Ratchet up 4% SWR amount with portfolio growth?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
foursix
Posts: 89
Joined: Wed Dec 24, 2014 2:45 pm

Ratchet up 4% SWR amount with portfolio growth?

Post by foursix »

I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
No one said the 4% rule never fails.

That said, what you're doing would add a flavor of constant-% on top of constant-dollar. There are implications of doing this.
sailaway
Posts: 8216
Joined: Fri May 12, 2017 1:11 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by sailaway »

Even assuming the 4% never fails (a huge assumption), some sequences require that growth to make it through to the end.

Of course, other sequences end up with multiples of the original number....
SilverSmurfer
Posts: 163
Joined: Sun Jan 16, 2022 7:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by SilverSmurfer »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
Then you're no longer using the 4% rule!

Logically:
if A) the 4% rule never fails (which may not be true, but we'll ignore that)
then B) you adjust your WR and are no longer using the 4% rule
why would you assume A still holds true or is applicable to your new WR?
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

sailaway wrote: Wed Jan 26, 2022 6:09 pm Even assuming the 4% never fails (a huge assumption), some sequences require that growth to make it through to the end.

Of course, other sequences end up with multiples of the original number....
This is correct.

If they're on the path to having $5 at the end of 4% SWR after 30 years (a "success" case), ratcheting up on good years puts them on the path of running out of money.
Topic Author
foursix
Posts: 89
Joined: Wed Dec 24, 2014 2:45 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by foursix »

Lets say investor A retires in 2020 with $1M and grows his portfolio to $2M by 2030. At that point, he recalculates his 4% SWR and starts over.

Investor B retires in 2030 with $2M. It seems to me the end result and odds of failure are exactly the same.

I guess Investor A just increased his SoRR as well though.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

foursix wrote: Wed Jan 26, 2022 6:21 pm Lets say investor A retires in 2020 with $1M and grows his portfolio to $2M by 2030. At that point, he recalculates his 4% SWR and starts over.

Investor B retires in 2030 with $2M. It seems to me the end result and odds of failure are exactly the same.

I guess Investor A just increased his SoRR as well though.
No, because the SWR is good for 30 years. For investor A, their retirement ends in 2050. Investor B, 2060. The situations are different.
User avatar
nisiprius
Advisory Board
Posts: 52216
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by nisiprius »

The 4% rule is what it is. The studies and failure rates are based on the assumption that you start with 4% of the initial portfolio value and increase it by the CPI. The numbers used depend only on the CPI, not on portfolio growth.
Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
That's not a reasonable assumption. In the original Trinity study, for a 30-year retirement the 4% rule had failures with all asset allocations:

Image

Vanguard's calculator shows a 10% failure rate for 4% withdrawals over 30 years using a 75/25 allocation.

Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Wed Jan 26, 2022 6:32 pm
foursix wrote: Wed Jan 26, 2022 6:21 pm Lets say investor A retires in 2020 with $1M and grows his portfolio to $2M by 2030. At that point, he recalculates his 4% SWR and starts over.

Investor B retires in 2030 with $2M. It seems to me the end result and odds of failure are exactly the same.

I guess Investor A just increased his SoRR as well though.
No, because the SWR is good for 30 years. For investor A, their retirement ends in 2050. Investor B, 2060. The situations are different.
In the OP's example, it's the same retiree. The retirees' remaining lifespan is ten years shorter in 2030 than it was in 2020, so it's fine to have a higher withdrawal rate in 2030.

If we make the admittedly huge assumption that the '4% rule' will never fail, then it's perfectly reasonable to take the greater of 4% of the current portfolio balance OR last year's withdrawal, adjusted for inflation.

Otherwise, we would be saying that retiree A with 29 years of planned retirement left must have a lower withdrawal rate than retiree B with 30 years of planned retirement left, all else being equal, which is obviously fallacious.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Wed Jan 26, 2022 6:39 pm In the OP's example, it's the same retiree. The retirees' remaining lifespan is ten years shorter in 2030 than it was in 2020, so it's fine to have a higher withdrawal rate in 2030.

If we make the admittedly huge assumption that the '4% rule' will never fail, then it's perfectly reasonable to take the greater of 4% of the current portfolio balance OR last year's withdrawal, adjusted for inflation.

Otherwise, we would be saying that retiree A with 29 years of planned retirement left must have a lower withdrawal rate than retiree B with 30 years of planned retirement left, all else being equal, which is obviously fallacious.
Well as I said initially, if this retiree was on a path such that they would have $5 after 30 years withdrawing 4%...ratcheting up during good years would put them on a path to run out of money, assuming the same sequence of returns. There really is no free lunch here.

If they weren't on such a path then of course ratcheting up can work out just fine.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Wed Jan 26, 2022 6:42 pm
willthrill81 wrote: Wed Jan 26, 2022 6:39 pm In the OP's example, it's the same retiree. The retirees' remaining lifespan is ten years shorter in 2030 than it was in 2020, so it's fine to have a higher withdrawal rate in 2030.

If we make the admittedly huge assumption that the '4% rule' will never fail, then it's perfectly reasonable to take the greater of 4% of the current portfolio balance OR last year's withdrawal, adjusted for inflation.

Otherwise, we would be saying that retiree A with 29 years of planned retirement left must have a lower withdrawal rate than retiree B with 30 years of planned retirement left, all else being equal, which is obviously fallacious.
Well as I said initially, if this retiree was on a path such that they would have $5 after 30 years withdrawing 4%...ratcheting up during good years would put them on a path to run out of money, assuming the same sequence of returns. There really is no free lunch here.

If they weren't on such a path then of course ratcheting up can work out just fine.
I don't believe that any of the historically poor periods that led to the '4% rule' being 4% started off with significant growth. Virtually all of them started with poor returns and, as a result, their effective withdrawal rate was almost immediately well above 4%.

There's never been a historic instance where the '4% rule' worked for 30 years where a retiree would have prematurely depleted their portfolio by increasing their amount withdrawn to 4% of the current portfolio balance. Mathematically, it would be impossible for this to prematurely deplete the portfolio IF rigid adherence to the '4% rule' would have worked.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Wed Jan 26, 2022 6:48 pm I don't believe that any of the historically poor periods that led to the '4% rule' being 4% started off with significant growth. Virtually all of them started with poor returns and, as a result, their effective withdrawal rate was almost immediately well above 4%.

There's never been a historic instance where the '4% rule' worked for 30 years where a retiree would have prematurely depleted their portfolio by increasing their amount withdrawn to 4% of the current portfolio balance. Mathematically, it would be impossible for this to prematurely deplete the portfolio IF rigid adherence to the '4% rule' would have worked.
I don't think we're on the same page. The OP is describing modified-SWR where you ratched up your withdrawal to 4% of current portfolio balance *only* after a good year.

Maybe an example is easier to see:
- 1M portfolio; you withdraw 40K.

a) after a bad year, your portfolio is now 500K -> you still withdraw 40K (assuming 0% inflation). This is the same as Bengen SWR.
b) after a good year, your portfolio is now 1.5M -> you now withdraw 60K.

I was describing issues with case b). Basically what can happen is the opposite of SORR, where good returns early on can lead to withdrawing too much early and run out of money, when you would have been fine hadn't you ratcheted up.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Wed Jan 26, 2022 6:57 pm
willthrill81 wrote: Wed Jan 26, 2022 6:48 pm I don't believe that any of the historically poor periods that led to the '4% rule' being 4% started off with significant growth. Virtually all of them started with poor returns and, as a result, their effective withdrawal rate was almost immediately well above 4%.

There's never been a historic instance where the '4% rule' worked for 30 years where a retiree would have prematurely depleted their portfolio by increasing their amount withdrawn to 4% of the current portfolio balance. Mathematically, it would be impossible for this to prematurely deplete the portfolio IF rigid adherence to the '4% rule' would have worked.
I don't think we're on the same page. The OP is describing modified-SWR where you ratched up your withdrawal to 4% of current portfolio balance *only* after a good year.

Maybe an example is easier to see:
- 1M portfolio; you withdraw 40K.

a) after a bad year, your portfolio is now 500K -> you still withdraw 40K (assuming 0% inflation). This is the same as Bengen SWR.
b) after a good year, your portfolio is now 1.5M -> you now withdraw 60K.

I was describing issues with case b). Basically what can happen is the opposite of SORR, where good returns early on can lead to withdrawing too much early and run out of money, when you would have been fine hadn't you ratcheted up.
But a new retiree with a 30 year retirement horizon could start by withdrawing $60k from their $1.5m portfolio, right? By the very same logic, the retiree whose portfolio has grown to that amount can now withdraw at least that much. What they've withdrawn in the past is completely irrelevant. The portfolio is 'memoryless'.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Wed Jan 26, 2022 7:25 pm But a new retiree with a 30 year retirement horizon could start by withdrawing $60k from their $1.5m portfolio, right? By the very same logic, the retiree whose portfolio has grown to that amount can now withdraw at least that much.
The outcome would be unknown. Again, I was describing a scenario where someone doing 4% SWR exactly as is would barely reach the finish line. If you ratchet up your withdrawals along the way, you would run out of money, assuming the same sequence of returns.

*I'm not talking about what the SWR methodology lets you do. I'm talking about the retirement path this hypothetical retiree is on.
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by vanbogle59 »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
I went through this with some very helpful folks here
viewtopic.php?f=10&t=358659&p=6252534&h ... t#p6252534

Bottom line:
If you choose a rule that never fails in a historical data set, you can ratchet up without risk of increased failure in that historical dataset.
If you choose a rule that sometimes fails in a historical data set, ratcheting will eventually increase your failure rate.
ITRW, if you can, flexibility of WR (up AND down) is advisable (most robust).
:sharebeer
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Wed Jan 26, 2022 7:34 pm
willthrill81 wrote: Wed Jan 26, 2022 7:25 pm But a new retiree with a 30 year retirement horizon could start by withdrawing $60k from their $1.5m portfolio, right? By the very same logic, the retiree whose portfolio has grown to that amount can now withdraw at least that much.
The outcome would be unknown. Again, I was describing a scenario where someone doing 4% SWR exactly as is would barely reach the finish line. If you ratchet up your withdrawals along the way, you would run out of money, assuming the same sequence of returns.

*I'm not talking about what the SWR methodology lets you do. I'm talking about the retirement path this hypothetical retiree is on.
That's why I said that we must preface all of this with the admittedly very big assumption that the '4% rule' won't fail in fewer than 30 years.

But if that assumption holds, then anywhere along the way, a retiree must be able to withdraw the greater of 4% of the current portfolio balance OR last year's withdrawal plus inflation.

For instance, let's assume that person A retired in 2010 with $1m and started by withdrawing $40k. 5 years later, person B retired with $1.5m and started by withdrawing $60k. If person A's portfolio has now also grown to $1.5m, s/he can also withdraw $60k. And in reality, person A can do so with even greater security than can person B because person A's retirement is now five years shorter than is person B's.
The Sensible Steward
FactualFran
Posts: 2776
Joined: Sat Feb 21, 2015 1:29 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by FactualFran »

willthrill81 wrote: Wed Jan 26, 2022 8:11 pm That's why I said that we must preface all of this with the admittedly very big assumption that the '4% rule' won't fail in fewer than 30 years.

But if that assumption holds, then anywhere along the way, a retiree must be able to withdraw the greater of 4% of the current portfolio balance OR last year's withdrawal plus inflation.
That leads to a question: how much greater than 4% can the re-retirement withdrawal rate be and have the portfolio last at least the number of remaining years?

With the investment returns that William Bengen used that resulted in a 50:50 stock:bond portfolio supporting, for all starting years, 30 years of annual inflation-adjusted withdrawals when the initial withdrawal rate was 4%, the initial withdrawal rate for 25 years was 4.48%.
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
The 4% "rule" is not 100% guaranteed to succeed. As long as there is a non-zero probability of failure, then the "ratchet up" approach increases the probability of failure. There was a long discussion on this not to long ago. Here is an explanation of why, and a long discussion that followed in more detail
viewtopic.php?p=6252502#p6252502

A more detailed analysis of what actually would have happened to some retirees if they had employed the "ratchet up" scheme:
viewtopic.php?p=6252752#p6252752
Once in a while you get shown the light, in the strangest of places if you look at it right.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Wed Jan 26, 2022 8:11 pm That's why I said that we must preface all of this with the admittedly very big assumption that the '4% rule' won't fail in fewer than 30 years.

But if that assumption holds, then anywhere along the way, a retiree must be able to withdraw the greater of 4% of the current portfolio balance OR last year's withdrawal plus inflation.

For instance, let's assume that person A retired in 2010 with $1m and started by withdrawing $40k. 5 years later, person B retired with $1.5m and started by withdrawing $60k. If person A's portfolio has now also grown to $1.5m, s/he can also withdraw $60k. And in reality, person A can do so with even greater security than can person B because person A's retirement is now five years shorter than is person B's.
But you're assuming Person B's 60K/year plan works, which we don't know. While at it, we don't know about Person A either, since they are only 12 years into their retirement since 2010.

This is the tricky thing about this - we are always receiving new data and the table is constantly getting updated. For example, with 2022 not so good so far, it's possible some cells on the table would change their value from 100% to 96% or what have you.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Wed Jan 26, 2022 11:51 pm
willthrill81 wrote: Wed Jan 26, 2022 8:11 pm That's why I said that we must preface all of this with the admittedly very big assumption that the '4% rule' won't fail in fewer than 30 years.

But if that assumption holds, then anywhere along the way, a retiree must be able to withdraw the greater of 4% of the current portfolio balance OR last year's withdrawal plus inflation.

For instance, let's assume that person A retired in 2010 with $1m and started by withdrawing $40k. 5 years later, person B retired with $1.5m and started by withdrawing $60k. If person A's portfolio has now also grown to $1.5m, s/he can also withdraw $60k. And in reality, person A can do so with even greater security than can person B because person A's retirement is now five years shorter than is person B's.
But you're assuming Person B's 60K/year plan works, which we don't know. While at it, we don't know about Person A either, since they are only 12 years into their retirement since 2010.

This is the tricky thing about this - we are always receiving new data and the table is constantly getting updated. For example, with 2022 not so good so far, it's possible some cells on the table would change their value from 100% to 96% or what have you.
Right, we don't know the future, which is why rigidly following the '4% rule' or any other SWR-style method is bogus. Everyone understands that flexibility is needed, especially if the portfolio suffers along the way.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Wed Jan 26, 2022 11:56 pm Right, we don't know the future, which is why rigidly following the '4% rule' or any other SWR-style method is bogus. Everyone understands that flexibility is needed, especially if the portfolio suffers along the way.
Right, although it's one thing to follow the SWR-style as written, it's another to ratchet up and increase the risk of failure, which is sort of the point I'm making here.

It sounds good on paper but there's real risk of reverse-SORR where getting good sequence of returns early on would elevate your withdrawals and impact the bottom line later.

Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
MikeG62
Posts: 5065
Joined: Tue Nov 15, 2016 2:20 pm
Location: New Jersey

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by MikeG62 »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?
Would you follow the same line of thinking if the opposite happened (i.e., your portfolio declined in value such that your "new 4% calculation" yields a smaller withdrawal than you are currently taking)?

There are withdrawal methodologies which attempt to take into account what you are describing (but on both the upside and downside). One comes from the guardrails built into Guyton & Kilnger's dynamic spending methodology. Another comes from Kitces' ratcheting methodology. Maybe one of these is better than adjusting in the manner you are suggesting?

FWIW, I retired in 2016 and am following a modified version of G&K's dynamic spending method.
Real Knowledge Comes Only From Experience
tvubpwcisla
Posts: 1167
Joined: Sat Nov 09, 2019 9:09 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by tvubpwcisla »

3% is the new 4%

:sharebeer
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by vanbogle59 »

MikeG62 wrote: Thu Jan 27, 2022 6:54 am Would you follow the same line of thinking if the opposite happened (i.e., your portfolio declined in value such that your "new 4% calculation" yields a smaller withdrawal than you are currently taking)?
Bingo. My preferred way of looking at it:
During your working career, when you got a raise, did you think it was OK to maybe spend a little more?
Or if you had unexpected major expenses hit your budget, or maybe even a job loss, did you tighten your belt?
So, if flexibility was just BAU for 40+ years, why would you expect it to go away in retirement?

Things like the 4% rule are just rules of thumb that allow you to evaluate if your plan is sane.
Valuable, but not precise.
VtDoc
Posts: 9
Joined: Wed Nov 29, 2017 5:52 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by VtDoc »

To the OP, yes this approach is correct. Assuming the 4% rule still applies, there is no way it could interpreted any other way. The goal of a fixed (inflation adjusted) WR is so that spending doesn't need to be reduced after bad returns. You could, however, recalculate your WR every year if you want, based on 4% of the new value, which would have essentially zero failure rate, but at a cost of wild swings in spending. Now, if you only adjust the WR when the market is up, it is intuitive that your personal chances of falling into a "failure" may increase from where it would have been if you hadn't increased your WR, but the new calculation and results will still be within the parameters of the original rule. In other words, you can recalculate when the market is up and still have a 3-5% risk of failing that a new retiree with your current account balance/allocations will have on that same day. In fact, since your duration of spending is lower every time you recalculate, your failure rate will decrease relative to the new retiree.
User avatar
martincmartin
Posts: 900
Joined: Wed Jul 02, 2014 3:04 pm
Location: Boston, MA USA

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by martincmartin »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
Depending on which historical data you use for "stock" and "bonds", it does fail. Even in the original Trinity study that gave us the 4% rule, it failed some small percentage of the time.

These failures happen after a long bull run, when stocks prices have been increasing and are overvalued. So by "reseting your retirement start date" whenever stocks go up, you make it more likely you end up in one of these market peaks that turn out to be one of the few start years where you run out of money.

Also, in the UK, the safe withdrawal rate is something like 3%. What if the US in the 21st century has stock market returns like the UK in the 20th century? At the start of the 20th century, the UK was the global superpower, "the sun never set on the British Empire" and all that. I'm not making a prediction, of course, it's just something that gives me pause.

Another thing that gives me pause: someone retiring in 1966, if they had done the historical analysis up to that point, would have come up with a SWR higher than 4%. If they then retired, they would run out of money, because 1966 is historically the worst year to retire in the US. The worst retirement period going forward could be worse than the worst retirement period in the past.

So personally, I'd stick with 4%, or even a little lower, so I'm only betting the actual year I retire isn't as bad as 1966, not every "restarted retirement" year. If stock valuations (e.g. CAPE10) were lower, I'd be less worried. But with valuations so high, it's quite possible this is a bad year to retire...
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
The Sensible Steward
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by vanbogle59 »

willthrill81 wrote: Thu Jan 27, 2022 9:34 am
Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
Yes, and yes.
Pick any dataset, pick any rule, pick any duration, pick any WR.... if you success rate is 100%, ratcheting "works".
Pick any dataset, pick any rule, pick any duration, pick any WR.... if you success rate is < 100%, ratcheting may lead to more failures.
As for the future, IDK.
User avatar
firebirdparts
Posts: 4414
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by firebirdparts »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
You've already been trolled furiously with this, but I would say 10 years in, you'd know how you were affected by "sequence of returns". Right? The dangerous part of the sequence already occurred. So you could draw some conclusion from that. In my opinion, if you want to draw conclusions, just let VPW be your conclusion and then you don't get trolled.

That said, we had a thread here by someone who was trying to specifically demonstrate that a 4% SWR would apply historically to any point before or after retirement (the maximum portfolio value of course being the point of interest), and they were satisfied with that. I didn't do the work, but it was discussed here. They gave it a name too, and if I think of the name (or any of you do, post it) I'll find that that thread and link it here.

Edit here it is
viewtopic.php?t=329473
Last edited by firebirdparts on Sat Jan 29, 2022 2:43 pm, edited 1 time in total.
This time is the same
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

firebirdparts wrote: Thu Jan 27, 2022 9:47 am
foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
You've already been trolled furiously with this, but I would say 10 years in, you'd know how you were affected by "sequence of returns". Right? The dangerous part of the sequence already occurred.
Yes and no. The early stages of any retirement are the most dangerous, but the danger of a poor sequence of returns continues throughout retirement.
The Sensible Steward
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by vanbogle59 »

firebirdparts wrote: Thu Jan 27, 2022 9:47 am we had a thread here by someone who was trying to specifically demonstrate that a 4% SWR would apply historically to any point before or after retirement (the maximum portfolio value of course being the point of interest), and they were satisfied with that. I didn't do the work, but it was discussed here. They gave it a name too, and if I think of the name (or any of you do, post it) I'll find that that thread and link it here.
Are you thinking of this?
Kitces guard rail approach to safe withdrawal

If yes, you might be thinking of me. :confused
And I was just wrong. :shock:
:sharebeer
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Thu Jan 27, 2022 9:34 am If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.
Yeah maybe it'd work out fine. Something feels odd but not sure what it is :D
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Marseille07 wrote: Thu Jan 27, 2022 10:20 am
willthrill81 wrote: Thu Jan 27, 2022 9:34 am If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.
Yeah maybe it'd work out fine. Something feels odd but not sure what it is :D
It must work out. The '4% rule' cannot be safe for later retirees but be unsafe for earlier retirees if both groups have the same withdrawal rate.
The Sensible Steward
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Thu Jan 27, 2022 9:34 am
Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
The 4% rule is not 100% safe. It failed for the 1966 retiree.
Thus, the 95% success rate.
Ratcheting up causes 1963, 1964, and 1965 retirees to go from a success to failure scenario.

See my link in post further above for details.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
vanbogle59
Posts: 1314
Joined: Wed Mar 10, 2021 7:30 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by vanbogle59 »

marcopolo wrote: Thu Jan 27, 2022 10:26 am
willthrill81 wrote: Thu Jan 27, 2022 9:34 am
Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
The 4% rule is not 100% safe. It failed for the 1966 retiree.
Thus, the 95% success rate.
Ratcheting up causes 1963, 1964, and 1965 retirees to go from a success to failure scenario.

See my link in post further above for details.
I think you 2 agree :beer
willthrill81 wrote: Wed Jan 26, 2022 6:39 pm If we make the admittedly huge assumption that the '4% rule' will never fail, then it's perfectly reasonable to take the greater of 4% of the current portfolio balance OR last year's withdrawal, adjusted for inflation.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Thu Jan 27, 2022 10:26 am
willthrill81 wrote: Thu Jan 27, 2022 9:34 am
Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
The 4% rule is not 100% safe. It failed for the 1966 retiree.
Thus, the 95% success rate.
Ratcheting up causes 1963, 1964, and 1965 retirees to go from a success to failure scenario.

See my link in post further above for details.
According to the Simba backtesting spreadsheet, the 30 year SWR for a 60/40 was never below 4%.

The precise assets and/or data used at the margins impact this significantly.

If it was safe for 1964 retirees to begin their 30 year retirement by withdrawing 4% of their starting balance, then it mathematically must have been safe for 1963 retirees with 29 years of retirement remaining to withdraw the exact same percentage of their then current portfolio balance.
The Sensible Steward
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

willthrill81 wrote: Thu Jan 27, 2022 10:38 am
marcopolo wrote: Thu Jan 27, 2022 10:26 am
willthrill81 wrote: Thu Jan 27, 2022 9:34 am
Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
The 4% rule is not 100% safe. It failed for the 1966 retiree.
Thus, the 95% success rate.
Ratcheting up causes 1963, 1964, and 1965 retirees to go from a success to failure scenario.

See my link in post further above for details.
According to the Simba backtesting spreadsheet, the 30 year SWR for a 60/40 was never below 4%.

The precise assets and/or data used at the margins impact this significantly.

If it was safe for 1964 retirees to begin their 30 year retirement by withdrawing 4% of their starting balance, then it mathematically must have been safe for 1963 retirees with 29 years of retirement remaining to withdraw the exact same percentage of their then current portfolio balance.
I used cFireSim, which shows several failures for starting in late 1960s. I have also always thought 4% rule was a 95% success rate, not 100%, consistent with the cFireSim results.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Thu Jan 27, 2022 10:43 am I used cFireSim, which shows several failures for starting in late 1960s. I have also always thought 4% rule was a 95% success rate, not 100%, consistent with the cFireSim results.
Again, the specific assets and/or data used have a significant impact on whether a couple of cohorts made it to the 30 year mark.

That said, it doesn't really matter in this instance and is only academic.
The Sensible Steward
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by randomguy »

marcopolo wrote: Thu Jan 27, 2022 10:43 am I used cFireSim, which shows several failures for starting in late 1960s. I have also always thought 4% rule was a 95% success rate, not 100%, consistent with the cFireSim results.
It depends on the methodology. Bengens paper had 100% because he was using yearly data (starting in Sept 1929 was much worse than starting in Jan 1929) and didn't have 1966 (want to say he stoped at 64) data yet Minor difference in what you use (S&P vs total market, 10 years versus commercial paper) also change the numbers up a bit. The later research (monthly starts was the big change) give us the 5% failure cases...
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by marcopolo »

randomguy wrote: Thu Jan 27, 2022 10:49 am
marcopolo wrote: Thu Jan 27, 2022 10:43 am I used cFireSim, which shows several failures for starting in late 1960s. I have also always thought 4% rule was a 95% success rate, not 100%, consistent with the cFireSim results.
It depends on the methodology. Bengens paper had 100% because he was using yearly data (starting in Sept 1929 was much worse than starting in Jan 1929) and didn't have 1966 (want to say he stoped at 64) data yet Minor difference in what you use (S&P vs total market, 10 years versus commercial paper) also change the numbers up a bit. The later research (monthly starts was the big change) give us the 5% failure cases...
Thanks for the clarification, that makes sense.

My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
Once in a while you get shown the light, in the strangest of places if you look at it right.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

marcopolo wrote: Thu Jan 27, 2022 10:51 am Thanks for the clarification, that makes sense.

My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
I think this poster is spot on. The risk of ratcheting is that when you cross a bad year anywhere during your retirement horizon, your (otherwise good) retirement start year might turn south. And you don't know that that has happened.
Alpha4
Posts: 176
Joined: Tue Apr 17, 2012 8:47 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Alpha4 »

willthrill81 wrote: Thu Jan 27, 2022 10:38 am
marcopolo wrote: Thu Jan 27, 2022 10:26 am
willthrill81 wrote: Thu Jan 27, 2022 9:34 am
Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
The 4% rule is not 100% safe. It failed for the 1966 retiree.
Thus, the 95% success rate.
Ratcheting up causes 1963, 1964, and 1965 retirees to go from a success to failure scenario.

See my link in post further above for details.

According to the Simba backtesting spreadsheet, the 30 year SWR for a 60/40 was never below 4%.

The precise assets and/or data used at the margins impact this significantly.

If it was safe for 1964 retirees to begin their 30 year retirement by withdrawing 4% of their starting balance, then it mathematically must have been safe for 1963 retirees with 29 years of retirement remaining to withdraw the exact same percentage of their then current portfolio balance.
I am only going off my own memory here but I could've swore that either Trinity or Bengen had stated that certain retirement sequences that started in 1907, 1929, 1937, 1965, and 1966 all failed (albeit in some cases just barely) at a 4% inflation adjusted withdrawal rate. There were various sequences starting at points in 1906, 1912, 1969, 1972, and 1973 that would've come pretty close to failure but that still left you with at least something (even if just a few dollars) at the end of thirty years of CPI-adjusted 4% withdrawals so those were counted as successes.

The first thing to note is that if you use ITTs instead of LTTs then portfolio survival will (obviously) be helped for the years 1965 and 1966 since rates rose sharply over the next 15 years or so. Whether this difference would be enough to account for Bengen calling 1966 a failure I do not know (if I recall correctly Mr. Bengen used LTTs in his study because the only bond data he had that went back to 1926 was the Ibbotson Long Term Government Bond total return index data).

Second, I could be wrong here (and quite possibly am) but my understanding of this is that there were no failures shown using the data from the Simba spreadsheet because that spreadsheet by definition only lets you start each 30-year sequence on Jan 1st of a given year (since said spreadsheet only has annual data). If you instead obtain monthly return data and then run the simulation twelve separate times for each year starting at the beginning of each month (i.e. the first one would start with, say, Jan 1st 1965 and go to Dec 31st 1994; the next one would start with Feb 1st 1965 and go to Jan 31st 1995, etc) IIRC you will for the 1965 and 1966 period find failures starting on November 1st 1965, February 1st 1966, and May 1st 1966 and perhaps on Oct 1st or Dec 1st 1965 as well.

1929, though...I was under the impression while you did OK (ended up with about 15% more in nominal terms than the starting amount at the end of 30 years of withdrawals....albeit with some rather hairy moments along the way in mid 1932 or early 1933 when it would've looked for sure that your portfolio was headed for depletion in a lot less than thirty years) starting Jan 1st 1929 but that if you started on, say, Sept 1 or Oct 1st then one either ran out of money or got this close (<<holds up thumb and forefinger about a millimeter apart>>).

1937 probably worked the same way; I couldn't (using data from the Simba spreadsheet) get it to fail starting Jan 1st no matter whether I used a 65/35, 60/40, or 50/50 stock/bond mix and no matter whether I used LTTs or ITTs as the bond portion. I never tried testing it starting the 1st of every month in 1937, though; if you started on March 1st or August 1st 1937 you would likely have faced portfolio depletion a good bit sooner than if you had started on January 1st, 1937.
User avatar
HomerJ
Posts: 21282
Joined: Fri Jun 06, 2008 12:50 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by HomerJ »

foursix wrote: Wed Jan 26, 2022 6:02 pm I've been looking at various retirement withdrawal methods and have a question regarding the standard 4% method. If your portfolio value grows in retirement such that a new 4% calculation yields a larger withdrawal than you're currently taking, is it OK to reset your WR to the new value?

It seems to me that's the same as basically restarting your retirement, for withdrawal purposes. Assuming the 4% rule never fails, it shouldn't fail with the new amount either, right?
Yes, restarting your retirement with a new 4% number should work...

I'll be doing something close to that.

But I might not go the full 4%. But if I retire, pulling 4%, and 5 years later my money has increased 40%, I'll probably start pulling more. Maybe not the full 4% of the new amount, but definitely more.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
FactualFran
Posts: 2776
Joined: Sat Feb 21, 2015 1:29 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by FactualFran »

Alpha4 wrote: Thu Jan 27, 2022 8:50 pm I am only going off my own memory here but I could've swore that either Trinity or Bengen had stated that certain retirement sequences that started in 1907, 1929, 1937, 1965, and 1966 all failed (albeit in some cases just barely) at a 4% inflation adjusted withdrawal rate. There were various sequences starting at points in 1906, 1912, 1969, 1972, and 1973 that would've come pretty close to failure but that still left you with at least something (even if just a few dollars) at the end of thirty years of CPI-adjusted 4% withdrawals so those were counted as successes.
It wasn't Bengen. According to his article Determining Withdrawal Rates Using Historical Data:
Assuming a minimum requirement of 30 years of portfolio longevity, a first-year withdrawal of 4 percent [Figure 1(B)], followed by inflation-adjusted withdrawals in subsequent years, should be safe. In no past case has it caused a portfolio to be exhausted before 33 years, and in most cases it will lead to portfolio lives of 50 years or longer. By comparison, a 4.25-percent first-year withdrawal could exhaust a portfolio in as little as 28 years, were past conditions to repeat themselves.
The portfolio used for Figure 1(B) was 50% stocks and 50% Intermediate-Term Government Bonds. The no past cases went back to 1926 as the starting year. The worst case was for 1966 as the starting year.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

Alpha4 wrote: Thu Jan 27, 2022 8:50 pm
willthrill81 wrote: Thu Jan 27, 2022 10:38 am
marcopolo wrote: Thu Jan 27, 2022 10:26 am
willthrill81 wrote: Thu Jan 27, 2022 9:34 am
Marseille07 wrote: Thu Jan 27, 2022 12:04 am Even if we look at the historical data, it's not hard to see those "barely" succeeding 4% SWR cases that would run out of money had ratcheting been used.
But that's just my point: there haven't been any such instances because there mathematically could not have been.

If the '4% rule' was safe for 1970 retirees, then it had to be safe for 1969 (or any prior year) retirees to 'ratchet up' their withdrawals to no higher than 4% of their new portfolio balance.

If a retiree started by withdrawing $40k from a $1m portfolio, and a year later that portfolio went up to $1.2m, that retiree or new retirees just now retiring could both withdraw 4% of their current portfolio balance with perfectly equal safety.

The catch is that we don't know in advance that the '4% rule' will continue to hold up. But in no historical case would ratcheting up the withdrawals in the fashion described by the OP have resulted in premature portfolio failure.
The 4% rule is not 100% safe. It failed for the 1966 retiree.
Thus, the 95% success rate.
Ratcheting up causes 1963, 1964, and 1965 retirees to go from a success to failure scenario.

See my link in post further above for details.

According to the Simba backtesting spreadsheet, the 30 year SWR for a 60/40 was never below 4%.

The precise assets and/or data used at the margins impact this significantly.

If it was safe for 1964 retirees to begin their 30 year retirement by withdrawing 4% of their starting balance, then it mathematically must have been safe for 1963 retirees with 29 years of retirement remaining to withdraw the exact same percentage of their then current portfolio balance.
I am only going off my own memory here but I could've swore that either Trinity or Bengen had stated that certain retirement sequences that started in 1907, 1929, 1937, 1965, and 1966 all failed (albeit in some cases just barely) at a 4% inflation adjusted withdrawal rate. There were various sequences starting at points in 1906, 1912, 1969, 1972, and 1973 that would've come pretty close to failure but that still left you with at least something (even if just a few dollars) at the end of thirty years of CPI-adjusted 4% withdrawals so those were counted as successes.

The first thing to note is that if you use ITTs instead of LTTs then portfolio survival will (obviously) be helped for the years 1965 and 1966 since rates rose sharply over the next 15 years or so. Whether this difference would be enough to account for Bengen calling 1966 a failure I do not know (if I recall correctly Mr. Bengen used LTTs in his study because the only bond data he had that went back to 1926 was the Ibbotson Long Term Government Bond total return index data).

Second, I could be wrong here (and quite possibly am) but my understanding of this is that there were no failures shown using the data from the Simba spreadsheet because that spreadsheet by definition only lets you start each 30-year sequence on Jan 1st of a given year (since said spreadsheet only has annual data). If you instead obtain monthly return data and then run the simulation twelve separate times for each year starting at the beginning of each month (i.e. the first one would start with, say, Jan 1st 1965 and go to Dec 31st 1994; the next one would start with Feb 1st 1965 and go to Jan 31st 1995, etc) IIRC you will for the 1965 and 1966 period find failures starting on November 1st 1965, February 1st 1966, and May 1st 1966 and perhaps on Oct 1st or Dec 1st 1965 as well.

1929, though...I was under the impression while you did OK (ended up with about 15% more in nominal terms than the starting amount at the end of 30 years of withdrawals....albeit with some rather hairy moments along the way in mid 1932 or early 1933 when it would've looked for sure that your portfolio was headed for depletion in a lot less than thirty years) starting Jan 1st 1929 but that if you started on, say, Sept 1 or Oct 1st then one either ran out of money or got this close (<<holds up thumb and forefinger about a millimeter apart>>).

1937 probably worked the same way; I couldn't (using data from the Simba spreadsheet) get it to fail starting Jan 1st no matter whether I used a 65/35, 60/40, or 50/50 stock/bond mix and no matter whether I used LTTs or ITTs as the bond portion. I never tried testing it starting the 1st of every month in 1937, though; if you started on March 1st or August 1st 1937 you would likely have faced portfolio depletion a good bit sooner than if you had started on January 1st, 1937.
As you note, seemingly small details like monthly vs. annual data or ITT vs. LTT (vs. 10 year bonds vs. corporate bonds) have made the difference in 'success' or 'failure' at the limits, but these are, IMHO, not very consequential. Does it really matter if the '4% rule' lasted 29.5 years vs. 30.1 in the worst historic periods?
The Sensible Steward
heyyou
Posts: 4461
Joined: Tue Feb 20, 2007 3:58 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by heyyou »

Why bother with using only the very first well-researched spending method, which has now been eclipsed by other methods? Salute Bengen's pioneering research, then move on to a better plan.

Any slightly variable method is likely better than SWR. My major criticism of the SWR method is how often the elderly retirees could have spent more since their specific retirements did not have near the worst case results of SWR, so they did not spend as much as they could have spent if they had used a slightly variable spending method.

Big ERN and McClung and others have charts of remaining portfolio values after 30 years of withdrawals for SWR and some other methods. SWR can have more assets unspent after 30 years, than at the start of retirement! That's fine for the heirs, but not for the retirees, unless that is your choice.

The solution is simple, use some other method that has variable income. Note how your SS+COLA and rising RMD amounts would partially buffer reduced income from a variable spending method based on recent portfolio value, such as the RMD method that spends your RMD % of the entire annual portfolio value, plus your annual interest and dividends. That method was developed by academics at Boston College's Center for Retirement Research.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

marcopolo wrote: Thu Jan 27, 2022 10:51 am My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
Yes, ratcheting up as the OP has described it, which is just 'restarting' the '4% rule', is no less safe than the '4% rule' itself. It's actually safer because the retiree presumably has fewer than 30 years of retirement remaining. But if the '4% rule' is in danger, so is ratcheting up.
The Sensible Steward
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by willthrill81 »

FactualFran wrote: Wed Jan 26, 2022 10:10 pm
willthrill81 wrote: Wed Jan 26, 2022 8:11 pm That's why I said that we must preface all of this with the admittedly very big assumption that the '4% rule' won't fail in fewer than 30 years.

But if that assumption holds, then anywhere along the way, a retiree must be able to withdraw the greater of 4% of the current portfolio balance OR last year's withdrawal plus inflation.
That leads to a question: how much greater than 4% can the re-retirement withdrawal rate be and have the portfolio last at least the number of remaining years?

With the investment returns that William Bengen used that resulted in a 50:50 stock:bond portfolio supporting, for all starting years, 30 years of annual inflation-adjusted withdrawals when the initial withdrawal rate was 4%, the initial withdrawal rate for 25 years was 4.48%.
Yes, the SWR for retirements shorter than 30 years has certainly been higher than 4%. The 20 year SWR for a 50/50 AA has been about 4.6%, and the 10 year SWR for the same has been about 7.5%.
The Sensible Steward
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Marseille07 »

willthrill81 wrote: Fri Jan 28, 2022 10:19 am
marcopolo wrote: Thu Jan 27, 2022 10:51 am My main point is that soon as you get to a situation where the success rate is less that 100%, ratcheting up increases risk of failure.
Yes, ratcheting up as the OP has described it, which is just 'restarting' the '4% rule', is no less safe than the '4% rule' itself. It's actually safer because the retiree presumably has fewer than 30 years of retirement remaining. But if the '4% rule' is in danger, so is ratcheting up.
This is not true. Suppose 1966 was a year that the 4% rule failed. The problem with ratcheting up is that it would cause earlier years (1963~1965) to fail, when the 4% rule itself would have worked for the 1963~1965 retirees.
Doc7
Posts: 1122
Joined: Fri Aug 24, 2007 12:55 pm

Re: Ratchet up 4% SWR amount with portfolio growth?

Post by Doc7 »

4 page thread on this topic:

viewtopic.php?t=352703
Post Reply