** on several other boards in the past so I guess I am not learning

 

[John Galt]

There is no absolute SWR, just like there is no Santa Claus. For those who have to get 100% comfortable first,
SWR might as well be STILL WISHING for
RETIREMENT. It's fun to debate though.

John Galt

 

[hocus]

if there is a true swr number aren't the people using the "other" number wrong?

The conventional methodology tells you the true historical surviving withdrawal rate (HSWR) and the data-based methodology tells you the true safe withdrawal rate (SWR). Which number is right or wrong in a given circumstance depends on the purpose for which you are making use of an analytical tool to inform your investment strategies in that particular circumstance.

That's my take on it, ataloss.

 

[arrete]

The conventional methodology tells you the true historical surviving withdrawal rate (HSWR) and the data-based methodology tells you the true safe withdrawal rate (SWR).

I think the word "true" in this quote is terribly misleading. Some withdrawal rates have been determined from past data, but they are only based on that data. Others are determined from Monte Carlo studies and so on. There isn't anything about any of these that makes one of them "true" in the sense that the other SWR are "false" SWR. People look at the data and that's what they come up with.

I personally take SWR with a grain of salt - a guideline, since no one can predict the future. One SWR for everyone under any circumstance makes even less sense.

arrete

 

[Lancelot]

Why do something that sounds harder than work? The other day someone posted that he was sitting in Thailand by a fan in 100 degree weather. Wouldn't most people rather get a job?

Oh, I dunno, I like my fan alot Sure, I have a/c, but I've acclimated to the hot weather. Maybe its not for everyone, but it works for me. I really have no problem staying busy: lots of reading, swimming and some traveling around Thailand. Time seems to go by pretty fast.

If I have a job, I'm back to spending my time according to someone else's dictates. Naaa, no thanks.

Still chilling by the fan

 

[hocus]

I personally take SWR with a grain of salt - a guideline, since no one can predict the future. One SWR for everyone under any circumstance makes even less sense.

I don't think there is too much difference in our viewpoints as a practical matter, arrete. I certainly do not believe that all aspiring early retirees should be using the same withdrawal rate. The SWR tool is a mathematical construct, so I believe that the calculation should be done objectively. But I also believe that each aspiring early retiree should be subjectively determining his or her own personal withdrawal rate (PWR).

At the calculation stage, you just report what the data with a bearing on the question of what is safe says; there are limits on what reasonable people should come up with at that stage. But at the application stage, it is perfectly appropriate for optimists to employ a higher number than pessimists. Optimists might take SWR Plus One as their PWR and pessimists might take SWR Minus One. The optimists know that it is always possible that a worst-case scenario will not pop up in one's particular retirement, and the pessimists know that it is always possible that we will see an economic event in the future worse than any we have seen in the past.

SWR analysis is an exercise in data collection and mathematics at the calculation stage. It is an exercise in the use of subjective judgment and the assessment of personal risk-tolerance at the application stage.

 

[Hyperborea]

I personally take SWR with a grain of salt - a guideline, since no one can predict the future.

Yup, the future is essentially unknowable so the best we can do is use history as a guide. Whether that means using straight historical data (intercst, Trinity, etc.) or extracting the statistical properties of the data (Monte Carlo) all of them use historical data. It is entirely possible that the future will be worse than the past and it is also possible that the historical asset balance won't work in the future as it did in the past (i.e. loss of US dominance, maturing of the US economy and slower growth).

It is extremely prudent for any early retiree to have some fallbacks in their plans such as being willing and able to scale back withdrawals, being more diversified than the "standard" portfolio, a house that can be downsized or reverse mortgaged, etc.

 

[comixfan]

If you retire now (at S&P around 1100), your SWR should be about 6.21%, given the assumption that the 30 yr period 2000-2029 would not be worse than the worst 30 year period from the past.

Reason:  If you had $1000000 at the beginning of 2000, FIRECalc will tell you that your SWR for 30yrs is 4.19%, using the defaults.  If you stick to 4.19% withdrawal, at the end of 2003, your nest egg will be about $674000.  Yet, you should be OK to continue withdrawing $41900, plus inflation adjustment til 20209.  Hence, if you had $1mil at the end of 2003, then you should be able to withdraw $41900*1000000/674000=$62166, or about 6.22% annually, inflation adjusted.  Of course, if you  believe that 2000-2029 period will be worse than the worst we've ever experienced then, your personal SWR will be less than that.

Ok, I read about 7 or 8 pages and then stopped so I could collect my thoughts and respond. If anyone has already made this point then I take it all back!

Here's the key to the puzzler above: the calculator assumes that you're going to withdraw the same amount every year (after adjusting for inflation). This means that some years, you're potentially going to be taking out more than the initial % in year 1 and some years you're going to be withdrawing less than the initial % than you took out in year 1.

The flaw in the logic as explained by amt is when you're taking a % of the CURRENT VALUE of the portfolio rather than the initial value of portfolio. 6.22% this year, could be 2.5% next year and 8% the year after that. That's the beauty of SWR in the first place when stacked up against historical data.

In other words, amt is correct in saying 6.22% can be withdrawn for the next 26 years, but wrong in comparing the % withdrawal to the CURRENT VALUE of the portfolio. AND this is only if you go by the SWR model at all.

Please correct me if my logic went south.

 

[JWR1945]

In other words, amt is correct in saying 6.22% can be withdrawn for the next 26 years, but wrong in comparing the % withdrawal to the CURRENT VALUE of the portfolio. AND this is only if you go by the SWR model at all.

Please correct me if my logic went south.

amt is saying that 6.22% (plus inflation) is totally safe for any portfolio with 26 years to go, not just the current one. [IIRC, amt was talking about 29 years fo go. Don't worry about that.]

This is an artifact of the assertion that the original 4% [or 3.9%?] withdrawal rate is truly 100% safe. It was safe for 100% of the 30-year historical sequences. But that does not guarantee total safety looking forward. Instead, it is a good idea to introduce probability and statistics into our analysis.

Doing so automatically rejects the notion that any withdrawal rate (other than zero) can be 100% safe.

Have fun.

John R.

 

[comixfan]

amt is saying that 6.22% (plus inflation) is totally safe for any portfolio with 26 years to go, not just the current one. [IIRC, amt was talking about 29 years fo go. Don't worry about that.]

This is an artifact of the assertion that the original 4% [or 3.9%?] withdrawal rate is truly 100% safe. It was safe for 100% of the 30-year historical sequences. But that does not guarantee total safety looking forward. Instead, it is a good idea to introduce probability and statistics into our analysis.

Doing so automatically rejects the notion that any withdrawal rate (other than zero) can be 100% safe.

Have fun.

John R.

Well, nothing in life is 100% safe, especially not future performance based on past results. What I'm saying is that firecalc bases withdrawals amounts on the INITIAL VALUE of the portfolio, not ongoing values. So in year 10, if the value is less than the initial value, the % withdrawn may be higher, but it doesn't make the SWR any more or less valid.

Bottom line, either go with the SWR based on historical performance or not. Or better yet, do as some do and take the SWR based on current value, WHICH IS NOT what firecalc does!

I actually am wondering how that much that changes the outcome. I'm betting that if you put a cap on the withdrawal amount and then decreased it in down years, your probability of success would increase tremendously. That's all this is, probability, with that 100% chance scenario in firecalc being a VERY deceiving concept, since, of course, there is no guarantee that you'll have any funds left after x number of years with a 4% SWR.

 

[Hyperborea]

That's all this is, probability, with that 100% chance scenario in firecalc being a VERY deceiving concept, since, of course, there is no guarantee that you'll have any funds left after x number of years with a 4% SWR.

It's not "deceiving".  What it may be is a mismatch of expectations.  The conditions that FIRECalc uses for success is that you can make every withdrawal over the time period specified using historical data.  There are studies out there that add to the success conditions.  Some of these are things like nominal portfolio value maintained at end of period or inflation adjusted value maintained at end of period.  You can make your definition of success anything that you want but the more conditions that you need then it is likely that the SWR will be lower.

It will also mean that the results become less "representative" as you slice the historical data set into smaller and smaller piles - we've only got somewhere between ~4 and 100 30 year data sets depending on how you count the overlap.  With the size of the current data at some point you have to add some other "things" to the mix.  That can be logic or reasoning though you have to be careful with that as there are very few things in financial theory that aren't also using the same limited data set.  One other thing that can be added is feedback or adaptation in the withdrawal rate.  This doesn't require as exact an understanding of the "one true blessed" SWR as these algorithms will adjust to the actual portfolio results.  There are a number of proposals floating around about this - everything from drawing a fixed percentage of the current value of the portfolio to more complicated base amounts plus percentages of the gains to just adjusting by the "seat of the pants".

 

[charlie]

Hyper pretty well nailed it.

IMHO, one should just take what they need
to maintain a comfortable but LYBM lifestyle.

SWR calculations are just a sanity check at
best.

Cheers,

Charlie

 

[Mr._johngalt]

I enjoy fiddling with the SWR numbers, even if the
endless debate gets a bit tedious. Charles'
"sanity check" point is well taken. What I do is try to imagine all of the worst-case scenarios that could possibly befall us and then have some idea how we
would deal with them. And here (I apologize as I
repeat myself).................if I lost 100% of my fixed income
portfolio (that's half of our net worth), I still would not have to go back to work. We would sell/lease one residence,
reverse mortgage the other one and collect our SS.
It wouldn't be pretty but it wouldn't derail the ER train.

JG

 

[Martha]

John, I know your circumstances and that they are different than mine, but I found your last post very comforting. Somehow I have this basic insecurity that we will "lose it all" and will be in no position to go back to work.

 

[unclemick]

Running out of retirement funds and having the health to enjoy it are the two emotional touchstones - probably for all retiree's. They are always in the background - I won't use JG's Al Pacino quote - but managing risk is part of life - easy to say - so each ER has to come to his/her own version.

 

[Eagle43]

I like JG's attitude. To philosophize, right attitude will trump bad circumstances most of the time and bad attitude can defeat good circumstances or at least keep you from enjoying your good fortune.

That said, one of the reasons I stayed at MegaCorp was to get that other pension; one of the reasons I stayed in military was to cover healthcare. I remember a friend of the family during the 50s lose everything to the doctor bills, and he was a hard-working individual. I am an exercise fanatic and although that's no guarantee, it doesn't hurt.

I've lost in the Index Funds this year, but can remain where I am because of the pensions, healthcare, and a good supply of cash. If all of that goes, then JG's attitude and maybe even his prescriptions are the way to go. And to paraphrase Martha, I'll go back to work if necessary. I worked for 43 years, so, if necessary another couple should be easy.