Variable Percentage Withdrawal (VPW)

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
longinvest
Posts: 5672
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest »

Pappy205 wrote: Sun Sep 25, 2022 1:00 pm Quick question - I am setting up the worksheet and I can’t seem to eliminate the work pension (amount $1,000)… I do not have a work pension. Any help on how to fix this? Many thanks!
Pappy205, try removing the name of the pension. All entries of Defined Benefit Pension #2 (cells E17 to E21) must be empty to remove it from calculations.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Pappy205
Posts: 2
Joined: Sun Sep 25, 2022 12:20 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Pappy205 »

Thank you, longinvest!
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Sat Sep 24, 2022 10:44 am
nigel_ht wrote: Fri Sep 23, 2022 11:20 am Skimming this thread I notice this question...isn't there both downturn AND inflation effects...in July 2019 the required flexibility was to be able to drop income to $61,286.
I don't believe percentage-based methods account for inflation. If you need 50K/year in today's dollars, simply prepare enough for 100K/year to provide 50% of flexibility, regardless of inflation. And hopefully market returns would be good enough to fund your retirement.

Personally I don't think you need 50% of flexibility, something like 20~30% should do.
I don’t know but in the forward test thread the required flexibility increased $5K in two years…I don’t think folks’ expenses dropped that much in real terms.

So is required flexibility only relevant in the first year or if you recalculate every period like in the forward test?

And what should you do if you no longer pass the flexibility test?
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Sun Sep 25, 2022 4:28 pm I don’t know but in the forward test thread the required flexibility increased $5K in two years…I don’t think folks’ expenses dropped that much in real terms.

So is required flexibility only relevant in the first year or if you recalculate every period like in the forward test?

And what should you do if you no longer pass the flexibility test?
You bring up excellent points. If I calculate the required flexibility at the time of retirement, I'd imagine I have to keep adjusting it for inflation, perhaps annually, since that's essentially your minimum level of spending required in retirement.

If you don't pass the flexibility test, you'd have to go back to work - at least that should be the implication, not sure how else to interpret the situation.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Sun Sep 25, 2022 4:43 pm
nigel_ht wrote: Sun Sep 25, 2022 4:28 pm I don’t know but in the forward test thread the required flexibility increased $5K in two years…I don’t think folks’ expenses dropped that much in real terms.

So is required flexibility only relevant in the first year or if you recalculate every period like in the forward test?

And what should you do if you no longer pass the flexibility test?
You bring up excellent points. If I calculate the required flexibility at the time of retirement, I'd imagine I have to keep adjusting it for inflation, perhaps annually, since that's essentially your minimum level of spending required in retirement.

If you don't pass the flexibility test, you'd have to go back to work - at least that should be the implication, not sure how else to interpret the situation.
I don’t think you necessarily have to back to work…there’s this fuzzy region between “required flexibility” which I believe is defined as no cuts needed to maintain a “comfortable lifestyle” and “the amount I actually require before I’m not just cutting discretionary but actual needs”.

At some point that fuzzy band disappears but since I don’t believe that actual needs (aka spending floor) are tracked in the spreadsheet I don’t think you know what happens unless you use a calculator that lets you set a floor.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Sun Sep 25, 2022 5:01 pm I don’t think you necessarily have to back to work…there’s this fuzzy region between “required flexibility” which I believe is defined as no cuts needed to maintain a “comfortable lifestyle” and “the amount I actually require before I’m not just cutting discretionary but actual needs”.

At some point that fuzzy band disappears but since I don’t believe that actual needs (aka spending floor) are tracked in the spreadsheet I don’t think you know what happens unless you use a calculator that lets you set a floor.
I'm confused now. If you read this piece: viewtopic.php?p=4623315#p4623315
The retiree is fully aware that, within the spending budget, at least $1,250/month must be reserved for totally optional expenses that could easily be cut at any point.
And this reads like, if your budget reduction is greater than 1250/mo then you have to start cutting your actual needs, no? And this is the situation we're talking about here when you use up your required flexibility.

I also haven't studied deep enough to understand where -300K on 1M came from, as it seems like 30% not 50%.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Sun Sep 25, 2022 5:57 pm
nigel_ht wrote: Sun Sep 25, 2022 5:01 pm I don’t think you necessarily have to back to work…there’s this fuzzy region between “required flexibility” which I believe is defined as no cuts needed to maintain a “comfortable lifestyle” and “the amount I actually require before I’m not just cutting discretionary but actual needs”.

At some point that fuzzy band disappears but since I don’t believe that actual needs (aka spending floor) are tracked in the spreadsheet I don’t think you know what happens unless you use a calculator that lets you set a floor.
I'm confused now. If you read this piece: viewtopic.php?p=4623315#p4623315
The retiree is fully aware that, within the spending budget, at least $1,250/month must be reserved for totally optional expenses that could easily be cut at any point.
And this reads like, if your budget reduction is greater than 1250/mo then you have to start cutting your actual needs, no? And this is the situation we're talking about here when you use up your required flexibility.

I also haven't studied deep enough to understand where -300K on 1M came from, as it seems like 30% not 50%.
Presumably longinvest will clarify.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

longinvest wrote: Wed Aug 17, 2022 7:01 pm While our hypothetical retiree is enjoying a comfortable retirement in the forward test, the race to the lowest possible SWR rages on! This shouldn't be a surprise; those who stand to make a profit from bigger portfolios or from selling retirement solutions love SWR because fear sells.++
Heh…missed this post…

SWR hasn’t changed because we haven’t seen an outcome worse than 1929 or 1966 yet.

Morningstar may be fear mongering but SWR itself hasn’t changed because it’s based on the historical worst case.
SWR is down to 1.9% according to a recent study.

Here's what I think. Maybe the sky will fall during retirement. Maybe not. Nobody knows. The difference between VPW and SWR is that VPW won't cut expenses before it needs to do so, and if it does, it will be due to a severe crisis where others in society have to cut their expenses, too.^^ SWR, on the other hand, will let the retiree severely underspend all retirement long and die with a gigantic unspent portfolio just in case something bad was to happen in the future, yet if things turn bad, it can leave the retiree penniless. SWR almost guarantees an undesirable outcome (underspend or overspend), and thus triggers fear.
Nope.

In fact for the forward test scenario SWR has caught up to and passed VPW in annual spend if you carve out a 10 year go-go years budget to front load spending.

SWR + Go-go initially was $6,000/month in 2019 dollars.
VPW initially was $6,356/month in 2019 dollars

SWR + Go-go currently is still $6,000/month in 2019 dollars ($6926 in 2022 dollars).
VPW currently is $5,940/month in 2019 dollars ($6,857 in 2022 dollars).

So VPW has already cut expenses in the forward test to below what SWR can produce…and the required flexibility target is $5K worse than it was in 2019.

And the SWR+Gogo budget has a larger remaining portfolio than VPW today so less fear of running out…

Calculations here:

viewtopic.php?p=6884072
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Sun Sep 25, 2022 6:04 pm Presumably longinvest will clarify.
I figured out why it was -30%; they assume -50% on 100/0; since the modeled portfolio is 60/40, it's -30%.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Sun Sep 25, 2022 6:50 pm
nigel_ht wrote: Sun Sep 25, 2022 6:04 pm Presumably longinvest will clarify.
I figured out why it was -30%; they assume -50% on 100/0; since the modeled portfolio is 60/40, it's -30%.
Yes, but the interesting thing is what you noticed above: the required flexibility shifts in lockstep with portfolio value...to the point that eventually it could cut into actual needs and not flexibility.

The spreadsheet doesn't necessarily warn you of this because the required flexibility number get recomputed...
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Mon Sep 26, 2022 9:34 am Yes, but the interesting thing is what you noticed above: the required flexibility shifts in lockstep with portfolio value...to the point that eventually it could cut into actual needs and not flexibility.

The spreadsheet doesn't necessarily warn you of this because the required flexibility number get recomputed...
Right. As you mentioned, the "required flexibility" business has to be calculated once when you retire, then has to be COLA'ed from that point on.
justgary
Posts: 37
Joined: Sat Jun 27, 2020 12:24 pm

Re: Variable Percentage Withdrawal (VPW)

Post by justgary »

nigel_ht wrote: Sun Sep 25, 2022 6:27 pm So VPW has already cut expenses in the forward test to below what SWR can produce…and the required flexibility target is $5K worse than it was in 2019.
The VPW forward test is working its way through a situation where the market has, since the test started in 2019 (using the DJIA for calculations):
Gained 10%
Lost 30%
Gained 80%
Lost 20% and still going down.

The five month smoothing filter has done a great job of keeping the retiree's monthly income from varying wildly. The required flexibility target is set in the spreadsheet as a 50% loss in securities value. Both times that the market dropped during the test, the required flexibility target dipped with it. The question is, once the market dips 20% to 30%, what is the probability that the market will then drop another 50%? In other words, is the 50% loss predictor a good indicator of what might happen next?

To me, the answer is that I don't care what the odds are, I don't want to run out of money before my wife and I die (we are both expecting to live well into our 90s, so we are financially planning for 100). Having a gloom-and-doom worst case indicator staring at me is exactly what I need to make sure that I quit buying tomahawk ribeyes and start buying chopped beef steaks instead. I would rather take a short-term punishment while the market is down so that I can live it up again when the market returns. It is always possible that some big event happens and the market will never return in my lifetime. This has to be more realistic than simply declaring in an ad-hoc manner that I'm going to live it up for five years no matter what the market does, then check to see if I can stay retired.

Obviously, we could study this and come up with a better required flexibility indicator. I'm OK with leaving it alone and knowing that the deeper the market tanks, the more pressure it has to stop tanking and turn around. Nothing is guaranteed, though, so watching the indicator that portends complete doom seems prudent while the market is heading down.

I would also like to admit that I take the VPW recommendation as a "that would be great" suggestion since we are putting off our big "go-go" spending until my only pension (Social Security) kicks in eight years from now. We have always lived within our means, so our VPW worksheet exceeds what we need to live. We'll come closer to the VPW recommendation later as we really live it up.
Last edited by justgary on Mon Sep 26, 2022 9:58 am, edited 1 time in total.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Mon Sep 26, 2022 9:43 am
nigel_ht wrote: Mon Sep 26, 2022 9:34 am Yes, but the interesting thing is what you noticed above: the required flexibility shifts in lockstep with portfolio value...to the point that eventually it could cut into actual needs and not flexibility.

The spreadsheet doesn't necessarily warn you of this because the required flexibility number get recomputed...
Right. As you mentioned, the "required flexibility" business has to be calculated once when you retire, then has to be COLA'ed from that point on.
I don't think longinvest wants that because then you have a hard spending floor.

That does beg the question of whether the "required flexibility" test actually works...because while it might give you an idea about whether you have the required flexibility for a 50% market drop I don't think it tells you anything about whether you have the "required flexibility" needed for high(er) inflation like the 1966 scenario.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Mon Sep 26, 2022 9:55 am I don't think longinvest wants that because then you have a hard spending floor.

That does beg the question of whether the "required flexibility" test actually works...because while it might give you an idea about whether you have the required flexibility for a 50% market drop I don't think it tells you anything about whether you have the "required flexibility" needed for high(er) inflation like the 1966 scenario.
But you do have a hard spending floor. A spending floor is necessarily hard, otherwise it's not a floor.

When you start the forward test on Sun Jun 30, 2019 using 1M running 60/40 and if you say this retiree expects to lose 30% (1M - 300K = 700K portfolio and $49,286 after loss), the $49,286 has to stick around.

Otherwise, what's the point? When this portfolio crashes to 500K, do you recalculate -30% and say this retiree, who initially thought $49,286 was the floor, is somehow OK with $25,000 now?
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

justgary wrote: Mon Sep 26, 2022 9:54 am
nigel_ht wrote: Sun Sep 25, 2022 6:27 pm So VPW has already cut expenses in the forward test to below what SWR can produce…and the required flexibility target is $5K worse than it was in 2019.
The VPW forward test is working its way through a situation where the market has, since the test started in 2019 (using the DJIA for calculations):
Gained 10%
Lost 30%
Gained 80%
Lost 20% and still going down.
The question is that in just three years that required flexibility test has gotten around $5K worse. The 2022 downturn is still pretty mild...
The five month smoothing filter has done a great job of keeping the retiree's monthly income from varying wildly. The required flexibility target is set in the spreadsheet as a 50% loss in securities value. Both times that the market dropped during the test, the required flexibility target dipped with it. The question is, once the market dips 20% to 30%, what is the probability that the market will then drop another 50%? In other words, is the 50% loss predictor a good indicator of what might happen next?
The historical worse case dropped to -89% so the probability is low but possible. I think one of the mantras around here is that the market can drop 50% at any time, even after a pretty big drop. I believe in this thread the position is that you should always be ready to be flexible...so I guess the position is that the new flexibility test number is the relevant one and not the one you started with.
To me, the answer is that I don't care what the odds are, I don't want to run out of money before my wife and I die (we are both expecting to live well into our 90s, so we are financially planning for 100). Having a gloom-and-doom worst case indicator staring at me is exactly what I need to make sure that quit buying tomahawk ribeyes and start buying chopped beef steaks instead. I would rather take a short-term punishment while the market is down so that I can live it up again when the market returns. It is always possible that some big event happens and the market will never return in my lifetime. This has to be more realistic than simply declaring in an ad-hoc manner that I'm going to live it up for five years no matter what the market does, then check to see if I can stay retired.
Not running out of money doesn't matter if your portfolio can no longer support your minimum expenses.

SWR + Social Security never runs out of money either because of Social Security.
Obviously, we could study this and come up with a better required flexibility indicator. I'm OK with leaving it alone and knowing that the deeper the market tanks, the more pressure it has to stop tanking and turn around. Nothing is guaranteed, though, so watching the indicator that portends complete doom seems prudent while the market is heading down.
Inflation has been the driving factor of historical portfolio failure...
I would also like to admit that I take the VPW recommendation as a "that would be great" suggestion since we are putting off our big "go-go" spending until my only pension (Social Security) kicks in eight years from now. We have always lived within our means, so our VPW worksheet exceeds what we need to live. We'll come closer to the VPW recommendation later as we really live it up.
One of the reasons I carved out a Go-Go years budget that ignores market conditions is that the probability is high that I run out of life before I run out of money. 8 years is a long time to wait on the off chance that today is 1966 or 1929. Plus I know I can get by on the inflation adjusted SWR value + Social Security after my 10 years of Go-go money runs out.

After 5 years I can assess where we are and have an idea if SORR really did bite me or not. If it did I can decide how to spend that last $100K of my Go-Go years budget...
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

justgary wrote: Mon Sep 26, 2022 9:54 am The required flexibility target is set in the spreadsheet as a 50% loss in securities value. Both times that the market dropped during the test, the required flexibility target dipped with it. The question is, once the market dips 20% to 30%, what is the probability that the market will then drop another 50%?
If the required flexibility target keeps shifting, it's the same as not having one.

Think about it, required flexibility is a level you can't compromise. When you embark on VPW with 75K/year of spending and 50K/year would be your spending floor, that 50K/year level would have to stay or you can't finance your essentials.

If the market crashes and VPW only withdraws 45K/year, it is nonsensical to recalculate "flexibility" and argue that the same retiree is fine to live on 30K/year. They still need 50K/year (and whether the 50K/year should be COLA'ed or not is debatable).
User avatar
canadianbacon
Posts: 676
Joined: Sun Nov 10, 2019 9:04 pm

Re: Variable Percentage Withdrawal (VPW)

Post by canadianbacon »

Some thoughts from longinvest earlier in this (long) thread, and then my own.
What's important to pay close attention to is the Required Flexibility section. The portfolio loss value (cell B17), the portfolio balance after loss (cell B18), and the portfolio withdrawal reduction (cell B19) are critical values. These are estimates of what could easily happen given the portfolio's asset allocation. It's important for the retiree to have sufficient flexibility in the budget to easily deal (if necessary, in the future) with the calculated income reduction (cell B19).
As I said in my previous post, one of the most important things is to maintain, at all times, the required flexibility suggested by the VPW Retirement Worksheet. In other words, a big enough part of the spending budget must be budgeted for optional discretionary spending that could be eliminated without affecting the retiree's comfort. That's a comfortable spending budget after the reduction. It's not a bare-bones budget after the cut. If that's not possible, it's probably too early to retire.
Both the suggested portfolio withdrawal amount and the required flexibility go hand in hand. A -50% stock loss shouldn't be a cause for reducing comfort (just a reduction of optional discretionary spending). If the post-loss portfolio isn't sufficient for great comfort, one isn't yet ready to retire.
To me, the function of the Required Flexibility section is as a thought exercise to help people not prematurely pull the plug on saving, and longinvest's quotes appear to back that up. I think there is potential for someone to fill in the worksheet for the first time, and say "I currently spend $50000 a year, and this spreadsheet says I can withdraw $64000 next year, so I don't have to work anymore." The required flexibility section is a reminder that the portfolio income is variable (thus the name) and that you need to realize what is, frankly over a long enough retirement period, fairly likely to take place at some point. In the example above, the person should not be retiring immediately, but should be making sure they have savings that allow for a $50000 spend after loss.

You could perform the same thought exercise in other ways, too, like using the Backtesting spreadsheet and entering 1966 as the starting year and seeing how you enjoy the sixteen years that follow. This spreadsheet shows both nominal and real values by year, and in my opinion does a better job of showing potential long term impacts of a bad period in the markets.

longinvest has never said that the required flexibility amount is a floor. Someone implementing their own drawdown plan is welcome to add a floor, and I have considered one myself, but it's no longer really his system at that point. If stocks fall 50% and you are now looking at a new withdrawal amount and a new required flexibility amount, I think it makes sense to still consider both. Yes, it's likely that the new required flexibility amount no longer satisfies the definition of "comfortable" -- if it had, the original required flexibility amount could consider a 75% market drop, right? But I think you should still look at that figure, however much lower it is than the current spend, and try to form some kind of contingency plan for how to handle it, like what to cut, or whether to sell the stamp collection to preserve the portfolio, or whatever. Hopefully such a plan never has to be executed, but you're happy you had it if you need it.
Bulls make money, bears make money, pigs get slaughtered.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

canadianbacon wrote: Mon Sep 26, 2022 4:40 pm I think there is potential for someone to fill in the worksheet for the first time, and say "I currently spend $50000 a year, and this spreadsheet says I can withdraw $64000 next year, so I don't have to work anymore." The required flexibility section is a reminder that the portfolio income is variable (thus the name) and that you need to realize what is, frankly over a long enough retirement period, fairly likely to take place at some point. In the example above, the person should not be retiring immediately, but should be making sure they have savings that allow for a $50000 spend after loss.
I'm not sure what the issue is.

If someone's essentials are covered by 50K/year, they include 30% discretionary (flexibility) (71K/year), and divide that by the percentage VPW says (say, 5%) and retire on a 1.4M portfolio running 60/40 @ 5%...I thought this is exactly how you'd retire with VPW.

And 50K/year would be your floor because that's the level you can't breach to cover your essentials (and it's not even COLA'ed so more reasons you can't breach the floor).
Wrench
Posts: 1046
Joined: Sun Apr 28, 2019 10:21 am

Re: Variable Percentage Withdrawal (VPW)

Post by Wrench »

Marseille07 wrote: Mon Sep 26, 2022 5:02 pm
canadianbacon wrote: Mon Sep 26, 2022 4:40 pm I think there is potential for someone to fill in the worksheet for the first time, and say "I currently spend $50000 a year, and this spreadsheet says I can withdraw $64000 next year, so I don't have to work anymore." The required flexibility section is a reminder that the portfolio income is variable (thus the name) and that you need to realize what is, frankly over a long enough retirement period, fairly likely to take place at some point. In the example above, the person should not be retiring immediately, but should be making sure they have savings that allow for a $50000 spend after loss.
I'm not sure what the issue is.

If someone's essentials are covered by 50K/year, they include 30% discretionary (flexibility) (71K/year), and divide that by the percentage VPW says (say, 5%) and retire on a 1.4M portfolio running 60/40 @ 5%...I thought this is exactly how you'd retire with VPW.

And 50K/year would be your floor because that's the level you can't breach to cover your essentials (and it's not even COLA'ed so more reasons you can't breach the floor).
I may be out in left field here, but the way I would use VPW (I'm not there yet), is that the recommended withdrawal would be entirely discretionary. My floor will come from social security, an inflation adjusted SPIA, and a stable value fund in a 403B. Those will cover >100% of my expected expenses. VPW gives me a number of how much to spend beyond my floor or base income.

Wrench
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Wrench wrote: Mon Sep 26, 2022 6:10 pm
Marseille07 wrote: Mon Sep 26, 2022 5:02 pm
canadianbacon wrote: Mon Sep 26, 2022 4:40 pm I think there is potential for someone to fill in the worksheet for the first time, and say "I currently spend $50000 a year, and this spreadsheet says I can withdraw $64000 next year, so I don't have to work anymore." The required flexibility section is a reminder that the portfolio income is variable (thus the name) and that you need to realize what is, frankly over a long enough retirement period, fairly likely to take place at some point. In the example above, the person should not be retiring immediately, but should be making sure they have savings that allow for a $50000 spend after loss.
I'm not sure what the issue is.

If someone's essentials are covered by 50K/year, they include 30% discretionary (flexibility) (71K/year), and divide that by the percentage VPW says (say, 5%) and retire on a 1.4M portfolio running 60/40 @ 5%...I thought this is exactly how you'd retire with VPW.

And 50K/year would be your floor because that's the level you can't breach to cover your essentials (and it's not even COLA'ed so more reasons you can't breach the floor).
I may be out in left field here, but the way I would use VPW (I'm not there yet), is that the recommended withdrawal would be entirely discretionary. My floor will come from social security, an inflation adjusted SPIA, and a stable value fund in a 403B. Those will cover >100% of my expected expenses. VPW gives me a number of how much to spend beyond my floor or base income.

Wrench
I would assume that the forward test is the expected way VPW would most likely get used and even there the bulk of the expenses is covered by the portfolio rather than pension in the early years.

If I recall correctly pension + SS is $3K a month or $36K which is likely lower than the unstated minimum expenses.

I don’t believe there are SPIA with full inflation adjusted COLA. A couple years ago 2% was what folks were saying you could get.

If you have a 2% SPIA you probably aren’t terribly happy at the moment…
Wrench
Posts: 1046
Joined: Sun Apr 28, 2019 10:21 am

Re: Variable Percentage Withdrawal (VPW)

Post by Wrench »

nigel_ht wrote: Mon Sep 26, 2022 7:21 pm
Wrench wrote: Mon Sep 26, 2022 6:10 pm
Marseille07 wrote: Mon Sep 26, 2022 5:02 pm
canadianbacon wrote: Mon Sep 26, 2022 4:40 pm I think there is potential for someone to fill in the worksheet for the first time, and say "I currently spend $50000 a year, and this spreadsheet says I can withdraw $64000 next year, so I don't have to work anymore." The required flexibility section is a reminder that the portfolio income is variable (thus the name) and that you need to realize what is, frankly over a long enough retirement period, fairly likely to take place at some point. In the example above, the person should not be retiring immediately, but should be making sure they have savings that allow for a $50000 spend after loss.
I'm not sure what the issue is.

If someone's essentials are covered by 50K/year, they include 30% discretionary (flexibility) (71K/year), and divide that by the percentage VPW says (say, 5%) and retire on a 1.4M portfolio running 60/40 @ 5%...I thought this is exactly how you'd retire with VPW.

And 50K/year would be your floor because that's the level you can't breach to cover your essentials (and it's not even COLA'ed so more reasons you can't breach the floor).
I may be out in left field here, but the way I would use VPW (I'm not there yet), is that the recommended withdrawal would be entirely discretionary. My floor will come from social security, an inflation adjusted SPIA, and a stable value fund in a 403B. Those will cover >100% of my expected expenses. VPW gives me a number of how much to spend beyond my floor or base income.

Wrench
I would assume that the forward test is the expected way VPW would most likely get used and even there the bulk of the expenses is covered by the portfolio rather than pension in the early years.

If I recall correctly pension + SS is $3K a month or $36K which is likely lower than the unstated minimum expenses.

I don’t believe there are SPIA with full inflation adjusted COLA. A couple years ago 2% was what folks were saying you could get.

If you have a 2% SPIA you probably aren’t terribly happy at the moment…
I DO have a true inflation adjusted annuity that I purchased before they disappeared. But for others, there are SPIAs with fixed annual adjustments, or one can build inflation adjustment for a nominal income (there are multiple threads on this in BH). As I said, I may be the exception rather than the rule, but it works for me. YMMV.

Wrench
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Wrench wrote: Tue Sep 27, 2022 8:12 am
nigel_ht wrote: Mon Sep 26, 2022 7:21 pm
Wrench wrote: Mon Sep 26, 2022 6:10 pm
Marseille07 wrote: Mon Sep 26, 2022 5:02 pm
canadianbacon wrote: Mon Sep 26, 2022 4:40 pm I think there is potential for someone to fill in the worksheet for the first time, and say "I currently spend $50000 a year, and this spreadsheet says I can withdraw $64000 next year, so I don't have to work anymore." The required flexibility section is a reminder that the portfolio income is variable (thus the name) and that you need to realize what is, frankly over a long enough retirement period, fairly likely to take place at some point. In the example above, the person should not be retiring immediately, but should be making sure they have savings that allow for a $50000 spend after loss.
I'm not sure what the issue is.

If someone's essentials are covered by 50K/year, they include 30% discretionary (flexibility) (71K/year), and divide that by the percentage VPW says (say, 5%) and retire on a 1.4M portfolio running 60/40 @ 5%...I thought this is exactly how you'd retire with VPW.

And 50K/year would be your floor because that's the level you can't breach to cover your essentials (and it's not even COLA'ed so more reasons you can't breach the floor).
I may be out in left field here, but the way I would use VPW (I'm not there yet), is that the recommended withdrawal would be entirely discretionary. My floor will come from social security, an inflation adjusted SPIA, and a stable value fund in a 403B. Those will cover >100% of my expected expenses. VPW gives me a number of how much to spend beyond my floor or base income.

Wrench
I would assume that the forward test is the expected way VPW would most likely get used and even there the bulk of the expenses is covered by the portfolio rather than pension in the early years.

If I recall correctly pension + SS is $3K a month or $36K which is likely lower than the unstated minimum expenses.

I don’t believe there are SPIA with full inflation adjusted COLA. A couple years ago 2% was what folks were saying you could get.

If you have a 2% SPIA you probably aren’t terribly happy at the moment…
I DO have a true inflation adjusted annuity that I purchased before they disappeared. But for others, there are SPIAs with fixed annual adjustments, or one can build inflation adjustment for a nominal income (there are multiple threads on this in BH). As I said, I may be the exception rather than the rule, but it works for me. YMMV.

Wrench
Having a true inflation adjusted annuity is great!

Pretty sure those aren’t around anymore based on comments here and elsewhere…we’re not quite at the age where SPIAs are likely cost effective and we haven’t retired yet anyway.

You may not be as big an exception around here…the percentage of folks who either have over saved (ie can retire on a lot less than 4% of their portfolio) or have a pension seems higher than you would expect in the general population.

The thing is that pretty much any approach works in this case. If your portfolio is completely discretionary then does it really matter how you choose to spend it or in what increments?
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Wed Sep 28, 2022 9:22 am The thing is that pretty much any approach works in this case. If your portfolio is completely discretionary then does it really matter how you choose to spend it or in what increments?
It can actually matter, depending on your objective.

a) depleting everything in the end
b) trying to maximize the amount you withdraw
c) leaving legacy

Your withdrawal method still matters in this sense, even if your portfolio is discretionary.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Wed Sep 28, 2022 11:09 am
nigel_ht wrote: Wed Sep 28, 2022 9:22 am The thing is that pretty much any approach works in this case. If your portfolio is completely discretionary then does it really matter how you choose to spend it or in what increments?
It can actually matter, depending on your objective.

a) depleting everything in the end
b) trying to maximize the amount you withdraw
c) leaving legacy

Your withdrawal method still matters in this sense, even if your portfolio is discretionary.
Well a) is pretty easy. Buy anything you want whenever you want with any or all of your discretionary portfolio. Don’t go past $0.

It won’t impact your retirement in a negative way.

c) is pretty easy too. Spend very little out of your discretionary portfolio.

It won’t impact your retirement in a negative way.

b) is a little harder since you’re not going to know when the game ends for you. I assume that simply giving it all away to charity at the end isn’t a valid “withdrawal” since that’s just scenario c.

The strategy changes based on health and behavior. A 65 year old smoker with a bad heart that likes street racing cars and solo climbing mountains should probably front load withdrawals…of course you know that’s going to be the fellow who will dies peacefully in his mistresses bed at 103…
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Wed Sep 28, 2022 12:19 pm Well a) is pretty easy. Buy anything you want whenever you want with any or all of your discretionary portfolio. Don’t go past $0.

It won’t impact your retirement in a negative way.

c) is pretty easy too. Spend very little out of your discretionary portfolio.

It won’t impact your retirement in a negative way.

b) is a little harder since you’re not going to know when the game ends for you. I assume that simply giving it all away to charity at the end isn’t a valid “withdrawal” since that’s just scenario c.

The strategy changes based on health and behavior. A 65 year old smoker with a bad heart that likes street racing cars and solo climbing mountains should probably front load withdrawals…of course you know that’s going to be the fellow who will dies peacefully in his mistresses bed at 103…
Well my point was, hitting any of those goals is "easy," but they are conflicting. Obviously if you aim for a) then you can't aim for c), and I wasn't talking about doing c) and donate everything when you pass to achieve a). I was talking more about your decade+ retirement planning, and this is where withdrawal methods matter.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Wed Sep 28, 2022 12:22 pm
nigel_ht wrote: Wed Sep 28, 2022 12:19 pm Well a) is pretty easy. Buy anything you want whenever you want with any or all of your discretionary portfolio. Don’t go past $0.

It won’t impact your retirement in a negative way.

c) is pretty easy too. Spend very little out of your discretionary portfolio.

It won’t impact your retirement in a negative way.

b) is a little harder since you’re not going to know when the game ends for you. I assume that simply giving it all away to charity at the end isn’t a valid “withdrawal” since that’s just scenario c.

The strategy changes based on health and behavior. A 65 year old smoker with a bad heart that likes street racing cars and solo climbing mountains should probably front load withdrawals…of course you know that’s going to be the fellow who will dies peacefully in his mistresses bed at 103…
Well my point was, hitting any of those goals is "easy," but they are conflicting. Obviously if you aim for a) then you can't aim for c), and I wasn't talking about doing c) and donate everything when you pass to achieve a). I was talking more about your decade+ retirement planning, and this is where withdrawal methods matter.
I guess my point is, other than b) you can just wing it to achieve a) or c)…no spreadsheets or withdrawal strategy required…
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

nigel_ht wrote: Wed Sep 28, 2022 12:25 pm I guess my point is, other than b) you can just wing it to achieve a) or c)…no spreadsheets or withdrawal strategy required…
They all require some strategy imo. a) generally requires 1/N (VPW is one flavor). c) means low-% SWR or FWR.
Wrench
Posts: 1046
Joined: Sun Apr 28, 2019 10:21 am

Re: Variable Percentage Withdrawal (VPW)

Post by Wrench »

Marseille07 wrote: Wed Sep 28, 2022 12:30 pm
nigel_ht wrote: Wed Sep 28, 2022 12:25 pm I guess my point is, other than b) you can just wing it to achieve a) or c)…no spreadsheets or withdrawal strategy required…
They all require some strategy imo. a) generally requires 1/N (VPW is one flavor). c) means low-% SWR or FWR.
There is a (d), too. Providing sufficient funds to cover long term care should it be needed before end of life (assuming one does not have LTC insurance). (d) is my situation. Since it is unknown whether it will be needed or not, and in my case, I have to plan for both me and my spouse, it can be a bit tricky. So absolutely, one needs a strategy.

Oh, and by the way, our resources are rather modest by BH standards. But, we live frugally and expect to continue to do so. We have enough to live comfortably, and cover LTC (hopefully not for decades yet). Our ability to cover expenses from fixed income was a carefully executed plan over many years, NOT a result of an ultra high net worth.

Wrench
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

Wrench wrote: Wed Sep 28, 2022 5:59 pm There is a (d), too. Providing sufficient funds to cover long term care should it be needed before end of life (assuming one does not have LTC insurance). (d) is my situation. Since it is unknown whether it will be needed or not, and in my case, I have to plan for both me and my spouse, it can be a bit tricky. So absolutely, one needs a strategy.

Oh, and by the way, our resources are rather modest by BH standards. But, we live frugally and expect to continue to do so. We have enough to live comfortably, and cover LTC (hopefully not for decades yet). Our ability to cover expenses from fixed income was a carefully executed plan over many years, NOT a result of an ultra high net worth.

Wrench
There is a (d). It's a tricky situation because frontloaders like VPW might not provide enough toward the end (which is by design, can't backload and deplete-in-the-end at the same time).

This is why I'm sold on backloading personally even if it meant leaner "go-go" years.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Marseille07 wrote: Wed Sep 28, 2022 6:12 pm
Wrench wrote: Wed Sep 28, 2022 5:59 pm There is a (d), too. Providing sufficient funds to cover long term care should it be needed before end of life (assuming one does not have LTC insurance). (d) is my situation. Since it is unknown whether it will be needed or not, and in my case, I have to plan for both me and my spouse, it can be a bit tricky. So absolutely, one needs a strategy.

Oh, and by the way, our resources are rather modest by BH standards. But, we live frugally and expect to continue to do so. We have enough to live comfortably, and cover LTC (hopefully not for decades yet). Our ability to cover expenses from fixed income was a carefully executed plan over many years, NOT a result of an ultra high net worth.

Wrench
There is a (d). It's a tricky situation because frontloaders like VPW might not provide enough toward the end (which is by design, can't backload and deplete-in-the-end at the same time).

This is why I'm sold on backloading personally even if it meant leaner "go-go" years.
They say hope isn’t a plan but our hope is that with home equity we can afford the buy in for a CCRC.

Part of the analysis is whether the monthly income covers the monthly fee of a few representative examples…then we have to hope the monthly cost doesn’t increase much faster than inflation…

Not worried at all about having too much money when we’re older…I’ll just buy into a better CCRC…
SnowBog
Posts: 4678
Joined: Fri Dec 21, 2018 10:21 pm

Re: Variable Percentage Withdrawal (VPW)

Post by SnowBog »

Marseille07 wrote: Wed Sep 28, 2022 12:30 pm
nigel_ht wrote: Wed Sep 28, 2022 12:25 pm I guess my point is, other than b) you can just wing it to achieve a) or c)…no spreadsheets or withdrawal strategy required…
They all require some strategy imo. a) generally requires 1/N (VPW is one flavor). c) means low-% SWR or FWR.
I think these recent posts are getting off-topic from this thread... I'll try to bring it back around...

Arguably most retirees "just wing it"... On the one end of the spectrum you have Mr. Larimore's withdrawal strategy: viewtopic.php?t=73249 Granted in his case, he retired before SWR and VPW existed. On the other end are probably the majority of retirees who never read BH, don't know what SWR, VPW, 1/N, or the countless other withdrawal strategies. The key to their success isn't having boatloads of excess funds (although for some that might be the case). But it's the realization that they can manage their expenses to stay within their means.

During our working years, we have a variable amount of disposable income, which we prioritize on how we spend/save/etc. Even if our "income" isn't variable, we still learn to prioritize that "we need to replace X this year, so we don't have money for Y this year". In other words, we (most of us on BH anyway) have learned to live within our means.

I don't see why retirement should be different... Mechanically, one difference is we no longer have a "paycheck" to use as a reference point for our "net income". Brining us back on topic, to me that's the benefit, beauty, and simplicity of VPW, it attempts to provide that "reference" in as simple of a manner as possible, while recognizing that the amount itself is variable - just as it was in our working years. So, in many ways, VPW is just an extension of how we've lived during our working years - we prioritize our spending based on the funds available - which may vary from year to year.

This is arguably true of the "required flexibility" concept as well... IMHO it's not intended as a "set in stone" viewpoint that your portfolio could only ever suffer a single 50% loss... It's a reminder that life is unpredictable, and you should be mindful that life doesn't always go according to plan. So, it would be a bad plan to have your "bare essential" amount be just barely met by the "income after loss" amount - as in fact it could be lower... But again, just like during working years - we have a set of essential expenses and then a range of discretionary expenses that we prioritize based on the funds available. Well, IMHO that's exactly the intent of the "range" between the "suggested" and "after loss" amount. Ideally, we can afford everything on our list every year, but if we can't, we cut back just like we did during working years, and we reassess and reprioritize as circumstances change. But we aren't in a position to retire until that "income after loss" amount covers "enough" of our retirement that we'd still be "comfortable" if we needed to scale back. In my personal case, VPW aligns with FireCalc and other retirement calculators when viewed this way - we are close ["suggested" is more than expenses] - but not yet to the point we should retire ["income after loss" is less than expenses]. My working theory is when we get success from FireCalc/etc. - we'll see the VPW "income after loss" showing adequate coverage of "enough" of our estimated expenses as well.

Other [recent] posts have attempted to portray VPW as "front-loading" spending, with specific examples such as not having enough left for long-term care. To me, I think this is a false portrayal on several fronts... The simplest of which is VPW is not forcing you to treat 100% of your savings the same way. For example, if a retiree has an intention of funding college for a child/grandchild, providing for long-term care for themselves/spouse, buying their "dream home", RV, boat, car, etc. - they can - and arguably should - feel free to "set aside" what they feel is appropriate to meet those needs separate from VPW. IMHO VPW is for "annual expenses" - not for large "one-time" expenses.

Obviously, no withdrawal approach is "perfect", each has their advantages and disadvantages. To me, VPW continues to shine with its simplicity, including its alignment to the reality of our life from working years through retirement (where again we have a variable amount of disposable income that we prioritize how it's used). The fact that the spreadsheet makes it simple to deal with the complexities of different "income" stages of retirement (with various pensions/SSN kicking in years later), is essentially "self-correcting" (if you over/underspend) and is something my non-financial spouse could manage if I'm no longer around just sweetening the deal.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

SnowBog wrote: Thu Sep 29, 2022 1:45 am
Marseille07 wrote: Wed Sep 28, 2022 12:30 pm
nigel_ht wrote: Wed Sep 28, 2022 12:25 pm I guess my point is, other than b) you can just wing it to achieve a) or c)…no spreadsheets or withdrawal strategy required…
They all require some strategy imo. a) generally requires 1/N (VPW is one flavor). c) means low-% SWR or FWR.
I think these recent posts are getting off-topic from this thread... I'll try to bring it back around...
The recent question is about the flexibility test and how it can change to be below the value of the original test as seen in the forward test.

These other topics have occurred kinda waiting for an official answer.
During our working years, we have a variable amount of disposable income, which we prioritize on how we spend/save/etc. Even if our "income" isn't variable, we still learn to prioritize that "we need to replace X this year, so we don't have money for Y this year". In other words, we (most of us on BH anyway) have learned to live within our means.
Except that the test to determine if you have the flexibility to live within your potential means was recalculated at every step in the forward test.

It’s now less than it was in 2019.

So what does failing the flexibility test mean in 2022 and how is it different than failing in 2019?
This is arguably true of the "required flexibility" concept as well... IMHO it's not intended as a "set in stone" viewpoint that your portfolio could only ever suffer a single 50% loss... It's a reminder that life is unpredictable, and you should be mindful that life doesn't always go according to plan.
Then what’s the point of the required flexibility test at all? If you pass in 2019 the implication is you have enough flexibility to retire using VPW.

If you can fail in subsequent years and no longer have sufficient flexibility then either

A) you never had sufficient flexibility to begin with and probably shouldn’t have retired or something other than VPW.

Or

B) VPW allowed you to spend too much and now you’re in potential trouble if the economy continues to worsen and the market drops more.
So, it would be a bad plan to have your "bare essential" amount be just barely met by the "income after loss" amount - as in fact it could be lower...
How much lower? And why not make the “bare essential” amount be the threshold for required flexibility? Or “bare essentials + 10% reserve” or whatever?

Of course, that becomes an income floor and should be COLA’d.
Ideally, we can afford everything on our list every year, but if we can't, we cut back just like we did during working years, and we reassess and reprioritize as circumstances change.

But we aren't in a position to retire until that "income after loss" amount covers "enough" of our retirement that we'd still be "comfortable" if we needed to scale back.
In July 2019 the required flexibility was to be able to drop income to $61,286.

If I spend $60,000 a year I pass with an extra 1,200 a year buffer.

The test tells me I have sufficient flexibility to retire, no?

However, in September 2022 the required flexibility is $64,894...or $56,217 in July 2019.

My expenses, due to inflation, is now $69,260.

So I should go back to work as Marseille suggests or should I just keep using VPW and hope that things don’t get worse and cut into meat as opposed to the fat between whatever that gray zone is between “required flexibility” and “comfort”?

Or do you continue to try to use VPW but only spend the bare minimum because you’re out of the required flexibility zone?

Currently VPW doesn’t tell you what you should do does it? Other than hand waving “be flexible”.
Other [recent] posts have attempted to portray VPW as "front-loading" spending, with specific examples such as not having enough left for long-term care. To me, I think this is a false portrayal on several fronts... The simplest of which is VPW is not forcing you to treat 100% of your savings the same way. For example, if a retiree has an intention of funding college for a child/grandchild, providing for long-term care for themselves/spouse, buying their "dream home", RV, boat, car, etc. - they can - and arguably should - feel free to "set aside" what they feel is appropriate to meet those needs separate from VPW. IMHO VPW is for "annual expenses" - not for large "one-time" expenses.
This is different issue and separate from using VPW “as intended” where the required flexibility threshold can shift to the point where you no longer have the required flexibility to use VPW for “annual expenses”.
The fact that the spreadsheet makes it simple to deal with the complexities of different "income" stages of retirement (with various pensions/SSN kicking in years later), is essentially "self-correcting" (if you over/underspend) and is something my non-financial spouse could manage if I'm no longer around just sweetening the deal.
Is it sufficiently self correcting if you can go from having the “required flexibility” to NOT having the “required flexibility” without changing your expenses while following the spreadsheet?

What should your non-financial spouse do if the spreadsheet tells her she no longer passes the “required flexibility” test? Just ignore it?
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

SnowBog wrote: Thu Sep 29, 2022 1:45 am This is arguably true of the "required flexibility" concept as well... IMHO it's not intended as a "set in stone" viewpoint that your portfolio could only ever suffer a single 50% loss... It's a reminder that life is unpredictable, and you should be mindful that life doesn't always go according to plan. So, it would be a bad plan to have your "bare essential" amount be just barely met by the "income after loss" amount - as in fact it could be lower...
But you have to draw a line somewhere, right?

If I determined that my bare essentials are 50K/year, I can't go lower than that. Now, it's debatable if the 30% flexibility is enough or not - that's a discussion for another day.

However, if I pad 30% of buffer and retire on 70K/year initial withdrawal, the market crashes 30% and I use up my flexibility, it is unreasonable for VPW to tell me that I now need to be able to cut another 30% and live off of 35K/year. That wasn't a deal I signed up for.
Lastrun
Posts: 1505
Joined: Wed May 03, 2017 6:46 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Lastrun »

Marseille07 wrote: Thu Sep 29, 2022 8:52 am That wasn't a deal I signed up for.
Help me with my confusion as I admit I may be wrong. If I look at the way VPW works in the calculation after loss detail in the spreadhseet, there is only a reduction in the portfolio amount available for withdrawals, NOT the other income sources (SSDI, pension, annuity, temporary income).

Longinvest has consistently said:
The retiree is supposed to already have ample flexibility to reduce expenses in bad markets, taking into account stable lifelong income and asset allocation.
(emphasis mine).

It seems to me the retiree should be in a situation where the stable lifelong income is sufficient to meet basic needs. So I am not sure the retiree is not getting what they "signed up for." As Longinvest has said in this thread:
Our wiki's VPW page strongly suggests to combine stable lifelong non-portfolio income with variable withdrawals from a balanced (stocks and bonds) portfolio which adapt to actual returns during retirement.
But I think it is a legitimate concern if folks are simply plugging in the numbers and assuming things could not be worse than the initial 50% flexibility number, as would be a person blindly starting a 4% SWR and adjusting forward into a possible abyss.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Lastrun wrote: Thu Sep 29, 2022 9:54 am
Marseille07 wrote: Thu Sep 29, 2022 8:52 am That wasn't a deal I signed up for.
Help me with my confusion as I admit I may be wrong. If I look at the way VPW works in the calculation after loss detail in the spreadhseet, there is only a reduction in the portfolio amount available for withdrawals, NOT the other income sources (SSDI, pension, annuity, temporary income).
Based on the forward test thread it includes everything.

“After such a loss, total annual retirement income would be reduced to $64,820”
Longinvest has consistently said:
The retiree is supposed to already have ample flexibility to reduce expenses in bad markets, taking into account stable lifelong income and asset allocation.
(emphasis mine).

It seems to me the retiree should be in a situation where the stable lifelong income is sufficient to meet basic needs. So I am not sure the retiree is not getting what they "signed up for."
The question is whether the intended function of the “required flexibility” test equation in the spreadsheet is to tell you whether or not you have “ample flexibility to reduce expenses in bad markets”.

The forward test includes both a pension and social security.

If that’s not the intended function, what’s it for?
But I think it is a legitimate concern if folks are simply plugging in the numbers and assuming things could not be worse than the initial 50% flexibility number, as would be a person blindly starting a 4% SWR and adjusting forward into a possible abyss.
The failure mode for constant dollar based on historical outcomes is inflation driven (1966) rather than market crash driven (1929).

The flexibility test doesn’t seem to account for inflation…so I don’t believe that VPW flexibility test provides much visibility as to the likelihood that VPW can support your estimated baseline spending requirements…ie you don’t really know what your required flexibility number really is…and even if the spreadsheet produces a workable number at start of retirement that number get recomputed every cycle so you may not pass in the future.

What does that mean? And what should you do?
SnowBog
Posts: 4678
Joined: Fri Dec 21, 2018 10:21 pm

Re: Variable Percentage Withdrawal (VPW)

Post by SnowBog »

Lastrun wrote: Thu Sep 29, 2022 9:54 am
Marseille07 wrote: Thu Sep 29, 2022 8:52 am That wasn't a deal I signed up for.
Help me with my confusion as I admit I may be wrong. If I look at the way VPW works in the calculation after loss detail in the spreadhseet, there is only a reduction in the portfolio amount available for withdrawals, NOT the other income sources (SSDI, pension, annuity, temporary income).

Longinvest has consistently said:
The retiree is supposed to already have ample flexibility to reduce expenses in bad markets, taking into account stable lifelong income and asset allocation.
(emphasis mine).

It seems to me the retiree should be in a situation where the stable lifelong income is sufficient to meet basic needs. So I am not sure the retiree is not getting what they "signed up for." As Longinvest has said in this thread:
Our wiki's VPW page strongly suggests to combine stable lifelong non-portfolio income with variable withdrawals from a balanced (stocks and bonds) portfolio which adapt to actual returns during retirement.
But I think it is a legitimate concern if folks are simply plugging in the numbers and assuming things could not be worse than the initial 50% flexibility number, as would be a person blindly starting a 4% SWR and adjusting forward into a possible abyss.
Exactly!

VPW isn't magical - and it can't control the markets. If you get hit by a bad sequence of return, all VPW will do is the mathematically responsible thing of continuing to adjust the "suggested" amount accordingly. If that results in your withdrawal being below your "minimal acceptable level" - then that likely means you lacked the required flexibility to begin with. If you mistakenly thought the projected 50% loss was the "worst possible scenario" - then I think you missed the point of the exercise...

As you aptly note, VPW is intended to be used with a lifelong income stream. In an ideal situation, you have pensions/social security/SPIA's that cover all "essential" expenses for life - and thus anything the portfolio can provide covers the difference from being able to survive to enjoying retirement.

But that may not be realistic for everyone, given that very few jobs have pensions anymore, and SPIA's aren't inflation adjusted (or at least no longer available tied to inflation). But again, VPW doesn't "require" you use these - you just need to have an overall plan that accounts for them.

For myself, I address this via my portfolio construction, especially as we are targeting an early retirement with delayed social security/etc. We'll have 10-15+ years with "no income" (from pensions/social security/SPIA/etc.), meaning we are fully dependent on our portfolio from the day we retire until those delayed - but "lifelong" income streams kick in as we get closer to 70. Our approach is a DIY Annuity using Savings Bonds - EE Bonds as the backbone with I Bonds filling in the years we didn't buy EE Bonds 20 years prior plus adding inflation protection (I wrote about this here: viewtopic.php?t=358793). My goal is to have roughly $40k/year nominal (which is 1 - $10k EE Bond purchased for each spouse per year, or an equivalent amount of I Bonds) from the year we retire until our delayed "income" kicks in; this amount should be adequate to ensure all our "essential" expenses are covered - no matter what the market does. Specific to the EE Bonds - I actually treat them in the VPW Spreadsheet as "temporary retirement income", with a start date of when the first 20-year bond matures and an end date of the last one (I'm tempted to try to do the I Bonds the same way - but for now I just leave the current value in my "portfolio balance").

In so doing, I've provided a structure that all our "essential" expenses should be covered from the day we retire until our deployed income kicks in - which will cover them for the rest of our live(s). I'll add that we are naturally risk adverse, and in constructing our portfolio this way we accept that we'll have a smaller "final balance" via this quasi-LMP portfolio built on savings bonds combined with an AA of 60/40 while still in our accumulating years. But our primary goal in doing so was to ensure we could meet our essential expenses at all times - which no withdrawal method can guarantee on its own.

Again, what's left is the "variable disposable income" that our portfolio makes available to us, which IMHO VPW is perfectly suited to provide a beautifully simple model to follow. We know our "essentials" are covered; we just don't know from year to year how much of the things we'd like to do are covered. With VPW, we have an easy-to-follow path to figure that out.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

SnowBog wrote: Thu Sep 29, 2022 10:51 am VPW isn't magical - and it can't control the markets. If you get hit by a bad sequence of return, all VPW will do is the mathematically responsible thing of continuing to adjust the "suggested" amount accordingly. If that results in your withdrawal being below your "minimal acceptable level" - then that likely means you lacked the required flexibility to begin with. If you mistakenly thought the projected 50% loss was the "worst possible scenario" - then I think you missed the point of the exercise...
Help me understand how estimating a 50% loss was a mistake. That's the number longinvest gave on the spreadsheet.

If 50% isn't enough, what number should we use?

Lastrun: please see nigel_ht's response, covered better than I could.
Zeno
Posts: 1042
Joined: Wed Sep 12, 2018 10:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by Zeno »

(deleted. And longinvest, thank you again for VPW.)
Last edited by Zeno on Sat Oct 01, 2022 2:40 pm, edited 1 time in total.
SnowBog
Posts: 4678
Joined: Fri Dec 21, 2018 10:21 pm

Re: Variable Percentage Withdrawal (VPW)

Post by SnowBog »

Zeno wrote: Thu Sep 29, 2022 11:20 am
nigel_ht wrote: Thu Sep 29, 2022 10:49 am ....
What does that mean? And what should you do?
To me, the beauty of VPW is that it reinserts the role of human judgement in retirement planning when done through online calculators. Judgment is deeply embedded in VPW. Indeed, even the terminology it uses -- e.g., ""Suggestion for [insert year]", "Flexibility" -- confirms that the future is unknowable. VPW is very savvy in reminding the user that the user, at the end of the day, is responsible.

If the success of somebody's retirement is dependent upon VPW's "suggestion" and/or "flexibility" then, with respect, they aren't prepared for retirement. They haven't saved enough. They are hoping beyond hope that a spreadsheet is going to give them an answer about the future that is simply unknowable. And that's the inadvertent trap of many formulas -- folks are looking to input a few entries, click "enter" then be provided with an answer that says "you can retire" or "you can spend $200K this year."

Life, alas, doesn't work that way.

So we will use VPW as a check on our retirement withdrawals. Nothing more, nothing less. Our withdrawals will be based on our best and independent judgement under all scenarios.

VPW is a wonderful tool. I'm still amazed that it is free, and that longinvest remains engaged here. The forward test thread is fantastic, too. These VPW threads -- and others like it -- provide the guts of the forum's intellectual and substantive foundation, in my humble opinion. VPW remains, however, just a tool. Nothing more, nothing less. It's the carpenter who builds the house, not the hammer. Some folks seem to be asking too much of their hammers, or not appreciating that even hammers have limitations.

Folks are free to develop their own retirement tools, post them here, crowd-source them, peer-review them, defend them. Maybe get them published in the academic literature, too. Or really fine tune and sell access to them. I can't do those things, nor would I want to even if I had the skills. I'm just deeply appreciative of what longinvest and the other BH's have done here. It's wonderful and I'm deeply appreciative. Throwing pebbles at those folks from the cheap seats does not, on the other hand, interest me.
:beer Well stated!!
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

Zeno wrote: Thu Sep 29, 2022 11:20 am If the success of somebody's retirement is dependent upon VPW's "suggestion" and/or "flexibility" then, with respect, they aren't prepared for retirement. They haven't saved enough. They are hoping beyond hope that a spreadsheet is going to give them an answer about the future that is simply unknowable.
Then please explain the point of the required flexibility. Or to put it differently, how do you know you are prepared for retirement?
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

SnowBog wrote: Thu Sep 29, 2022 10:51 am
Lastrun wrote: Thu Sep 29, 2022 9:54 am
Longinvest has consistently said:
The retiree is supposed to already have ample flexibility to reduce expenses in bad markets, taking into account stable lifelong income and asset allocation.
(emphasis mine).

It seems to me the retiree should be in a situation where the stable lifelong income is sufficient to meet basic needs. So I am not sure the retiree is not getting what they "signed up for." As Longinvest has said in this thread:
Our wiki's VPW page strongly suggests to combine stable lifelong non-portfolio income with variable withdrawals from a balanced (stocks and bonds) portfolio which adapt to actual returns during retirement.
But I think it is a legitimate concern if folks are simply plugging in the numbers and assuming things could not be worse than the initial 50% flexibility number, as would be a person blindly starting a 4% SWR and adjusting forward into a possible abyss.
Exactly!

VPW isn't magical - and it can't control the markets. If you get hit by a bad sequence of return, all VPW will do is the mathematically responsible thing of continuing to adjust the "suggested" amount accordingly. If that results in your withdrawal being below your "minimal acceptable level" - then that likely means you lacked the required flexibility to begin with. If you mistakenly thought the projected 50% loss was the "worst possible scenario" - then I think you missed the point of the exercise...
If 50% loss isn't a good threshold to determine if you don't lack the required flexibility why bother doing it vs a more realistic number?

How do you know you have the required flexibility? Perhaps very few folks have the "required flexibility" to retire using VPW...except for those where 80%+ of their retirement needs are covered by SS and pensions...

Nobody thinks that 50% loss is the "worst possible scenario". Certainly not folks that use SWR which you guys seem to like bringing up. The worst historical scenario was 89% loss. Even then THAT wasn't the worst historical case.

If folks who use 4% SWR, which tests itself against an 89% loss, are "blindly staring into the abyss" what does it mean when the VPW flexibility test is only testing against a 50% loss?
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by nigel_ht »

Zeno wrote: Thu Sep 29, 2022 11:20 am
nigel_ht wrote: Thu Sep 29, 2022 10:49 am ....
What does that mean? And what should you do?
To me, the beauty of VPW is that it reinserts the role of human judgement in retirement planning when done through online calculators. Judgment is deeply embedded in VPW. Indeed, even the terminology it uses -- e.g., ""Suggestion for [insert year]", "Flexibility" -- confirms that the future is unknowable. VPW is very savvy in reminding the user that the user, at the end of the day, is responsible.
I agree that human judgement is very important in retirement planning and part of that is questioning our sacred cows to make sure there aren't issues that might come back and bite us in the rear...

Is VPW "savvy" in reminding the user that they are responsible or is it simply passing the buck and saying "well, if you fail, its because you lacked the required flexibility to make VPW work" without providing tools to determine if you had the required flexibility to begin with?

So, in your human judgement, what does it mean if you fail the VPW required flexibility test midway in retirement and what should you do?

And, in your human judgement, is there an issue if there's no advice on what to do if this happens because it is already possible in the forward test? Someone who passed the required flexibility test in July 2019 could fail the required flexibility test in September 2022...
SnowBog
Posts: 4678
Joined: Fri Dec 21, 2018 10:21 pm

Re: Variable Percentage Withdrawal (VPW)

Post by SnowBog »

Marseille07 wrote: Thu Sep 29, 2022 11:29 am
Zeno wrote: Thu Sep 29, 2022 11:20 am If the success of somebody's retirement is dependent upon VPW's "suggestion" and/or "flexibility" then, with respect, they aren't prepared for retirement. They haven't saved enough. They are hoping beyond hope that a spreadsheet is going to give them an answer about the future that is simply unknowable.
Then please explain the point of the required flexibility. Or to put it differently, how do you know you are prepared for retirement?
Personally, I think this comes back to Zeno's comments about "judgement". Nothing is guaranteed. Even someone with a 2% SWR could end up running into failure if the future looks nothing like the past. Again, it's about judgement (and not about the withdrawal method).

If it helps, here's how I'm looking at things...

For simple math, let's say our estimated annual retirement expenses (before taxes) are $100k, which is based on our current annual expenses with adjustments related to retirement (like having to get private health insurance vs. employer provided health insurance).

If we break that down further, our "essential" expenses are maybe $40k - mostly for the private health insurance, but also for property tax, insurance, and utilities on our house and enough for "beans and rice" to live off of. But even here, you'll notice we could actually cut back further if needed, as we could engineer our income to qualify for ACA subsidies to bring down the cost of health insurance... This is true for us with taxes as well, as we could engineer income during initial retirement to have $0 in taxes [and get ACA subsidies], or keep our plan of doing Roth conversions to spread out taxes over our lifetime.

The difference between the $40k and $100k are the "nice to have's", things like having an occasional steak for dinner, eating out every once in a while, going to a concert/event/etc., taking a vacation or two, doing some home projects, etc. We could "live without" these things if we had to, but that isn't our "plan". Your situation might have less - or more - between these endpoints. But I highly doubt many people on BH are building plans to retire on the bare minimum it would take them to survive... Their budget likely has some "fat" (buffer, fudge factor, etc.) that could be trimmed if needed...

As I noted before, when I look at VPW, I think about the "required" flexibility" not as a "set in stone - it will never be worse than" 50% reduction, but as a "reminder to remain flexible and use good judgement." If you want a more "practical" answer, I would never retire if the "income after loss" was within the margin of error of our $40k "essentials" - as we know it can always be worse... I'd want that number to be somewhere comfortably between our $40k "essential" mark and our $100k "planned" mark. Exactly where? That's back to judgement - where I might be comfortable isn't necessarily where you'd be comfortable.

For us personally, I want the majority of $100k "planned expenses" (plus estimated taxes due to Roth conversions) to fit in the "required flexibility", but then again, we are mid-40's planning an "early" retirement in our 50's, and so we have the luxury of being "greedy" in working/saving until we are confident that we can meet our known goals as we realize those goals might actually change over time. We'd feel differently if we were nearing retirement "age" and had more solidified plans at that point (vs. the unknowns of a child that hasn't yet launched, which could mean extra travel costs to visit if they move away from home, and trying to estimate 40+ years of expenses vs. 25+ years, etc.).

We aren't quite there yet... Our $100k (plus taxes) fits in the "suggested" withdrawal, but not the "income after loss"...

But again as Zeno said, VPW is "just a tool"... I use other tools as well, such as FireCalc, Flexible Retirement Planner, Parlana Gold, Fidelity's retirement calculator. They all tell me similar things - I'm close, but not quite there...

Which brings things back to Zeno's exceptional perspective on "judgement." Forget VPW for a second, and just think of any retirement calculator that provides results in a % success output. What % gives you the confidence you can successfully retire? Again, for each of us, we might have a different perspective. Some might only feel confident if all "calculators" under all conditions always report 100%. Some might accept say 95% success when viewed in the historically worst time periods. Others might accept 90% (or less) across all scenarios. Viewed in the context of VPW's "required flexibility" some might be comfortable if it covered 100% of their planned expenses, others 80%, maybe some 50% (or less). There isn't a "correct" answer - it's all about using our own judgement. But those with less tolerance (aka want 100% success in all scenarios) are likely going to "over save" and those with a large tolerance ("anything over 60% is 'passing' right?") are at risk for having "under saved". The withdrawal method they use (or don't use if they "wing it") isn't going to change that...

I'm 99% sure that when FireCalc, Fidelity, FRP, etc. all show the level of success we feel comfortable with, so too will VPW. And if one shows success and the others don't - it will give me pause to understand why and determine if my assumptions are flawed.

Returning us to VPW, I don't view it (individually) as my "validation I'm ready to retire", I don't think that was its point... IMHO its point is to help provide guidance to how much to save during working years, and during retirement years help provide guidance on how much you could spend in an unpredictable world combined with the reminder to remain flexible and use good judgement. If you are looking for precision, or prediction of all possible future outcomes, you aren't going to find it in VPW or any other method/tool...
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

SnowBog wrote: Thu Sep 29, 2022 12:40 pm Personally, I think this comes back to Zeno's comments about "judgement". Nothing is guaranteed. Even someone with a 2% SWR could end up running into failure if the future looks nothing like the past. Again, it's about judgement (and not about the withdrawal method).

If it helps, here's how I'm looking at things...

For simple math, let's say our estimated annual retirement expenses (before taxes) are $100k, which is based on our current annual expenses with adjustments related to retirement (like having to get private health insurance vs. employer provided health insurance).

If we break that down further, our "essential" expenses are maybe $40k - mostly for the private health insurance, but also for property tax, insurance, and utilities on our house and enough for "beans and rice" to live off of. But even here, you'll notice we could actually cut back further if needed, as we could engineer our income to qualify for ACA subsidies to bring down the cost of health insurance... This is true for us with taxes as well, as we could engineer income during initial retirement to have $0 in taxes [and get ACA subsidies], or keep our plan of doing Roth conversions to spread out taxes over our lifetime.

The difference between the $40k and $100k are the "nice to have's", things like having an occasional steak for dinner, eating out every once in a while, going to a concert/event/etc., taking a vacation or two, doing some home projects, etc. We could "live without" these things if we had to, but that isn't our "plan".

As I noted before, when I look at VPW, I think about the "required" flexibility" not as a "set in stone - it will never be worse than" 50% reduction, but as a "reminder to remain flexible and use good judgement." If you want a more "practical" answer, I would never retire if the "income after loss" was within the margin of error of our $40k "essentials" - as we know it can always be worse... I'd want that number to be somewhere comfortably between our $40k "essential" mark and our $100k "planned" mark. Exactly where? That's back to judgement - where I might be comfortable isn't necessarily where you'd be comfortable.

For us personally, I want the majority of $100k "planned expenses" (plus estimated taxes due to Roth conversions) to fit in the "required flexibility", but then again, we are mid-40's planning an "early" retirement in our 50's, and so we have the luxury of being "greedy" in working/saving until we are confident that we can meet our known goals as we realize those goals might actually change over time. We'd feel differently if we were nearing retirement "age" and had more solidified plans at that point (vs. the unknowns of a child that hasn't yet launched, which could mean extra travel costs to visit if they move away from home, and trying to estimate 40+ years of expenses vs. 25+ years, etc.).

We aren't quite there yet... Our $100k (plus taxes) fits in the "suggested" withdrawal, but not the "income after loss"...

But again as Zeno said, VPW is "just a tool"... I use other tools as well, such as FireCalc, Flexible Retirement Planner, Parlana Gold, Fidelity's retirement calculator. They all tell me similar things - I'm close, but not quite there...

Which brings things back to Zeno's exceptional perspective on "judgement." Forget VPW for a second, and just think of any retirement calculator that provides results in a % success output. What % gives you the confidence you can successfully retire? Again, for each of us, we might have a different perspective. Some might only feel confident if all "calculators" under all conditions always report 100%. Some might accept say 95% success when viewed in the historically worst time periods. Others might accept 90% (or less) across all scenarios. There isn't a "correct" answer - it's all about using our own judgement.

I'm 99% sure that when FireCalc, Fidelity, FRP, etc. all show the level of success we feel comfortable with, so too will VPW. And if one shows success and the others don't - it will give me pause to understand why and determine if my assumptions are flawed.

Returning us to VPW, I don't view it (individually) as my "validation I'm ready to retire", I don't think that was its point... IMHO its point is to help provide guidance to how much to save during working years, and during retirement years help provide guidance on how much you could spend combined with the reminder to remain flexible and use good judgement. If you are looking for precision, or prediction of all possible future outcomes, you aren't going to find it in VPW or any other method/tool...
So if we apply your situation to VPW-speak, I read that:
- your bare essentials: 40K/year
- the original "required flexibility" based on a 30% buffer (I am assuming 60/40): 57K/year
- your target withdrawal to start retirement: 100K/year

In other words, you are going above the required flexibility of 30% and instead, preparing for 60% (40K / (1 - 0.6) ) = 100K. There is nothing wrong with your plan. That said, there are a couple of follow-ups:

a) What's wrong with someone in your situation to retire with 57K/year, following the required flexibility specified by the spreadsheet?
b) Let's say you retire with 100K/year but the markets crash really hard and VPW says you can only withdraw 40K/year. While it is still enough to cover your essentials, the VPW also demands that you need to be able to live off of 28K/year. How do you generate this extra flexibility? If you are unable, is it fair if someone claims later that you weren't prepared to retire?
SnowBog
Posts: 4678
Joined: Fri Dec 21, 2018 10:21 pm

Re: Variable Percentage Withdrawal (VPW)

Post by SnowBog »

Marseille07 wrote: Thu Sep 29, 2022 1:12 pm
SnowBog wrote: Thu Sep 29, 2022 12:40 pm Personally, I think this comes back to Zeno's comments about "judgement". Nothing is guaranteed. Even someone with a 2% SWR could end up running into failure if the future looks nothing like the past. Again, it's about judgement (and not about the withdrawal method).
...
So if we apply your situation to VPW-speak, I read that:
- your bare essentials: 40K/year
- the original "required flexibility" based on a 30% buffer (I am assuming 60/40): 57K/year
- your target withdrawal to start retirement: 100K/year

In other words, you are going above the required flexibility of 30% and instead, preparing for 60% (40K / (1 - 0.6) ) = 100K. There is nothing wrong with your plan. That said, there are a couple of follow-ups:

a) What's wrong with someone in your situation to retire with 57K/year, following the required flexibility specified by the spreadsheet?
It's a judgement call. I'm not one to tell someone what's "wrong" for them - only they know their finances, needs, and risk tolerances.

But if I'm following your question, someone with a "can't survive on less than $40k/year" minimum with the "required flexibility" showing $57k/year, and then presumably the "suggested" amount showing $74k (130% of $57k) is obviously short of their target of $100k/year. Either that target isn't right for them, or they aren't yet ready to retire IMHO... (And I suspect they'd see similar results in FireCalc, etc.)
Marseille07 wrote: Thu Sep 29, 2022 1:12 pm b) Let's say you retire with 100K/year but the markets crash really hard and VPW says you can only withdraw 40K/year. While it is still enough to cover your essentials, the VPW also demands that you need to be able to live off of 28K/year. How do you generate this extra flexibility? If you can't, is it reasonable for someone to say you weren't prepared to retire?
Again, if I'm following you, VPW initially reported the "suggested" amount meeting my $100k target, and with a 60/40 AA the required flexibility was initially $70k. I'd personally feel fairly comfortable with that... And I'd suspect FireCalc/etc. would show the same... Although in my personal case, since we are considering retiring early with more unknowns, we are actually waiting for the entire $100k to be covered by the "required flexibility" - to each their own...

But if I'm following you, due to a horrible set of market conditions, the portfolio loses roughly 60% of its value (which would be something like two consecutive 50% market drops - or something like that), and as a result the "suggested" amount now barely meets my "essential" amount... And your concern is if yet a 3rd 50% market crash happens to bring the amount down to below my "essential" amount...

Let's turn this question around and face reality... Do you think during the great depression, or in the future if we were to have as horrible markets as this would suggest (stocks down something like 87%), that people didn't adapt their spending patterns? If things were that bad, they didn't do things like try to find work [even as a day laborer], rent out rooms in their house, sell household items, go to breadlines, or otherwise find ways to stretch their funds as far as possible? I sure the heck would...

And as a reminder, a large chuck of my $40k "essentials" was health care costs which could be drastically lowered by stopping Roth conversions, which would bring my taxes to $0 (especially with the TLH that would be going on) and make me eligible for ACA subsidies. So, I'd pull that thread before I start hunting for food or making moonshine in my basement to sell...

But no withdrawal approach will "save you from doom and gloom." You can always create a scenario that is going to expose hardship and force you to use your judgement. At least with variable approaches like VPW, you can hopefully "see this coming" - as when you do your updates (be it annual or monthly) you'll notice "hey my income is going down" and can start to - you guessed it - use your judgement to think about how you'll adapt.

You can also partially mitigate this by using a "dampening" approach to your withdrawals, assuming at least that such massive market crashes aren't decade long events, you hopefully only need to weather a few years of the worst before things start to improve. The forward thread uses monthly withdrawals with a 6-month buffer, some people might want to make annual withdrawals and use a few years of cash as a buffer. Again - judgement call...

While there's obviously a risk in doing so, and it would be better to find ways such isn't required (see above and below), another great thing about VPW is to the extent possible its "self-correcting". If the "suggested amount" after a nearly 87% crash in the stock market said I could only take out $28k a year - and I was completely unable to do anything to bridge the gap (which I don't think is reality) to my $40k "essential" expenses - if I had to - to keep our house and stay alive - I'd absolutely take out the "extra" $13k needed to live and pray that the economy got better in the future... That obviously means I'm depleting the portfolio quicker than I should - hence the risk - and I'd continue to do everything in my power to reduce expenses or increase income, but some number on a spreadsheet is not going to put my life or house at risk without a fight...

And you can help try to mitigate this via how you've setup your income streams and retirement portfolio. As I noted in a prior post, once we hit 70 our "essential" expenses are more than covered by our lifelong retirement "income" (aka social security, etc.) - with the vast majority (social security) indexed with inflation. To attempt to do the same between when we retire and when those "lifelong" income streams kick in, I'm building up a DIY Annuity using EE Bonds & I Bonds, which I expect to provide a minimum of $40k a year. So, in theory, even if my entire portfolio is wiped out where VPW says we can withdraw $0 from our portfolio, I've constructed things such that "we'll survive" until such our retirement income streams kick in. Now, that's not the retirement we "want", and again as per above, we'd absolutely be looking at ways to reduce spending and/or find new sources of income should such a scenario be realized, but baring the US defaulting on its debt obligations, I feel like I've done everything I can to ensure that we are as well setup for the most realistic outcomes that we can expect.

If someone has done nothing but blindly follow a withdrawal method into their own financial ruin, and does nothing to attempt to adapt or prepare for negative outcomes, then yes - I don't think they were ready to retire... But that may have had nothing to do with how much they "saved" and everything to do with their not being mentally prepared and/or using bad judgement in making their long-term plans (AA, portfolio construction, flexibility, etc.). My working theory (hope?) is the majority of BH folks who retire have a plan they feel confident about, and that plan reflects the reality that markets can and at some point will go down - and they are ready to adjust as needed.

I just happen to think VPW helps them do so more "visibly" by adjusting as their portfolio changes, and reminding them "it can get worse - have a plan"... But ultimately, VPW is just a tool - we have other tools including our judgement - don't give up one for another...
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

SnowBog wrote: Thu Sep 29, 2022 4:22 pm It's a judgement call. I'm not one to tell someone what's "wrong" for them - only they know their finances, needs, and risk tolerances.

But if I'm following your question, someone with a "can't survive on less than $40k/year" minimum with the "required flexibility" showing $57k/year, and then presumably the "suggested" amount showing $74k (130% of $57k) is obviously short of their target of $100k/year. Either that target isn't right for them, or they aren't yet ready to retire IMHO... (And I suspect they'd see similar results in FireCalc, etc.)
I think our maths are different. I am talking about a situation where:
- "can't survive on less than": 40K
- The VPW "suggested" amount, including flexibility: 57K since 40K / (1 - 0.3) = 57K. There is nothing to 1.3x on top of 57K, the 57K includes the 30% required flexibility already.

Check out how longinvest describes the required flexibility: viewtopic.php?p=6767277#p6767277

"The retiree must maintain the flexibility to easily cut spending by up to -$1,195/month because stocks could easily lose -50% of their value within a short time period. In other words, at least $1,195 must be budgeted for optional discretionary spending that could be eliminated without affecting the retiree's comfort."

IOW the "-1.195/mo" is basically the delta of 57K vs 40K already accounted for (the numbers are different but you get the idea).
Again, if I'm following you, VPW initially reported the "suggested" amount meeting my $100k target, and with a 60/40 AA the required flexibility was initially $70k. I'd personally feel fairly comfortable with that... And I'd suspect FireCalc/etc. would show the same... Although in my personal case, since we are considering retiring early with more unknowns, we are actually waiting for the entire $100k to be covered by the "required flexibility" - to each their own...

But if I'm following you, due to a horrible set of market conditions, the portfolio loses roughly 60% of its value (which would be something like two consecutive 50% market drops - or something like that), and as a result the "suggested" amount now barely meets my "essential" amount... And your concern is if yet a 3rd 50% market crash happens to bring the amount down to below my "essential" amount...

Let's turn this question around and face reality... Do you think during the great depression, or in the future if we were to have as horrible markets as this would suggest (stocks down something like 87%), that people didn't adapt their spending patterns? If things were that bad, they didn't do things like try to find work [even as a day laborer], rent out rooms in their house, sell household items, go to breadlines, or otherwise find ways to stretch their funds as far as possible? I sure the heck would...

And as a reminder, a large chuck of my $40k "essentials" was health care costs which could be drastically lowered by stopping Roth conversions, which would bring my taxes to $0 (especially with the TLH that would be going on) and make me eligible for ACA subsidies. So, I'd pull that thread before I start hunting for food or making moonshine in my basement to sell...

But no withdrawal approach will "save you from doom and gloom." You can always create a scenario that is going to expose hardship and force you to use your judgement. At least with variable approaches like VPW, you can hopefully "see this coming" - as when you do your updates (be it annual or monthly) you'll notice "hey my income is going down" and can start to - you guessed it - use your judgement to think about how you'll adapt.

You can also partially mitigate this by using a "dampening" approach to your withdrawals, assuming at least that such massive market crashes aren't decade long events, you hopefully only need to weather a few years of the worst before things start to improve. The forward thread uses monthly withdrawals with a 6-month buffer, some people might want to make annual withdrawals and use a few years of cash as a buffer. Again - judgement call...

While there's obviously a risk in doing so, and it would be better to find ways such isn't required (see above and below), another great thing about VPW is to the extent possible its "self-correcting". If the "suggested amount" after a nearly 87% crash in the stock market said I could only take out $28k a year - and I was completely unable to do anything to bridge the gap (which I don't think is reality) to my $40k "essential" expenses - if I had to - to keep our house and stay alive - I'd absolutely take out the "extra" $13k needed to live and pray that the economy got better in the future... That obviously means I'm depleting the portfolio quicker than I should - hence the risk - and I'd continue to do everything in my power to reduce expenses or increase income, but some number on a spreadsheet is not going to put my life or house at risk without a fight...

And you can help try to mitigate this via how you've setup your income streams and retirement portfolio. As I noted in a prior post, once we hit 70 our "essential" expenses are more than covered by our lifelong retirement "income" (aka social security, etc.) - with the vast majority (social security) indexed with inflation. To attempt to do the same between when we retire and when those "lifelong" income streams kick in, I'm building up a DIY Annuity using EE Bonds & I Bonds, which I expect to provide a minimum of $40k a year. So, in theory, even if my entire portfolio is wiped out where VPW says we can withdraw $0 from our portfolio, I've constructed things such that "we'll survive" until such our retirement income streams kick in. Now, that's not the retirement we "want", and again as per above, we'd absolutely be looking at ways to reduce spending and/or find new sources of income should such a scenario be realized, but baring the US defaulting on its debt obligations, I feel like I've done everything I can to ensure that we are as well setup for the most realistic outcomes that we can expect.

If someone has done nothing but blindly follow a withdrawal method into their own financial ruin, and does nothing to attempt to adapt or prepare for negative outcomes, then yes - I don't think they were ready to retire... But that may have had nothing to do with how much they "saved" and everything to do with their not being mentally prepared and/or using bad judgement in making their long-term plans (AA, portfolio construction, flexibility, etc.). My working theory (hope?) is the majority of BH folks who retire have a plan they feel confident about, and that plan reflects the reality that markets can and at some point will go down - and they are ready to adjust as needed.

I just happen to think VPW helps them do so more "visibly" by adjusting as their portfolio changes, and reminding them "it can get worse - have a plan"... But ultimately, VPW is just a tool - we have other tools including our judgement - don't give up one for another...
Right. People would certainly make adjustments, and I agree estimating -60% is plenty. Imo even -30% is plenty for 60/40. The 2 points nigel_ht and I are making are:

a) It isn't helpful to call a retiree that they shouldn't have retired in a hindsighted fashion after their 30% flexibility runs out in retirement.
b) The retiree does not have extra flexibility should they run out of their initial flexibility they considered at the time of retirement.

The point b) is especially important because this is an area of improvement to the methodology to better estimate the flexibility.
SnowBog
Posts: 4678
Joined: Fri Dec 21, 2018 10:21 pm

Re: Variable Percentage Withdrawal (VPW)

Post by SnowBog »

Marseille07 wrote: Thu Sep 29, 2022 4:42 pm I think our maths are different. I am talking about a situation where:
- "can't survive on less than": 40K
- The VPW "suggested" amount, including flexibility: 57K since 40K / (1 - 0.3) = 57K. There is nothing to 1.3x on top of 57K, the 57K includes the 30% required flexibility already.

...

Right. People would certainly make adjustments, and I agree estimating -60% is plenty. Imo even -30% is plenty for 60/40. The 2 points nigel_ht and I are making are:

a) It isn't helpful to call a retiree that they shouldn't have retired in a hindsighted fashion after their 30% flexibility runs out in retirement.
b) The retiree does not have extra flexibility should they run out of their initial flexibility they considered at the time of retirement.

The point b) is especially important because this is an area of improvement to the methodology to better estimate the flexibility.
Personally, I think this is a flawed test... But to play it out...

If someone really, truly can't survive if they have less than $x, and that is the very bottom of the "required flexibility" range, than IMHO they lack the required flexibility. Or started more simply, they haven't saved enough to retire. And other retirement tools/calculators will likely tell them the same.

"All models are wrong, some models are useful"

I don't think (ok, hope) most people who use VPW blindly assume that the "worst that could ever happen is the 50% reduction shown". It can always be worse... . So to retire with $0 margin for error is not something I'd ever be comfortable with, nor would ever recommend.

But someone who "wants" $57k, can live "comfortably" on $40k, but if the "poop hits the fan" knows they can get by on less... That's the intent!!!

As Zeno stated, VPW is just a tool that's to be used with good judgement. I don't think someone accepting a $0 margin if error is someone using good judgement.

But I also think it's a fool's errand to try to "pick a better number than 50%"... VPW doesn't attempt to predict the future... It's not a guarantee that things will not be worse... Arguably it does not matter much if instead of 50% a number like 40% or 60% was used. IMHO the whole point of having "a number" is to remind the user they need to use their judgement to decide what's right for them.

Again, to me I see no difference with Firecalc giving me a 99%, or even a 100%, chance of success. That does not mean that my plan will never fail (if I blindly follow it). In fact, the future could look nothing like the past and my "true" success rate was maybe more like 70% if I had the luxury of time travel to see how the future unfolds.

Ultimately, we'll each pull the plug and retire when, in our own judgement, we feel that we've adequately saved and are adequately prepared. (Or when our employer or life circumstances force us to do so, and in such circumstances we adapt as best as we can.)

And what I feel keeps getting missed is VPW accepts that the future is unknown and unknowable. The whole point is to continue to exercise good judgment and adapt to the situation as it unfolds. Not to blindly assume that your income will always fall between the high and low points of the spreadsheet...

VPW is just a tool... IMHO a great tool. But it's not going to give you your "if you meet this criteria under no circumstances can you ever not fail" - such does not exist...
SnowBog
Posts: 4678
Joined: Fri Dec 21, 2018 10:21 pm

Re: Variable Percentage Withdrawal (VPW)

Post by SnowBog »

Here a tl;dr version of what I wrote above...

No tool is a replacement for good judgement. But I find VPW is a tool that can help me practice good judgment. And if the "poop hits the fan", I trust my good judgement and human ability to adapt, to see me through.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Marseille07 »

SnowBog wrote: Thu Sep 29, 2022 6:03 pm Personally, I think this is a flawed test... But to play it out...

If someone really, truly can't survive if they have less than $x, and that is the very bottom of the "required flexibility" range, than IMHO they lack the required flexibility. Or started more simply, they haven't saved enough to retire. And other retirement tools/calculators will likely tell them the same.
Well, it sounds like you're saying that -30% (-50% x 0.6 since this portfolio is 60/40) might not be enough, and that's fair. Imo -30% on 60/40 is plenty, but it is just an estimate nonetheless.
But I also think it's a fool's errand to try to "pick a better number than 50%"... VPW doesn't attempt to predict the future... It's not a guarantee that things will not be worse... Arguably it does not matter much if instead of 50% a number like 40% or 60% was used. IMHO the whole point of having "a number" is to remind the user they need to use their judgement to decide what's right for them.
Then perhaps VPW should rip out the required flexibility section, since I don't see an added value here, if ultimately we simply say use your best judgment.
SevenBridgesRoad
Posts: 1179
Joined: Sat Jul 07, 2018 12:14 am

Re: Variable Percentage Withdrawal (VPW)

Post by SevenBridgesRoad »

Marseille07 wrote: Thu Sep 29, 2022 6:16 pm
Then perhaps VPW should rip out the required flexibility section, since I don't see an added value here, if ultimately we simply say use your best judgment.
What stops you from offering up your version of a variable percentage withdrawal system with no flexibility section? Start a new thread and go for it.

I'm retired and using the version of VPW as described in the BH wiki and perfectly comfortable with it. But the great thing about a forum like BH, is you are free to post your plan and develop a following in the free market of ideas. Innovation? Let's see what you've got.
Post Reply