I was just agreeing with SnowBog, basically "use your judgment," rather than depending on the VPW spreadsheet "income after loss" to calculate the required flexibility that's allSevenBridgesRoad wrote: ↑Thu Sep 29, 2022 6:29 pm 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.
Variable Percentage Withdrawal (VPW)
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
-
- Founder
- Posts: 11589
- Joined: Fri Feb 23, 2007 12:06 pm
- Location: Chicago
- Contact:
Re: Variable Percentage Withdrawal (VPW)
[This is temporarily locked for moderator review. We will be reopening it, please hold tight. - Alex Frakt]
Update: See below.
Update: See below.
Re: Variable Percentage Withdrawal (VPW)
After further review, a nonproductive interchange has been removed.
This thread is now unlocked to continue the discussion.
This thread is now unlocked to continue the discussion.
Re: Variable Percentage Withdrawal (VPW)
I don't see how relying on the historical worst case is better than using a doomsday estimator to see if a retiree needs to change something. Suppose a retiree in 1966 had used SWR, noting that 1929 was the current "historical worst case," thus proving that nothing could go wrong. The retiree would eventually figure out that plenty could go wrong. If the retiree didn't have enough "flexibility" to stay retired as inflation and interest rates kept rising, he would have to find another income source. That generally means un-retire and go back to work.
As for demanding to know how the "required flexibility" number is allowed to change, I think it is absolutely correct to do so. Expecting to retire with a simple number that represents the bottom line income for the rest of your life is rather simplistic. Suppose you retire comfortably and then your rich aunt dies and leaves you $1M. Your financial picture will change at that point, and any "required flexibility" number you used when you retired is worthless. Similarly, what if you retire and then your poor aunt comes for a visit and accidentally runs you down when she pulls into your driveway. You now require a ventilator and full-time nursing care to live. She is poor and has no insurance to help make you whole. How is the previous "required flexibility" number useful now? The point is that your basement number will never stay the same, because you don't know what might happen to change it.
I think of the "required flexibility" number as simply a "retirement status indicator." When I look at it, I do a quick mental calculation and decide whether I need to stop buying tomahawk ribeyes or if I am even still retired. My financial guy says I can't possibly spend it all. I don't want to get thirty years down the road and find out that he was wrong because nobody knew that inflation would be 15% for ten years, assisted living would cost $500,000 per year, and Social Security is now "means tested." Relying on the historical worst case doesn't help at all when history decides to create a new worst case.
I decided to go back to work part time last month (after two years of not working) because even though I probably have enough to stay retired, I would rather take my chances on earning a little more while I am still at my maximum earning rate than waiting until I can only get a job greeting people at the local store. If inflation gets tamed quickly (I'm not holding my breath) and I can retire again soon, great. A little more money for more "go-go" years is fine with me.
The one thing I can't know right now is what my required spending will be when I'm 95. How is it possible for me to fault a spreadsheet for giving me a warning to quit buying ribeyes as the market tanks and trailing inflation is 8%? That's as close as I can get to a crystal ball to know how to plan for the future. Thanks again, @longinvest, for adding this feature to your great spreadsheet.
- canadianbacon
- Posts: 677
- Joined: Sun Nov 10, 2019 9:04 pm
Re: Variable Percentage Withdrawal (VPW)
Your posts keep making me want steak.
But yeah, I think having a reminder that even where you currently are, it can get worse, and asking yourself if you have a plan for that, is valuable.
Bulls make money, bears make money, pigs get slaughtered.
Re: Variable Percentage Withdrawal (VPW)
I rarely eat a steak, but when I do I generally go for a good one. I don't care about the tomahawk bone, but very good marbling on a thick ribeye is important to me. That usually means one that is graded Prime by a USDA inspector. I don't know what the Canadian equivalent is, but I suspect it may also be named prime.
The whole steak thing is a good example of "required flexibility" to me. I like a good steak, but I'm ready to eat a Choice one if the market dips. I'm also ready to adapt to further market dips by switching to good hamburger, then so-so hamburger or stew meat, then canned stew, and so on. I really can't think of anything I consume that can't be downgraded in steps to meet some other income level.
I can also think of many ways to meet an increased income level if I need to....
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
So maybe you could help us consider the following scenario:justgary wrote: ↑Mon Oct 03, 2022 6:35 pm I rarely eat a steak, but when I do I generally go for a good one. I don't care about the tomahawk bone, but very good marbling on a thick ribeye is important to me. That usually means one that is graded Prime by a USDA inspector. I don't know what the Canadian equivalent is, but I suspect it may also be named prime.
The whole steak thing is a good example of "required flexibility" to me. I like a good steak, but I'm ready to eat a Choice one if the market dips. I'm also ready to adapt to further market dips by switching to good hamburger, then so-so hamburger or stew meat, then canned stew, and so on. I really can't think of anything I consume that can't be downgraded in steps to meet some other income level.
I can also think of many ways to meet an increased income level if I need to....
- SWR 4%/year, 50K/year "comfortable" retirement on a 1.25M portfolio running 60/40
Assuming the same person considers VPW, what would be a comparable portfolio? I can provide one example:
- VPW, 71.4K/year to start their retirement (30% flexibility added on top of 50K/year "comfortable" retirement) on a 1.43M portfolio running 60/40, 5% WR
Do they seem comparable to you? If not, why not?
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
Dear Marseille07,Marseille07 wrote: ↑Mon Oct 03, 2022 6:53 pm So maybe you could help us consider the following scenario:
- SWR 4%/year, 50K/year "comfortable" retirement on a 1.25M portfolio running 60/40
Assuming the same person considers VPW, what would be a comparable portfolio? I can provide one example:
- VPW, 71.4K/year to start their retirement (30% flexibility added on top of 50K/year "comfortable" retirement) on a 1.43M portfolio running 60/40, 5% WR
First, I suggest reading this post.
Second, a similar scenario to yours was analyzed just a few weeks back: In summary, a double -50% stock loss (a -75% stock loss with rebalancing beteen the two -50% losses) depleted the SWR retiree's portfolio within only 16 years. In contrast, the VPW retiree left a moderate bequest behind at age 88.
Third, here are excerpts from earlier posts:
longinvest wrote: ↑Sun May 30, 2021 8:00 am There has never been a version of VPW (in terms of retirement approach) where we ignore pensions or don't plan to replace missing payments for a delayed pension. There has never been a version of VPW where it wasn't combined with sufficient stable lifelong inflation-indexed guaranteed income. There has never been a version of VPW where the retiree assumed that there is a floor on how low withdrawals from a portfolio of fluctuating assets can get. There has never been a version of VPW where the retiree didn't live like most humans adapting spending to available income.
Advisory board member Nisiprius explained in 2011, before VPW was created, the philosophy on which the VPW retirement approach is based:nisiprius wrote: ↑Fri Apr 22, 2011 6:24 am In reality, we spend our preretirement lives dealing with fluctuating incomes and fluctuating expenses, and we adapt our lifestyle to our income. At age 30, we don't sit down with some Monte Carlo simulator and say "based on the statistical average of all careers in the United States, there's a 94.3% chance that I can spend $3141.59 a month, indexed by inflation, and never overdraw my checking account."
I don't see why things should change in retirement. We're counting more on our adaptability than on the predictability of our investment returns.
longinvest wrote: ↑Sun May 30, 2021 8:58 am No portfolio withdrawal method can create returns that markets don't deliver. No amount of wishful thinking can change this. As a result, there cannot be a floor on how low withdrawals from a portfolio of fluctuating assets can get.
Some approaches attempt to withdraw a constant inflation-indexed amount from a portfolio of fluctuating assets, but this can prematurely deplete the portfolio and cause the seemingly constant withdrawals to drop to zero (and then remain constant there).
VPW, instead, withdraws a sensible return-adjusted amount and, as a consequence, never prematurely depletes the portfolio.
longinvest wrote: ↑Mon Oct 05, 2020 7:01 pm There's no floor to how low VPW withdrawals can get, because there's no limit to how low future market returns (for both stocks and bonds) can get. That's the reality with have to live with.
Lastly, forum members Justgary, Canadianbacon, and SnowBog (and a few others) have already provided you with insightful and detailed answers during the last few days. (I thank them for these awesome posts!)
I think that the subject has been sufficiently explored for now in this thread. If you wish to pursue seeking an "income floor" using withdrawals from a portfolio of fluctuating assets (without considering non-portfolio income such as Social Security and pensions), VPW isn't what you seek. You're free to start a new thread to discuss other withdrawal methods which claim to provide such a floor. This thread is about VPW.
Best regards,
longinvest
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
Thanks longinvest.longinvest wrote: ↑Mon Oct 03, 2022 8:27 pmDear Marseille07,Marseille07 wrote: ↑Mon Oct 03, 2022 6:53 pm So maybe you could help us consider the following scenario:
- SWR 4%/year, 50K/year "comfortable" retirement on a 1.25M portfolio running 60/40
Assuming the same person considers VPW, what would be a comparable portfolio? I can provide one example:
- VPW, 71.4K/year to start their retirement (30% flexibility added on top of 50K/year "comfortable" retirement) on a 1.43M portfolio running 60/40, 5% WR
First, I suggest reading this post.
Second, a similar scenario to yours was analyzed just a few weeks back: In summary, a double -50% stock loss (a -75% stock loss with rebalancing beteen the two -50% losses) depleted the SWR retiree's portfolio within only 16 years. In contrast, the VPW retiree left a moderate bequest behind at age 88.
Third, here are excerpts from earlier posts:longinvest wrote: ↑Sun May 30, 2021 8:00 am There has never been a version of VPW (in terms of retirement approach) where we ignore pensions or don't plan to replace missing payments for a delayed pension. There has never been a version of VPW where it wasn't combined with sufficient stable lifelong inflation-indexed guaranteed income. There has never been a version of VPW where the retiree assumed that there is a floor on how low withdrawals from a portfolio of fluctuating assets can get. There has never been a version of VPW where the retiree didn't live like most humans adapting spending to available income.
Advisory board member Nisiprius explained in 2011, before VPW was created, the philosophy on which the VPW retirement approach is based:nisiprius wrote: ↑Fri Apr 22, 2011 6:24 am In reality, we spend our preretirement lives dealing with fluctuating incomes and fluctuating expenses, and we adapt our lifestyle to our income. At age 30, we don't sit down with some Monte Carlo simulator and say "based on the statistical average of all careers in the United States, there's a 94.3% chance that I can spend $3141.59 a month, indexed by inflation, and never overdraw my checking account."
I don't see why things should change in retirement. We're counting more on our adaptability than on the predictability of our investment returns.longinvest wrote: ↑Sun May 30, 2021 8:58 am No portfolio withdrawal method can create returns that markets don't deliver. No amount of wishful thinking can change this. As a result, there cannot be a floor on how low withdrawals from a portfolio of fluctuating assets can get.
Some approaches attempt to withdraw a constant inflation-indexed amount from a portfolio of fluctuating assets, but this can prematurely deplete the portfolio and cause the seemingly constant withdrawals to drop to zero (and then remain constant there).
VPW, instead, withdraws a sensible return-adjusted amount and, as a consequence, never prematurely depletes the portfolio.longinvest wrote: ↑Mon Oct 05, 2020 7:01 pm There's no floor to how low VPW withdrawals can get, because there's no limit to how low future market returns (for both stocks and bonds) can get. That's the reality with have to live with.
Lastly, forum members Justgary, Canadianbacon, and SnowBog (and a few others) have already provided you with insightful and detailed answers during the last few days. (I thank them for these awesome posts!)
I think that the subject has been sufficiently explored for now in this thread. If you wish to pursue seeking an "income floor" using withdrawals from a portfolio of fluctuating assets (without considering non-portfolio income such as Social Security and pensions), VPW isn't what you seek. You're free to start a new thread to discuss other withdrawal methods which claim to provide such a floor. This thread is about VPW.
Best regards,
longinvest
I wasn't aware VPW must consider SS or pensions or what have you. I don't see a reason why it must, but if you say so then obviously we're going in parallel and there's little point continuing this discussion.
Re: Variable Percentage Withdrawal (VPW)
Why would one not consider other income sources when determining a withdrawal rate?Marseille07 wrote: ↑Mon Oct 03, 2022 8:41 pmThanks longinvest.longinvest wrote: ↑Mon Oct 03, 2022 8:27 pmDear Marseille07,Marseille07 wrote: ↑Mon Oct 03, 2022 6:53 pm So maybe you could help us consider the following scenario:
- SWR 4%/year, 50K/year "comfortable" retirement on a 1.25M portfolio running 60/40
Assuming the same person considers VPW, what would be a comparable portfolio? I can provide one example:
- VPW, 71.4K/year to start their retirement (30% flexibility added on top of 50K/year "comfortable" retirement) on a 1.43M portfolio running 60/40, 5% WR
First, I suggest reading this post.
Second, a similar scenario to yours was analyzed just a few weeks back: In summary, a double -50% stock loss (a -75% stock loss with rebalancing beteen the two -50% losses) depleted the SWR retiree's portfolio within only 16 years. In contrast, the VPW retiree left a moderate bequest behind at age 88.
Third, here are excerpts from earlier posts:longinvest wrote: ↑Sun May 30, 2021 8:00 am There has never been a version of VPW (in terms of retirement approach) where we ignore pensions or don't plan to replace missing payments for a delayed pension. There has never been a version of VPW where it wasn't combined with sufficient stable lifelong inflation-indexed guaranteed income. There has never been a version of VPW where the retiree assumed that there is a floor on how low withdrawals from a portfolio of fluctuating assets can get. There has never been a version of VPW where the retiree didn't live like most humans adapting spending to available income.
Advisory board member Nisiprius explained in 2011, before VPW was created, the philosophy on which the VPW retirement approach is based:nisiprius wrote: ↑Fri Apr 22, 2011 6:24 am In reality, we spend our preretirement lives dealing with fluctuating incomes and fluctuating expenses, and we adapt our lifestyle to our income. At age 30, we don't sit down with some Monte Carlo simulator and say "based on the statistical average of all careers in the United States, there's a 94.3% chance that I can spend $3141.59 a month, indexed by inflation, and never overdraw my checking account."
I don't see why things should change in retirement. We're counting more on our adaptability than on the predictability of our investment returns.longinvest wrote: ↑Sun May 30, 2021 8:58 am No portfolio withdrawal method can create returns that markets don't deliver. No amount of wishful thinking can change this. As a result, there cannot be a floor on how low withdrawals from a portfolio of fluctuating assets can get.
Some approaches attempt to withdraw a constant inflation-indexed amount from a portfolio of fluctuating assets, but this can prematurely deplete the portfolio and cause the seemingly constant withdrawals to drop to zero (and then remain constant there).
VPW, instead, withdraws a sensible return-adjusted amount and, as a consequence, never prematurely depletes the portfolio.longinvest wrote: ↑Mon Oct 05, 2020 7:01 pm There's no floor to how low VPW withdrawals can get, because there's no limit to how low future market returns (for both stocks and bonds) can get. That's the reality with have to live with.
Lastly, forum members Justgary, Canadianbacon, and SnowBog (and a few others) have already provided you with insightful and detailed answers during the last few days. (I thank them for these awesome posts!)
I think that the subject has been sufficiently explored for now in this thread. If you wish to pursue seeking an "income floor" using withdrawals from a portfolio of fluctuating assets (without considering non-portfolio income such as Social Security and pensions), VPW isn't what you seek. You're free to start a new thread to discuss other withdrawal methods which claim to provide such a floor. This thread is about VPW.
Best regards,
longinvest
I wasn't aware VPW must consider SS or pensions or what have you. I don't see a reason why it must, but if you say so then obviously we're going in parallel and there's little point continuing this discussion.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
I just presented an SWR scenario and asked if a VPW scenario would be comparable, that's all.
Also, VPW does not consider other income sources when determining a withdrawal rate...it just uses your N and your AA as far as I understand.
Re: Variable Percentage Withdrawal (VPW)
It does if you use the worksheet where you can input various other income streamsMarseille07 wrote: ↑Tue Oct 04, 2022 10:17 amI just presented an SWR scenario and asked if a VPW scenario would be comparable, that's all.
Also, VPW does not consider other income sources when determining a withdrawal rate...it just uses your N and your AA as far as I understand.
It does not if you simply use the table.
Don't ask me why they are different
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
I don't ask you why they are different, but I still find it strange that considering SS / pensions is a requirement for VPW.
Those retiring early do not necessarily want to include SS even though they are going to receive some dollars 20 years later.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
Balbrec2, here's what the wiki page says about the difference:
How to use variable percentage withdrawals during retirement
-- With the VPW table
1. To simplify calculations when using the VPW Table with a pen and paper, missing payments between retirement and the start of a pension such as Social Security (possibly delayed to age 70[6]) can be provided by using a simple high-interest savings account or a certificate of deposit (CD) ladder. For the purposes of VPW calculations, the money set aside in this savings account or CD ladder should not be considered as part of the portfolio.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
Marseille07, both the Accumulation sheet and the Retirement sheet of the VPW Worksheet have entries for pensions. The wiki page is clear:Marseille07 wrote: ↑Tue Oct 04, 2022 1:05 pm I don't ask you why they are different, but I still find it strange that considering SS / pensions is a requirement for VPW.
Those retiring early do not necessarily want to include SS even though they are going to receive some dollars 20 years later.
VPW is best used in conjunction with guaranteed base income from Social Security, a pension (if any), and (if necessary) an inflation-indexed Single Premium Immediate Annuity (SPIA).*
* A SPIA indexed to the consumer price index for all urban consumers (CPI-U) might be difficult to find, but a 2%-indexed SPIA is easy to find and it's indexed to promised inflation given that the Federal Reserve has adopted a 2% inflation target since 2012 and has clearly expressed its intention to keep inflation on a 2% target to allow the public "to make accurate longer-term economic and financial decisions".
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Variable Percentage Withdrawal (VPW)
We are back to where we were a week ago.Lastrun wrote: ↑Thu Sep 29, 2022 9:54 am....
Longinvest has consistently said:
(emphasis mine).The retiree is supposed to already have ample flexibility to reduce expenses in bad markets, taking into account stable lifelong income and asset allocation.
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.
To me, in a using variable withdrawal strategy the retiree has chosen income variability as a preferred risk over sequence of return risk (portfolio depletion) that is present with SWR approaches.
With the VPW approach, the retiree must mitigate the negative consequences of the income variability and this accomplished with a stable floor of income that meets basic needs (SSDI, pensions, LMP) and required flexibility in the variable portfolio withdrawals. Other forms of mitigation discussed in this thread include annuitization, smoothing techniques, and more conservative asset allocations.
This mitigation is not unique to VPW, although some posters attempt to use this mitigation as a flaw and are surprised that it is a design.
SWR advocates have chosen their preferred risk, portfolio depletion, in favor of more stable income through inflation adjusted withdrawals. But they also employ mitigation techniques as well (many the same as VPW) including more conservative withdrawal rates, more conservative asset allocations, annuitization, LMPs, bond tents, guard rails, and over-saving (e.g. 50X). Many of these methods have the impact of reducing retirement income, the same as the required flexibility does with VPW.
No free lunch here, and the benefit to me personally of VPW is that it is more mathematically driven and is better than a simple RMD approach, and easier for my DW to understand than an ABW approach. It is also more certain to me compared to the "self-adjustment" approach some use on here (Taylor's posts come to mind). I am sympathetic to those that have lived off a salary for their working careers and need to emulate that in retirement. I personally have lived with a variable income for most of my life and am comfortable with the approach.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
Well...I want to make it clear that I was merely tabling comparable retirement scenarios of SWR and VPW. I wasn't adding a floor to VPW.longinvest wrote: ↑Tue Oct 04, 2022 1:17 pm Marseille07, both the Accumulation sheet and the Retirement sheet of the VPW Worksheet have entries for pensions. The wiki page is clear:VPW is best used in conjunction with guaranteed base income from Social Security, a pension (if any), and (if necessary) an inflation-indexed Single Premium Immediate Annuity (SPIA).*
* A SPIA indexed to the consumer price index for all urban consumers (CPI-U) might be difficult to find, but a 2%-indexed SPIA is easy to find and it's indexed to promised inflation given that the Federal Reserve has adopted a 2% inflation target since 2012 and has clearly expressed its intention to keep inflation on a 2% target to allow the public "to make accurate longer-term economic and financial decisions".
As far as guaranteed base income, I see your language saying "VPW is best used in conjunction with," but it sounds like it is more like a hard requirement. In that case, we can still craft comparable retirement scenarios by adding the same set of guaranteed base income on both sides.
I think it's a bit dismissive to try shutting down SWR vs VPW conversations by citing a lack of guaranteed base income. We don't have to continue with the comparison, just making my points clear here.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
Dear Marseille07,
Luckily, there's no need to derail this thread to launch a long discussion about the "SWR vs VPW" topic. You can simply start a new thread about it!
Best regards,
longinvest
Yesterday, I wrote:Marseille07 wrote: ↑Tue Oct 04, 2022 4:05 pm Well...I want to make it clear that I was merely tabling comparable retirement scenarios of SWR and VPW. I wasn't adding a floor to VPW.
This was a comparable retirement scenario for SWR and VPW where Social Security was left out of calculations.longinvest wrote: ↑Mon Oct 03, 2022 8:27 pm Second, a similar scenario to yours was analyzed just a few weeks back: In summary, a double -50% stock loss (a -75% stock loss with rebalancing beteen the two -50% losses) depleted the SWR retiree's portfolio within only 16 years. In contrast, the VPW retiree left a moderate bequest behind at age 88.
VPW is meant to be used by real people. People get Social Security; there's no reason to keep that out of calculations during accumulation and retirement. Social Security payments can be increased by delaying them; the VPW approach to accumulation and retirement plans for replacing missing payments between retirement and the start of a delayed pension.Marseille07 wrote: ↑Tue Oct 04, 2022 4:05 pm As far as guaranteed base income, I see your language saying "VPW is best used in conjunction with," but it sounds like it is more like a hard requirement. In that case, we can still craft comparable retirement scenarios by adding the same set of guaranteed base income on both sides.
The title of this thread isn't "SWR vs VPW". It's "Variable Percentage Withdrawal (VPW)".Marseille07 wrote: ↑Tue Oct 04, 2022 4:05 pm I think it's a bit dismissive to try shutting down SWR vs VPW conversations by citing a lack of guaranteed base income.
Luckily, there's no need to derail this thread to launch a long discussion about the "SWR vs VPW" topic. You can simply start a new thread about it!
Best regards,
longinvest
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
It's not the scenario I wanted to discuss a couple of days ago, but let's use your example then, since you wrote it and it's on-topic.longinvest wrote: ↑Tue Oct 04, 2022 4:42 pm This was a comparable retirement scenario for SWR and VPW where Social Security was left out of calculations.
viewtopic.php?p=6808605#p6808605
In this example you crafted, you conveniently assumed that the SWR retiree had already determined their 4% SWR before the crash (40K) and kept drawing the same amount every year.
Well, let's set up reasonable assumptions for the VPW retiree: this retiree had expected to withdraw 50K to start their retirement (VPW% is 5%), and their initial "required flexibility" after loss was -30%, or 35K/year (700K * 5%) before the crash. Fair?
If so, then how could you explain that this retiree managed to pay their bills on only 24,483/year of spending, well below their initial required flexibility that's not even COLA'ed? This is basically the point nigel_ht and I brought up - that you can't keep recalculating the required flexibility, since doing so essentially means you assume a retiree has unlimited flexibility.
Re: Variable Percentage Withdrawal (VPW)
It is very clear to me that the terrible market return in that example changed the whole game, so expected outcomes are no longer expectable. One method gave much less than initially promised, and the other made good on the promise until it ran out of money early. My personal requirement is that I never run out of money while my wife and/or myself are alive. SWR cannot guarantee that.Marseille07 wrote: ↑Tue Oct 04, 2022 10:18 pm ... how could you explain that this retiree managed to pay their bills on only 24,483/year of spending, well below their initial required flexibility that's not even COLA'ed? This is basically the point nigel_ht and I brought up - that you can't keep recalculating the required flexibility, since doing so essentially means you assume a retiree has unlimited flexibility.
Like I mentioned before, it is unreasonable to expect that having half of your retirement savings removed during your retirement should continue to give you the income that the whole retirement would have. At some point the retiree has to face reality and draw a new baseline for his standard of living. Even @nigel_ht indicated that he planned to use SWR (with a "go-go" addon; woo-hoo!) for five years then reassess the financial situation after the party was over. To me, that means he plans to draw a new baseline (and consider setting a new rate) after five years of living it up on a presumed safe withdrawal rate.
The method used in VPW does exactly that, but it happens to recalculate that baseline after every withdrawal. With the market we have today, that's a feature, not a flaw.
Re: Variable Percentage Withdrawal (VPW)
I forgot to mention that my 95-year-old mother has been living for over 50 years on an income that is currently up to nearly that amount (and any increases have been due to inflation adjustments after they started happening). Her taxes and insurance eat up a huge portion of that, yet she still plugs along. She always has good homemade food to offer visitors and a smile on her face. I see no reason why any of us couldn't do the same if faced with the reality of suddenly needing to make a big change in expenditures. Would it stink to have to do it? Absolutely. Could I do it? Absolutely. I have seen it done.Marseille07 wrote: ↑Tue Oct 04, 2022 10:18 pm ... this retiree managed to pay their bills on only 24,483/year of spending....
Re: Variable Percentage Withdrawal (VPW)
You seem to lack the understanding of the word variable in the Variable Withdrawal Method.Marseille07 wrote: ↑Tue Oct 04, 2022 10:18 pmIt's not the scenario I wanted to discuss a couple of days ago, but let's use your example then, since you wrote it and it's on-topic.longinvest wrote: ↑Tue Oct 04, 2022 4:42 pm This was a comparable retirement scenario for SWR and VPW where Social Security was left out of calculations.
viewtopic.php?p=6808605#p6808605
In this example you crafted, you conveniently assumed that the SWR retiree had already determined their 4% SWR before the crash (40K) and kept drawing the same amount every year.
Well, let's set up reasonable assumptions for the VPW retiree: this retiree had expected to withdraw 50K to start their retirement (VPW% is 5%), and their initial "required flexibility" after loss was -30%, or 35K/year (700K * 5%) before the crash. Fair?
If so, then how could you explain that this retiree managed to pay their bills on only 24,483/year of spending, well below their initial required flexibility that's not even COLA'ed? This is basically the point nigel_ht and I brought up - that you can't keep recalculating the required flexibility, since doing so essentially means you assume a retiree has unlimited flexibility.
It's not "variable - but only until some arbitrary threshold"...
It's not "variable - but never lower than the initial projected 'loss after income'"...
It's not "variable - but only in the 1st and 2nd withdrawal period"...
It's not "variable - but only upwards as I can't afford to go less"...
It's simply variable. VPW is designed to not prematurely deplete your portfolio. It is does so by using a variable withdrawal, which makes no guarantees as to how much that will actually be in any given year, since the amount is based on the cold hard reality of your portfolio - be that good or bad.
Asking your question in countless ways using different scenerios doesn't change the reality of the situation. If your portfolio is decimated, that's reality...
In your example, if the retiree assumed that VPW would never go below the $35k "income after loss" amount, then they assumed wrong. And it doesn't matter if instead of a 50% drop it showed a 60% or 70% or X% drop. There is no assertion, no guarantee, no "future projection" made by VPW as to the lowest income their portfolio might produce. You can backtest and see what the lowest has been historically, and use your judgement to decide what that means to your scenario and your assumptions about the future. This is essentially similar to looking at SWR with different withdrawal rates. You can accept as little or as much risk as you want, and as a result attempt to maximize your years in retirement or your years saving for retirement.
Again, I think you miss the point of the "required flexibility" intent. It isn't to - in your example - say "make sure you can live off of $35k.". The "number" is simply one of infinite possibilities, one that is commonly accepted to be a possible reality (50% stock crash) within a planning cycle, as remember the spreadsheet is intended to be updated regularly, such as annually if you are making annual withdrawals. Which means if you suffer a 50% loss in year 1, and a subsequent 50% loss in year 2, the numbers reflect that and show the potential of yet another 50% loss. The "income after loss" is simply to remind you that VPW is a variable approach, which requires that you remain flexible throughout retirement. If we suffer 3 consecutive drops of 50% each, then that's all that's left of the portfolio. That's reality. VPW might "hurt" with low withdrawals, but it won't deplete your portfolio - that's the tradeoff.
If you can't accept that variability, than either VPW isn't for you, or you simply aren't ready to retire. Maybe you'll feel differently when you've saved more. Maybe you'll feel differently when you get a better perspective on your expenses, and how to distinguish what's essential versus something that can be scaled back if needed. Maybe you'll realize that one can only do so much "doomsday" planning before realizing that people find ways to make things work when they need to. Maybe you'll realize that the recommendation to use VPW with adequate lifelong income sources such as social security (SPIA, TIPS, Savings Bonds, etc.) is to help address these concerns by providing enough income to meet all essential needs regardless of what the markets do (or don't do),
But nothing you'll do, no scenario you create, no matter how you ask the question, you cannot make a variable withdrawal method all of a sudden not be variable because you don't like it. As you've been told, if you want to go invent your "variable with a fixed floor" method - go for it. But it won't be VPW anymore. VPW requires a retiree to remain flexible at all times. . It's that simple.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
First, let's not start calling names. Actually, the word "variable" refers to the fact that your WR percentage changes year over year, not that your withdrawal amount is variable, which is already known and obvious.
Then what's the point of calculating "income after loss"? That's all I'm asking.In your example, if the retiree assumed that VPW would never go below the $35k "income after loss" amount, then they assumed wrong. And it doesn't matter if instead of a 50% drop it showed a 60% or 70% or X% drop.
Last edited by Marseille07 on Wed Oct 05, 2022 12:16 am, edited 2 times in total.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
I appreciate your responses. They are the kind of responses I would expect at a forum of this quality.justgary wrote: ↑Tue Oct 04, 2022 11:32 pmI forgot to mention that my 95-year-old mother has been living for over 50 years on an income that is currently up to nearly that amount (and any increases have been due to inflation adjustments after they started happening). Her taxes and insurance eat up a huge portion of that, yet she still plugs along. She always has good homemade food to offer visitors and a smile on her face. I see no reason why any of us couldn't do the same if faced with the reality of suddenly needing to make a big change in expenditures. Would it stink to have to do it? Absolutely. Could I do it? Absolutely. I have seen it done.Marseille07 wrote: ↑Tue Oct 04, 2022 10:18 pm ... this retiree managed to pay their bills on only 24,483/year of spending....
Re: Variable Percentage Withdrawal (VPW)
To add to my prior post, and give you "something to look at"...
Shows:
But if it helps, the "comparison" is SWR completely failing in 1990, where there was only $4k of the $40k available to withdraw...
So, the retiree is going to have to pick... Do they want to:
Assumptions:Marseille07 wrote: ↑Tue Oct 04, 2022 10:18 pm Well, let's set up reasonable assumptions for the VPW retiree: this retiree had expected to withdraw 50K to start their retirement (VPW% is 5%), and their initial "required flexibility" after loss was -30%, or 35K/year (700K * 5%) before the crash. Fair?
- 60/40 portfolio
- Retiree is 65 (age of 5% WR for 60/40: https://www.bogleheads.org/wiki/Variabl ... withdrawal)
- Planning for 35 year retirement (FICalc "assumes" age 100 at end of plan for VPW, so for age 65 @ 5%, must plan for 35 years
- Starting portfolio of $1M ($50k / 5%)
Shows:
- Successful 100% of time (again this is by design, as VPW won't prematurely deplete your portfolio)
- 50% of historical withdrawals would have been between $47k - $80k
- 90% of historical withdrawals would have been between $31k - $121k
- Out of 112 historical periods if you artificially constrained VPW to withdraw a minimum of $31k, only once would withdrawals have been below $31k, which would have occurred for someone who retired in 1887, where their average withdrawal would have been $55k, before dropping dramatically from 1916 ($60k) to $39k in 1918, declining until the final withdrawal of $30.5k in the last year.
- Out of 112 historical periods if you artificially constrained VPW to withdraw a minimum of $35k, 19 periods would have failed: 1886, 1887 (above), 1890, 1892, 1899, 1946, 1956-7, and 1959-69.
- A retiree in 1966 would have seen their portfolio fall from $1M to $0.44M in 9 years (1975) and down to $0.27M in another 7 years (1982). Their income would fall in a similar fashion, starting around $50k and reaching a low of $20k in 1982, before recovering to average $39k annually.
- 90
- Out of 112 historical periods, 9 failures: 1906, 1962, 1964-9, 1973
- Out of 112 historical periods, 16 "near" failures: 1899, 1902, 1907, 1909-13, 1937, 1956, 1959-61, 1963, and 1971-2
- Out of 112 historical periods, 28 ending portfolios more than 3x larger than initial portfolio: 1871-8, 1919-27, 1932, and 1978-87
- 90% of historical withdrawals would have been between $31k - $121k
- A retiree in 1966 would have depleted their portfolio after 24 years (1990)
Per my prior post, VPW is variable. This is simple math. Using the 1966 example, if the portfolio is only 27% of its initial balance after 16 years - then that's what it is... And 7.2% (age 82 WR for 60/40) of 27% of the original balance is obviously going to be far less than the original withdrawal and/or the initial "reminder" to remain flexible which only showed a 30% reduction (50% stock crash = 30% portfolio loss with 60/40). I'm not sure what's hard to understand with this...Marseille07 wrote: ↑Tue Oct 04, 2022 10:18 pm If so, then how could you explain that this retiree managed to pay their bills on only 24,483/year of spending, well below their initial required flexibility that's not even COLA'ed? This is basically the point nigel_ht and I brought up - that you can't keep recalculating the required flexibility, since doing so essentially means you assume a retiree has unlimited flexibility.
But if it helps, the "comparison" is SWR completely failing in 1990, where there was only $4k of the $40k available to withdraw...
So, the retiree is going to have to pick... Do they want to:
- Save more such that both VPW and SWR show better results
- Annuitize part of their portfolio (or recognize social security/pension/etc. income) such that they have a lifelong income source separate from their portfolio
- Determine that they can't live on less than $40k/year, and are willing to accept SWR's 92% success rate based on historical results
- Determine that they can accept that the future isn't predictable and may look nothing like the past, and thus focus on continuing to adapt their expenses, and if needed income, to align with what their portfolio (and lifelong time) provides
- Find another withdrawal method - which again is going to have trade-offs and decisions with no guarantees of future success (without such drastic over saving that at that point any plan - including "winging it" works)...
Re: Variable Percentage Withdrawal (VPW)
I didn't call anyone names, I'm just pointing out the obvious.Marseille07 wrote: ↑Wed Oct 05, 2022 12:04 amFirst, let's not start calling names. Actually, the word "variable" refers to the fact that your WR percentage changes year over year, not that your withdrawal amount is variable, which is already known and obvious.
And to restate the obvious, variable absolutely applies to the withdrawal amount. The whole point of VPW is to take the withdrawal against the current portfolio balance. (Which is different than SWR which only applies the WR rate against the initial portfolio balance, and never again looks to the actual portfolio to see how things are going.)
Let's consider a 65-year-old with a 60/40 portfolio making quarterly withdrawals, which the initial VPW WR is 5% of their current portfolio balance. They will absolutely have a different portfolio balance each and every quarter they make a withdrawal, even though the WR is 5% each time. Meaning even with the same withdrawal rate that year, their actual withdrawals will be variable, as each reflects the then current portfolio balance.
When they are 66, yes the WR is now 5.1%, but again it's against whatever their current portfolio balance is at that point.
So, I'm left scratching my head when you continue to ask questions like "explain that this retiree managed to pay their bills on only 24,483/year of spending", with the implication that somehow VPW "promised" they'd have $35k or more... It can't do that... It's just applying a % to whatever the current portfolio balance is, which is absolutely going to result in potentially vastly different withdrawals as the portfolio changes - good or bad...
As a reminder that VPW requires you to remain flexible. Again, central to the whole premise of VPW is the results will be variable based on simple math applied to the current portfolio balance. If the portfolio is decimated by horrific sequence of returns, such as a 50% drop, followed by another 50% drop, and then another 50% drop, yes VPW expects that you find a way to make things work - as it can't force your portfolio to give you money that your portfolio doesn't have.Marseille07 wrote: ↑Wed Oct 05, 2022 12:04 amThen what's the point of calculating "income after loss"? That's all I'm asking.In your example, if the retiree assumed that VPW would never go below the $35k "income after loss" amount, then they assumed wrong. And it doesn't matter if instead of a 50% drop it showed a 60% or 70% or X% drop.
That's true of all portfolio methods - ultimately, they are bound by the reality of the portfolio itself, and if the reality is being down to 12.5% of your original portfolio - there's no silver bullet - that's a horrific and likely unprecedented bad sequence of returns - but it is what it is... In my case, as I've stated before, I'm creating a DIY Annuity (using savings bonds) to coincide with our delayed social security, providing "lifelong income" from when we retire until we die adequate to cover our essential expenses, so in this horrific scenario - I'd still be able to meet all my essential expenses, but I'd likely have quit all discretionary spending long ago to protect the portfolio until it recovered...
IMHO you are trying to read too much into the 50% drop modeled in the spreadsheet... That's not a commitment, promise, prediction, or limit... It's just a "reminder"...
Likewise, you seem to be forgetting that VPW is an "ongoing approach"... Each and every time you make the withdrawal, you should be mindful that the next one could be less (or more)... Even if the one you are making is far less than the one prior, the next one could still be less... It might not be as bad as the 50% drop modeled in the spreadsheet, or it might be much worse... IMHO the point isn't the specific $ (or specific % drop). IMHO it's just the reminder to stay flexible.
And as to "why have it then", because humans like to see a "comparison point." Putting a number in front of them, even if that number isn't "perfect" or "guaranteed", it helps the brain wrap around the idea of "it could be worse." And again, when they make their next withdrawal, they'll be reminded "it could be worse". And hopefully doing that throughout, they won't be blindsided when they eventually hit a "bad year" like this year, and they see their income falling... To my view, that again is by design and is a "feature" of VPW.
As I've stated before, my spouse is not financially inclined, couldn't care less about the S&P 500, treasury rates, and would be ill prepared to re-assess things if we had a horrific market crash that put our retirement plans in jeopardy if they continued to blindly spend the same every year going forward. But with VPW, my hope is they'll understand "hmm... this is less than I could withdraw last period... And that 'required flexibility' section reminds me that it could always be worse... So maybe we'll scale back expenses this year and wait for things to start looking better, just to be safe..." If so, then the odds of our plan working even if I'm not here to shepherd it just went up significantly.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
Dear Marseille07,Marseille07 wrote: ↑Wed Oct 05, 2022 12:04 am Actually, the word "variable" refers to the fact that your WR percentage changes year over year, not that your withdrawal amount is variable, which is already known and obvious.
The method isn't called variable-percentage withdrawal; it's called variable percentage withdrawal. For clarity, here's a mathematical-like notation to explain the difference in semantics:
- variable-percentage withdrawal = (variable percentage) withdrawal
- variable percentage withdrawal = variable (percentage withdrawal)
Best regards,
longinvest
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
SnowBog, this isn't how VPW works. If FICalc allows for setting a withdrawal amount floor, then it doesn't implement the Variable Percentage Withdrawal (VPW) method discussed in this thread.
And where is the guaranteed income? The retiree of the simulated retirement has no Social Security!
What matters to the retiree isn't the withdrawal amount, it's total retirement income, including Social Security, pension (if any), and (if necessary) cost-of-living indexed SPIA*.
* Single Premium Immediate Annuity.
There's no floor to how low VPW withdrawals can get, because there's no limit to how low future market returns (for both stocks and bonds) can get. That's the reality with have to live with. But reasonable certainty for a total retirement income floor can be bought. It's expensive, but it's available. Delaying Social Security to age 70 is probably the cheapest way to buy additional guaranteed lifelong income.
The VPW approach to retirement described in this thread and implemented in the VPW Accumulation And Retirement Worksheet takes into account current and future pensions, with and without cost-of-living adjustments.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Moderator
- Posts: 961
- Joined: Fri Oct 16, 2009 10:49 pm
Re: Variable Percentage Withdrawal (VPW)
First longinvest thank you very much for your detailed work and the spreadsheets. We plan to use VPW knowing it is variable and having relative small social security payments (being in 3 countries and not creating larger social security balances does not help). A lot of our savings are in taxable and some in 401k plans.
We are ok with a 50% drop and will then act accordingly and reduce our spending. Obviously this only works if a lot of our spending is variable. If the market drops another 50% then we will have to drastically downsize (sell house, cars, move to a lower cost of living area into a cheap rental) but we still will eat more than Ramen every day. If it drops another 50% we all have a different issue anyway.
I am aware that no withdrawal method offers perfection and will pick the best of the (imperfect) alternatives and this for us is VPW. I understand that various methods have their own advantages and disadvantages and the future is never certain. VPW is attractive for us because it enables us not to underspend in our early retirement years where we will be able to travel more but still will leave us with money in our later years. It is also easy to use and has a sound mathematical background.
We are ok with a 50% drop and will then act accordingly and reduce our spending. Obviously this only works if a lot of our spending is variable. If the market drops another 50% then we will have to drastically downsize (sell house, cars, move to a lower cost of living area into a cheap rental) but we still will eat more than Ramen every day. If it drops another 50% we all have a different issue anyway.
I am aware that no withdrawal method offers perfection and will pick the best of the (imperfect) alternatives and this for us is VPW. I understand that various methods have their own advantages and disadvantages and the future is never certain. VPW is attractive for us because it enables us not to underspend in our early retirement years where we will be able to travel more but still will leave us with money in our later years. It is also easy to use and has a sound mathematical background.
-
- Posts: 226
- Joined: Tue Dec 27, 2016 11:59 am
Re: Variable Percentage Withdrawal (VPW)
I find VPW to be an attractive tool for guiding retirement planning and spending...and thank longinvest once again for all the work...but man, this thread has become a disaster with all the bickering, recalcitrance, and unwillingness to listen to others. And it just doesn't stop.
No way I'm going to keep wading through this...sad that a couple bad apples have spoiled this barrel...
No way I'm going to keep wading through this...sad that a couple bad apples have spoiled this barrel...
-
- Posts: 40
- Joined: Thu Jan 03, 2019 10:43 pm
Re: Variable Percentage Withdrawal (VPW)
A friendly suggestion to longinvest and any other folks involved in VPW: I would recommend updating the "wiki" explainer for VPW to include more disclosures about the proper uses and limitations of the "required flexibility" calculation. I shared Marseille07's confusion about what this parameter was conveying when I read through the explainer, and I think it would be helpful for your potential "consumer base" if this part of the VPW method is clarified (or, if you like, clarified further!)
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
Chicagoprof, thanks for the suggestion. To address your concern, starting this month I'll include a link to this post which explains the required flexibility in details (and provides an illustrative scenario) in the monthly detailed calculation post of the forward test thread. The wiki page already points readers to that thread as a detailed example of how to use the VPW Accumulation And Retirement Worksheet during retirement.Chicagoprof wrote: ↑Wed Oct 05, 2022 9:14 am A friendly suggestion to longinvest and any other folks involved in VPW: I would recommend updating the "wiki" explainer for VPW to include more disclosures about the proper uses and limitations of the "required flexibility" calculation. I shared Marseille07's confusion about what this parameter was conveying when I read through the explainer, and I think it would be helpful for your potential "consumer base" if this part of the VPW method is clarified (or, if you like, clarified further!)
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 40
- Joined: Thu Jan 03, 2019 10:43 pm
Re: Variable Percentage Withdrawal (VPW)
@Longinvest: Great!
Re: Variable Percentage Withdrawal (VPW)
Let's flip the story to the accumulation side. You buy a few stocks and bonds, or maybe a fund or two as you accumulate wealth. Your goal is to keep your portfolio balanced at 60%-40%, but after five years you begin to notice that it is nowhere near that balance any more. The result is that your portfolio is either much more exposed to loss or far too conservative for good gains. In order to fix it, you examine the status of each item in your portfolio in order to rebalance it (buy and sell some of the things, add more saved cash, etc.) to get back to the goal.Marseille07 wrote: ↑Wed Oct 05, 2022 12:04 am Then what's the point of calculating "income after loss"? That's all I'm asking.
Failing to do so can have enormous consequences down the road, so rebalancing on a regular basis is very important. Similarly, the retiree must rebalance his withdrawals in order to meet the stated goal of not running out of money. SWR on its face offers no rebalancing at all. VPW rebalances at every withdrawal. As it does so, it offers the "income after loss" calculation so the retiree can be ready to make changes (rebalance expenses) as needed in the case of a sudden market drop.
I think it is reasonable to assume that anyone reading Bogleheads is aware of the overall status of the market. I reflexively quit buying steaks when the market dips just to protect my retirement resources, and I don't need VPW to tell me. The question is whether my spouse notices at all. When I'm gone, she can use VPW to calculate her spending and use the "income after loss" as her warning lamp to spend modestly when it is required.
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
I see, I think we should leave it at that. I don't see much utility in re-calculating -30% every month that's not even COLA'ed, but if you find it useful then I'm in no position to tell you otherwise.justgary wrote: ↑Wed Oct 05, 2022 9:52 am Let's flip the story to the accumulation side. You buy a few stocks and bonds, or maybe a fund or two as you accumulate wealth. Your goal is to keep your portfolio balanced at 60%-40%, but after five years you begin to notice that it is nowhere near that balance any more. The result is that your portfolio is either much more exposed to loss or far too conservative for good gains. In order to fix it, you examine the status of each item in your portfolio in order to rebalance it (buy and sell some of the things, add more saved cash, etc.) to get back to the goal.
Failing to do so can have enormous consequences down the road, so rebalancing on a regular basis is very important. Similarly, the retiree must rebalance his withdrawals in order to meet the stated goal of not running out of money. SWR on its face offers no rebalancing at all. VPW rebalances at every withdrawal. As it does so, it offers the "income after loss" calculation so the retiree can be ready to make changes (rebalance expenses) as needed in the case of a sudden market drop.
I think it is reasonable to assume that anyone reading Bogleheads is aware of the overall status of the market. I reflexively quit buying steaks when the market dips just to protect my retirement resources, and I don't need VPW to tell me. The question is whether my spouse notices at all. When I'm gone, she can use VPW to calculate her spending and use the "income after loss" as her warning lamp to spend modestly when it is required.
Thanks, I enjoyed your responses.
Re: Variable Percentage Withdrawal (VPW)
Agreed! 100%!longinvest wrote: ↑Wed Oct 05, 2022 8:02 amSnowBog, this isn't how VPW works. If FICalc allows for setting a withdrawal amount floor, then it doesn't implement the Variable Percentage Withdrawal (VPW) method discussed in this thread.
This was done solely to try to identify years that showed a historical example that seemed to fit the question being proposed, and show how VPW handled that year.
As I had noted in my prior post.
.SnowBog wrote: ↑Tue Oct 04, 2022 11:53 pm ... you cannot make a variable withdrawal method all of a sudden not be variable because you don't like it. ... if you want to go invent your "variable with a fixed floor" method - go for it. But it won't be VPW anymore. VPW requires a retiree to remain flexible at all times. . It's that simple.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
SnowBog, thanks for the clarification.SnowBog wrote: ↑Wed Oct 05, 2022 11:18 amAgreed! 100%!longinvest wrote: ↑Wed Oct 05, 2022 8:02 amSnowBog, this isn't how VPW works. If FICalc allows for setting a withdrawal amount floor, then it doesn't implement the Variable Percentage Withdrawal (VPW) method discussed in this thread.
This was done solely to try to identify years that showed a historical example that seemed to fit the question being proposed, and show how VPW handled that year.
As I had noted in my prior post.
SnowBog wrote: ↑Tue Oct 04, 2022 11:53 pm ... you cannot make a variable withdrawal method all of a sudden not be variable because you don't like it. ... if you want to go invent your "variable with a fixed floor" method - go for it. But it won't be VPW anymore. VPW requires a retiree to remain flexible at all times. . It's that simple.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Variable Percentage Withdrawal (VPW)
Quick question about V2.7 on the accumulation sheet. I noticed when I updated it at my birthday this year using the new version, I was getting the new warning message that my portfolio was too small to bridge pensions (I forget exactly what it said - I'm at work atm), when I used the Age 70 estimate for social security (delay as long as possible, which is what Mike Piper's calculator recommends I do). But if i changed it to the age 67 estimate (" full retirement age") the warnings dropped off. Doing so only raised my recommended monthly savings by a modest sum.
So if I'm doing as recommended by VPW, I take it then that I should use the age 67 estimate since it doesn't have the warning messages? What would longinvest do?
Thanks!
So if I'm doing as recommended by VPW, I take it then that I should use the age 67 estimate since it doesn't have the warning messages? What would longinvest do?
Thanks!
Re: Variable Percentage Withdrawal (VPW)
You should model your particular scenario as best as you can...azanon wrote: ↑Thu Oct 06, 2022 12:30 pm Quick question about V2.7 on the accumulation sheet. I noticed when I updated it at my birthday this year using the new version, I was getting the new warning message that my portfolio was too small to bridge pensions (I forget exactly what it said - I'm at work atm), when I used the Age 70 estimate for social security (delay as long as possible, which is what Mike Piper's calculator recommends I do). But if i changed it to the age 67 estimate (" full retirement age") the warnings dropped off. Doing so only raised my recommended monthly savings by a modest sum.
So if I'm doing as recommended by VPW, I take it then that I should use the age 67 estimate since it doesn't have the warning messages? What would longinvest do?
Thanks!
The idea of the "bridge" is to "set aside" the gap between retirement and those sources. For simple math, let's say social security will provide $10k/year (regardless of when you claim for a simple example). If you delay to 70 and plan to retire at say age 55, then VPW needs to "set aside" 15 years * $10k = $150k. If claim at 67, then VPW only needs to set aside 12 years * $10k = $120k.
So I wouldn't change an input to avoid the warning message - that sort of defeats the purpose...
But if your particular scenario is showing you the warning message, and presumably that your bridge is not 100% funded, then you need to understand the implications. That might mean you are better off claiming SS early and/or delaying retirement as it means there is less years to "bridge."
But again, the default should be to model your particular scenario, and then work through the warnings/messages/etc. it provides.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
Azanon, the "WARNING: Portfolio too small to fully bridge pensions" warning appears in the Accumulation worksheet when the cost to fully bridge pensions is bigger than 50% of the projected retirement portfolio balance.azanon wrote: ↑Thu Oct 06, 2022 12:30 pm Quick question about V2.7 on the accumulation sheet. I noticed when I updated it at my birthday this year using the new version, I was getting the new warning message that my portfolio was too small to bridge pensions (I forget exactly what it said - I'm at work atm), when I used the Age 70 estimate for social security (delay as long as possible, which is what Mike Piper's calculator recommends I do). But if i changed it to the age 67 estimate (" full retirement age") the warnings dropped off. Doing so only raised my recommended monthly savings by a modest sum.
So if I'm doing as recommended by VPW, I take it then that I should use the age 67 estimate since it doesn't have the warning messages? What would longinvest do?
The Accumulation worksheet only displayed a warning. It's a human who attempted to change the Social Security claiming age to eliminate the warning. The human could have attempted other changes, instead. The worksheet didn't give any indication of what to attempt changing because addressing the bridge funding deficit is beyond the ability of such a simple spreadsheet. Human knowledge is required to properly analyze the situation and address the problem, if there's one.
What would I do?
I would try to understand the situation. I would look at the grey sections in the lower part of the Accumulation worksheet. They're very similar to those of the Retirement worksheet which are explained in the monthly detailed calculations post of the forward test (like this recent post). There are four sections: Ratios, Income Streams, Calculations For 2022, and Calculations After Loss.
The Income Streams section calculates the total cost to fully bridge pensions. The Calculations For 2022 and Calculations After Loss sections include a Funding Ratio column. A 100% ratio indicates that a pension bridge is fully funded. A lower ratio, such as 30%, indicates that only 30% of the pension's Adjusted Annual Payment is considered in calculations so that bridging it will only cost 30% of the full bridge cost. These lower amounts are displayed in the Effective Annual Payment and Effective Bridge Cost columns. The worksheet adjusts funding ratios such that the Total Effective Cost of Bridges represents no more than 50% of the Projected Retirement Portfolio Balance.
As my primary objective would probably be financial independence (not necessarily retirement), I would also consider the Accumulation worksheet's projected income Available For Taxes And Expenses. If I could live with it, I would probably ignore the warning and go with the suggested Portfolio Contribution for 2022.
If the income available for taxes and expenses is too low, I would probably change my plan to address the problem based on my preferences.
As I consider that increasing Social Security inflation-adjusted payments, by delaying them to age 70, provides valuable lifelong financial protection to the retiree, I wouldn't consider claiming at an earlier age.
Instead, I would probably consider increasing Retirement Age by one or more years. This has many impacts: it adds accumulation years increasing the projected retirement portfolio balance, it possibly increases Social Security primary insurance amount (PIA), and it reduces the number of years of pension bridges.
As I said, I would only do such a change if the income available for taxes and expenses (after savings) is too low without making changes. If it's sufficient, I would most probably ignore the warning to keep saving more, and just wait until the projected retirement age is reached to switch to "one more year" mode (delay retirement by one year, possibly multiple times), if the warning is still present at that time.
But, this is based on personal values, such as primarily aiming for financial independence. Someone with different values could choose differently.
I wouldn't retire if the Retirement worksheet displayed the same warning, because this would indicate the the current portfolio balance, at the time of retirement, is too low.
It's difficult to discuss solutions (if they're necessary) without first understanding the investor's situation.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
SnowBog, thanks for this simple explanation of the general idea.
For interested readers, the actual worksheet calculation is slightly different. It uses the VPW Table taking into account that the portfolio is used for bridging pensions. This is illustrated monthly in the forward test thread (see this post).
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Variable Percentage Withdrawal (VPW)
Thanks so much Snowbog and longinvest, taking the time to answer my question. I'm at work now, but this is hugely helpful, and I'll look at it closer this weekend based on that guidance to plan accordingly and adjust where I need to!
Re: Variable Percentage Withdrawal (VPW)
It might be useful as you dig into this to recognize that your portfolio is no doubt less now than it was a year ago. I would not rely on the market going in any particular direction before you retire, but if retirement is later than sooner, perhaps the odds are on your side that the market will recover and improve your outlook.
It is good that you are looking ahead and planning now, though, since you still have time to take action to protect your retirement.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
In the forward test thread, forum member AlohaBill asked:
In real life, I would calibrate the size of the withdrawal cushion according to the couple's comfort, and:
AlohaBill, the monthly process (updating the worksheet, taking a withdrawal, adjusting it with the withdrawal cushion) would have been identical. But because Vanguard’s Target Retirement Income Fund (VTINX) has a 30/70 stocks/bonds allocation, it would have resulted into slightly smaller and slightly less-volatile total retirement income than using Vanguard's LifeStrategy Moderate Growth Fund (VSMGX) because the percentages in the VPW Table are lower for a 30/70 stocks/bonds allocation than for a 60/40 one, and a 30/70 portfolio is less volatile than a 60/40 one.
I think that monthly withdrawals are an awesome opportunity to help one's spouse learn the process and commit it to memory, when participating to the process every month. It's likely more effective than a less-frequent annual exposure to the process.AlohaBill wrote: ↑Sun Nov 13, 2022 10:53 am We have been retired for 5 years now and the most we’ve withdrawn in a year from our savings is about .7%. We are rather conservative investors with no more than 75% stock/ 25% bond in the past and no less than Vtinx’s 30/70 today. At this point in time for us, I don’t really think it matters. Thank you.
In real life, I would calibrate the size of the withdrawal cushion according to the couple's comfort, and:
- I would keep the withdrawal cushion separate from the portfolio. (I wouldn't include it as part of the worksheet's "Portfolio Balance" entry).
- I would make the calculations (using the worksheet) and trades on a regular schedule convenient to both spouses, like "the 3rd Wednesday of the month", something that can easily be added to one's agenda (instead of the forward test's unrealistic midnight of the last day of the month).
- Once trades are settled, I'd transfer the proceeds into the withdrawal cushion savings account. (See below about taxes).
- On the first day of next month, I'd transfer 1/X of the withdrawal cushion savings account into the spending account, where X (equal to 6 or more) is the chosen target number of months in the withdrawal cushion.
- Note that I would handle the withdrawal cushion on an after-tax basis in a taxable savings account. Money for taxes would be taken off the proceeds of trades and directly sent into a savings account for taxes. This is simpler because sometimes some taxes are already withheld by the brokerage, like when taking money out of a traditional IRA account.
Thanks for the nice words.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 1184
- Joined: Sat Jul 07, 2018 12:14 am
Re: Variable Percentage Withdrawal (VPW)
I plugged in the same yellow cell numbers as the most recent Forward Test retirement worksheet, only changing the portfolio allocation to 30/70. It changes the withdrawal to $4,942 from $5,397.longinvest wrote: ↑Sun Nov 13, 2022 2:11 pm In the forward test thread, forum member AlohaBill asked:
AlohaBill, the monthly process (updating the worksheet, taking a withdrawal, adjusting it with the withdrawal cushion) would have been identical. But because Vanguard’s Target Retirement Income Fund (VTINX) has a 30/70 stocks/bonds allocation, it would have resulted into slightly smaller and slightly less-volatile total retirement income than using Vanguard's LifeStrategy Moderate Growth Fund (VSMGX) because the percentages in the VPW Table are lower for a 30/70 stocks/bonds allocation than for a 60/40 one, and a 30/70 portfolio is less volatile than a 60/40 one.
-
- Posts: 1184
- Joined: Sat Jul 07, 2018 12:14 am
Re: Variable Percentage Withdrawal (VPW)
What great advice. I need to do this with my wife.longinvest wrote: ↑Sun Nov 13, 2022 2:11 pm
I think that monthly withdrawals are an awesome opportunity to help one's spouse learn the process and commit it to memory, when participating to the process every month. It's likely more effective than a less-frequent annual exposure to the process.
In real life, I would calibrate the size of the withdrawal cushion according to the couple's comfort, and:
- I would keep the withdrawal cushion separate from the portfolio. (I wouldn't include it as part of the worksheet's "Portfolio Balance" entry).
- I would make the calculations (using the worksheet) and trades on a regular schedule convenient to both spouses, like "the 3rd Wednesday of the month", something that can easily be added to one's agenda (instead of the forward test's unrealistic midnight of the last day of the month).
- Once trades are settled, I'd transfer the proceeds into the withdrawal cushion savings account. (See below about taxes).
- On the first day of next month, I'd transfer 1/X of the withdrawal cushion savings account into the spending account, where X (equal to 6 or more) is the chosen target number of months in the withdrawal cushion.
- Note that I would handle the withdrawal cushion on an after-tax basis in a taxable savings account. Money for taxes would be taken off the proceeds of trades and directly sent into a savings account for taxes. This is simpler because sometimes some taxes are already withheld by the brokerage, like when taking money out of a traditional IRA account.
Thanks longinvest. Your hard work is making my retirement much simpler. It's interesting how "good enough" is so satisfying. Toast to you, my friend.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
SevenBridgesRoad, thanks. I hadn't thought of changing asset allocation only for this latest withdrawal as quick illustration.SevenBridgesRoad wrote: ↑Sun Nov 13, 2022 5:20 pm I plugged in the same yellow cell numbers as the most recent Forward Test retirement worksheet, only changing the portfolio allocation to 30/70. It changes the withdrawal to $4,942 from $5,397.
Comparing total retirement income, including the $1,000 pension payment, tells us that income would have been slightly lower in this case or, more precisely (($5,942 / $6,397) - 1) = -7% lower.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Variable Percentage Withdrawal (VPW)
Thank you all.
-
- Posts: 5682
- Joined: Sat Aug 11, 2012 8:44 am
Re: Variable Percentage Withdrawal (VPW)
In another thread, I explained how VPW is, by design, immune to sequence of returns risk but remains exposed to market risk. Here's the main text of my post (with minor modifications):
The Variable Percentage Withdrawal (VPW) method has been specifically designed to mathematically guarantee that the portfolio will never be prematurely depleted.
By design, VPW withdrawals fluctuate; they're variable. It's in the name of the method.
The Nth VPW withdrawal is mathematically determined by the cumulative time-weighted return of the portfolio between retirement and the withdrawal, regardless of the sequence of (N - 1) annual returns that precedes the withdrawal. Said differently, any sequence of annual returns that delivers the same cumulative time-weighted return will result into the same withdrawal (and the same portfolio balance before and after withdrawal).
In other words, VPW is mathematically immune to sequence of returns risk (SORR).
But, VPW withdrawals fluctuate because they're exposed to market risk (MR): lower cumulative time-weighted returns between retirement and the Nth withdrawal will result into a smaller Nth withdrawal than higher cumulative time-weighted returns over the same period.
In contrast, the so-called "safe" withdrawal rate (SWR) method is subject to SORR. Changing the sequence of (N - 1) annual returns that precedes the Nth withdrawal, while keeping the same cumulative time-weighted return, can change the seemingly constant inflation-adjusted withdrawal to zero, due to premature portfolio depletion.
The Variable Percentage Withdrawal (VPW) method has been specifically designed to mathematically guarantee that the portfolio will never be prematurely depleted.
By design, VPW withdrawals fluctuate; they're variable. It's in the name of the method.
The Nth VPW withdrawal is mathematically determined by the cumulative time-weighted return of the portfolio between retirement and the withdrawal, regardless of the sequence of (N - 1) annual returns that precedes the withdrawal. Said differently, any sequence of annual returns that delivers the same cumulative time-weighted return will result into the same withdrawal (and the same portfolio balance before and after withdrawal).
In other words, VPW is mathematically immune to sequence of returns risk (SORR).
But, VPW withdrawals fluctuate because they're exposed to market risk (MR): lower cumulative time-weighted returns between retirement and the Nth withdrawal will result into a smaller Nth withdrawal than higher cumulative time-weighted returns over the same period.
In contrast, the so-called "safe" withdrawal rate (SWR) method is subject to SORR. Changing the sequence of (N - 1) annual returns that precedes the Nth withdrawal, while keeping the same cumulative time-weighted return, can change the seemingly constant inflation-adjusted withdrawal to zero, due to premature portfolio depletion.
Last edited by longinvest on Sun Dec 11, 2022 11:40 am, edited 1 time in total.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 16054
- Joined: Fri Nov 06, 2020 12:41 pm
Re: Variable Percentage Withdrawal (VPW)
Correct. For some reason even some long-timers here do not understand this. I find it very bizarre.longinvest wrote: ↑Sun Dec 11, 2022 11:35 am In another thread, I explained how VPW is, by design, immune to sequence of returns risk but remains exposed to market risk. Here's the main text of my post (with minor modifications):
The Variable Percentage Withdrawal (VPW) method has been specifically designed to mathematically guarantee that the portfolio will never be prematurely depleted.
By design, VPW withdrawals fluctuate; they're variable. It's in the name of the method.
The Nth VPW withdrawal is mathematically determined by the cumulative time-weighted return of the portfolio between retirement and the withdrawal, regardless of the sequence of (N - 1) annual returns that precedes the withdrawal. Said differently, any sequence of annual returns that delivers the same cumulative time-weighted return will result into the same withdrawal (and the same portfolio balance before and after withdrawal).
In other words, VPW is mathematically immune to sequence of returns risk (SORR).
But, VPW withdrawals fluctuate because they're exposed to market risk (MR): lower cumulative time-weighted returns between retirement and the Nth withdrawal will result into a smaller Nth withdrawal than higher cumulative time-weighted returns over the same period.
In contrast, the so-called "safe" withdrawal rate (SWR) method is subject to SORR. Changing the sequence of (N - 1) annual returns that precedes the Nth withdrawal, while keeping the same cumulative time-weighted return, can change its seemingly constant inflation-adjusted withdrawal it to zero, due to premature portfolio depletion.