Variable Percentage Withdrawal (VPW)

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
smectym
Posts: 845
Joined: Thu May 26, 2011 5:07 pm

Re: Variable Percentage Withdrawal (VPW)

Post by smectym » Wed Feb 12, 2020 8:11 am

longinvest wrote:
Sun Feb 09, 2020 10:25 am
Maybe it's a good time to remind readers that one of the primary objectives of VPW is to spend the money without any risk of prematurely depleting the portfolio, instead of dying with a huge unspent portfolio!

Here's what I wrote in the first post of this thread, in July 2013, before the later improvements to the overall VPW accumulation and retirement approach to address various concerns like longevity:
longinvest wrote:
Thu Jul 25, 2013 11:12 pm
I'm not satisfied with the methods I've seen :
  • Constant-Dollar : [...] and (2) there is a very high probability that I will leave much money unspent (if the bad sequence doesn't show up). I might be too old to spend it, once I realize that I have too big a pile of money.
  • Constant-Percentage : [...] The problem is that I am, again, quite likely to underspend and leave a lot of money unspent.
  • Spend Only the Dividends :[...] Yet, I would still be very likely to leave a lot of money unspent.
Here's my idea. It is a Variable Percentage Withdrawal (VPW) method. [...]

[...]
And so on, until year 30, where I withdraw 100% of my remaining portfolio. [...]

So, even if the famous bad sequence of return was to happen, I wouldn't run out of money before the end of 30 years. As a bonus, I am sure to have spent it all; there's no risk of having a single dime left for year 31 (unless I died before year 30, of course).
I see no reason to wait until I'm dead to give money to loved ones and charities. I can do it while alive using part of the excess money VPW gives me. This way, I get to enjoy the process and also observe if they are responsible with it, making sure my money is really helping them instead of hurting them.

The reason I live below my means and invest the difference in a portfolio, today, is to delay consumption and, most likely, increase future consumption and giving. It isn't in the hope of looking at a huge number on a financial statement when I'll be on my deathbed.
longinvest, I quite agree. Moreover, having accumulated according to semi-paranoid “fear of running out” parameters over the decades, and retiring in the midst of this ridiculous permabullmarket, it seems like no matter what we withdraw, the balances keep rising. (The money worries that keep one up at night...) Nonetheless, we have the money to give to relatives and others who could use it, as well as to charitable, and we’re doing that—though still not departing from VPW guidelines

dpotter
Posts: 7
Joined: Mon Nov 19, 2018 12:09 pm

Re: Variable Percentage Withdrawal (VPW)

Post by dpotter » Sat Feb 15, 2020 5:54 pm

Rolyatroba wrote:
Sat Feb 08, 2020 7:27 pm
P.S. I also use i-orp.com for this kind of planning.
Wow, I am really impressed with this tool. First time I've heard of it, thanks for mentioning it! It's providing strong support for my "what-if" scenario analyses. And I really enjoyed its recommendations for Roth conversions!

I have a question about the VPW Retirement worksheet:

I noticed that worksheet uses long-term equity performance of 5.0%, consistent with CSRI's long-term returns for World stocks. If I wanted to explore performance at 6.4% (CSRI's long-term return for US stocks), is this as simple as modifying the single cell in the Lists worksheet (B132, "Growth Trends, Stocks")? Or are additional modifications necessary?

Prebuttal: Yes, I understand that 6.4% might not be a good measure of future US stock returns. I'm asking a technical question about the worksheet.

User avatar
Peculiar_Investor
Posts: 1574
Joined: Thu Oct 20, 2011 12:23 am
Location: Staying home - Calgary, AB
Contact:

Re: Variable Percentage Withdrawal (VPW)

Post by Peculiar_Investor » Sun Feb 16, 2020 9:08 am

Following through the connecting dots in an article in my newsfeed I found One Portfolio Risk To Rule Them All which specifically mentions
Strategy #1: Dynamic Withdrawals
Flexible spending strategies decrease withdrawals after a portfolio falls in value. Two widespread methods are Guyton-Klinger’s guardrails and the Bogleheads VPW.
Normal people… believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet. – Scott Adams

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sun Feb 16, 2020 2:05 pm

Peculiar_Investor wrote:
Sun Feb 16, 2020 9:08 am
Following through the connecting dots in an article in my newsfeed I found One Portfolio Risk To Rule Them All which specifically mentions
Strategy #1: Dynamic Withdrawals
Flexible spending strategies decrease withdrawals after a portfolio falls in value. Two widespread methods are Guyton-Klinger’s guardrails and the Bogleheads VPW.
Thanks for the link.

Unfortunately, it's another misleading article that fails to disclose that it back-tested VPW with a high stock allocation and without any pension income:
For example, the graph below shows the inflation-adjusted withdrawals of a 1966 retiree. The orange and grey lines show that these strategies led to a 60% decrease in withdrawals:

[graph]

Cutting expenses by 60% in retirement is not a realistic option for most people.
That's fear mongering.

Here's a more complete picture. It includes both the accumulation and retirement phases, putting the much milder reduction in total retirement income into context using VPW on a balanced 60/40 stock/bond portfolio along with a Social Security pension. It's an earlier post of this thread where I was illustrating the VPW accumulation and retirement approach and comparing it to an inflexible approach (fixed 10% of income savings rate, constant inflation-adjusted withdrawal amount equal to 4% of the initial retirement portfolio):
longinvest wrote:
Mon Jun 17, 2019 11:55 am
And, here's the retirement in 1966 scenario:

Image

Note that the person using the VPW worksheet stopped portfolio contributions during the last decade of work, from 1956 to 1965, except for a $1,392 contribution in 1958.
We see that the 1966 VPW retiree (blue line) had been used to living on ($60,000 salary - $3,180 Social Security contribution) = $56,820 during the last few years before retirement (when savings were stopped because of high market returns). After retirement, nominal income kept mostly increasing (not shown on the graph), but inflation-adjusted income dropped, mostly due to inflation, and oscillated between $45,000 and $50,000 for approximately 10 years. That's a 20% drop in inflation-adjusted income relative to what the retiree was used to during the last working years, in one of the worst historical retirement scenarios of US history.

Note that the fixed 4% approach (red line) increased spending during retirement and then bankrupted the retiree before death. It's really a bad idea to target a portfolio equal to 25 times expenses and retire as soon as it's reached with a so-called fixed 4% "Safe" Withdrawal Rate (SWR) approach. It makes the retiree vulnerable to retiring with an illogical withdrawal method at the top of a bubble.

What the article doesn't discuss, either, is other scenarios, like retiring in 1982 (one of the best retirement scenarios*) , or retiring in 1989 (a more normal scenario). The linked posts include proper overall accumulation and retirement charts of available (pre-tax) income, after savings during accumulation, and including the Social Security pension during retirement.

* It's also one of the dumbest scenarios for a 4% SWR retirement!
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

travelintime
Posts: 7
Joined: Sun Mar 15, 2020 7:04 pm

Re: Variable Percentage Withdrawal (VPW)

Post by travelintime » Tue Mar 17, 2020 10:34 pm

I like the smoothing out idea of the monthly withdrawal approached combined with a dampening account.

I also like the idea of starting each year with a one- time, annual amount.

Do you see any drawback to the following approach:

To create the annual amount for January 2021, start a year early, and in January 2020, withdraw the monthly amount suggested by the VPR spreadsheet and it run through the dampening account. Add that monthly amount into a 2021 Income holding account.

Rinse and repeat monthly for February through December.

Then in January, the amount available for use in 2021, is the total sum of the January 2020 – December 2020 monthly withdrawals.

My goal would be two-fold
  • 1) Even out the yearly withdrawal amount, based on the monthly withdrawals, instead of the performance on the market on one day – January 1st 2020

    2) Instead of waiting until January 1st, 2021 to find out our yearly income, we would see it building over the course of the year.
Would this make sense?

Is there a better way to accomplish this?

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Wed Mar 18, 2020 8:31 am

travelintime wrote:
Tue Mar 17, 2020 10:34 pm
I like the smoothing out idea of the monthly withdrawal approached combined with a dampening account.

I also like the idea of starting each year with a one- time, annual amount.

Do you see any drawback to the following approach:

To create the annual amount for January 2021, start a year early, and in January 2020, withdraw the monthly amount suggested by the VPR spreadsheet and it run through the dampening account. Add that monthly amount into a 2021 Income holding account.

Rinse and repeat monthly for February through December.

Then in January, the amount available for use in 2021, is the total sum of the January 2020 – December 2020 monthly withdrawals.

My goal would be two-fold
  • 1) Even out the yearly withdrawal amount, based on the monthly withdrawals, instead of the performance on the market on one day – January 1st 2020

    2) Instead of waiting until January 1st, 2021 to find out our yearly income, we would see it building over the course of the year.
Would this make sense?

Is there a better way to accomplish this?
Travelintime,

Taking annual withdrawals would reduce the effectiveness the the withdrawal cushion. Here's what I previously wrote about this:

(Note that this original post about monthly withdrawals used more complex "monthly silos" as withdrawal cushion. This has been simplified with the help of Bogleheads members.)
longinvest wrote:
Sat Jun 29, 2019 11:58 pm
I've simulated annual and monthly VPW withdrawals on a 50/50 stocks/bonds portfolio, with stocks divided evenly between domestic (US) and international (ex US) from 1998 to 2018. The initial withdrawal is taken on December 31, 1997 at age 65 from a $1,000,000 portfolio. I've used cash monthly returns from Portfolio Visualizer as savings account returns.

Source data for simulations: Portfolio Visualizer

In the following figure, the black line represents monthly income (including interest) derived from annual VPW withdrawals taken on December 31. The red line represents monthly VPW withdrawals, and the green line represents monthly income from the withdrawal cushion.

Image

While monthly income from the withdrawal cushion is smoother than the other two, it remains obvious that the major fluctuations (up in the late 1990s, down in the early 2000s, up in the mid 2000s, significantly down in 2008-2009, and then up) are determined by portfolio returns. It's the smaller short-term fluctuations that are significantly dampened by the withdrawal cushion.

It's worth zooming on the 2008-2010 period:

Image

The drop in monthly income from annual VPW withdrawals was 24% from $5,829 in December 2008 to $4,416 in January 2009. The most severe part of portfolio losses happened in September and October 2008, just a few months before the next VPW withdrawal. The retiree was somehow lucky; had the next withdrawal happened at the end of February 2009, it would have been lower.

That's were the withdrawal cushion shines:
  1. It reduces the level of "luck" involved. By taking smaller monthly withdrawals and averaging them, retirement incomes fluctuations get less dependent on the luck of the portfolio's balance on withdrawal day, and mostly dependent on the more fundamental trend of market returns.
  2. [...]
I think that the volatility of current markets easily justify why we wouldn't want to take the entire annual withdrawal in a single transaction on one specific day of the year. There's safety in breaking the annual withdrawal into 12 smaller distinct withdrawals on 12 different days of the year.

I've illustrated in this thread my preferred approach to generate monthly income. One of the most important parts is to maintain at all times the Required Flexibility suggested by the worksheet:
longinvest wrote:
Sat Mar 07, 2020 8:47 am
Forward test as of February 29, 2020
[...]
We update the Portfolio Balance cell in the Retirement sheet of the VPW Accumulation And Retirement Worksheet. No other entry needs updating. We get:

Image
longinvest wrote:
Sat Mar 14, 2020 8:55 am
The VPW worksheet also calculates a Required Flexibility that must be maintained by the retiree. To do so, it first applies a -50% loss to the stocks allocation and then repeats its calculations. With 60/40 stocks/bonds allocation, this results into a (-50% X 60%) = -30% portfolio loss. That's a (-30% X $977,833) = -$293,350 portfolio loss, reducing the portfolio to ($977,833 - $293,350) = $684,483 after the loss. This implies a (($24,384 - $4,116 + (($684,483 - $113,414) X 5.0%)) / 12) = $4,068 monthly portfolio withdrawal which represents a ($4,068 - $5,291) = -$1,223 reduction after the loss.

The retiree must maintain the flexibility to easily cut spending by up to -$1,223/month because stocks could easily lose -50% of their value within a short time period. In other words, at least $1,223 must be budgeted for optional discretionary spending that could be eliminated without affecting the retiree's comfort.
In summary, I think that your proposal of a single annual withdrawal would lose some significant advantages of taking 12 smaller monthly withdrawals and, in the process, reduce the effectiveness of the withdrawal cushion. I think that, instead, the retiree should maintain the required flexibility suggested by the VPW Retirement Worksheet.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

travelintime
Posts: 7
Joined: Sun Mar 15, 2020 7:04 pm

Re: Variable Percentage Withdrawal (VPW)

Post by travelintime » Wed Mar 18, 2020 8:43 pm

Thank you for the thoughtful response. I may be thinking of this incorrectly, but in my above scenario, I will be taking out monthly withdrawals from the portfolio. I will just be saving the monthly withdrawals up and spending them a year later.

Meaning, a year prior to commencement of retirement, I will begin taking monthly withdrawals, and holding the accumulated money to form my annual income for the next year. Starting in January, I would slowly be building up my annual income for the next year. As I rinse and repeat, this would result in an ongoing, monthly withdrawal from a year prior to retirement, onward.

It seems that the strength of the variable percentage withdrawal method is in the withdrawal. One wouldn't be required to spend the money immediately in order to experience the benefit of the approach. Of course, if one wasn't going to spend the money, the question would be, why withdraw it in the first place. But it seems that at some conceptual level, the withdrawal is disconnected from the spending of the withdrawal. I get that the goal is to even out cash flow, and end result is paying what will likely be monthly bills.

We find a benefit to knowing our annual income prior to the start of the year. Short of picking one random day to withdraw an annual amount, it seems like creating a yearly amount, out of the averaging effect of monthly withdrawals would be helpful. But people on this site have way more experience on this, so I am willing to learn and immediately disassociate myself my comments :happy

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Thu Mar 19, 2020 7:16 am

travelintime wrote:
Wed Mar 18, 2020 8:43 pm
Thank you for the thoughtful response. I may be thinking of this incorrectly, but in my above scenario, I will be taking out monthly withdrawals from the portfolio. I will just be saving the monthly withdrawals up and spending them a year later.

Meaning, a year prior to commencement of retirement, I will begin taking monthly withdrawals, and holding the accumulated money to form my annual income for the next year. Starting in January, I would slowly be building up my annual income for the next year. As I rinse and repeat, this would result in an ongoing, monthly withdrawal from a year prior to retirement, onward.

It seems that the strength of the variable percentage withdrawal method is in the withdrawal. One wouldn't be required to spend the money immediately in order to experience the benefit of the approach. Of course, if one wasn't going to spend the money, the question would be, why withdraw it in the first place. But it seems that at some conceptual level, the withdrawal is disconnected from the spending of the withdrawal. I get that the goal is to even out cash flow, and end result is paying what will likely be monthly bills.

We find a benefit to knowing our annual income prior to the start of the year. Short of picking one random day to withdraw an annual amount, it seems like creating a yearly amount, out of the averaging effect of monthly withdrawals would be helpful. But people on this site have way more experience on this, so I am willing to learn and immediately disassociate myself my comments :happy
(I added the emphasis).

Travelintime, I think that you've correctly identified that your question isn't a withdrawal approach question; it's a budgeting question.

I suggest thinking of monthly income generated from monthly VPW withdrawals with a withdrawal cushion as being similar to a monthly pre-tax employment income (including occasional bonuses and payments for overtime). During accumulation, we don't know one year in advance our exact upcoming inflation-adjusted income payments due to various uncertainties:
  • uncertainty about inflation in the upcoming months
  • uncertainty about upcoming overtime hours worked
  • uncertainty about future bonuses
  • uncertainty about the outcome of future salary negotiations
  • uncertainty about unplanned downtime (like many employees are currently experiencing in various businesses)
  • etc.
I suggest to budget similarly to the way it was done during accumulation. If additional certainty is required, it can be bought. It's called a (joint, for a couple) cost-of-living adjusted life SPIA*. It's expensive! It's an insurance product that exchanges liquidity for stable cost-of-living adjusted lifelong income. Buying it is an irreversible transaction. One should never buy more of it than strictly necessary. It's usually suggested to wait until age 80 and (when necessary) only buy as little of it as required to dampen the financial risk associated with living beyond age 100 (when the investment portfolio could be almost depleted). There's a lot of value to keeping as much liquidity as possible (e.g. keeping an investment portfolio) because of the flexibility it provides, especially during the earlier and most active years of retirement.

* Single Premium Immediate Annuity.

Stable cost-of-living adjusted lifelong income can be increased by delaying Social Security until age 70. The calculations of the VPW Accumulation And Retirement Worksheet take into account current and delayed pensions with and without cost-of-living adjustments.

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.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

travelintime
Posts: 7
Joined: Sun Mar 15, 2020 7:04 pm

Re: Variable Percentage Withdrawal (VPW)

Post by travelintime » Fri Mar 20, 2020 12:25 am

Thank you for the responses.

Perhaps I can get at something from a different perspective. Assuming one wanted to take out an annual withdrawal (which I think is how the VPW Worksheet was originally set up). Besides picking one random day of the year, is there a better way to do it

For example, let's say I pick July 1st as the start of my annual budget year.

A) I could take 100% of my annual withdrawal on July 1st of each year. Everything will ride or fall with the portfolio balance of that day. I'm guessing the singular withdrawal was the practice of those following the original spreadsheet.

B) Another option is, I could take half the withdrawal on June 30th and half the withdrawal a day later on July 1st. I might get an ever so tiny, perhaps statistically insignificant diversification of connecting the withdrawals to two random steps along the random walk of Wall Street, not one random step. On a practical matter, I’m guessing some people may need to take their annual withdrawal amount over the course of several financial transactions, perhaps extending over the course of several days)

C) I could take an amount equal to one 1/4 of the annual withdrawal percentage on June 1st, an amount equal to one 1/4 of the annual withdrawal percentage on June 8th, an amount equal to one 1/4 of the annual withdrawal percentage on June 15th, an amount equal to one 1/4 of the annual withdrawal percentage on June 22nd, and now I have money in the bank and I'm ready to go on July 1st. Would these four separate withdrawals provide any diversification benefit towards the idea of averaging out the effect of one singular withdrawal based on one singular day?

D) If these withdrawals became more numerous and more spread out, we could get to the 12 monthly withdrawals I originally asked about.

So if one wanted to have an annual amount to spend, maybe the best is option A) Just pick a date and go for it. If not, is there a better way to get at the annual withdrawal?

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Fri Mar 20, 2020 6:23 am

travelintime wrote:
Fri Mar 20, 2020 12:25 am
A) I could take 100% of my annual withdrawal on July 1st of each year. Everything will ride or fall with the portfolio balance of that day. I'm guessing the singular withdrawal was the practice of those following the original spreadsheet.
...
So if one wanted to have an annual amount to spend, maybe the best is option A) Just pick a date and go for it.
Many VPW users do exactly that. :D
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

retiringwhen
Posts: 1743
Joined: Sat Jul 08, 2017 10:09 am
Location: New Jersey, USA

Re: Variable Percentage Withdrawal (VPW)

Post by retiringwhen » Fri Mar 20, 2020 6:36 am

Longinvest will surely respond, but I will comment that am not sure how you handled the response to market conditions part of the equation in options B and C...

VPW looks simple because the execution model has been cleverly simplified, but the underlying principle is somewhat complex and very powerful.

To summarize, VPW assumes you can take out more (as in a higher portfolio percentage) now than later because you are taking the risk of lower withdrawals in the future if the market goes down. This is very different that the SWR approach were you essentially lock in a rate for the duration that is calculated to be essentially worst case (if anyone really does that!).

I believe you are simply asking just how often you can respond to that market fluctuations. If you reset the withdrawal rate at each withdrawal point, you are in the spirit of the VPW model (heck in theory, you could do daily withdrawals for that matter if you divide the rate by 365!!).

But, if you set the value once for a year and take the money out over 12 mos (in any partial increments), you are now in uncharted territory since you have disconnected withdrawals with valuations. I would hazard a guess that if you do that and model it, the results will be a much LOWER withdrawal rate since you will in times like this month be withdrawing much reduced value funds based upon a previously higher valuation calculation.

The smoothing approach in the VPW sheet is a happy middle where you use the previous 6 mos. valuations to set today's withdrawal rate to help with smoothing without impacting the core principle of the VPW calculations (every month you respond to the current valuations.)

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Fri Mar 20, 2020 8:11 am

Retiringwhen, I suggest that you open a thread in the Personal Investments forum using the format suggested in this post. Other forum members will be able to help you identify the best approach for using VPW within your personal retirement plan.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

dknightd
Posts: 2164
Joined: Wed Mar 07, 2018 11:57 am

Re: Variable Percentage Withdrawal (VPW)

Post by dknightd » Fri Mar 20, 2020 9:54 am

There are many ways to slice a loaf of bread. What I like about the various VPW spread sheets is they estimate what my future SS and other future income might be worth in today's dollars. There are a lot of assumptions involved in doing this, and there a lot of assumptions in what future growth of our portfolios might be.

What I want to do is once a year try to figure out what I can spend that year. This is my first year being retired, so my method might change. So I ran some numbers in the VPW 1.4 spreadsheet (among others).

I do not want my nominal income to go down, ever. I hope it goes up to keep up with inflation. I figure inflation varies slowly, so I can adjust, hopefully, maybe.

On jan 1 2020 I picked my withdrawal number for this year. I used just above the "required flexibility" number. So essentially I assumed stocks will drop 50% this year, instead of going up as usual. This will be enough for us to be comfortable. I actually have a range since I did many different ways of doing this, I'm OK planning to take the low of this range out this year. It is not likely we'll be taking a world cruise this year . . .

By withdrawing at the lower number now, I'm less likely to have to one year take out less in the future. Next year I'll reassess.
If you value a bird in the hand, pay off the loan. If you are willing to risk getting two birds (or none) from the market, invest the funds.

ososnilknarf
Posts: 72
Joined: Mon Feb 18, 2019 7:08 pm

Re: Variable Percentage Withdrawal (VPW)

Post by ososnilknarf » Fri Mar 20, 2020 8:48 pm

I've been reading up on this VPW approach, and I think it is an ideal retirement strategy for our situation, as we do plan to have a significant SS income at 70 to use a base income. I am wondering though about the background of the spreadsheet, and how it comes up with its number and how that differs from the simple table based approach.
This is all really good information and I am super appreciative of the work that went into it. It has given me a lot of confidence in my plan, and that my retirement looks like it will be feasible.
One thing though is that it seems that the ideas have evolved and there are so many posts and so much information scattered everywhere throughout the thread and in the wiki, it is somewhat hard to really get a good grasp of what is current and what is outdated and has been improved upon.
This should really be organized into a book or something.

User avatar
canadianbacon
Posts: 127
Joined: Sun Nov 10, 2019 10:04 pm

Re: Variable Percentage Withdrawal (VPW)

Post by canadianbacon » Sat Mar 21, 2020 12:52 pm

This recent period has been a good stress test for the approach. My annual withdrawal was just over 29K a month ago. It's now sitting at 22.8K. Were I retired, I would be able to stick to 22.8K but it would certainly be tight. My original "target" was 35K to feel financially independent, so I'm a ways off on that now.
Bulls make money, bears make money, pigs get slaughtered.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Wed Apr 08, 2020 5:52 pm

In the thread "The One-Fund Portfolio as a default suggestion", forum member Atticus713 asked questions related to VPW. I'll answer them here.

Atticus713 wrote:
Wed Apr 08, 2020 6:29 am
Would you be so kind as to further explain in more detail the mechanics and thought process behind the “***footnote” you mentioned in your initial post?
In the first post of that thread, I wrote the following:
longinvest wrote:
Mon Aug 12, 2019 8:10 am
My personal preference is for portfolio 1, representing a globally-diversified lifelong 60/40 stocks/bonds allocation because I consider that all investment assets are risky, but in different ways. I think that it's best to broadly diversify across them all lifelong***.

*** In retirement, combining variable portfolio withdrawals with Social Security (possibly delayed to age 70) and a pension (if any) often results into mild total income fluctuations. When necessary, Total Retirement Income fluctuations can be further dampened by using a small part of the portfolio to buy an inflation-indexed Single Premium Immediate Annuity (SPIA) instead of increasing the bond allocation above 40%.

Atticus713, if you're unfamiliar with the variable percentage withdrawal (VPW) retirement approach, I suggest to first read a summary in this post.

Once the basics of the VPW retirement approach are understood, I suggest reading two posts I wrote in the "60/40 VS 30/70 -- Bernstein VS Ferri" thread:
  • a first post which compares Bernstein's 60/40 with Ferri's 30/70 allocations while using VPW during retirement, and
  • a second post which answers your question:
    longinvest wrote:
    Wed Jul 24, 2019 10:04 am
    I think that the magnitude of portfolio withdrawal fluctuations should be considered in context of the size of total retirement income including Social Security (possibly delayed), pension (if any), and (if necessary) inflation-indexed SPIA* relative to the retiree's financial needs for a comfortable retirement.

    * Single Premium Immediate Annuity.

    Note that the baseline, here, isn't a bare-bones retirement but a comfortable one!

    Let's pick the example of a retiree who owns her home free and clear who could live comfortably on $80,000 (including taxes and expenses). By comfortable, I mean expenses that include travel and other fun things.

    The retiree has just turned 65. She worked from age 25 to 64. Her salary started at $11,532 in 1979 ($40,687 in today's dollars) and gradually increased to $115,560 in 2018 ($117,878 in today's dollars). She has decided to delay Social Security until age 70 to receive $3,334/month (in today's dollars) and has accumulated a $1,500,000 portfolio through investing and inheritance. She has no work pension.

    She is used to living below her means and could live on less than $80,000 (including taxes), if it comes to it, but it wouldn't be as comfortable.

    With a 60/40 stocks/bonds portfolio, the VPW Accumulation And Retirement Worksheet tells us that she could withdraw $105,661 from the portfolio in 2019 (taking into account the future $40,008 annual income from Social Security). A -50% stocks loss would lower this to $83,176, still above her comfortable retirement needs.

    With a 30/70 stocks/bonds portfolio, she could withdraw $97,864 from the portfolio in 2019 and a -50% stocks loss would lower this to $87,932.

    The difference, in 2019, is ($105,661 - $97,864) = $7,797 (including taxes), which could allow for quite some additional fun while she is younger and still healthy. That's 8% more than with a 30/70 stocks/bonds portfolio.

    The difference after a -50% stocks loss would be ($83,176 - $87,932) = -$4,756 (including taxes). That's -5% less than with a 30/70 stocks/bonds portfolio. As I wrote earlier, it's still within the comfortable retirement zone.

    Let's also investigate the doomsday scenario of a -75% stocks loss (e.g. two 50% stocks losses). With a 60/40 stocks/bonds portfolio (reduced to $735,000) total retirement income would be $67,436. With a 30/70 stocks/bonds portfolio (reduced to $1,083,750) total retirement income would be $79,490. There's no surprise, here. The 30/70 stocks/bonds portfolio is more resilient to doomsday scenarios, yet, the 60/40 stocks/bonds portfolio still delivers sufficient retirement income, even though it's -15% below full comfort level. Remember that we're in a Great Depression (1930s) kind of situation, here. The rest of society is suffering.

    All assets are risky. Stocks are significantly more volatile than bonds, but bonds have their own risks. A good argument could be made that the world bond stock portfolio is probably one of the most resilient portfolios in the long term. It currently sits at approximately 57/43 stocks/bonds according to the Bill Sharpe's preferred portfolio thread. A retiree could easily prefer this portfolio over one concentrated into the bonds asset class. I personally have a liking for Vanguard's LifeStrategy Moderate Growth Fund (VSMGX) which provides a good enough approximation of this stock-and-bond market portfolio with a moderate home bias.**

    ** A moderate home bias is justified for investors, across the world, by the additional frictions of international investing.

    Instead of using a 30/70 stocks/bonds portfolio to reduce income fluctuations (if she wanted to reduce them), our retiree could use part of her portfolio to buy an inflation-indexed SPIA. With a 30/70 stocks/bonds allocation, her $1,500,000 portfolio would have $450,000 allocated to stocks. If she wanted to keep $450,000 in stocks with a 60/40 stocks/bonds allocation, she could retain a $750,000 portfolio and use the other $750,000 to buy an inflation-indexed SPIA which should pay her approximately $2,553/month. She would be reducing her liquidity by 50% but significantly increasing he guaranteed lifelong income. Her 2019 income would be $98,582 and a -50% stocks loss would reduce this to $87,340. The a doomsday -75% stocks loss would result into a $79,740 income; still comfortable.

    One could easily argue that it's more efficient to annuitize and keep a market-like portfolio than adopting an allocation concentrated into bonds. Yet, our retiree could prefer to keep both the additional liquidity and higher income in 2019 with a 60/40 portfolio and wait until age 80 to annuitize part of her portfolio to dampen the financial risk of living beyond age 100.

    "There is more than one road to Dublin." -- Taylor Larimore

As for your other question:
Atticus713 wrote:
Wed Apr 08, 2020 6:29 am
Can you also talk about what an SPIA is and why, later in life, you might use some of the fund to purchase that?—in contrast with exchanging the 60/40 LifeStrategy Fund for one of the more conservative (40/60 or 20/80) iterations at around age 70?
An inflation-indexed Single Premium Immediate Annuity (SPIA) is a product which can be bought from an insurance company. It's like buying a pension. In exchange for a lump sum of money, the insurance company pays the buyer (and their spouse, if it's a joint-life annuity) a monthly amount for life (until both spouses are dead, if it's a joint-life annuity). Assuming a 2%-indexed SPIA was bought, the monthly payment increases by 2% every year to compensate for the increase in cost of living.

Earlier in this thread, I wrote a detailed post to explain why it's suggested to buy an inflation-indexed SPIA at age 80 (when necessary) to dampen the financial risk associated with living beyond age 100.

A SPIA indexed to the consumer price index for all urban consumers (CPI-U) might be difficult to find. This post suggests that an easy-to-find 2%-indexed SPIA is good enough as it's indexed to promised inflation.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

Atticus713
Posts: 15
Joined: Sat May 25, 2019 9:24 am

Re: Variable Percentage Withdrawal (VPW)

Post by Atticus713 » Wed Apr 08, 2020 10:28 pm

Thank you for your helpful and detailed response to my questions—very much appreciated. I’ll go to the other threads and links you provided and will post any follow-up questions. Stay well. Best,
Atticus713

User avatar
KEotSK66
Posts: 340
Joined: Wed Mar 11, 2020 7:03 pm

longinvest

Post by KEotSK66 » Thu Apr 09, 2020 8:49 am

hi

re constant dollars, if you don't assume high volatility and get reasonable income i don't think srr is high, it's certain you'll have to sell low at points throughout retirement but that comes with the territory, you have to feel you'll recover in the future. getting income >= draw is important

i plan to use constant dollars, with a needed return (positive cash flow) target sufficient to grow the portfolio with inflation and so get my real draw. i'm still 4 years from retiring so my plan could change
"i just got fluctuated out of $1,500", jerry

jeroly
Posts: 101
Joined: Wed Jan 24, 2018 2:07 pm

Re: Variable Percentage Withdrawal (VPW)

Post by jeroly » Thu Apr 09, 2020 11:04 am

First of all, thanks to longinvest and everybody that helped develop the VPW toolset. It's a very cool spreadsheet in its current form.

Two separate points in this post:

1.

I'd like to add (? didn't read every post in this seven year long thread so maybe this is redundant) that I understand the theory behind buying an inflation-indexed annuity at around age 80 but think in practice it's not that great an idea.

It makes sense to pay more for an inflation indexed annuity so you don't wind up eating dog food in your 90's should inflation rear its ugly head. However, doesn't the risk really kick in when inflation gets much bigger than it currently is? To me, buying an ordinary annuity seems fine when inflation is under 3% as data seems to show that annual spending in that age bracket drops about 2% annually anyway. Moreover, these days, aren't inflation indexed annuities capped at something like 3% annual inflation adjustments? So they don't really protect you against the real risk anyway...

___________________________________________________
2.

I have started to use the VPW methodology as a way to reassure myself about my annual withdrawals being sustainable, and as a gauge to whether I could splurge on a big irregular purchase such as a car, or a large gift to my daughter, without feeling I'm putting my retirement security at risk.

Here's how I'm approaching it: I use the VPW methodology to calculate what my annual withdrawal could safely be. Let's say for example that it comes up with $110,000 as the annual withdrawal amount. I wind up spending say $90,000 on living expenses over the course of the year. I then consider myself to have a 'slush fund' of $20,000 which I could, if I wanted to, spend at will, but for the sake of argument let's assume I don't. So for the next year's calculations I subtract that $20k from the amount used to calculate the VPW (let's assume it comes up with $115,000) and thus have up to $135k to spend in year 2, and so forth, with the goals being to smooth out withdrawals in the event of a market crash as well as to feel at liberty to spend a chunk of cash when necessary without worrying.

Comments and criticisms appreciated!

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat Apr 11, 2020 10:36 am

jeroly wrote:
Thu Apr 09, 2020 11:04 am
It makes sense to pay more for an inflation indexed annuity so you don't wind up eating dog food in your 90's should inflation rear its ugly head. However, doesn't the risk really kick in when inflation gets much bigger than it currently is? To me, buying an ordinary annuity seems fine when inflation is under 3% as data seems to show that annual spending in that age bracket drops about 2% annually anyway. Moreover, these days, aren't inflation indexed annuities capped at something like 3% annual inflation adjustments? So they don't really protect you against the real risk anyway...
I've just discussed this three days ago:
longinvest wrote:
Wed Apr 08, 2020 5:52 pm
Earlier in this thread, I wrote a detailed post to explain why it's suggested to buy an inflation-indexed SPIA at age 80 (when necessary) to dampen the financial risk associated with living beyond age 100.

A SPIA indexed to the consumer price index for all urban consumers (CPI-U) might be difficult to find. This post suggests that an easy-to-find 2%-indexed SPIA is good enough as it's indexed to promised inflation.
I suggest reading the linked detailed post.

Also, in its explanations of how to use variable percentage withdrawals during retirement with the VPW Accumulation And Retirement Worksheet, our wiki says:
3. At age 80, if you're still alive, it's important to consider using part (but not all) of your remaining portfolio to buy an inflation-indexed Single Premium Immediate Annuity (SPIA), so that the estimated Income Floor After 100 is sufficient to live comfortably, independently of future portfolio withdrawals. This aims to reduce the financial risks associated with living past age 100.
The inflation-indexed SPIA addresses the financial risk associated with living beyond age 100, when the retiree's portfolio will be mostly depleted.

If the retiree buys a 2%-indexed SPIA (indexed to promised inflation) and the Federal Reserve keeps its promise to maintain inflation on a 2% target during the retiree's lifetime, the purchase power of 2%-indexed SPIA payments will be preserved.

If inflation runs higher, let's say 3%, the purchase power of 2%-indexed SPIA payments at age 100 will be reduced by (((1.02 / 1.03)^20) - 1) = -18%. Payments will have preserved 82% of their purchase power. If inflation persists around 3%, payments will continue losing a small -1% per year in purchase power. The retiree won't end up eating cat food under a bridge (figuratively) while still alive.

If the retiree had done as you're suggesting, instead, and bought a fixed SPIA with no indexing, the purchase power would be reduced by -45% at age 100, that's almost half, and would continue falling by an additional -3% every year!

I think that it's pretty obvious that even if SPIA indexed to promised inflation (2%) might not always perfectly match future CPI-U inflation, it's still much better than a SPIA with no indexing at dampening the financial risk associated with living beyond age 100.
jeroly wrote:
Thu Apr 09, 2020 11:04 am
Let's say for example that it comes up with $110,000 as the annual withdrawal amount. I wind up spending say $90,000 on living expenses over the course of the year. I then consider myself to have a 'slush fund' of $20,000 which I could, if I wanted to, spend at will, but for the sake of argument let's assume I don't. So for the next year's calculations I subtract that $20k from the amount used to calculate the VPW (let's assume it comes up with $115,000) and thus have up to $135k to spend in year 2, and so forth, with the goals being to smooth out withdrawals in the event of a market crash as well as to feel at liberty to spend a chunk of cash when necessary without worrying.
Let me illustrate what you're suggesting in a way that might help you consider not doing it.

At the top of a bubble, when VPW delivers $20,000 more than you need, you take the suggested portfolio withdrawal and reinvest $20,000 into the bubbly portfolio (that's mathematically identical to not having withdrawn the $20,000 in the first place). Then, the following year, after the portfolio has lost 50% in a severe crash (and the reinvested $20,000 has shrunk along with it to $10,000) you take the suggested VPW withdrawal based on the after-crash balance (which includes the $10,000) and also make an additional $20,000 withdrawal, just because you had reinvested $20,000 the previous year. Do you see the problem, now?

Here's what I intend to do: If VPW delivers more money than I need, I'll put some of it aside in a high-interest savings account for unspecified future spending, use some of it for unplanned fun (extra travel, etc.), and take advantage of the opportunity to give money in a helping way to loved ones and causes (education. etc.), enjoying the process while still alive.

The best time to withdraw more money from a portfolio is when the portfolio is bigger. The best time to withdraw less money from a portfolio is when the portfolio is smaller. This is exactly what VPW naturally does.

Most of the time, when a user feels the need to smooth VPW withdrawals, it's because the portfolio is too volatile or because there is insufficient stable non-portfolio income. The solution to this problem is obvious, it's to increase the allocation to bonds or to increase the amount of stable non-portfolio income (e.g. delay Social Security or buy a small inflation-indexed SPIA).

I suggest to read this thread which illustrates how to create monthly retirement income using the VPW worksheet while dampening short-term income fluctuations* using a small withdrawal cushion. Note that VPW portfolio withdrawals aren't smoothed; it's the resulting monthly income which is smoothed.

* The withdrawal cushion doesn't smooth long-term income fluctuations which are caused by large portfolio fluctuations.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

Willmunny
Posts: 99
Joined: Mon Jun 16, 2014 6:35 pm

Re: Variable Percentage Withdrawal (VPW)

Post by Willmunny » Sat Apr 11, 2020 7:40 pm

longinvest - I am at least a decade and a half away from retirement, but this system appeals to me and I am going to keep it in mind as I get closer to planning for the withdrawal phase. Thanks for being so generous with your knowledge. Cheers :sharebeer

jeroly
Posts: 101
Joined: Wed Jan 24, 2018 2:07 pm

Re: Variable Percentage Withdrawal (VPW)

Post by jeroly » Wed Apr 15, 2020 8:14 am

longinvest wrote:
Sat Apr 11, 2020 10:36 am
jeroly wrote:
Thu Apr 09, 2020 11:04 am
It makes sense to pay more for an inflation indexed annuity so you don't wind up eating dog food in your 90's should inflation rear its ugly head. However, doesn't the risk really kick in when inflation gets much bigger than it currently is? To me, buying an ordinary annuity seems fine when inflation is under 3% as data seems to show that annual spending in that age bracket drops about 2% annually anyway. Moreover, these days, aren't inflation indexed annuities capped at something like 3% annual inflation adjustments? So they don't really protect you against the real risk anyway...
I've just discussed this three days ago:
longinvest wrote:
Wed Apr 08, 2020 5:52 pm
Earlier in this thread, I wrote a detailed post to explain why it's suggested to buy an inflation-indexed SPIA at age 80 (when necessary) to dampen the financial risk associated with living beyond age 100.

A SPIA indexed to the consumer price index for all urban consumers (CPI-U) might be difficult to find. This post suggests that an easy-to-find 2%-indexed SPIA is good enough as it's indexed to promised inflation.
I suggest reading the linked detailed post.

Also, in its explanations of how to use variable percentage withdrawals during retirement with the VPW Accumulation And Retirement Worksheet, our wiki says:
3. At age 80, if you're still alive, it's important to consider using part (but not all) of your remaining portfolio to buy an inflation-indexed Single Premium Immediate Annuity (SPIA), so that the estimated Income Floor After 100 is sufficient to live comfortably, independently of future portfolio withdrawals. This aims to reduce the financial risks associated with living past age 100.
The inflation-indexed SPIA addresses the financial risk associated with living beyond age 100, when the retiree's portfolio will be mostly depleted.

If the retiree buys a 2%-indexed SPIA (indexed to promised inflation) and the Federal Reserve keeps its promise to maintain inflation on a 2% target during the retiree's lifetime, the purchase power of 2%-indexed SPIA payments will be preserved.

If inflation runs higher, let's say 3%, the purchase power of 2%-indexed SPIA payments at age 100 will be reduced by (((1.02 / 1.03)^20) - 1) = -18%. Payments will have preserved 82% of their purchase power. If inflation persists around 3%, payments will continue losing a small -1% per year in purchase power. The retiree won't end up eating cat food under a bridge (figuratively) while still alive.

If the retiree had done as you're suggesting, instead, and bought a fixed SPIA with no indexing, the purchase power would be reduced by -45% at age 100, that's almost half, and would continue falling by an additional -3% every year!

I think that it's pretty obvious that even if SPIA indexed to promised inflation (2%) might not always perfectly match future CPI-U inflation, it's still much better than a SPIA with no indexing at dampening the financial risk associated with living beyond age 100.
First of all I had read your earlier post and it did not address the specific points that (a) inflation indexed SPIAs don't address the real inflation risk of runaway inflation (thus providing a false sense of security), and (b) you need less at age 100 than you do at age 80 if you are the typical senior.

You pay more for an inflation indexed annuity than an ordinary annuity, so the higher payment at later ages is offset by having to pay more for it. In fact, it's more than offset as you are paying more to the insurance company for their profit (in compensation for the [limited] risk they are taking on), as well as for their expenses such as commissions. Given that (in the low-inflation environment that inflation indexing provides cover for) the requirements for spending (down around 2% a year due to lower spending at higher ages) come reasonably close to matching the discounted payouts of an ordinary annuity (down around 2% a year due to inflation), I see no reason to go for the inflation indexing.
jeroly wrote:
Thu Apr 09, 2020 11:04 am
Let's say for example that it comes up with $110,000 as the annual withdrawal amount. I wind up spending say $90,000 on living expenses over the course of the year. I then consider myself to have a 'slush fund' of $20,000 which I could, if I wanted to, spend at will, but for the sake of argument let's assume I don't. So for the next year's calculations I subtract that $20k from the amount used to calculate the VPW (let's assume it comes up with $115,000) and thus have up to $135k to spend in year 2, and so forth, with the goals being to smooth out withdrawals in the event of a market crash as well as to feel at liberty to spend a chunk of cash when necessary without worrying.
Let me illustrate what you're suggesting in a way that might help you consider not doing it.

At the top of a bubble, when VPW delivers $20,000 more than you need, you take the suggested portfolio withdrawal and reinvest $20,000 into the bubbly portfolio (that's mathematically identical to not having withdrawn the $20,000 in the first place). Then, the following year, after the portfolio has lost 50% in a severe crash (and the reinvested $20,000 has shrunk along with it to $10,000) you take the suggested VPW withdrawal based on the after-crash balance (which includes the $10,000) and also make an additional $20,000 withdrawal, just because you had reinvested $20,000 the previous year. Do you see the problem, now?

Here's what I intend to do: If VPW delivers more money than I need, I'll put some of it aside in a high-interest savings account for unspecified future spending, use some of it for unplanned fun (extra travel, etc.), and take advantage of the opportunity to give money in a helping way to loved ones and causes (education. etc.), enjoying the process while still alive.

The best time to withdraw more money from a portfolio is when the portfolio is bigger. The best time to withdraw less money from a portfolio is when the portfolio is smaller. This is exactly what VPW naturally does.

Most of the time, when a user feels the need to smooth VPW withdrawals, it's because the portfolio is too volatile or because there is insufficient stable non-portfolio income. The solution to this problem is obvious, it's to increase the allocation to bonds or to increase the amount of stable non-portfolio income (e.g. delay Social Security or buy a small inflation-indexed SPIA).

I suggest to read this thread which illustrates how to create monthly retirement income using the VPW worksheet while dampening short-term income fluctuations* using a small withdrawal cushion. Note that VPW portfolio withdrawals aren't smoothed; it's the resulting monthly income which is smoothed.

* The withdrawal cushion doesn't smooth long-term income fluctuations which are caused by large portfolio fluctuations.
For me and I believe a large chunk of the Boglehead population, the 'problem' of needing less than the VPW calculations indicate our portfolios could provide is not based on a 'bubbly' portfolio. The stock portion of my portfolio could drop another 80% from where it is now and I'd still get a recommended withdrawal amount in excess of my needs. I definitely see where the VPW approach could be fantastic for somebody in the accumulation stage, and for overall planning purposes for someone in the decumulation phase, for help in putting all the pieces of pensions, social security, and withdrawals together. (A side note: a very useful addition would be to explicitly take end-of-life care costs into account.) For me, though, it's somewhat of a pacifier, a tool that tells me, "Jeroly, it's okay to make that gift / buy that car / take that trip ... you will still have enough left over." So while I may not spend/withdraw as much as I could, I'm definitely either going to spend more or feel better about what I do spend, thanks to the VPW methodology.

Thank you for pointing out the following:
- If I were taking an investment approach where I were withdrawing funds and then rebalancing to reflect my target asset allocation, keeping the accounting in what amounts to an escrow account while in fact keeping it invested would distort the results from the VPW methodology.

(Side question... when you're coming up with the reserve amount for leveling out income prior to Social Security and/or pensions kicking in, aren't you doing exactly that - treating it like an escrow account while keeping it invested? You are not having people actually withdraw those funds and put them in some investment that pays a guaranteed 5%/year (if only we could!), but they keep that money as part of their overall investments.)

However, I currently am taking a reverse glidepath approach where I do not rebalance and am letting the stock portion of my portfolio 'ride' for the next nine years while taking my withdrawals from the cash and fixed income portions of my portfolio. So your recommendation for putting the VPW method's recommended withdrawal amount into a separate savings account amounts to more or less the same thing, but with tax consequences for taking withdrawals from retirement accounts in excess of the RMDs.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Wed Apr 15, 2020 7:23 pm

Dear Jeroly,
jeroly wrote:
Wed Apr 15, 2020 8:14 am
First of all I had read your earlier post and it did not address the specific points that (a) inflation indexed SPIAs don't address the real inflation risk of runaway inflation (thus providing a false sense of security), and (b) you need less at age 100 than you do at age 80 if you are the typical senior.
Runaway inflation (like in Venezuela) tends to be caused by deep societal and political turmoil. When this happens, it's difficult to protect one's assets from direct or indirect* confiscation, whether financial assets or any other type of assets. You might want to read William Bernstein's Deep Risk for an awesome investigation of the subject.

* Inflation, default on government bonds, understatement of inflation in inflation-linked bonds, taxflation, nationalization of big companies, making foreign investment illegal, etc.

Let's be realistic, here. The US Federal Reserve has proven under Volcker that it can tame inflation when it wishes to do so. Reading about inflation and inflation targeting, one learns that the Federal Reserve is way more worried about deflation!

Lastly, the inflation-indexed SPIA's goal it to dampen the financial risk of living beyond age 100. That's 100, not 80 or 90. So, one calibrates the amount one buys (at age 80) based on needs at age 100 and beyond, not based on needs at age 80 or 90. One buys as little as necessary to dampen, not eliminate, the financial risk. It's a risk that is unlikely to show up for most people. One shouldn't put more money than strictly necessary into an insurance product.

Of course, I'm assuming that the retiree has first delayed Social Security to age 70 to maximize the amount of its inflation-indexed payments. Some retirees won't need to buy additional income. The VPW Retirement Worksheet can conveniently help with filling the gap in payments between retirement and age 70.

You'll have a very hard time convincing me that financial needs keep going down 2% to 3%, relative to inflation, every year from 100 to 101, to 102, to 103, ..., to 110. Yes, there are women who live to that age. Its likelihood for any individual is very low. But, it would be foolish to pay for an insurance product which fails to achieve the protection we're looking for (providing sufficient income) exactly when the risk shows up (the retiree survives beyond age 100).

I'm repeating myself, though. I've explained all this in the posts I've linked to in my previous reply.
jeroly wrote:
Wed Apr 15, 2020 8:14 am
The stock portion of my portfolio could drop another 80% from where it is now and I'd still get a recommended withdrawal amount in excess of my needs.
Congratulations.

Best regards,

longinvest
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

ososnilknarf
Posts: 72
Joined: Mon Feb 18, 2019 7:08 pm

Re: Variable Percentage Withdrawal (VPW)

Post by ososnilknarf » Fri Apr 17, 2020 5:54 pm

I have a question about how to handle a mortgage being paid off using the VPW spreadsheet. Sorry if this has been addressed before, but I didn't see anything about it.
Say I'm retiring at 60, and have a mortgage payment of $2200/month, and it will be paid off at age 65. Is it reasonable to handle this by considering it another Defined Benefit Pension of $2200 that starts at age 65 and isn't indexed to inflation? I'm thinking that it is $2200 that I will no longer need from the portfolio, so it is in effect like another SS or pension payment.

ososnilknarf
Posts: 72
Joined: Mon Feb 18, 2019 7:08 pm

Re: Variable Percentage Withdrawal (VPW)

Post by ososnilknarf » Tue Apr 21, 2020 10:01 pm

ososnilknarf wrote:
Fri Apr 17, 2020 5:54 pm
I have a question about how to handle a mortgage being paid off using the VPW spreadsheet. Sorry if this has been addressed before, but I didn't see anything about it.
Say I'm retiring at 60, and have a mortgage payment of $2200/month, and it will be paid off at age 65. Is it reasonable to handle this by considering it another Defined Benefit Pension of $2200 that starts at age 65 and isn't indexed to inflation? I'm thinking that it is $2200 that I will no longer need from the portfolio, so it is in effect like another SS or pension payment.
Bump.
Can anyone confirm my question? Just would like to know if I'm thinking about this correctly, or if there is some flaw with using the VPW spreadsheet in this way.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Wed Apr 22, 2020 6:34 am

ososnilknarf wrote:
Tue Apr 21, 2020 10:01 pm
ososnilknarf wrote:
Fri Apr 17, 2020 5:54 pm
I have a question about how to handle a mortgage being paid off using the VPW spreadsheet. Sorry if this has been addressed before, but I didn't see anything about it.
Say I'm retiring at 60, and have a mortgage payment of $2200/month, and it will be paid off at age 65. Is it reasonable to handle this by considering it another Defined Benefit Pension of $2200 that starts at age 65 and isn't indexed to inflation? I'm thinking that it is $2200 that I will no longer need from the portfolio, so it is in effect like another SS or pension payment.
Bump.
Can anyone confirm my question? Just would like to know if I'm thinking about this correctly, or if there is some flaw with using the VPW spreadsheet in this way.
Dear Ososnilknarf,

This is an indirect question about leveraged investing; that's why it's not attracting answers. It isn't a VPW question. The obvious answer is to withdraw the money from the portfolio and immediately pay off the mortgage. If that's not technically possible due to the mortgage contract (couldn't one just pay a penalty and pay off the mortgage?), one could just put the money aside into a high-interest savings account from which mortgage payments will be taken during the 4 or 5 years left on the mortgage.

Just in case this isn't obvious to you, let me show the situation from a different angle. A person owns a home free and clear (no debt) and retires. Why would that person go to the bank and take a mortgage on the home? That would be strange, right? Yet, after having taken the mortgage, the new retiree would end up in the same situation you're describing.

The usual approach is to pay off the mortgage before retirement. When that's not possible (like in case of a forced retirement with too small a portfolio), it's often because the retiree is house poor. The solution to being house poor is to sell the home and buy a much less expensive home or rent a place one can easily afford (below one's means).

Best regards,

longinvest
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

KarenC
Posts: 141
Joined: Mon Apr 27, 2015 7:25 am

Re: Variable Percentage Withdrawal (VPW)

Post by KarenC » Wed Apr 22, 2020 7:27 am

ososnilknarf wrote:
Fri Apr 17, 2020 5:54 pm
I have a question about how to handle a mortgage being paid off using the VPW spreadsheet. Sorry if this has been addressed before, but I didn't see anything about it.
Say I'm retiring at 60, and have a mortgage payment of $2200/month, and it will be paid off at age 65. Is it reasonable to handle this by considering it another Defined Benefit Pension of $2200 that starts at age 65 and isn't indexed to inflation? I'm thinking that it is $2200 that I will no longer need from the portfolio, so it is in effect like another SS or pension payment.
I've also run into situations where it's difficult to model spending like you describe using the VPW spreadsheet. It's definitely more complex, but you might want to look into the spreadsheet provided by Siamond in association with the article Early Retirement and Time Value of Money. (There's also a longish thread about this.)
"The first principle is that you must not fool yourself, and you are the easiest person to fool." — Richard P. Feynman

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Wed Apr 22, 2020 7:55 am

No spreadsheet trick will eliminate the fact that a lump sum of money is borrowed on a home and invested into a portfolio of fluctuating assets hoping that the constant regular withdrawals taken to make mortgage payments will result into a positive outcome. It's leveraged investing.

I personally think that there's no need to take unnecessary risk with leverage. A simple and safe solution is to withdraw the money from the portfolio, pay off the mortgage, and retire debt-free.

This question has nothing to do with VPW and would be best discussed in its own thread about leveraged investing.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

MrDrinkingWater
Posts: 166
Joined: Tue Dec 23, 2014 11:30 am

Re: Variable Percentage Withdrawal (VPW)

Post by MrDrinkingWater » Wed Apr 22, 2020 10:12 am

When I read member ososnilknarf's question, I read it this way: At age 65, he does not want to replace $2200/month of allowable spending on a mortgage with $2200/month of spending on other things.

If we assume that he is currently spending at his highest highest allowed VPW-calculated withdrawal rate each year, what he is asking for is some mechanism or approach to not spend at his highest allowed VPW-calculated withdrawal rate at age 65, but spend $2200/month less, and have that money stay invested per his asset allocation, rather than withdrawn by rule and parked in, say, a savings account.

I agree that his question is not a VPW question. Maybe there is a Change Request embedded in his question. He's asking for another few lines of code to be added to VPW so that he might mechanize returning $26,400 ($2200 X 12) back into the Portfolio Balance on the Retirement spreadsheet within the VPW Excel Workbook, rather than withdrawing it and not spending it during the year he turns 65 and thereafter.

I think it is better to keep a spend plan on a separate spread sheet and not accept this Change Request to the VPW workbook. If he is so inclined, member osonilkarf can choose to include the "unspent" $26,400 in his Portfolio Balance for his next year's calculation or not, or leave it fenced off as "unspent" Age 65 Year money.

User avatar
SevenBridgesRoad
Posts: 710
Joined: Sat Jul 07, 2018 12:14 am
Location: Within Pacific Ring of Fire

Re: Variable Percentage Withdrawal (VPW)

Post by SevenBridgesRoad » Wed Apr 22, 2020 10:33 am

MrDrinkingWater wrote:
Wed Apr 22, 2020 10:12 am
...I agree that his question is not a VPW question...
...I think it is better to keep a spend plan on a separate spread sheet and not accept this Change Request to the VPW workbook. If he is so inclined, member osonilkarf can choose to include the "unspent" $26,400 in his Portfolio Balance for his next year's calculation or not, or leave it fenced off as "unspent" Age 65 Year money.
Agree. There's already plenty of budgeting software. The VPW Retirement sheet is simple to use and sophisticated in doing what it is designed to do.
Retired 2018 age 61/Variable Percentage Withdrawal method/One fund: VTINX all accounts/No mortgage,debt/Good enough | "Not using an alarm is one of the great glories of my life." Robert Greene

ososnilknarf
Posts: 72
Joined: Mon Feb 18, 2019 7:08 pm

Re: Variable Percentage Withdrawal (VPW)

Post by ososnilknarf » Wed Apr 22, 2020 4:55 pm

longinvest,
First off thank you for your reply, and thank you for all of the work that you've put into the VPW method. I'm still learning but the more I learn the more it looks like the best approach to drawing from retirement accounts.
longinvest wrote:
Wed Apr 22, 2020 6:34 am
This is an indirect question about leveraged investing; that's why it's not attracting answers. It isn't a VPW question. The obvious answer is to withdraw the money from the portfolio and immediately pay off the mortgage.
I see what you mean. I guess there are two interpretations of my question: whether one *should* do this, and whether one *could* do this. I wasn't intending to ask about whether one *should*, but I appreciate you calling my attention to this as I now see that it is a questionable approach (or at least that is a different debate unrelated to VPW).
However, the question of whether one *could* do this is what I intended with my question, and is definitely a VPW question. I intended the question purely as an accounting question relating to the existing implementation of the VPW spreadsheet. I simply wanted to know if the accounting for it would be correct. Set aside the leveraged investing question and instead just think of it as a monthly fixed expense that will end at some point in time.
longinvest wrote:
Wed Apr 22, 2020 6:34 am
... one could just put the money aside into a high-interest savings account from which mortgage payments will be taken during the 4 or 5 years left on the mortgage.
Yes, but that is also the approach once suggested for bridging the gap until SS kicks in, but was abandoned in favor of leaving it in the portfolio and using the spreadsheet to do it "virtually" for you, was it not? I was trying to see if this same approach worked for a fixed expense that would be ending at a known point in the future. The thinking was that an expense ending is the same as a new income stream, in effect, so you would just draw less from the portfolio at that point.
And just to be clear, I wasn't requesting an enhancement to the spreadsheet. I was looking for confirmation that the existing SS/pension calculations would work out correctly for this case.

Sincerely,
--Pete

User avatar
siamond
Posts: 5442
Joined: Mon May 28, 2012 5:50 am

Re: Variable Percentage Withdrawal (VPW)

Post by siamond » Thu Apr 23, 2020 9:33 am

ososnilknarf wrote:
Wed Apr 22, 2020 4:55 pm
I see what you mean. I guess there are two interpretations of my question: whether one *should* do this, and whether one *could* do this. [...]
Yes, but that is also the approach once suggested for bridging the gap until SS kicks in, but was abandoned in favor of leaving it in the portfolio and using the spreadsheet to do it "virtually" for you, was it not? I was trying to see if this same approach worked for a fixed expense that would be ending at a known point in the future. The thinking was that an expense ending is the same as a new income stream, in effect, so you would just draw less from the portfolio at that point. [...]
Indeed. Your reasoning is absolutely correct. This is all a present-value calculation. As an early retiree, real life tends to present many such situations (mix of expenses and income going beyond one's regular budget) which can't be modeled as a simplistic pension/SS cash flow or pushed under the rug. I've been using such PV/PMT model for the past few years and this has proved very helpful. I have a few friends in the same situation, including one with a mortgage like you. Once you get the hang of a couple of handy spreadsheet formulas, this is actually very simple to do.

So yes, you "could". As to the "should you", this is of course your decision. If you're reasonably conversant with spreadsheets, please come back to Karen's post and you'll find one possible answer. And I'll be happy to swap private messages to share my experience.

ososnilknarf
Posts: 72
Joined: Mon Feb 18, 2019 7:08 pm

Re: Variable Percentage Withdrawal (VPW)

Post by ososnilknarf » Thu Apr 23, 2020 5:15 pm

KarenC wrote:
Wed Apr 22, 2020 7:27 am
I've also run into situations where it's difficult to model spending like you describe using the VPW spreadsheet. It's definitely more complex, but you might want to look into the spreadsheet provided by Siamond in association with the article Early Retirement and Time Value of Money. (There's also a longish thread about this.)
siamond wrote:
Thu Apr 23, 2020 9:33 am
Indeed. Your reasoning is absolutely correct. This is all a present-value calculation. As an early retiree, real life tends to present many such situations (mix of expenses and income going beyond one's regular budget) which can't be modeled as a simplistic pension/SS cash flow or pushed under the rug. I've been using such PV/PMT model for the past few years and this has proved very helpful. I have a few friends in the same situation, including one with a mortgage like you. Once you get the hang of a couple of handy spreadsheet formulas, this is actually very simple to do.

So yes, you "could". As to the "should you", this is of course your decision. If you're reasonably conversant with spreadsheets, please come back to Karen's post and you'll find one possible answer. And I'll be happy to swap private messages to share my experience.
Thank you KarenC and siamond. I'll take a look at this. It does look like good stuff relevant to the question I was asking. Siamond, thank you for your offer to private message. I'll let you know if I have questions after reviewing this information.

--Pete

travelintime
Posts: 7
Joined: Sun Mar 15, 2020 7:04 pm

Re: Variable Percentage Withdrawal (VPW)

Post by travelintime » Sun Apr 26, 2020 12:32 am

What would the math look like if the withdrawal table was extended to 100 years? Is there a quick way to figure that out?

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sun Apr 26, 2020 8:17 am

travelintime wrote:
Sun Apr 26, 2020 12:32 am
What would the math look like if the withdrawal table was extended to 100 years? Is there a quick way to figure that out?
Dear Travelintime,

The VPW Table is already set to hypothetically deliver its last withdrawal at age 99. But, that's not exactly how it's used. Here's why.

The VPW retirement approach is more than a simple table; it's an overall approach to create retirement income for as long as one lives, like a lifelong retirement paycheck (especially with monthly withdrawals as illustrated in this thread). It takes into account that we don't know how long we'll live. So, the VPW Table is calibrated such that there will be sufficient money left in the portfolio, at age 80, to buy (when necessary) sufficient income in the form of an inflation-indexed SPIA* to dampen the financial risk associated with long life (longer than 100 years), when the portfolio will be mostly depleted. The VPW approach also aims at keeping liquidity (e.g. a portfolio) all lifelong by capping the withdrawal percentage to at most 10% of the remaining portfolio (starting around age 90), which is after the inflation-indexed SPIA was potentially bought.

* Single Premium Immediate Annuity.

This is explained in details in this post. Here's a part of the post:
longinvest wrote:
Wed Oct 17, 2018 11:04 pm
According to United States Life Tables, 2013 (page 6), 1.0% of males and 2.8% of females reach age 100. In particular, (1,023 / 51,252) = 2.0% of 80 years old males and (2,754 / 64,427) = 4.3% of 80 years old females make it to age 100. The chance of one spouse surviving to age 100, in a man and woman couple (both of age 80) is ((2.0% + 4.3%) - (2.0% X 4.3%)) = 6.2%. Note that this is for the general population, without accounting for socioeconomic status.
[...]
According to life tables, one spouse statistically reaches the venerable age of 100 out of every 16 couples when both spouses are still alive at age 80. That's for the general population, without accounting for socioeconomic status. That's why the VPW approach insists on dampening at age 80 the financial risk related to long life.

If your objective is to study the impact of various settings, to better understand how the VPW Table parameters were chosen, our wiki provides a VPW Backtesting Spreadsheet. It's the spreadsheet that I and other Bogleheads members used to learn about how VPW would have reacted in various historical market environments with various settings, helping to better understand the amount of uncertainty and the relation between asset allocation and withdrawal fluctuations. But, backtesting is a tricky tool; too many people use it in a predictive manner, which is a mistake. Backtesting isn't a predictive tool; it's a tool to demolish faulty assumptions by finding counterexamples. I'll note that the confidence that VPW will never prematurely deplete a portfolio isn't based on backtesting, it's based on rigorous mathematics.

Best regards,

longinvest
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

travelintime
Posts: 7
Joined: Sun Mar 15, 2020 7:04 pm

Re: Variable Percentage Withdrawal (VPW)

Post by travelintime » Sun Apr 26, 2020 11:45 am

Thank you for your thorough response and the valuable work you're doing.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat May 16, 2020 12:23 pm

In the VPW forward test thread, forum member JamesJonesJrJr asked:
JamesJonesJrJr wrote:
Sat May 16, 2020 11:23 am
Sorry if this is a stupid question, but how does VPW account for inflation?
JamesJonesJrJr, the variable percentage withdrawal method is reactive to market returns; withdrawals vary according the the variations of the portfolio from which they are taken. Some years, withdrawals will increase more than inflation. Other years, withdrawals will fail to match inflation and might even decrease. That's why the VPW wiki page says that 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).

Stable inflation-indexed income can be bought indirectly (cheapest) be delaying Social Security to age 70 and directly (more expensive) by buying an inflation-indexed SPIA.*

* Like a 2%-indexed SPIA which is indexed to promised inflation, as the Federal Reserve has a 2% inflation target.

JamesJonesJrJr wrote:
Sat May 16, 2020 11:23 am
The growth trend assumptions indicate 5% growth trend for stocks and 1.9% for bonds. Are these 'real', post-inflation returns or nominal returns?
For the benefit of readers, I'll copy the explanation of what a growth trend is and its purpose from the List sheet of the VPW Accumulation and Retirement Worksheet:
  • A growth trend is a timeless wild-ass guess that aims to represent an annual return that is lower than a high annual return and higher than a low annual return.
  • Its role is to distinguish between a high annual return and a low annual return.
  • It's not a prediction of future returns. It is fixed and must never be changed.
  • At the top of a bubble, it is likely to be higher than future returns. At the bottom of a crash it is likely to lower than future returns.
  • The selected growth trends are based on the long-term returns of world stocks and bonds from 1900 to 2018 according the Summary Edition of the Credit Suisse Global Investment Returns Yearbook 2019.
A link to the Summary Edition of the Credit Suisse Global Investment Returns Yearbook 2019 can be found on this page: The title of Figure 25 on page 37 indicates that they are annualized real returns. The bond and stock growth trends are thus based on historical global inflation-adjusted returns over a period of 119 years.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

JamesJonesJrJr
Posts: 63
Joined: Sun Apr 19, 2020 3:00 pm

Re: Variable Percentage Withdrawal (VPW)

Post by JamesJonesJrJr » Sat May 16, 2020 12:37 pm

longinvest wrote:
Sat May 16, 2020 12:23 pm
JamesJonesJrJr wrote:
Sat May 16, 2020 11:23 am
The growth trend assumptions indicate 5% growth trend for stocks and 1.9% for bonds. Are these 'real', post-inflation returns or nominal returns?
... they are annualized real returns. The bond and stock growth trends are thus based on historical global inflation-adjusted returns over a period of 119 years.
Have you considered market-based measures of future real returns on treasuries are deeply negative? The current rate on 10-year TIPS is -0.4%.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat May 16, 2020 12:44 pm

JamesJonesJrJr wrote:
Sat May 16, 2020 12:37 pm
longinvest wrote:
Sat May 16, 2020 12:23 pm
JamesJonesJrJr wrote:
Sat May 16, 2020 11:23 am
The growth trend assumptions indicate 5% growth trend for stocks and 1.9% for bonds. Are these 'real', post-inflation returns or nominal returns?
... they are annualized real returns. The bond and stock growth trends are thus based on historical global inflation-adjusted returns over a period of 119 years.
Have you considered market-based measures of future real returns on treasuries are deeply negative? The current rate on 10-year TIPS is -0.4%.
JamesJonesJrJr, I've repeatedly explained in this thread why it would be a bad idea to use prediction metrics to calibrate withdrawals. You'll find a detailed explanation in this previous post (and many other posts):
longinvest wrote:
Mon May 21, 2018 7:01 am
...
using valuations to set withdrawal rates would lead to a "sell less when high, sell more when low" behavior. That would be a pretty bad investing approach.
...
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

JamesJonesJrJr
Posts: 63
Joined: Sun Apr 19, 2020 3:00 pm

Re: Variable Percentage Withdrawal (VPW)

Post by JamesJonesJrJr » Sat May 16, 2020 1:31 pm

longinvest wrote:
Sat May 16, 2020 12:44 pm
JamesJonesJrJr wrote:
Sat May 16, 2020 12:37 pm
longinvest wrote:
Sat May 16, 2020 12:23 pm
JamesJonesJrJr wrote:
Sat May 16, 2020 11:23 am
The growth trend assumptions indicate 5% growth trend for stocks and 1.9% for bonds. Are these 'real', post-inflation returns or nominal returns?
... they are annualized real returns. The bond and stock growth trends are thus based on historical global inflation-adjusted returns over a period of 119 years.
Have you considered market-based measures of future real returns on treasuries are deeply negative? The current rate on 10-year TIPS is -0.4%.
JamesJonesJrJr, I've repeatedly explained in this thread why it would be a bad idea to use prediction metrics to calibrate withdrawals. You'll find a detailed explanation in this previous post (and many other posts):
longinvest wrote:
Mon May 21, 2018 7:01 am
...
using valuations to set withdrawal rates would lead to a "sell less when high, sell more when low" behavior. That would be a pretty bad investing approach.
...
What if markets don't eventually return to the mean, as you're tacitly assuming? In that case you're simply withdrawing more than you should and you will have less retirement income later on in life, no? And the solution, if this was a concern, would be to simply anticipate lower market returns. Perhaps this isn't a 'bad investing approach' so much as a conservative one. Just some food for thought.

Also, I noticed the retirement spreadsheet discounts non-COLA pension cash flows by an inflation rate of 2%. It will be interesting to see how treasuries, which are currently yielding 1.1%, will somehow produce 1.9% real returns after 2% inflation.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat May 16, 2020 2:13 pm

JamesJonesJrJr wrote:
Sat May 16, 2020 1:31 pm
What if markets don't eventually return to the mean, as you're tacitly assuming? In that case you're simply withdrawing more than you should and you will have less retirement income later on in life, no? And the solution, if this was a concern, would be to simply anticipate lower market returns. Perhaps this isn't a 'bad investing approach' so much as a conservative one. Just some food for thought.
Dear JamesJonesJrJr,

Let me repeat what I wrote:
  • A growth trend is a timeless wild-ass guess that aims to represent an annual return that is lower than a high annual return and higher than a low annual return.
  • Its role is to distinguish between a high annual return and a low annual return.
  • It's not a prediction of future returns. It is fixed and must never be changed.
  • At the top of a bubble, it is likely to be higher than future returns. At the bottom of a crash it is likely to lower than future returns.
  • The selected growth trends are based on the long-term returns of world stocks and bonds from 1900 to 2018 according the Summary Edition of the Credit Suisse Global Investment Returns Yearbook 2019.
If I look, for example, at some of the past (nominal) annual returns of a TIPS fund, the Vanguard Inflation-Protected Securities Fund (VIPSX), I can use the VPW worksheet growth trend to grossly estimate (it's not precise) if they were low or high (in what was more-or-less a 2% inflation environment):
  • 2007: 11.59% (high)
  • 2008: -2.85% (low)
  • 2009: 10.80% (high)
  • 2011: 13.24% (high)
  • 2013: -8.92% (low)
  • 2018: -1.49% (low)
  • 2019: 8.06% (high)
Returns fluctuate; they're not constant. The fluctuations are significant (but smaller than stock fluctuations). That's the main thing that affects withdrawals.

Nobody knows how much the TIPS market will return over my lifetime. Even if I could buy a zero-coupon 10-year TIPS (this doesn't exist), I would only know its exact time-weighted real return over the single specific period from now to its maturity date. I would not know its time-weighted return over any other future period of time as I couldn't predict how much a buyer would be willing to give me for it in the future before maturity and, of course, it wouldn't exist after maturity. Also, I won't be spending all of my portfolio in exactly 10 years. I don't even know exactly how much a Toyota Corolla will cost in inflation-adjusted dollars in exactly 10 years! I'm actually pretty sure that its price will still vary from car vendor to car vendor.

Our lives and markets are full of uncertainty. It's best to accept it, stop trying to predict future returns, and move on with our lives.

I strongly suggest to fully embrace our philosophy which says to never try to time the market.

Best regards,

longinvest
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

JamesJonesJrJr
Posts: 63
Joined: Sun Apr 19, 2020 3:00 pm

Re: Variable Percentage Withdrawal (VPW)

Post by JamesJonesJrJr » Sat May 16, 2020 2:37 pm

longinvest wrote:
Sat May 16, 2020 2:13 pm

I strongly suggest to fully embrace our philosophy which says to never try to time the market.
I think you misunderstood what I'm saying. I'm not suggesting to time the market, or to dynamically adjust the growth assumption or anything like that. I'm just saying that it's reasonable to have conservative assumptions about market returns so that you don't impair your ability maintain your standard of living later on in life. I don't think this is incompatible with VPW.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat May 16, 2020 4:35 pm

JamesJonesJrJr wrote:
Sat May 16, 2020 2:37 pm
longinvest wrote:
Sat May 16, 2020 2:13 pm

I strongly suggest to fully embrace our philosophy which says to never try to time the market.
I think you misunderstood what I'm saying. I'm not suggesting to time the market, or to dynamically adjust the growth assumption or anything like that. I'm just saying that it's reasonable to have conservative assumptions about market returns so that you don't impair your ability maintain your standard of living later on in life. I don't think this is incompatible with VPW.
Dear JamesJonesJrJr,

I don't think that I misunderstood anything. You're proposing that a -0.5% real growth trend would be a better wild ass guess (WAG) as a timeless neutral return (applicable in all my future years of life) to determine if bond fund returns are high or low. So, for example, a 1% real return would be a high return and a -0.5% real return would merely be normal. The reason you'd select this WAG is based on the current 10-year constant-maturity TIPS yield (https://fred.stlouisfed.org/series/DFII10). I'm saying that this would be a mistake because yields fluctuate over one's lifetime. A timeless WAG growth trend isn't a prediction forward from this point in time. It's timeless; it applies to all future points in time, including after negative bond returns when yields increase.

The stock and bond growth trends that the VPW worksheet uses as WAG are based on long historical global averages including countries where all stocks and bonds went to zero. They're pretty conservative, already, as estimates of whether stock and bond returns are high or low. Growth trends aren't future return predictions.

If I wanted certainty for maintaining a minimal standard of living, I would buy stable lifelong inflation-indexed income. It's expensive, but it provides certainty.

Most of us are flexible, in our lives. Everywhere I look, I see people adapting their spending to their income. Adapting my spending to my income is what I've done all lifelong, while living below my means and saving the difference into a balanced portfolio which will allow me to retire with dignity. I intend to continue adapting my spending to my income when I'll retire, and I'll use the VPW method to guide my portfolio withdrawals.

Best regards,

longinvest
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

deserat
Posts: 12
Joined: Fri Dec 27, 2019 5:08 am

Re: Variable Percentage Withdrawal (VPW)

Post by deserat » Mon Jun 08, 2020 3:35 pm

To longinvest and others who have contributed to this model, spreadsheet, the wiki, and the various threads.

Thank you!!!! I have been accumulating for years, will have two pensions and SS as well as a portfolio. I have lately been running many retirement calculators to determine how much I could spend in an early-ish retirement. The VPW spreadsheet is so simple and yet the UI of the spreadsheet belies the underlying thinking and years of work and testing underneath.

I want to make sure I understand the mechanics completely:

1) If one wants to smooth out monthly income or portfolio withdrawal, then one can change the portfolio balance monthly and use that to update the suggested monthly portfolio withdrawal
2) Age is only changed when you have a birthday :happy
3) Portfolio asset allocation is only changed when that is changed (one may wish to transition down to a lower stock/bond ratio as they age)
4) Pension amounts can be adjusted if they change. For example I am eligible for a military Reserve pension that will get COLA, but the base pension amount may be different based on the pay raises given to military personnel (pension is computed as percentage of base pay).
5) If guaranteed income (pension/annuities) covers your basic living costs, the SPIA at age 80 may not be required to establish a floor as you already have one.

Again - thanks so much for this simple, brilliant worksheet.

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Mon Jun 08, 2020 5:33 pm

deserat wrote:
Mon Jun 08, 2020 3:35 pm
To longinvest and others who have contributed to this model, spreadsheet, the wiki, and the various threads.

Thank you!!!!
Deserat, on behalf of the numerous direct contributors and indirect contributors who shared their suggestions, comments, and feedback helping to improve VPW: You're welcome!
deserat wrote:
Mon Jun 08, 2020 3:35 pm
I want to make sure I understand the mechanics completely:

1) If one wants to smooth out monthly income or portfolio withdrawal, then one can change the portfolio balance monthly and use that to update the suggested monthly portfolio withdrawal
When using VPW, there are two kinds of retirement income fluctuations: smaller and bigger.

Combining monthly withdrawals with a small withdrawal cushion (representing approximately 5 months of withdrawals) does an excellent job of smoothing smaller income fluctuations. This is illustrated in details in this thread. (Note that monthly withdrawals aren't smoothed; it's the monthly income which is smoothed using the cushion).

Sometimes, a picture is worth a thousand words. Here's a graph extracted from an earlier post of this thread:
longinvest wrote:
Sat Jun 29, 2019 11:58 pm
I've simulated annual and monthly VPW withdrawals on a 50/50 stocks/bonds portfolio, with stocks divided evenly between domestic (US) and international (ex US) from 1998 to 2018. The initial withdrawal is taken on December 31, 1997 at age 65 from a $1,000,000 portfolio. I've used cash monthly returns from Portfolio Visualizer as savings account returns.

Source data for simulations: Portfolio Visualizer

In the following figure, the black line represents monthly income (including interest) derived from annual VPW withdrawals taken on December 31. The red line represents monthly VPW withdrawals, and the green line represents monthly income from monthly silos (average of the last 12 monthly VPW withdrawals plus interest). [NOTE: "Monthly silos" were an earlier version of the now much simpler "withdrawal cushion" illustrated in the VPW forward test thread.]

Image

While monthly income from silos is smoother than the other two, it remains obvious that the major fluctuations (up in the late 1990s, down in the early 2000s, up in the mid 2000s, significantly down in 2008-2009, and then up) are determined by portfolio returns. It's the smaller short-term fluctuations that are significantly dampened by monthly silos.
Bigger retirement income fluctuations, due to significant stock losses or gains, can be dampened in two ways:
  • by increasing the allocation to bonds to dampen the volatility of the portfolio, or
  • (if necessary) by increasing the amount of stable lifelong non-portfolio income at the cost of reducing liquidity by buying a small inflation-indexed SPIA* with part of the portfolio.
* Single Premium Immediate Annuity.
deserat wrote:
Mon Jun 08, 2020 3:35 pm
2) Age is only changed when you have a birthday :happy
Yes.
deserat wrote:
Mon Jun 08, 2020 3:35 pm
3) Portfolio asset allocation is only changed when that is changed (one may wish to transition down to a lower stock/bond ratio as they age)
Yes. (I'm personally a fan of fixed balanced allocations, but it works with gliding allocations).
deserat wrote:
Mon Jun 08, 2020 3:35 pm
4) Pension amounts can be adjusted if they change. For example I am eligible for a military Reserve pension that will get COLA, but the base pension amount may be different based on the pay raises given to military personnel (pension is computed as percentage of base pay).
Yes.

On the other hand, for any pension that doesn't get such adjustments, it's important to select "No" for the "Cost of Living Adjustments" option when describing the pension in the worksheet.
deserat wrote:
Mon Jun 08, 2020 3:35 pm
5) If guaranteed income (pension/annuities) covers your basic living costs, the SPIA at age 80 may not be required to establish a floor as you already have one.
Yes, if the floor is high enough. After age 80 it's important for non-portfolio income (Social Security, pensions, ...) to allow the retiree to live comfortably independently of portfolio withdrawals. The idea is to dampen the financial risk associated with living past age 100, once the portfolio will be mostly depleted. Comfortable living means what it says; it's much better than bare-bones living. :happy

Enjoy!
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

Brit
Posts: 68
Joined: Wed Oct 10, 2012 6:33 am

Re: Variable Percentage Withdrawal (VPW)

Post by Brit » Sat Jun 27, 2020 9:26 am

If I may I have a AA question related to the VPW method.

In my current IPS I have the following statement:
I will rebalance portfolio every year on my birthday to increase bond holdings by 2% percent until age 62 to an AA of 40/60. Then I will increase stocks by 1% for ten years to an AA of 50/50 (retirement glide path). Reduces risk of portfolio failure in sequence of poor portfolio returns in the first (most critical) ten years of retirement.
However, I remember reading a statement in viewtopic.php?f=10&t=284519&sid=091f4ad ... c6cc5c2f07 that by definition the VPW method reduces (eliminates?) that risk and therefore the path that I am taking to a 40/60 to reduce risk, and then back to 50/50 over ten years, would be more likely to reduce my opportunity for growth. Thus if I do follow the VPW method strictly and am comfortable with an AA of 50/50 is there any reason why I wouldn't stay that course at a 50/50 AA once I reach it?

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat Jun 27, 2020 10:09 am

Brit wrote:
Sat Jun 27, 2020 9:26 am
If I may I have a AA question related to the VPW method.

In my current IPS I have the following statement:
I will rebalance portfolio every year on my birthday to increase bond holdings by 2% percent until age 62 to an AA of 40/60. Then I will increase stocks by 1% for ten years to an AA of 50/50 (retirement glide path). Reduces risk of portfolio failure in sequence of poor portfolio returns in the first (most critical) ten years of retirement.
However, I remember reading a statement in this thread that by definition the VPW method reduces (eliminates?) that risk and therefore the path that I am taking to a 40/60 to reduce risk, and then back to 50/50 over ten years, would be more likely to reduce my opportunity for growth. Thus if I do follow the VPW method strictly and am comfortable with an AA of 50/50 is there any reason why I wouldn't stay that course at a 50/50 AA once I reach it?
Brit, I personally think that a constant balanced allocation, all lifelong, is good enough. During accumulation, when using the VPW Accumulation Worksheet, a balanced portfolio dampens the impact of wild stock fluctuations on the suggested savings amount. During retirement, when using the VPW Retirement Worksheet, a balanced portfolio dampens the impact of wild stock fluctuations on the suggested withdrawal amount.

All of my life, I've adapted my spending to my financial means. My wife and I continue to adapt our standard of living to our salaries while living below our means and saving the difference into a balanced portfolio. When our family income gets reduced, we spend less and save less. When our family income increases, we spend more and save more. We always adapt to our means. We have no direct control on our means (our employers decide of our salaries, not the other way around). Of course, we made early carreer choices that had an impact, but this isn't a direct year-to-year control. I don't see any reaon why we wouldn't continue to do the same during retirement. We won't have complete control on our income; markets will determine part of it. But we can make choices today that will impact our future means (how much we reasonably save to balance quality of life now and later, avoiding speculation, and not concentrating our investments into a single asset class, single country, or single sector).

We've already simplified our investments to a One Fund Portfolio, a single identical all-in-one low-cost globally-diversified balanced index ETF in all of our accounts. Having a balanced portfolio in the latest months has significantly dampened the impact of stock fluctuations on the savings amount suggested by the VPW Accumulation Worksheet. We also duplicate the idea of a cash cushion* (of the foward test thread) to stabilize our spending budget** while letting our actual savings amount vary with monthly worksheet suggestions without any smoothing.

* The cash cushion somehow replaces the traditional emergency fund.
** We also use standard budgeting techniques (sinking funds, low-interest loans, etc.) to handle irregular bigger spending items.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

justgary
Posts: 3
Joined: Sat Jun 27, 2020 12:24 pm

Re: Variable Percentage Withdrawal (VPW)

Post by justgary » Sat Jun 27, 2020 4:10 pm

longinvest -

I wanted to take a moment and get in line to thank you and all of the others who have spent your time creating the VPW spreadsheet. I'm near retirement, and the primary reason I haven't quit working is my fear of knowing exactly how to make my investments pay me. Your spreadsheet has solved that problem for me, so I am much more confident in myself when considering a switch from accumulation to decumulation. Once I quit work, I probably can't go back.

I have followed this and your forward test thread for a year now, and the method makes sense. Simple is good, and you have hit the nail squarely on the head! If I ever meet you in person, I'll offer you a very decent rye whiskey (or your choice of beverages). Thanks again.

Regards,

- Just Gary

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat Jul 04, 2020 2:24 pm

On the VPW forward test thread, forum member Travelintime asked a theory question:
travelintime wrote:
Sat Jul 04, 2020 1:45 pm
The dampening account smooths out the month to month withdrawal amounts. What effect does it have for the twelve month total. Meaning, at the end of twelve months how might these two total withdrawal amounts compare:

A) The total dollar amount of twelve monthly withdrawals using a dampening account

B) The total dollar amount of twelve monthly withdrawals without a dampening account

Would the dollar amount at the finish line be similar, but "A" provides a smoother ride along the way, or would the dollar amounts be significantly different?
Travelintime, you're asking an interesting question, but I'll take a different perspective to provide a mathematical result that indirectly answers it.

The withdrawal cushion spreads the impact of a withdrawal over time. As the savings account contains approximately 5 months of withdrawal to which we add a the current month portfolio withdrawal amount and immediately remove 1/6 of the resulting account balance, the net impact is as follows:

The first month, when an amount M is withdrawn from the portfolio, 1/6 M = 17% M is spent. The remaining 5/6 M is added to the savings account (called withdrawal cushion) and starts accumulating interest.

In the second month, 1/6 of (the remaining 5/6 M with added interest) is spent. If I was to ignore interest (hoping interest matches inflation) and calculate in more-or-less inflation-indexed dollars, we could say that (1/6 X 5/6 M) = 5/36 M = 14% M is spent.

In the third month, 1/6 of the remaining (5/6)^2 M (more-or-less inflation-indexed) is spent. That's 25/216 M = 12% M.

And so on.

In other words, in the Nth month, (1/6 X (5/6)^(N-1)) M is spent. More importantly, after spending, ((5/6)^N) M remains unspent.

Here's the mathematical result:

Using the above formula we get that:
  • Half of a withdrawal is spent during the first 4 months because after 4 months 48% of the original withdrawal amount remains unspent.
  • 90% of a withdrawal is spent during the first year (that's an additional 40% during the next 8 months) because after 12 months 11% of the original withdrawal amount remains unspent.
  • A withdrawal is almost entirely spent during the first two years (that's most of the remaining 10% during the second year) because after 24 months only 1% of the original withdrawal amount remains unspent.
Visually:

Code: Select all

 - - - - - - - - - - - - - - - - - - - - - - - -
| 4 mths|     8 mths    |         1 year        |
|  50%  |      40%      |     remaining 10%     |
 - - - - - - - - - - - - - - - - - - - - - - - -

Note: When I say "spent", I more accurately mean "transformed into retirement income". Whether the money is actually spent (on taxes and expenses) or directed to a different personal budget category is out of scope for the VPW retirement income approach.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

Topic Author
longinvest
Posts: 4289
Joined: Sat Aug 11, 2012 8:44 am

Re: Variable Percentage Withdrawal (VPW)

Post by longinvest » Sat Jul 04, 2020 7:56 pm

Using formulas in my previous post, it's easy to calculate the weightings of withdrawals for a 12 month period:
  • 7.4% of annual income originates from month 1 withdrawal, because (1/12 X (1 - (5/6)^12)) = 7.4%
  • 7.2% of annual income originates from month 2 withdrawal, because (1/12 X (1 - (5/6)^11)) = 7.2%
  • 7.0% of annual income originates from month 3 withdrawal, ...
  • 6.7% of annual income originates from month 4 withdrawal
  • 6.4% of annual income originates from month 5 withdrawal
  • 6.0% of annual income originates from month 6 withdrawal
  • 5.5% of annual income originates from month 7 withdrawal
  • 5.0% of annual income originates from month 8 withdrawal
  • 4.3% of annual income originates from month 9 withdrawal
  • 3.5% of annual income originates from month 10 withdrawal
  • 2.5% of annual income originates from month 11 withdrawal
  • 1.4% of annual income originates from month 12 withdrawal
In total 63% of annual income originates from the last 12 withdrawals and 37% originates from earlier withdrawals.

So, the answer to Travelintime's question
travelintime wrote:
Sat Jul 04, 2020 1:45 pm
The dampening account smooths out the month to month withdrawal amounts. What effect does it have for the twelve month total. Meaning, at the end of twelve months how might these two total withdrawal amounts compare:

A) The total dollar amount of twelve monthly withdrawals using a dampening account

B) The total dollar amount of twelve monthly withdrawals without a dampening account

Would the dollar amount at the finish line be similar, but "A" provides a smoother ride along the way, or would the dollar amounts be significantly different?
is this:

Only 63% of A is only made of parts of the twelve monthly withdrawals with unequal weightings*. B, on the other hand, is entirely made of the twelve monthly withdrawals with equal weightings. As a consequence, it's normal for A and B to differ, but there's a 63% overlap between them.

* A includes 89% of month 1 withdrawal, but only 17% of month 12 withdrawal. In contrast, B includes 100% of month 1 withdrawal and 100% of month 12 withdrawal.

Equivalently (for A), 37% of the 12 withdrawals will be spread over the next two years (50% over the first 4 months, 40% over the next 8 months, and 10% over the following year).

The composition of A is this (with unequal weightings of monthly withdrawals): 63% current year, (37% X 90%) = 33% previous year, and 4% 2 years ago.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

Post Reply