Author Topic: What the Market Swing Numbers May Actually Look Like in FIRE?  (Read 2346 times)

heybro

  • Stubble
  • **
  • Posts: 221
What the Market Swing Numbers May Actually Look Like in FIRE?
« on: January 05, 2020, 02:23:02 AM »
Can anyone post actual numbers of what a portfolio's value might look like throughout the years of FIRE?
These do not need to be your personal portfolio's real numbers.  I'm looking for the relationship between likely numbers more so.

For instance, if you are taking out 4% a year, we could say that is $4.00 and that you have $100 saved in the stock market at FIRE age.

Portfolio Value Each Year:
100
96
2000
10
60
10,000

etc.

I know for certain the numbers I just wrote out are completely wrong, but can someone put up likely scenarios (or the % swing of their actual numbers?).  It would help to visually see -in real numbers- just how far down a portfolio could go so that one can look at those numbers and not panic when the time comes.  Likewise, if a portfolio looks extremely high, it helps to know those actual numbers may be considered normal and thus one should actually NOT increase their spending even if that number looks very high. 
« Last Edit: January 05, 2020, 02:24:50 AM by heybro »

SwordGuy

  • Walrus Stache
  • *******
  • Posts: 8955
  • Location: Fayetteville, NC
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #1 on: January 05, 2020, 11:56:03 AM »
www.cfiresim.com has a good FI success rate calculator.  It graphs the market for each year for the last 130+ years given your starting assets.    Just click on each year's line and you'll get a good idea.



Mmm_Donuts

  • Bristles
  • ***
  • Posts: 410
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #2 on: January 05, 2020, 01:36:06 PM »
+1 to try cfiresim

You can also download a spreadsheet from there and analyze all the numbers, line by line, for each starting year of sample retirement throughout history. It's really helpful.

BTDretire

  • Magnum Stache
  • ******
  • Posts: 3074
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #3 on: January 05, 2020, 01:58:04 PM »
I could give you 10 years of portfolio growth, but I wasn't making withdrawals.
It was up Jan 1 of every year. As was almost everyone's, we are in the best of times!
 But let's face it, everything will be great until we have another 2008, then our portfolios will be down, some will reexamine, some will just shrug, some will cut back on spending, and some will get a job. And then a few years later things will be growing again.

John Galt incarnate!

  • Handlebar Stache
  • *****
  • Posts: 2038
  • Location: On Cloud Nine
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #4 on: January 05, 2020, 02:53:30 PM »

 But let's face it, everything will be great until we have another 2008, then our portfolios will be down, some will reexamine, some will just shrug, some will cut back on spending, and some will get a job. And then a few years later things will be growing again.

Right.

Stock-market historical performance and the eye-opening findings of behavioral finance align with your post.

businessinsider.com › what-you-can-learn-from-jp-morgan

Oct 17, 2016 -


In one famous incident, when asked by a passerby what the stock market was going to do next, Morgan responded simply: "It will fluctuate." Unfortunately, Morgan probably never quite said that — according to Barry Popik's research, the earliest appearance of the quote dates back to 1934, decades after the man's death.



Whoever said it was  right when they said the stock market "will fluctuate."

Eventually,  will the stock market zigzag downward or drop precipitously?

Yes.

And subsequently  zigzag upward  or undergo large, upside spurts that in short order lift it to new highs?

Yes, but no one can reliably predict  the timing, duration, and magnitude of stock-market upside or downside.
« Last Edit: January 05, 2020, 03:08:34 PM by John Galt incarnate! »

jeroly

  • Pencil Stache
  • ****
  • Posts: 606
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #5 on: January 05, 2020, 04:36:45 PM »
Anyone responding with actual numbers will give you a skewed perspective as the market has been very forgiving of high withdrawal rates etc. the past dozen or so years.

As noted above, you can see what your results would have been under whatever withdrawal scenarios you are interested in using cfiresim or other online calculators.

heybro

  • Stubble
  • **
  • Posts: 221
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #6 on: January 06, 2020, 03:12:36 AM »
Alright, I put in 1 million and said I'd spend $10,000 per year.
It tells me the lowest value for this portfolio is 1 million and the highest value is 12 million.
How the heck does a person decide not to spend more should their portfolio reach 12 million?

What I am asking is, what type of datum or data line should you use to compare wild swings against before you "react" to them.  At what point do you say, "I better go get a part time job" and at what point do you say "I better go on twelve trips this year to Disneyland."

I mean, what is a working system to keep yourself from over-reacting or over-correcting?

If you simply say, "I can spend 4% per year" do you take out 4% of it's current value.  4% of 12 million would be $480,000.  Should you take that out if that thing actually did 12 million?

DK

  • Stubble
  • **
  • Posts: 237
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #7 on: January 06, 2020, 05:47:49 AM »
I haven't completely fleshed out my plan, but I know what I was thinking tentatively given the example you put was something like:
- if current portfolio's 3% withdrawal amount is > initial 4% inflation adjusted withdrawal amount, start taking that instead
- if current portfolio is off stock market high >30% for >1 yr, cut back/get temp job

jeroly

  • Pencil Stache
  • ****
  • Posts: 606
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #8 on: January 06, 2020, 08:10:15 AM »
Alright, I put in 1 million and said I'd spend $10,000 per year.
It tells me the lowest value for this portfolio is 1 million and the highest value is 12 million.
So the simulator is telling you that not only does a 1% withdrawal rate have 100% success in past sequences of returns, but 100% of the time you would have ended up with at least as much as you started with (in nominal $$$)
Quote
How the heck does a person decide not to spend more should their portfolio reach 12 million?
By deciding that their goal is to maximize their estate to leave to children, charities, etc.
Quote
What I am asking is, what type of datum or data line should you use to compare wild swings against before you "react" to them.  At what point do you say, "I better go get a part time job" and at what point do you say "I better go on twelve trips this year to Disneyland."
I mean, what is a working system to keep yourself from over-reacting or over-correcting?
You did one simulation. Play with it some more and you may answer your own question.  This is a very personal issue and everybody’s thresholds are different.
Quote
If you simply say, "I can spend 4% per year" do you take out 4% of it's current value.  4% of 12 million would be $480,000.  Should you take that out if that thing actually did 12 million?
1.  The 4% “rule” was intended as a guideline to determine whether someone has enough assets to support an intended level of withdrawals.
In the classic Trinity study used to support the 4% rule, there are several considerations which tend to make it more conservative than is perhaps necessary, based on the historical data:
A.  The study assumed a 0.5% expense ratio, and most index funds charge far less.  So, all other things being equal, you should be able to use a 4.45% withdrawal rate in the current environment and get similar results.
B.  The study assumed that the retiree would blindly bump up their withdrawals to match inflation regardless of market results.  In the example you cite, the withdrawal when the account reached that $12 million would be the original $10,000 adjusted by however many years of inflation. It isis not a 4%-of-balances approach.  While that strategy will never fail, it could result in poverty-level-or-worse withdrawals if, for example, your accounts dwindle down to $10,000 and your formula tells you that you can only withdraw $400.

2.  So if you have some “fat” in your budget and have an ability to cut back when the market dives, you should be able to spend more in the good times. There are alternative formulas that can overcome some of this stuff.  The Bogleheads site has a Variable Percentage Withdrawal calculator that tells you to calculate your withdrawal, based on things like your age, your asset allocation, your pension/ social security amounts and effective dates, etc. 

heybro

  • Stubble
  • **
  • Posts: 221
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #9 on: January 08, 2020, 12:43:20 AM »
What a tricky little puzzle.  Awhile back of bunch of people posted that you should assume 500k in health expenses.  So there seems to be a huge potential for me to have way way way too much than I'll actually ever need.

vand

  • Handlebar Stache
  • *****
  • Posts: 2302
  • Location: UK
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #10 on: January 08, 2020, 04:10:14 AM »
Recent REs have known nothing but a rising market, and probably rising enough to more than offset any withdrawals that they would have made. They have literally been getting richer in spite of their spending.

It will be interesting to see how psychology changes when market conditions eventually shift and we get the first year or 2 when retirement portfolios get whacked by more than people have taken out in living expenses.

heybro

  • Stubble
  • **
  • Posts: 221
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #11 on: January 08, 2020, 03:17:28 PM »
Recent REs have known nothing but a rising market, and probably rising enough to more than offset any withdrawals that they would have made. They have literally been getting richer in spite of their spending.

It will be interesting to see how psychology changes when market conditions eventually shift and we get the first year or 2 when retirement portfolios get whacked by more than people have taken out in living expenses.

Right!  We have to be wise enough to only spend our little number no matter how high things go *as well as* to STILL spend our little number when our balances are going down.  It's a beautiful thing though if you understand how to do it.

Monkey Uncle

  • Handlebar Stache
  • *****
  • Posts: 1740
  • Location: West-by-god-Virginia
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #12 on: January 09, 2020, 06:50:12 AM »
Alright, I put in 1 million and said I'd spend $10,000 per year.
It tells me the lowest value for this portfolio is 1 million and the highest value is 12 million.
How the heck does a person decide not to spend more should their portfolio reach 12 million?

What I am asking is, what type of datum or data line should you use to compare wild swings against before you "react" to them.  At what point do you say, "I better go get a part time job" and at what point do you say "I better go on twelve trips this year to Disneyland."

I mean, what is a working system to keep yourself from over-reacting or over-correcting?

If you simply say, "I can spend 4% per year" do you take out 4% of it's current value.  4% of 12 million would be $480,000.  Should you take that out if that thing actually did 12 million?

I follow a version of the "retire again and again" strategy.  While the market is at or near a new high (like now), I run my numbers through cFiresim at the beginning of each calendar year and ask it to tell me the maximum amount I can spend with a 100% historical success rate.  I then deduct a 15% buffer (this is a personal fudge factor to account for who-knows-what might happen, either in my life or in the markets).  If the market goes down, I just use last year's spending number plus inflation.  This way, I get to take a raise in good years, I never have to take a cut, and I maintain a 100% historical success rate.

Last year this system gave me 46k to spend.  Because the markets were up this past year, I now get 48k to spend this year.  If the market goes up again this year, I'll recalculate a new, higher spending level next year.  If the market goes down, my spending amount next year will be 48k * (1+inflation rate).

rantk81

  • Pencil Stache
  • ****
  • Posts: 905
  • Age: 42
  • Location: Chicago
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #13 on: January 09, 2020, 12:36:20 PM »
Gah. I'm hostage to a OMY state of mind, of my own making.  Paid off home, 1.5M invested assets. Annual expenses about 50K/yr.  If I knew the ACA would last, I'd pull the trigger.  I think.

BostonBrit

  • 5 O'Clock Shadow
  • *
  • Posts: 63
Re: What the Market Swing Numbers May Actually Look Like in FIRE?
« Reply #14 on: January 09, 2020, 02:56:11 PM »
Not to be smart but lets just say it's 1/1/2000 when you FIRE (I'm sure someone out there thought that would be their perfect date).

$1m in SPXT (debate asset allocation).
$10k withdrawal then you'd of had ~$811k on 1/1/2010 and $2.66m on 1/1/2020 (this last decade has been insane). If you repeated '00-10 for the next decade then on 1/1/2030 you'd have $2.3m

If you ran the same numbers but withdrew 4% so $40k over the period with no flex.
You'd of had ~$492k on 1/1/2010 and $873k on 1/1/2020 (this last decade has been insane). If you repeated '00-10 for the next decade then on 1/1/2030 you'd have $376k

My view on the flexibility of the withdrawal rate comes down how much in absolute $$$ you expect to withdraw and is budgeted. It's obviously far more easy practically to take $10k of a $100k budget than a $25k budget.