Withdrawal rate for an early retirement

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
birdog
Posts: 888
Joined: Fri Apr 07, 2017 1:35 pm

Re: Withdrawal rate for an early retirement

Post by birdog »

EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

birdog wrote: Tue Nov 24, 2020 2:15 pm
EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
Kintora
Posts: 73
Joined: Sun Jan 12, 2020 5:21 pm
Location: Fargo, ND

Re: Withdrawal rate for an early retirement

Post by Kintora »

willthrill81 wrote: Tue Nov 24, 2020 2:33 pm
birdog wrote: Tue Nov 24, 2020 2:15 pm
EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.
When you say that the cash is only spent when the portfolio is down more than 20% in the prior year, how is that calculated exactly? Is it an ongoing check of the portfolio value against one year ago from current? Does it include reduction due to normal withdrawals? If it recovers back to above being 20% down before the 2 years of cash is depleted, you would go back to stock/bond withdrawals until it possibly drops 20% again and then go back to the cash well?

Sorry if any of these are stupid questions. Just trying to wrap my head around this one as it sounds quite interesting.
marcopolo
Posts: 3849
Joined: Sat Dec 03, 2016 10:22 am

Re: Withdrawal rate for an early retirement

Post by marcopolo »

willthrill81 wrote: Tue Nov 24, 2020 2:33 pm
birdog wrote: Tue Nov 24, 2020 2:15 pm
EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.

You have pointed to this article in the past as well.
When doing so, I think it would paint a clearer picture if you included the fact that the cash bucket was in addition to the portfolio.

Most of the stated benefit come from having 27x spending rather than 25x. The fact that the extra 2x happens to be in cash is of pretty marginal value. I believe the same ERN series had an earlier article with a pretty critical view of having a "cash cushion" that was taken from the portfolio rather than in addition to it.

EDIT:
Here is the link to ERN earlier article regarding the limits of a "cash cushion":
https://earlyretirementnow.com/2017/03/ ... sh-cushion
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

marcopolo wrote: Tue Nov 24, 2020 6:38 pm
willthrill81 wrote: Tue Nov 24, 2020 2:33 pm
birdog wrote: Tue Nov 24, 2020 2:15 pm
EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.

You have pointed to this article in the past as well.
When doing so, I think it would paint a clearer picture if you included the fact that the cash bucket was in addition to the portfolio.

Most of the stated benefit come from having 27x spending rather than 25x. The fact that the extra 2x happens to be in cash is of pretty marginal value. I believe the same ERN series had an earlier article with a pretty critical view of having a "cash cushion" that was taken from the portfolio rather than in addition to it.

EDIT:
Here is the link to ERN earlier article regarding the limits of a "cash cushion":
https://earlyretirementnow.com/2017/03/ ... sh-cushion
The 'additional' 2 years of cash means that the withdrawal rate is 1/27.5x instead of 1/25x (i.e. 3.64% vs. 4% WR).

IIRC, Karsten said that he was initially skeptical of this approach but was surprised by how well it worked historically. I'm not aware that any of the portfolios or approaches he's tested that used 'conventional' assets like stocks, bonds, and cash would have otherwise supported a 3.64% SWR for 50 years across all historic periods.
Last edited by willthrill81 on Tue Nov 24, 2020 6:53 pm, edited 1 time in total.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

Kintora wrote: Tue Nov 24, 2020 4:56 pm
willthrill81 wrote: Tue Nov 24, 2020 2:33 pm
birdog wrote: Tue Nov 24, 2020 2:15 pm
EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.
When you say that the cash is only spent when the portfolio is down more than 20% in the prior year, how is that calculated exactly? Is it an ongoing check of the portfolio value against one year ago from current?
Using this approach, you only make annual withdrawals, making it easy to determine if your portfolio was down more than 20% in the prior year.
Kintora wrote: Tue Nov 24, 2020 4:56 pm Does it include reduction due to normal withdrawals?
Karsten didn't make it clear how he tested it, but I believe that it would include the prior year's withdrawals (i.e. they are part of the portfolio reduction).
Kintora wrote: Tue Nov 24, 2020 4:56 pm If it recovers back to above being 20% down before the 2 years of cash is depleted, you would go back to stock/bond withdrawals until it possibly drops 20% again and then go back to the cash well?
Yes.
Kintora wrote: Tue Nov 24, 2020 4:56 pm Sorry if any of these are stupid questions. Just trying to wrap my head around this one as it sounds quite interesting.
I highly recommend that you read Karsten's post and test of the approach (link above).
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
marcopolo
Posts: 3849
Joined: Sat Dec 03, 2016 10:22 am

Re: Withdrawal rate for an early retirement

Post by marcopolo »

willthrill81 wrote: Tue Nov 24, 2020 6:47 pm
marcopolo wrote: Tue Nov 24, 2020 6:38 pm
willthrill81 wrote: Tue Nov 24, 2020 2:33 pm
birdog wrote: Tue Nov 24, 2020 2:15 pm
EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.

You have pointed to this article in the past as well.
When doing so, I think it would paint a clearer picture if you included the fact that the cash bucket was in addition to the portfolio.

Most of the stated benefit come from having 27x spending rather than 25x. The fact that the extra 2x happens to be in cash is of pretty marginal value. I believe the same ERN series had an earlier article with a pretty critical view of having a "cash cushion" that was taken from the portfolio rather than in addition to it.

EDIT:
Here is the link to ERN earlier article regarding the limits of a "cash cushion":
https://earlyretirementnow.com/2017/03/ ... sh-cushion
The 'additional' 2 years of cash means that the withdrawal rate is 1/27.5x instead of 1/25x (i.e. 3.64% vs. 4% WR).

IIRC, Karsten said that he was initially skeptical of this approach but was surprised by how well it worked historically.
Well sure, but that cash cushion does not come for free.
Following that logic, why not keep 10 years of cash cushion, I am sure 35x would be even safer than 27x. :oops:

It seems to me that the proper comparison is to consider the same size portfolio and see what happens when you take a slice out it and keep it in cash as cushion against SORR. That is what the earlier article I linked did, and the results were less than convincing.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

marcopolo wrote: Tue Nov 24, 2020 6:54 pm
willthrill81 wrote: Tue Nov 24, 2020 6:47 pm
marcopolo wrote: Tue Nov 24, 2020 6:38 pm
willthrill81 wrote: Tue Nov 24, 2020 2:33 pm
birdog wrote: Tue Nov 24, 2020 2:15 pm

Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.

You have pointed to this article in the past as well.
When doing so, I think it would paint a clearer picture if you included the fact that the cash bucket was in addition to the portfolio.

Most of the stated benefit come from having 27x spending rather than 25x. The fact that the extra 2x happens to be in cash is of pretty marginal value. I believe the same ERN series had an earlier article with a pretty critical view of having a "cash cushion" that was taken from the portfolio rather than in addition to it.

EDIT:
Here is the link to ERN earlier article regarding the limits of a "cash cushion":
https://earlyretirementnow.com/2017/03/ ... sh-cushion
The 'additional' 2 years of cash means that the withdrawal rate is 1/27.5x instead of 1/25x (i.e. 3.64% vs. 4% WR).

IIRC, Karsten said that he was initially skeptical of this approach but was surprised by how well it worked historically.
Well sure, but that cash cushion does not come for free.
Following that logic, why not keep 10 years of cash cushion, I am sure 35x would be even safer than 27x. :oops:

It seems to me that the proper comparison is to consider the same size portfolio and see what happens when you take a slice out it and keep it in cash as cushion against SORR. That is what the earlier article I linked did, and the results were less than convincing.
The 3.64% I quoted is taking the entire 27.5 years of expenses into account, not 3.64% of 25x with 2 more years of additional spending.

Karsten said this about it:
So, for the record, let me state that this cash bucket strategy seems to work pretty well, despite my previous doubts! It’s relatively inexpensive insurance against Sequence Risk! Think of it as a mini-glidepath during the first few years of retirement!
However, I just realized that Karsten didn't say how much residual funds were left for 1966 retirees after 50 years using this approach, just that it worked. Considering that the 50 year SWR for 1966 using an 80/20 was 3.60% already, I'm confident that the SWR of this approach would have been significantly higher than that.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
Kintora
Posts: 73
Joined: Sun Jan 12, 2020 5:21 pm
Location: Fargo, ND

Re: Withdrawal rate for an early retirement

Post by Kintora »

willthrill81 wrote: Tue Nov 24, 2020 6:52 pm
Kintora wrote: Tue Nov 24, 2020 4:56 pm
willthrill81 wrote: Tue Nov 24, 2020 2:33 pm
birdog wrote: Tue Nov 24, 2020 2:15 pm
EnjoyIt wrote: Tue Nov 24, 2020 1:56 am We don't have a dedicated emergency fund and see no need for one. To me an emergency fund is cash wasted away. Correction, our emergency fund is our bond portfolio followed by our equities portfolio.
Big ERN has talked about a cash position as a way to guard against SORR. He said that it could sit idle unless your portfolio drops by 20% and then your withdrawals would come from your cash account until it is depleted and then never refilled. I've always thought, like you, that cash sitting idle (.5% at Marcus right now) is a drag on your returns. But I think he has a point in that it could serve as a protection against SORR. I'm strongly considering a couple years expenses in a HYSA for that reason. Thoughts?
The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.
When you say that the cash is only spent when the portfolio is down more than 20% in the prior year, how is that calculated exactly? Is it an ongoing check of the portfolio value against one year ago from current?
Using this approach, you only make annual withdrawals, making it easy to determine if your portfolio was down more than 20% in the prior year.
Kintora wrote: Tue Nov 24, 2020 4:56 pm Does it include reduction due to normal withdrawals?
Karsten didn't make it clear how he tested it, but I believe that it would include the prior year's withdrawals (i.e. they are part of the portfolio reduction).
Kintora wrote: Tue Nov 24, 2020 4:56 pm If it recovers back to above being 20% down before the 2 years of cash is depleted, you would go back to stock/bond withdrawals until it possibly drops 20% again and then go back to the cash well?
Yes.
Kintora wrote: Tue Nov 24, 2020 4:56 pm Sorry if any of these are stupid questions. Just trying to wrap my head around this one as it sounds quite interesting.
I highly recommend that you read Karsten's post and test of the approach (link above).
Thanks for your replies. I will take a closer look.
marcopolo
Posts: 3849
Joined: Sat Dec 03, 2016 10:22 am

Re: Withdrawal rate for an early retirement

Post by marcopolo »

willthrill81 wrote: Tue Nov 24, 2020 7:06 pm
marcopolo wrote: Tue Nov 24, 2020 6:54 pm
willthrill81 wrote: Tue Nov 24, 2020 6:47 pm
marcopolo wrote: Tue Nov 24, 2020 6:38 pm
willthrill81 wrote: Tue Nov 24, 2020 2:33 pm

The idea was put forward by Fritz of The Retirement Manifesto, and Karsten of ERN analyzed it. The specific method tested was to keep two years of expenses in cash and the rest of the portfolio in an 80/20 AA. The cash is only spent when the portfolio is down more than 20% in the prior year, and if/when the cash is exhausted, it is never replenished (so this isn't 'mere mental accounting'). Karsten found that this would have supported a 3.64% SWR for 50 years in the worse historic periods, starting in 1929 and 1966, quite an impressive feat.

You have pointed to this article in the past as well.
When doing so, I think it would paint a clearer picture if you included the fact that the cash bucket was in addition to the portfolio.

Most of the stated benefit come from having 27x spending rather than 25x. The fact that the extra 2x happens to be in cash is of pretty marginal value. I believe the same ERN series had an earlier article with a pretty critical view of having a "cash cushion" that was taken from the portfolio rather than in addition to it.

EDIT:
Here is the link to ERN earlier article regarding the limits of a "cash cushion":
https://earlyretirementnow.com/2017/03/ ... sh-cushion
The 'additional' 2 years of cash means that the withdrawal rate is 1/27.5x instead of 1/25x (i.e. 3.64% vs. 4% WR).

IIRC, Karsten said that he was initially skeptical of this approach but was surprised by how well it worked historically.
Well sure, but that cash cushion does not come for free.
Following that logic, why not keep 10 years of cash cushion, I am sure 35x would be even safer than 27x. :oops:

It seems to me that the proper comparison is to consider the same size portfolio and see what happens when you take a slice out it and keep it in cash as cushion against SORR. That is what the earlier article I linked did, and the results were less than convincing.
The 3.64% I quoted is taking the entire 27.5 years of expenses into account, not 3.64% of 25x with 2 more years of additional spending.

Karsten said this about it:
So, for the record, let me state that this cash bucket strategy seems to work pretty well, despite my previous doubts! It’s relatively inexpensive insurance against Sequence Risk! Think of it as a mini-glidepath during the first few years of retirement!
However, I just realized that Karsten didn't say how much residual funds were left for 1966 retirees after 50 years using this approach, just that it worked. Considering that the 50 year SWR for 1966 using an 80/20 was 3.60% already, I'm confident that the SWR of this approach would have been significantly higher than that.
I think you may have a small error in your math, and perhaps a different understanding of what that ERN simulation did than the way i read it.

The way I understood it is he started with a $1M portfolio with 80/20 AA, and taking out $40k/yr (4%). This caused failures in 1929 and 1966 starting years. He then slowly added an additional cash amount until the portfolio just barely survived (which i think would answer your question about excess funds). What he found was the 1929 case needed and additional $100,078 in cash to succeed. The 1966 case needed and additional $115,290 in cash to succeed. So, the total portfolio size is $1,115,290, withdrawing $40k/yr. When I do the math, I get 3.59%. (how did you get 3.64?). That does not seem that far off from his result for a simple 80/20 portfolio. His base case chart does not show a 80/20 AA case, or a 3.6% WR, so it is hard to tell for sure, but I do believe the extra cash cushion case was maybe a small fraction of a percentage better. But, the impact is pretty small from my reading of the data. It seems like noise in the scheme of things.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

marcopolo wrote: Tue Nov 24, 2020 8:59 pm
willthrill81 wrote: Tue Nov 24, 2020 7:06 pm
marcopolo wrote: Tue Nov 24, 2020 6:54 pm
willthrill81 wrote: Tue Nov 24, 2020 6:47 pm
marcopolo wrote: Tue Nov 24, 2020 6:38 pm


You have pointed to this article in the past as well.
When doing so, I think it would paint a clearer picture if you included the fact that the cash bucket was in addition to the portfolio.

Most of the stated benefit come from having 27x spending rather than 25x. The fact that the extra 2x happens to be in cash is of pretty marginal value. I believe the same ERN series had an earlier article with a pretty critical view of having a "cash cushion" that was taken from the portfolio rather than in addition to it.

EDIT:
Here is the link to ERN earlier article regarding the limits of a "cash cushion":
https://earlyretirementnow.com/2017/03/ ... sh-cushion
The 'additional' 2 years of cash means that the withdrawal rate is 1/27.5x instead of 1/25x (i.e. 3.64% vs. 4% WR).

IIRC, Karsten said that he was initially skeptical of this approach but was surprised by how well it worked historically.
Well sure, but that cash cushion does not come for free.
Following that logic, why not keep 10 years of cash cushion, I am sure 35x would be even safer than 27x. :oops:

It seems to me that the proper comparison is to consider the same size portfolio and see what happens when you take a slice out it and keep it in cash as cushion against SORR. That is what the earlier article I linked did, and the results were less than convincing.
The 3.64% I quoted is taking the entire 27.5 years of expenses into account, not 3.64% of 25x with 2 more years of additional spending.

Karsten said this about it:
So, for the record, let me state that this cash bucket strategy seems to work pretty well, despite my previous doubts! It’s relatively inexpensive insurance against Sequence Risk! Think of it as a mini-glidepath during the first few years of retirement!
However, I just realized that Karsten didn't say how much residual funds were left for 1966 retirees after 50 years using this approach, just that it worked. Considering that the 50 year SWR for 1966 using an 80/20 was 3.60% already, I'm confident that the SWR of this approach would have been significantly higher than that.
I think you may have a small error in your math, and perhaps a different understanding of what that ERN simulation did than the way i read it.

The way I understood it is he started with a $1M portfolio with 80/20 AA, and taking out $40k/yr (4%). This caused failures in 1929 and 1966 starting years. He then slowly added an additional cash amount until the portfolio just barely survived (which i think would answer your question about excess funds). What he found was the 1929 case needed and additional $100,078 in cash to succeed. The 1966 case needed and additional $115,290 in cash to succeed. So, the total portfolio size is $1,115,290, withdrawing $40k/yr. When I do the math, I get 3.59%. (how did you get 3.64?). That does not seem that far off from his result for a simple 80/20 portfolio. His base case chart does not show a 80/20 AA case, or a 3.6% WR, so it is hard to tell for sure, but I do believe the extra cash cushion case was maybe a small fraction of a percentage better. But, the impact is pretty small from my reading of the data. It seems like noise in the scheme of things.
1/27.5x = 3.64%. That represents 25x (i.e. 4% SWR) plus 2x of additional cash.

Either way, we're splitting hairs. A 3.6% SWR-style withdrawal scheme having survived for 50 years, whether with a fixed AA or with the cash bucket strategy, in the worst historic scenarios is pretty impressive IMHO.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
luminous
Posts: 278
Joined: Sat Dec 30, 2017 10:28 pm

Re: Withdrawal rate for an early retirement

Post by luminous »

Rob1 wrote: Tue Nov 24, 2020 2:38 am A progression of withdrawal thinking
  • The 4% rule: it’s so simple!
  • SWR rabbit hole: hm maybe 5%, or 3%...no it’s 3.14159%
  • Dynamic withdrawal rates (Guyton-Klinger guard rails et al): it’s so obvious...flexibility is key!
  • Variable Percentage Withdrawal (VPW): simplicity is the ultimate sophistication
  • Amortization Based Withdrawal (ABW aka TVM): actually I’ll take a bit of complexity and customization with my VPW
ABW is enlightenment! Or is it?

Next? :D
Oh man, bingo. Too funny. :sharebeer
50/20/30 US stock/international stock/bonds. Hope to semi-retire in 2022.
Topic Author
nptit
Posts: 57
Joined: Sun Mar 29, 2015 5:51 pm

Re: Withdrawal rate for an early retirement

Post by nptit »

JoMoney wrote: Tue Nov 17, 2020 10:38 pm Personally, the methods I like to us is 1 divided your expected life expectancy
A life expectancy table would suggest a 45 yo has 38.8 years
https://www.fidelity.com/content/apps/m ... -Table.pdf

1/38.8 = 2.57%

Without expecting to earn anything on the money, one could reasonably expect to spend 2.57% a year for the remainder of their life without running out. If the money was put into TIPS they could expect to take that with an inflation adjustment. When they got towards the later end of their life they could likely take the remaining portfolio and purchase a Single Premium Immediate Annuity that would guarantee income the rest of their life (without an inflation adjustment) and possibly have Social Security income as well (SS would offer an inflation adjustment)

Another method would be to simply price out what an immediate annuity bought today could provide in guaranteed income
https://www.immediateannuities.com/
Not necessarily to actually buy one, but just to see ball-park what an insurance company believes they could safely guarantee you for the rest of your life (bear in mind SPIA doesn't offer any inflation adjustments).
Thanks for the recommendation, I asked this question on behalf of my wife. I am comfortable to retire with 3-3.5 withdraw rate, wife wants 2% so we don’t run out of money and have some left to the children and grand children.
Topic Author
nptit
Posts: 57
Joined: Sun Mar 29, 2015 5:51 pm

Re: Withdrawal rate for an early retirement

Post by nptit »

Marseille07 wrote: Tue Nov 17, 2020 10:43 pm Similar situation here, my approach would simply be 2~2.5% "constant-percentage." I'm not a big fan of SWR + inflation because (imo) it's nonsensical to withdraw the same amount after a year like 2008. Constant-percentage would automatically adjust for such events.
My wife wants keep it simple as well, she wants to be at the safe side so we don’t end up needing come back to work because of running out of money. We are planning to have 2 MM left for children and grandchildren.
Topic Author
nptit
Posts: 57
Joined: Sun Mar 29, 2015 5:51 pm

Re: Withdrawal rate for an early retirement

Post by nptit »

FrugalInvestor wrote: Mon Nov 23, 2020 7:42 pm Here's a data point for you....

When most people were perfectly comfortable with a SWR of 4% + inflation rate I retired early using a 3% guideline with no inflation adjustment (for the then foreseeable future). We were also prepared to pull in our horns should a black swan event occur, which happened with the financial crisis of '08-'09. So far all has gone well for us in retirement. We've stuck very closely to our original 3% guideline but as our portfolio has continued to grow the dollars available to us for spending has increased fairly substantially. Now that we are at normal retirement age we are in the process of ratcheting our withdrawal guideline upward.

If I were to retire today I would adopt an even lower withdrawal rate and remain prepared to adjust downward in difficult economic times as well. I don't know exactly what that number would be but likely 2.5% or less.
Thank you for sharing your experience
Marseille07
Posts: 1073
Joined: Fri Nov 06, 2020 1:41 pm

Re: Withdrawal rate for an early retirement

Post by Marseille07 »

nptit wrote: Tue Nov 24, 2020 11:10 pm
Marseille07 wrote: Tue Nov 17, 2020 10:43 pm Similar situation here, my approach would simply be 2~2.5% "constant-percentage." I'm not a big fan of SWR + inflation because (imo) it's nonsensical to withdraw the same amount after a year like 2008. Constant-percentage would automatically adjust for such events.
My wife wants keep it simple as well, she wants to be at the safe side so we don’t end up needing come back to work because of running out of money. We are planning to have 2 MM left for children and grandchildren.
If you already have some amount to leave behind, I believe ABW can factor that in and calculate WR, which will likely give you much higher numbers than 2~2.5%. I do not know if ABW accounts for inflation though.
YRT70
Posts: 801
Joined: Sat Apr 27, 2019 8:51 am

Re: Withdrawal rate for an early retirement

Post by YRT70 »

willthrill81 wrote: Tue Nov 24, 2020 10:11 pm A 3.6% SWR-style withdrawal scheme having survived for 50 years, whether with a fixed AA or with the cash bucket strategy, in the worst historic scenarios is pretty impressive IMHO.
Agreed. Could this rising equity glide path strategy be even better?

3.56% for 60 years retirement.

Image

Source: https://earlyretirementnow.com/2017/09/ ... lidepaths/
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

YRT70 wrote: Wed Nov 25, 2020 11:35 am
willthrill81 wrote: Tue Nov 24, 2020 10:11 pm A 3.6% SWR-style withdrawal scheme having survived for 50 years, whether with a fixed AA or with the cash bucket strategy, in the worst historic scenarios is pretty impressive IMHO.
Agreed. Could this rising equity glide path strategy be even better?

3.56% for 60 years retirement.

Image

Source: https://earlyretirementnow.com/2017/09/ ... lidepaths/
In that post, Karsten was looking at 60 year SWRs, not 50 years, which leads to different historic results.

Many here have pointed out that the rising equity glidepath didn't really do anything meaningful compared to static AAs. And in subsequent work, I believe that Pfau and Kitces have basically said as much.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
YRT70
Posts: 801
Joined: Sat Apr 27, 2019 8:51 am

Re: Withdrawal rate for an early retirement

Post by YRT70 »

willthrill81 wrote: Wed Nov 25, 2020 11:41 am
YRT70 wrote: Wed Nov 25, 2020 11:35 am
willthrill81 wrote: Tue Nov 24, 2020 10:11 pm A 3.6% SWR-style withdrawal scheme having survived for 50 years, whether with a fixed AA or with the cash bucket strategy, in the worst historic scenarios is pretty impressive IMHO.
Agreed. Could this rising equity glide path strategy be even better?

3.56% for 60 years retirement.

Image

Source: https://earlyretirementnow.com/2017/09/ ... lidepaths/
In that post, Karsten was looking at 60 year SWRs, not 50 years, which leads to different historic results.
I'm aware. that's why I wrote 60 years.
Many here have pointed out that the rising equity glidepath didn't really do anything meaningful compared to static AAs.
I'd say ERN's analysis is showing something different. In 1966 the glide path's SWR was ~0.33% higher than 60/40 fixed. That can be a lot of money.
And in subsequent work, I believe that Pfau and Kitces have basically said as much.
Having listened to several recent podcasts with Pfau and Kitces, I don't think they'd entirely agree on that.
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

YRT70 wrote: Wed Nov 25, 2020 11:53 am
willthrill81 wrote: Wed Nov 25, 2020 11:41 am Many here have pointed out that the rising equity glidepath didn't really do anything meaningful compared to static AAs.
I'd say ERN's analysis is showing something different. In 1966 the glide path's SWR was ~0.33% higher than 60/40 fixed. That can be a lot of money.
Perhaps the rising equity glidepath really did work better for 60 year retirements than for 30 year retirements. I haven't personally dug into it deeply enough to verify Karsten's finding.
YRT70 wrote: Wed Nov 25, 2020 11:53 am
willthrill81 wrote: Wed Nov 25, 2020 11:41 am And in subsequent work, I believe that Pfau and Kitces have basically said as much.
Having listened to several recent podcasts with Pfau and Kitces, I don't think they'd entirely agree on that.
Take a look at Kitces' original post on the topic, with the chart below. The difference in the historic likelihood of success for a rising equity glidepath from 30% stock to 70% stock was 95.1% but 94.6% for a static 40/60. There's no way that the .5% advantage would be statistically significant.
Image
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
geerhardusvos
Posts: 1516
Joined: Wed Oct 23, 2019 10:20 pm
Location: heavenlies

Re: Withdrawal rate for an early retirement

Post by geerhardusvos »

nptit wrote: Tue Nov 24, 2020 11:10 pm
Marseille07 wrote: Tue Nov 17, 2020 10:43 pm Similar situation here, my approach would simply be 2~2.5% "constant-percentage." I'm not a big fan of SWR + inflation because (imo) it's nonsensical to withdraw the same amount after a year like 2008. Constant-percentage would automatically adjust for such events.
My wife wants keep it simple as well, she wants to be at the safe side so we don’t end up needing come back to work because of running out of money. We are planning to have 2 MM left for children and grandchildren.
You will never need to go below 3% withdrawal rate in the most conservative scenario. Very likely your portfolio will continue to grow with a 3.5% withdrawal rate. 4% WR could end up being extremely conservative for you in the long run. You could easily leave much more than $2 million to your children with a 3.5% withdrawal rate. Don’t buy into the below 3% crowd, their answers are not grounded in reality or real historical data (which is the only reliable way of predicting the possibilities of the future). It’s very possible that the future is brighter and that markets will be more stable.

Think about it, in 2019 we had almost 30% returns if you had an all equity portfolio. So if you had $3 million, your portfolio went up $900,000. That’s just one year. Would someone who retired in 2018 really have to keep withdrawing just 4%? No way, they could very comfortably increase to 4.5%. And with a $3.9 million portfolio, that’s a brand new Tesla or some nice vacations.

Even if we have measly or negative real returns over the next 10 to 15 years, it’s followed by the next cycle of growth. Just keep tabs on the performance of the portfolio, and you will be fine. There are many withdrawal rate strategies that are very simple that take into account market factors, and only require a couple hours per year to manage. You will spend more time doing laundry in a month then you will managing your portfolio in a year. Don’t sell yourself short. Tell your wife that 3.5% withdrawal rate is extremely conservative. And if the market goes down 30%, you can go down to 3.2% and be safe.
VTSAX and chill
Marseille07
Posts: 1073
Joined: Fri Nov 06, 2020 1:41 pm

Re: Withdrawal rate for an early retirement

Post by Marseille07 »

willthrill81 wrote: Wed Nov 25, 2020 12:07 pm
YRT70 wrote: Wed Nov 25, 2020 11:53 am
willthrill81 wrote: Wed Nov 25, 2020 11:41 am Many here have pointed out that the rising equity glidepath didn't really do anything meaningful compared to static AAs.
I'd say ERN's analysis is showing something different. In 1966 the glide path's SWR was ~0.33% higher than 60/40 fixed. That can be a lot of money.
Perhaps the rising equity glidepath really did work better for 60 year retirements than for 30 year retirements. I haven't personally dug into it deeply enough to verify Karsten's finding.
YRT70 wrote: Wed Nov 25, 2020 11:53 am
willthrill81 wrote: Wed Nov 25, 2020 11:41 am And in subsequent work, I believe that Pfau and Kitces have basically said as much.
Having listened to several recent podcasts with Pfau and Kitces, I don't think they'd entirely agree on that.
Take a look at Kitces' original post on the topic, with the chart below. The difference in the historic likelihood of success for a rising equity glidepath from 30% stock to 70% stock was 95.1% but 94.6% for a static 40/60. There's no way that the .5% advantage would be statistically significant.
Image
I don't see the rising equity glidepath being realistic. Do folks really want to go from 60/40 at age 60 to 90/10 at age 80 (just as an example)?
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

Marseille07 wrote: Wed Nov 25, 2020 12:41 pm
willthrill81 wrote: Wed Nov 25, 2020 12:07 pm
YRT70 wrote: Wed Nov 25, 2020 11:53 am
willthrill81 wrote: Wed Nov 25, 2020 11:41 am Many here have pointed out that the rising equity glidepath didn't really do anything meaningful compared to static AAs.
I'd say ERN's analysis is showing something different. In 1966 the glide path's SWR was ~0.33% higher than 60/40 fixed. That can be a lot of money.
Perhaps the rising equity glidepath really did work better for 60 year retirements than for 30 year retirements. I haven't personally dug into it deeply enough to verify Karsten's finding.
YRT70 wrote: Wed Nov 25, 2020 11:53 am
willthrill81 wrote: Wed Nov 25, 2020 11:41 am And in subsequent work, I believe that Pfau and Kitces have basically said as much.
Having listened to several recent podcasts with Pfau and Kitces, I don't think they'd entirely agree on that.
Take a look at Kitces' original post on the topic, with the chart below. The difference in the historic likelihood of success for a rising equity glidepath from 30% stock to 70% stock was 95.1% but 94.6% for a static 40/60. There's no way that the .5% advantage would be statistically significant.
Image
I don't see the rising equity glidepath being realistic. Do folks really want to go from 60/40 at age 60 to 90/10 at age 80 (just as an example)?
That's a good point that others have made as well.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
vineviz
Posts: 8495
Joined: Tue May 15, 2018 1:55 pm

Re: Withdrawal rate for an early retirement

Post by vineviz »

willthrill81 wrote: Wed Nov 25, 2020 12:07 pm Take a look at Kitces' original post on the topic, with the chart below. The difference in the historic likelihood of success for a rising equity glidepath from 30% stock to 70% stock was 95.1% but 94.6% for a static 40/60. There's no way that the .5% advantage would be statistically significant.
Image
I think it's also worth noting that his conclusion was very sensitive to the return assumptions, especially the 2.4% real return that Kitces used for bond returns.

If you re-run his analysis but with more realistic return assumptions based on current conditions (e.g 5% real returns for VTSMX and 0% real returns for VBMFX) the conclusion flips: a constant 50/50 allocation survives more often (67.26%) than a gliding allocation from 30/70 to 70/30 (60.78%).
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
YRT70
Posts: 801
Joined: Sat Apr 27, 2019 8:51 am

Re: Withdrawal rate for an early retirement

Post by YRT70 »

willthrill81 wrote: Wed Nov 25, 2020 12:07 pm Take a look at Kitces' original post on the topic, with the chart below. The difference in the historic likelihood of success for a rising equity glidepath from 30% stock to 70% stock was 95.1% but 94.6% for a static 40/60. There's no way that the .5% advantage would be statistically significant.
Image
I'm aware of the article. I think you made a mistake in reading the data. The rising equity glide path got 95.1%, while the static 60/40 got 93.2% (not 94.6%) under these assumptions. So that's a 1.9% advantage under his assumptions (historical averages). The static 40/60 did get 94.6%.

Iirc it was in a relatively recent Rational Reminder podcast in which Kitces talked about bond tents. I did not get the impression he has changed his mind about it, at all. He still believes they have a benefit.

Here's another way of looking at Kitces' chart: for 9 of the 10 static AAs there was a rising equity glide path that did better. Sometimes a little, sometimes a lot.
luckyducky99
Posts: 163
Joined: Sun Dec 15, 2019 7:47 pm

Re: Withdrawal rate for an early retirement

Post by luckyducky99 »

geerhardusvos wrote: Wed Nov 25, 2020 12:22 pm You will never need to go below 3% withdrawal rate in the most conservative scenario. Very likely your portfolio will continue to grow with a 3.5% withdrawal rate. 4% WR could end up being extremely conservative for you in the long run. You could easily leave much more than $2 million to your children with a 3.5% withdrawal rate. Don’t buy into the below 3% crowd, their answers are not grounded in reality or real historical data (which is the only reliable way of predicting the possibilities of the future). It’s very possible that the future is brighter and that markets will be more stable.
"Never" is a strong word, and I think it's an unreasonable one to use when talking about an unknown future.

To say that people who think that the risk of the future being worse than the relatively short past is real, and who think that that risk is worth insuring against, are "not grounded in reality" is also unreasonable.
User avatar
geerhardusvos
Posts: 1516
Joined: Wed Oct 23, 2019 10:20 pm
Location: heavenlies

Re: Withdrawal rate for an early retirement

Post by geerhardusvos »

luckyducky99 wrote: Wed Nov 25, 2020 1:49 pm
geerhardusvos wrote: Wed Nov 25, 2020 12:22 pm You will never need to go below 3% withdrawal rate in the most conservative scenario. Very likely your portfolio will continue to grow with a 3.5% withdrawal rate. 4% WR could end up being extremely conservative for you in the long run. You could easily leave much more than $2 million to your children with a 3.5% withdrawal rate. Don’t buy into the below 3% crowd, their answers are not grounded in reality or real historical data (which is the only reliable way of predicting the possibilities of the future). It’s very possible that the future is brighter and that markets will be more stable.
"Never" is a strong word, and I think it's an unreasonable one to use when talking about an unknown future.

To say that people who think that the risk of the future being worse than the relatively short past is real, and who think that that risk is worth insuring against, are "not grounded in reality" is also unreasonable.
Is 50 years the relatively short past? How about 150 years?

Is being worried that you are going to die in a car accident every day reasonable? I am OK saying that if I drive safely it will never happen to me, in fact by being willing to drive in my car I am saying that I am OK with the level of risk of dying in a car crash. But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...

The definition of reasonable is the possibility of an event or events happening based on what we know to be possible given our history and current variables. In this sense, it is quite reasonable to say that you will never run out of money if you have a high equity portfolio and a 3% withdraw it.

If the 3% withdrawal rate doesn’t work over the next 60 years, your portfolio being depleted will not be your biggest concern.
VTSAX and chill
EnjoyIt
Posts: 5311
Joined: Sun Dec 29, 2013 8:06 pm

Re: Withdrawal rate for an early retirement

Post by EnjoyIt »

geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm
luckyducky99 wrote: Wed Nov 25, 2020 1:49 pm
geerhardusvos wrote: Wed Nov 25, 2020 12:22 pm You will never need to go below 3% withdrawal rate in the most conservative scenario. Very likely your portfolio will continue to grow with a 3.5% withdrawal rate. 4% WR could end up being extremely conservative for you in the long run. You could easily leave much more than $2 million to your children with a 3.5% withdrawal rate. Don’t buy into the below 3% crowd, their answers are not grounded in reality or real historical data (which is the only reliable way of predicting the possibilities of the future). It’s very possible that the future is brighter and that markets will be more stable.
"Never" is a strong word, and I think it's an unreasonable one to use when talking about an unknown future.

To say that people who think that the risk of the future being worse than the relatively short past is real, and who think that that risk is worth insuring against, are "not grounded in reality" is also unreasonable.
Is 50 years the relatively short past? How about 150 years?

Is being worried that you are going to die in a car accident every day reasonable? I am OK saying that if I drive safely it will never happen to me, in fact by being willing to drive in my car I am saying that I am OK with the level of risk of dying in a car crash. But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...

The definition of reasonable is the possibility of an event or events happening based on what we know to be possible given our history and current variables. In this sense, it is quite reasonable to say that you will never run out of money if you have a high equity portfolio and a 3% withdraw it.

If the 3% withdrawal rate doesn’t work over the next 60 years, your portfolio being depleted will not be your biggest concern.
I will agree with one thing. If returns are so poor that one feels they need less than 3% withdrawal rates to make that cash last, then it is just as likely that 2%, 1% or even less won’t be enough either. There will be a calamity that your money won’t solve. There are ways to protect yourself though. Instead of saving an extra 25 or 50 years worth of expenses, one can make their home off grid capable. They will need to have a farm and raise livestock. They can learn to make ammunition and stock pile plenty of resources. This will accomplish two things. 1) by creating a self sustaining living situation one’s fixed expenses go down significantly requiring a much smaller portfolio. 2) you actually protect yourself from an economic collapse where 3% withdrawal rate fails.
A time to EVALUATE your jitters: | https://www.bogleheads.org/forum/viewtopic.php?f=10&t=79939&start=400#p5275418
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

YRT70 wrote: Wed Nov 25, 2020 1:22 pm
willthrill81 wrote: Wed Nov 25, 2020 12:07 pm Take a look at Kitces' original post on the topic, with the chart below. The difference in the historic likelihood of success for a rising equity glidepath from 30% stock to 70% stock was 95.1% but 94.6% for a static 40/60. There's no way that the .5% advantage would be statistically significant.
Image
I'm aware of the article. I think you made a mistake in reading the data. The rising equity glide path got 95.1%, while the static 60/40 got 93.2% (not 94.6%) under these assumptions. So that's a 1.9% advantage under his assumptions (historical averages). The static 40/60 did get 94.6%.

Iirc it was in a relatively recent Rational Reminder podcast in which Kitces talked about bond tents. I did not get the impression he has changed his mind about it, at all. He still believes they have a benefit.

Here's another way of looking at Kitces' chart: for 9 of the 10 static AAs there was a rising equity glide path that did better. Sometimes a little, sometimes a lot.
I specifically referenced the 40/60 portfolio, not a 60/40.

In that table, the rising equity glidepath never improved the historical odds of success by more than .5%. As I said, that's not statistically significant, meaning that we cannot place any real confidence that there was any improvement at all by using the rising equity glidepath.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

EnjoyIt wrote: Wed Nov 25, 2020 2:20 pm
geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm
luckyducky99 wrote: Wed Nov 25, 2020 1:49 pm
geerhardusvos wrote: Wed Nov 25, 2020 12:22 pm You will never need to go below 3% withdrawal rate in the most conservative scenario. Very likely your portfolio will continue to grow with a 3.5% withdrawal rate. 4% WR could end up being extremely conservative for you in the long run. You could easily leave much more than $2 million to your children with a 3.5% withdrawal rate. Don’t buy into the below 3% crowd, their answers are not grounded in reality or real historical data (which is the only reliable way of predicting the possibilities of the future). It’s very possible that the future is brighter and that markets will be more stable.
"Never" is a strong word, and I think it's an unreasonable one to use when talking about an unknown future.

To say that people who think that the risk of the future being worse than the relatively short past is real, and who think that that risk is worth insuring against, are "not grounded in reality" is also unreasonable.
Is 50 years the relatively short past? How about 150 years?

Is being worried that you are going to die in a car accident every day reasonable? I am OK saying that if I drive safely it will never happen to me, in fact by being willing to drive in my car I am saying that I am OK with the level of risk of dying in a car crash. But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...

The definition of reasonable is the possibility of an event or events happening based on what we know to be possible given our history and current variables. In this sense, it is quite reasonable to say that you will never run out of money if you have a high equity portfolio and a 3% withdraw it.

If the 3% withdrawal rate doesn’t work over the next 60 years, your portfolio being depleted will not be your biggest concern.
I will agree with one thing. If returns are so poor that one feels they need less than 3% withdrawal rates to make that cash last, then it is just as likely that 2%, 1% or even less won’t be enough either. There will be a calamity that your money won’t solve. There are ways to protect yourself though. Instead of saving an extra 25 or 50 years worth of expenses, one can make their home off grid capable. They will need to have a farm and raise livestock. They can learn to make ammunition and stock pile plenty of resources. This will accomplish two things. 1) by creating a self sustaining living situation one’s fixed expenses go down significantly requiring a much smaller portfolio. 2) you actually protect yourself from an economic collapse where 3% withdrawal rate fails.
I've said in the past that a fruit tree in your backyard is like a little Roth IRA. And its performance will probably be uncorrelated with everything else in your portfolio to boot. Win win. 8-)
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
EnjoyIt
Posts: 5311
Joined: Sun Dec 29, 2013 8:06 pm

Re: Withdrawal rate for an early retirement

Post by EnjoyIt »

Vineviz,

My understanding is that you believe a 2% withdrawal rate is what is needed, correct? If that is the case have you considered spending money to decrease your cost of living?

For example, pay $30k for solar which will provide well over $600 worth of electricity and better than the 2% you discuss. Some more examples, build a well, geothermal heating, have a garden. Or better yet, buy farmland.

Actually, if all I needed was to outpace 2% real returns I would probably do everything in my power to cut my fixed expenses and then get involved in real estate rentals across the country with a hired manager for everything. After all, all I need is 2% real which is not that hard to accomplish with a diversified real estate portfolio.

Am I missing something?
A time to EVALUATE your jitters: | https://www.bogleheads.org/forum/viewtopic.php?f=10&t=79939&start=400#p5275418
EnjoyIt
Posts: 5311
Joined: Sun Dec 29, 2013 8:06 pm

Re: Withdrawal rate for an early retirement

Post by EnjoyIt »

willthrill81 wrote: Wed Nov 25, 2020 2:26 pm
EnjoyIt wrote: Wed Nov 25, 2020 2:20 pm
geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm
luckyducky99 wrote: Wed Nov 25, 2020 1:49 pm
geerhardusvos wrote: Wed Nov 25, 2020 12:22 pm You will never need to go below 3% withdrawal rate in the most conservative scenario. Very likely your portfolio will continue to grow with a 3.5% withdrawal rate. 4% WR could end up being extremely conservative for you in the long run. You could easily leave much more than $2 million to your children with a 3.5% withdrawal rate. Don’t buy into the below 3% crowd, their answers are not grounded in reality or real historical data (which is the only reliable way of predicting the possibilities of the future). It’s very possible that the future is brighter and that markets will be more stable.
"Never" is a strong word, and I think it's an unreasonable one to use when talking about an unknown future.

To say that people who think that the risk of the future being worse than the relatively short past is real, and who think that that risk is worth insuring against, are "not grounded in reality" is also unreasonable.
Is 50 years the relatively short past? How about 150 years?

Is being worried that you are going to die in a car accident every day reasonable? I am OK saying that if I drive safely it will never happen to me, in fact by being willing to drive in my car I am saying that I am OK with the level of risk of dying in a car crash. But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...

The definition of reasonable is the possibility of an event or events happening based on what we know to be possible given our history and current variables. In this sense, it is quite reasonable to say that you will never run out of money if you have a high equity portfolio and a 3% withdraw it.

If the 3% withdrawal rate doesn’t work over the next 60 years, your portfolio being depleted will not be your biggest concern.
I will agree with one thing. If returns are so poor that one feels they need less than 3% withdrawal rates to make that cash last, then it is just as likely that 2%, 1% or even less won’t be enough either. There will be a calamity that your money won’t solve. There are ways to protect yourself though. Instead of saving an extra 25 or 50 years worth of expenses, one can make their home off grid capable. They will need to have a farm and raise livestock. They can learn to make ammunition and stock pile plenty of resources. This will accomplish two things. 1) by creating a self sustaining living situation one’s fixed expenses go down significantly requiring a much smaller portfolio. 2) you actually protect yourself from an economic collapse where 3% withdrawal rate fails.
I've said in the past that a fruit tree in your backyard is like a little Roth IRA. And its performance will probably be uncorrelated with everything else in your portfolio to boot. Win win. 8-)
We have talked about installing solar just to decrease fixed expenses and minimize SORR. I’m all for spending money to cut those fixed expenses which should add flexibility. I am pretty sure you and I agree that flexibility is probably the most valuable item in any withdrawal strategy.

A fruit tree is better than an IRA because you can share it with people and make friends. One of my colleagues every year brings in a huge box of oranges to work. I usually grab a dozen or so with plenty left over for others. It’s awesome and makes a lot of people happy.
A time to EVALUATE your jitters: | https://www.bogleheads.org/forum/viewtopic.php?f=10&t=79939&start=400#p5275418
luckyducky99
Posts: 163
Joined: Sun Dec 15, 2019 7:47 pm

Re: Withdrawal rate for an early retirement

Post by luckyducky99 »

geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm Is 50 years the relatively short past? How about 150 years?
Yes and yes. This makes for a small sample size of past periods to try to make future predictions from.
geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...
I disagree that we can be confident enough of future returns to even be able to make that comparison given the miniscule magnitude of the probability of dying in a car crash.
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

EnjoyIt wrote: Wed Nov 25, 2020 2:33 pm Vineviz,

My understanding is that you believe a 2% withdrawal rate is what is needed, correct? If that is the case have you considered spending money to decrease your cost of living?

For example, pay $30k for solar which will provide well over $600 worth of electricity and better than the 2% you discuss. Some more examples, build a well, geothermal heating, have a garden. Or better yet, buy farmland.

Actually, if all I needed was to outpace 2% real returns I would probably do everything in my power to cut my fixed expenses and then get involved in real estate rentals across the country with a hired manager for everything. After all, all I need is 2% real which is not that hard to accomplish with a diversified real estate portfolio.

Am I missing something?
Your question is a very good one, and you aren't missing anything.

If I really thought that my portfolio would only support a 2% SWR (or provide only a 2% real return), I would do just as you suggest: make investments that reduce our cost of living and move into rental real estate.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

EnjoyIt wrote: Wed Nov 25, 2020 2:38 pm
willthrill81 wrote: Wed Nov 25, 2020 2:26 pm
EnjoyIt wrote: Wed Nov 25, 2020 2:20 pm
geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm
luckyducky99 wrote: Wed Nov 25, 2020 1:49 pm

"Never" is a strong word, and I think it's an unreasonable one to use when talking about an unknown future.

To say that people who think that the risk of the future being worse than the relatively short past is real, and who think that that risk is worth insuring against, are "not grounded in reality" is also unreasonable.
Is 50 years the relatively short past? How about 150 years?

Is being worried that you are going to die in a car accident every day reasonable? I am OK saying that if I drive safely it will never happen to me, in fact by being willing to drive in my car I am saying that I am OK with the level of risk of dying in a car crash. But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...

The definition of reasonable is the possibility of an event or events happening based on what we know to be possible given our history and current variables. In this sense, it is quite reasonable to say that you will never run out of money if you have a high equity portfolio and a 3% withdraw it.

If the 3% withdrawal rate doesn’t work over the next 60 years, your portfolio being depleted will not be your biggest concern.
I will agree with one thing. If returns are so poor that one feels they need less than 3% withdrawal rates to make that cash last, then it is just as likely that 2%, 1% or even less won’t be enough either. There will be a calamity that your money won’t solve. There are ways to protect yourself though. Instead of saving an extra 25 or 50 years worth of expenses, one can make their home off grid capable. They will need to have a farm and raise livestock. They can learn to make ammunition and stock pile plenty of resources. This will accomplish two things. 1) by creating a self sustaining living situation one’s fixed expenses go down significantly requiring a much smaller portfolio. 2) you actually protect yourself from an economic collapse where 3% withdrawal rate fails.
I've said in the past that a fruit tree in your backyard is like a little Roth IRA. And its performance will probably be uncorrelated with everything else in your portfolio to boot. Win win. 8-)
We have talked about installing solar just to decrease fixed expenses and minimize SORR. I’m all for spending money to cut those fixed expenses which should add flexibility. I am pretty sure you and I agree that flexibility is probably the most valuable item in any withdrawal strategy.

A fruit tree is better than an IRA because you can share it with people and make friends. One of my colleagues every year brings in a huge box of oranges to work. I usually grab a dozen or so with plenty left over for others. It’s awesome and makes a lot of people happy.
If your area is one where the projected real return of going solar is at least 5% real, I'd say go for it. I would be very reticent with a lesser return than that because I'm not convinced that solar increases property values by enough to offset their cost much at all, and the returns generally involve a lot of assumptions that may not hold up.

Also, when I start retirement, I want it to be with a brand new vehicle, all major home maintenance (e.g. roof) to be completely up to date, etc. in an effort to reduce SORR.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
EnjoyIt
Posts: 5311
Joined: Sun Dec 29, 2013 8:06 pm

Re: Withdrawal rate for an early retirement

Post by EnjoyIt »

willthrill81 wrote: Wed Nov 25, 2020 2:45 pm
EnjoyIt wrote: Wed Nov 25, 2020 2:38 pm
willthrill81 wrote: Wed Nov 25, 2020 2:26 pm
EnjoyIt wrote: Wed Nov 25, 2020 2:20 pm
geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm

Is 50 years the relatively short past? How about 150 years?

Is being worried that you are going to die in a car accident every day reasonable? I am OK saying that if I drive safely it will never happen to me, in fact by being willing to drive in my car I am saying that I am OK with the level of risk of dying in a car crash. But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...

The definition of reasonable is the possibility of an event or events happening based on what we know to be possible given our history and current variables. In this sense, it is quite reasonable to say that you will never run out of money if you have a high equity portfolio and a 3% withdraw it.

If the 3% withdrawal rate doesn’t work over the next 60 years, your portfolio being depleted will not be your biggest concern.
I will agree with one thing. If returns are so poor that one feels they need less than 3% withdrawal rates to make that cash last, then it is just as likely that 2%, 1% or even less won’t be enough either. There will be a calamity that your money won’t solve. There are ways to protect yourself though. Instead of saving an extra 25 or 50 years worth of expenses, one can make their home off grid capable. They will need to have a farm and raise livestock. They can learn to make ammunition and stock pile plenty of resources. This will accomplish two things. 1) by creating a self sustaining living situation one’s fixed expenses go down significantly requiring a much smaller portfolio. 2) you actually protect yourself from an economic collapse where 3% withdrawal rate fails.
I've said in the past that a fruit tree in your backyard is like a little Roth IRA. And its performance will probably be uncorrelated with everything else in your portfolio to boot. Win win. 8-)
We have talked about installing solar just to decrease fixed expenses and minimize SORR. I’m all for spending money to cut those fixed expenses which should add flexibility. I am pretty sure you and I agree that flexibility is probably the most valuable item in any withdrawal strategy.

A fruit tree is better than an IRA because you can share it with people and make friends. One of my colleagues every year brings in a huge box of oranges to work. I usually grab a dozen or so with plenty left over for others. It’s awesome and makes a lot of people happy.
If your area is one where the projected real return of going solar is at least 5% real, I'd say go for it. I would be very reticent with a lesser return than that because I'm not convinced that solar increases property values by enough to offset their cost much at all, and the returns generally involve a lot of assumptions that may not hold up.

Also, when I start retirement, I want it to be with a brand new vehicle, all major home maintenance (e.g. roof) to be completely up to date, etc. in an effort to reduce SORR.
I believe they say adding solar adds a small percentage to the price of the home. Though I would think for us who will live in the house for decades it may be inconsequential. The big thing is I see is, Will the panels and the roof survive long enough to pay for themselves and their replacement when the time comes. Last I checked though may be wrong, around here it takes about 7-8 years for solar to pay for itself. As long as they last for 14-16 years without issues, the decision is a no brainer.

I agree that it would be nice to retire in a freshly renovated house, new HVAC and with cars that would last another 10 years without much issue. I doubt we will be so lucky, to have it all but you never know. More than likely we will retire with a bit of extra cash invested to cover those upcoming events.
A time to EVALUATE your jitters: | https://www.bogleheads.org/forum/viewtopic.php?f=10&t=79939&start=400#p5275418
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

EnjoyIt wrote: Wed Nov 25, 2020 3:47 pm I believe they say adding solar adds a small percentage to the price of the home. Though I would think for us who will live in the house for decades it may be inconsequential. The big thing is I see is, Will the panels and the roof survive long enough to pay for themselves and their replacement when the time comes. Last I checked though may be wrong, around here it takes about 7-8 years for solar to pay for itself. As long as they last for 14-16 years without issues, the decision is a no brainer.
With a break-even point under ten years, it seems like an easy decision to go for it to me. For us, it would take at least 20 years to break-even, and there's no way we're going for that.
EnjoyIt wrote: Wed Nov 25, 2020 3:47 pm I agree that it would be nice to retire in a freshly renovated house, new HVAC and with cars that would last another 10 years without much issue. I doubt we will be so lucky, to have it all but you never know. More than likely we will retire with a bit of extra cash invested to cover those upcoming events.
You wouldn't have to pay for all of that from outside your portfolio. Even if you have to reduce your savings for a while before retiring or pay for the expenses from your portfolio right after retiring, it would still reduce SORR. Anything you can do that will help you to avoid making 'forced' withdrawals while your portfolio is doing poorly will help to reduce SORR, though the cost of reducing the risk may not be worth it in all cases, obviously.

When my parents retired last year, they moved into a newly built home and shouldn't have any significant maintenance expenses beyond replacing kitchen appliances and the like for at least a decade, maybe two. Their portfolio has done well enough in the mean time that they'll probably use next year's RMD to buy a new vehicle; my dad partly wants to do that to further reduce their SORR.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
EnjoyIt
Posts: 5311
Joined: Sun Dec 29, 2013 8:06 pm

Re: Withdrawal rate for an early retirement

Post by EnjoyIt »

willthrill81 wrote: Wed Nov 25, 2020 3:53 pm
EnjoyIt wrote: Wed Nov 25, 2020 3:47 pm I believe they say adding solar adds a small percentage to the price of the home. Though I would think for us who will live in the house for decades it may be inconsequential. The big thing is I see is, Will the panels and the roof survive long enough to pay for themselves and their replacement when the time comes. Last I checked though may be wrong, around here it takes about 7-8 years for solar to pay for itself. As long as they last for 14-16 years without issues, the decision is a no brainer.
With a break-even point under ten years, it seems like an easy decision to go for it to me. For us, it would take at least 20 years to break-even, and there's no way we're going for that.
EnjoyIt wrote: Wed Nov 25, 2020 3:47 pm I agree that it would be nice to retire in a freshly renovated house, new HVAC and with cars that would last another 10 years without much issue. I doubt we will be so lucky, to have it all but you never know. More than likely we will retire with a bit of extra cash invested to cover those upcoming events.
You wouldn't have to pay for all of that from outside your portfolio. Even if you have to reduce your savings for a while before retiring or pay for the expenses from your portfolio right after retiring, it would still reduce SORR. Anything you can do that will help you to avoid making 'forced' withdrawals while your portfolio is doing poorly will help to reduce SORR, though the cost of reducing the risk may not be worth it in all cases, obviously.

When my parents retired last year, they moved into a newly built home and shouldn't have any significant maintenance expenses beyond replacing kitchen appliances and the like for at least a decade, maybe two. Their portfolio has done well enough in the mean time that they'll probably use next year's RMD to buy a new vehicle; my dad partly wants to do that to further reduce their SORR.
Right. Cutting our fixed expenses down and having a very large discretionary budged compared to fixed expenses will allow for the flexibility needed to withstand just about any market turmoil. We are lucky enough to have hit our number and in professions we enjoy. We both work part time now. Our plan is to spend a bit extra on one off items and allow a bit more lifestyle creep as long as it fits into a 4% withdrawal rate scenario. For example while still employed if our portfolio goes up by $150k we can comfortably spend an additional $500/month on discretionary expenses taking potential taxes into account.
A time to EVALUATE your jitters: | https://www.bogleheads.org/forum/viewtopic.php?f=10&t=79939&start=400#p5275418
User avatar
geerhardusvos
Posts: 1516
Joined: Wed Oct 23, 2019 10:20 pm
Location: heavenlies

Re: Withdrawal rate for an early retirement

Post by geerhardusvos »

luckyducky99 wrote: Wed Nov 25, 2020 2:40 pm
geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...
I disagree that we can be confident enough of future returns to even be able to make that comparison given the miniscule magnitude of the probability of dying in a car crash.
I think you’re missing it. I am telling you reality. There has been no 60 year retirement cycle with available United States data that has required a withdrawal rate under 3%. It’s never happened. Meaning that you were overwhelmingly more likely to die in a car crash than you were to run out of money using a 3% withdrawal rate and having most of your portfolio in US index funds.

No one has provided evidence in academia or on this forum to suggest that the next 100 years will be much different than the last 100 years economically speaking for the United States. Will it be different? Yes, it’s the future and it’s unknown. Is it extremely probable that it will be some degree similar to the past? Yes. Reasonable people understand that things that have never happened before, both positive and negative, are in our future, just like new things both positive and negative have happened in our past. The only reasonable approach to looking at the future is to thoroughly analyze and understand our historical data and to look at our current variables. Despite issues with birth rates, interest rates, changing governments, or whatever else, the future is bright and is likely to be similar to the past. Will we go through all of the normal economic cycles like we always have? Yes. Does a 3 to 4% withdrawal rate account for those world wars, recessions, and everything else? Yes, and with enough historical probability and certainly for us to not even have to be debating these types of issues without pointing to specific variables that would necessitate a completely different outcome than what we have had the last 150 years.

I am all ears to any argument outside of, “well the future will be different than the past, and we don’t know the future, therefore we can’t be reasonably confident in what has worked in the past.” Our lives are short and when we start talking about things that are so improbable, like dying in a car crash, And yet people want to advise others to work an additional 10 to 20 years to account for the unknown, they are making a sacrifice of the most valuable asset they have: time. It’s extremely dangerous to not use your time wisely. I would prefer to use historical data, monitor current portfolio, and move on with my life. What’s the worst that can happen? I run out of money or I die.
Last edited by geerhardusvos on Wed Nov 25, 2020 4:59 pm, edited 1 time in total.
VTSAX and chill
7eight9
Posts: 1678
Joined: Fri May 17, 2019 7:11 pm

Re: Withdrawal rate for an early retirement

Post by 7eight9 »

geerhardusvos wrote: Wed Nov 25, 2020 4:52 pm
luckyducky99 wrote: Wed Nov 25, 2020 2:40 pm
geerhardusvos wrote: Wed Nov 25, 2020 2:09 pm But the likelihood of you dying in a car crash is higher than your portfolio being depleted if you are using a 3% withdrawal rate over 60 years... please let that sink in...
I disagree that we can be confident enough of future returns to even be able to make that comparison given the miniscule magnitude of the probability of dying in a car crash.
I think you’re missing it. I am telling you reality. There has been no 60 year retirement cycle with available United States data that has required a withdrawal rate under 3%. It’s never happened. Meaning that you were overwhelmingly more likely to die in a car crash than you were to run out of money using a 3% withdrawal rate and having most of your portfolio in US index funds.

No one has provided evidence in academia or on this forum to suggest that the next 100 years will be much different than the last 100 years economically speaking for the United States. Will it be different? Yes, it’s the future and it’s unknown. Is it extremely probable that it will be some degree similar to the past? Yes. Reasonable people understand that things that have never happened before, both positive and negative, are in our future, just like new things both positive and negative have happened in our past. The only reasonable approach to looking at the future is to thoroughly analyze and understand our historical data and to look at our current variables. Despite issues with birth rates, interest rates, changing governments, or whatever else, the future is bright and is likely to be similar to the past. Will we go through all of the normal economic cycles like we always have? Yes. Does a 3 to 4% withdrawal rate account for those world wars, recessions, and everything else? Yes, and with enough historical probability and certainly for us to not even have to be debating these types of issues without pointing to specific variables that would necessitate a completely different outcome than what we have had the last 150 years.

I am all ears to any argument outside of, “well the future will be different than the past, and we don’t know the future, therefore we can’t be reasonably confident in what has worked in the past.”
Die in a car crash - you have no problems.
Run out of money in retirement - you have problems.
I guess it all could be much worse. | They could be warming up my hearse.
User avatar
geerhardusvos
Posts: 1516
Joined: Wed Oct 23, 2019 10:20 pm
Location: heavenlies

Re: Withdrawal rate for an early retirement

Post by geerhardusvos »

7eight9 wrote: Wed Nov 25, 2020 4:58 pm Die in a car crash - you have no problems.
Run out of money in retirement - you have problems.
Right, and I won’t run out of money using between a 3 and 4% withdrawal rate. Can you give me a comprehensive reason why I would run out of money if I am actively monitoring my portfolio, my spending, market conditions, and withdrawing accordingly? Did you read any of my above post? Do you know something about the future that I don’t?
VTSAX and chill
klaus14
Posts: 603
Joined: Sun Nov 25, 2018 7:43 pm

Re: Withdrawal rate for an early retirement

Post by klaus14 »

vineviz wrote: Wed Nov 25, 2020 12:56 pm
I think it's also worth noting that his conclusion was very sensitive to the return assumptions, especially the 2.4% real return that Kitces used for bond returns.

If you re-run his analysis but with more realistic return assumptions based on current conditions (e.g 5% real returns for VTSMX and 0% real returns for VBMFX) the conclusion flips: a constant 50/50 allocation survives more often (67.26%) than a gliding allocation from 30/70 to 70/30 (60.78%).
I would guess this is due to having a lower bond alloc at the start (50 > 30). Glide should still help: 50->30 bond reduction should beat 50->50
User avatar
vineviz
Posts: 8495
Joined: Tue May 15, 2018 1:55 pm

Re: Withdrawal rate for an early retirement

Post by vineviz »

EnjoyIt wrote: Wed Nov 25, 2020 2:33 pm Vineviz,

My understanding is that you believe a 2% withdrawal rate is what is needed, correct? If that is the case have you considered spending money to decrease your cost of living?
No, I don’t think a 2% withdrawal rate would be necessary for any except maybe the very longest retirement period (eg 60 years with 3% real return). For a 30 year retirement using the conventional definition of “safe” (ie 90% probability of “success”) the sustainable withdrawal rate is likely between 3% and 3.25%.

Someone willing to adjust the withdrawal rate could start higher than that, of course.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
nigel_ht
Posts: 1558
Joined: Tue Jan 01, 2019 10:14 am

Re: Withdrawal rate for an early retirement

Post by nigel_ht »

willthrill81 wrote: Wed Nov 25, 2020 2:41 pm
EnjoyIt wrote: Wed Nov 25, 2020 2:33 pm Vineviz,

My understanding is that you believe a 2% withdrawal rate is what is needed, correct? If that is the case have you considered spending money to decrease your cost of living?

For example, pay $30k for solar which will provide well over $600 worth of electricity and better than the 2% you discuss. Some more examples, build a well, geothermal heating, have a garden. Or better yet, buy farmland.

Actually, if all I needed was to outpace 2% real returns I would probably do everything in my power to cut my fixed expenses and then get involved in real estate rentals across the country with a hired manager for everything. After all, all I need is 2% real which is not that hard to accomplish with a diversified real estate portfolio.

Am I missing something?
Your question is a very good one, and you aren't missing anything.

If I really thought that my portfolio would only support a 2% SWR (or provide only a 2% real return), I would do just as you suggest: make investments that reduce our cost of living and move into rental real estate.
Weirdly I am looking at doing this...not because of potential apocalypse or even SORR but mostly because power here sucks...so new, more efficient appliances, etc reduces energy needs and smaller costs for replacing the grid when it goes down...

But yeah, now that you mention it, the more self sufficient you are the lower the risks of SORR in some kind of apocalyptic financial event. This only works in early retirement though...after a point I’m not going to want to maintain even the most turnkey of (near) off grid systems and it’s unlikely that any of the kids will be willing or able to live in the middle of nowhere where the nearest store is 20m away.

Of course, what I describe could just be anywhere in Cali where they have blackouts when it gets windy and dry...
User avatar
vineviz
Posts: 8495
Joined: Tue May 15, 2018 1:55 pm

Re: Withdrawal rate for an early retirement

Post by vineviz »

geerhardusvos wrote: Wed Nov 25, 2020 4:52 pmNo one has provided evidence in academia or on this forum to suggest that the next 100 years will be much different than the last 100 years economically speaking for the United States. Will it be different? Yes, it’s the future and it’s unknown.
You don’t need to know the future in order to know that the next 30 years have a lower likelihood of supporting a 4% withdrawal rate than any 30 year period in the past.

The CURRENT real yield on bonds and the CURRENT valuation of equities tells you everything you need to know. No crystal ball required.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
willthrill81
Posts: 22549
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Withdrawal rate for an early retirement

Post by willthrill81 »

nigel_ht wrote: Wed Nov 25, 2020 7:21 pm But yeah, now that you mention it, the more self sufficient you are the lower the risks of SORR in some kind of apocalyptic financial event. This only works in early retirement though...after a point I’m not going to want to maintain even the most turnkey of (near) off grid systems and it’s unlikely that any of the kids will be willing or able to live in the middle of nowhere where the nearest store is 20m away.

Of course, what I describe could just be anywhere in Cali where they have blackouts when it gets windy and dry...
What kind of maintenance is involved with a grid-tied solar system? I thought that the only thing you might have to do is periodically clean the panels, and a teenager with a hose could probably do that for you. Maybe I'm completely wrong though.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
geerhardusvos
Posts: 1516
Joined: Wed Oct 23, 2019 10:20 pm
Location: heavenlies

Re: Withdrawal rate for an early retirement

Post by geerhardusvos »

vineviz wrote: Wed Nov 25, 2020 7:22 pm
geerhardusvos wrote: Wed Nov 25, 2020 4:52 pmNo one has provided evidence in academia or on this forum to suggest that the next 100 years will be much different than the last 100 years economically speaking for the United States. Will it be different? Yes, it’s the future and it’s unknown.
You don’t need to know the future in order to know that the next 30 years have a lower likelihood of supporting a 4% withdrawal rate than any 30 year period in the past.

The CURRENT real yield on bonds and the CURRENT valuation of equities tells you everything you need to know. No crystal ball required.
Those two variables have been discussed at length and nothing has been presented that would dissuade a 3 to 4% WR moving forward
VTSAX and chill
EnjoyIt
Posts: 5311
Joined: Sun Dec 29, 2013 8:06 pm

Re: Withdrawal rate for an early retirement

Post by EnjoyIt »

vineviz wrote: Wed Nov 25, 2020 7:22 pm
geerhardusvos wrote: Wed Nov 25, 2020 4:52 pmNo one has provided evidence in academia or on this forum to suggest that the next 100 years will be much different than the last 100 years economically speaking for the United States. Will it be different? Yes, it’s the future and it’s unknown.
You don’t need to know the future in order to know that the next 30 years have a lower likelihood of supporting a 4% withdrawal rate than any 30 year period in the past.

The CURRENT real yield on bonds and the CURRENT valuation of equities tells you everything you need to know. No crystal ball required.
People have been trying to predict future returns for decades. These have been very smart people using math and history to make those predictions. History has proven that they have been wrong over and over again. What makes you think that your predictions are somehow better or more accurate? I see you trying to predict 30+ years of returns and to be honest with you, it seems very nonsensical to even attempt anything more than just a few years ahead.

As a recent point of reference people all over were predicting that interest rates had only to go up. They were all wrong. People have been predicting muted returns for the last 10 years. So far they have all been wrong. People have been predicting that small caps will outperform the broad market and they have been wrong. People love to make predictions. I will make a prediction too because it’s fun. I’m going to predict that the not so distant future equities are going to drop by 20% or more and it will take more than just a few months to recover.

I do agree with one thing you say. If one is looking to retire with over 30 years they may want to start with a lower withdrawal rate than 4% or have some decent amount of flexibility in spending. I believe the sub 3% people are going way overboard. It’s one thing to retire at retirement age and then figuring out you can comfortably live on 2.5% withdrawals. It’s a complexity other thing to really want to retire but delay it for 5-10 years because of unrealistic fear.
A time to EVALUATE your jitters: | https://www.bogleheads.org/forum/viewtopic.php?f=10&t=79939&start=400#p5275418
Marseille07
Posts: 1073
Joined: Fri Nov 06, 2020 1:41 pm

Re: Withdrawal rate for an early retirement

Post by Marseille07 »

EnjoyIt wrote: Wed Nov 25, 2020 7:48 pm I believe the sub 3% people are going way overboard. It’s one thing to retire at retirement age and then figuring out you can comfortably live on 2.5% withdrawals.
Retirement age doesn't matter. If OP can hit 4M by 45 and figure out they can comfortably live on 2% (80K/yr) then nothing wrong with 2% or retiring early. Advocating such an idea isn't going way overboard.
User avatar
geerhardusvos
Posts: 1516
Joined: Wed Oct 23, 2019 10:20 pm
Location: heavenlies

Re: Withdrawal rate for an early retirement

Post by geerhardusvos »

Marseille07 wrote: Wed Nov 25, 2020 7:55 pm
EnjoyIt wrote: Wed Nov 25, 2020 7:48 pm I believe the sub 3% people are going way overboard. It’s one thing to retire at retirement age and then figuring out you can comfortably live on 2.5% withdrawals.
Retirement age doesn't matter. If OP can hit 4M by 45 and figure out they can comfortably live on 2% (80K/yr) then nothing wrong with 2% or retiring early. Advocating such an idea isn't going way overboard.
It’s not necessary for someone to own 7 vehicles. Some people like owning multiple cars (Collectors, off-roading, trucks, RVs, etc). Not everyone has that luxury. But they probably only need one car, maybe two. Advocating that people need 7 cars when no one has ever needed that many cars in the history of the US, is indeed, overboard. Advocating for a 2% wr is way overboard, despite some people having the ability to do it.
VTSAX and chill
Post Reply