sequence of withdrawals during retirement--differing philosophies

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
rs9876lg
Posts: 31
Joined: Wed Jul 03, 2024 8:48 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by rs9876lg »

Do your kids routinely turn down a promotion because it might put them in higher tax bracket? Do they negotiate with their manager to give them tax free raise?

That “prank” comment really gets to me. If I had kids like that I will donate everything to charity instead!
stay.the.course
Posts: 34
Joined: Thu Apr 04, 2024 4:29 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by stay.the.course »

rs9876lg wrote: Fri Jul 05, 2024 5:19 pm Do your kids routinely turn down a promotion because it might put them in higher tax bracket?
Couldn’t agree more.
If someone gives me $1 million, would I not be grateful, even if half goes to taxes?

I am in the 37% tax bracket. I have given this a lot of thought and have decided (having read many thoughtful people on this thread), to continue to tax-defer even though during retirement I will have an RMD “problem”. When I posted my portfolio, one respondent said (I am paraphrasing) “yes, you sure will have an RMD problem but that’s a rich man’s problem and as you make money you will need to pay taxes”. His words ring true in this conversation as well.

When may kids receive my tax-deferred assets, they better be thankful. Period.
rs9876lg
Posts: 31
Joined: Wed Jul 03, 2024 8:48 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by rs9876lg »

I find the obsession in this forum with having to extract the very last dollars from one’s multiple million dollar portfolio amusing. I understand they reached this stage by watching the expenses and saving aggressively. It is not end of the world if you end up giving little bit more to government in taxes than if you could optimize it proactively. We agree that taxes are necessary evil of living in a functioning society. I mean I no longer have kids in school but I still support the town by paying significant real estate taxes and supporting the school committee budget when it comes for voting
Exchme
Posts: 1401
Joined: Sun Sep 06, 2020 3:00 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by Exchme »

The answer is going to vary according to your circumstances. So use a good calculator program and try different withdrawal order options. Make sure to examine:
What happens in Bull & bear markets?
What happens at different lifespans?
What happens if you need Long Term Care?
Will tax laws revert on schedule in 2026 to pre-TCJA?
What will your heirs' tax brackets be?
Might you move to a state with different taxes?
Do you have a desire to give to charities either via QCD or bequests in your will?
Do you need to control income prior to Medicare to get ACA premium credits?
Do you mirror asset allocation in your accounts or hold bonds in tax deferred, stocks in taxable & Roth?

For us, the math favors using taxable first and doing Roth Conversions, but I would not want to generalize to others
VanGar+Goyle
Posts: 653
Joined: Sat May 29, 2021 1:31 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by VanGar+Goyle »

KlangFool wrote: Fri Jun 21, 2024 9:05 pm Philosophically, I do not associate withdrawal with spending needs.

A) The spending is from 2 to 3 years of expense in cash buffer.

B) The refilling of the cash buffer is from the taxable dividend and interest income. And, from the Roth account.

C) The tax management is from Roth conversion and taxable gain dependent on the amount of taxes I want to pay each year.

(A), (B), and (C) are independent from each other.
Do not confuse spending with distributions with taxation. There are several different but related issues going on here.

One is taxation of distributions from a tax deferred retirement account, with Roth Conversions, or RMDs. In general you want to avoid high ordinary marginal tax rates, so may strive to keep your tax rate under 13%, or 25%, or 33% for years. Distributions can be taxable or not taxable, and do not have to be spent.
A second is spending from accounts based on account location. Distributions from Roth accounts are taxed less, more tax efficient. Ordinary brokerage accounts are a combination of ordinary interest, dividends, qualified dividends, long term capital gains, short term capital gains ... It can be good to have spendable assets in any location, and even spend from a Roth account even if you still have money in other locations, but spending from an ordinary taxable account is more common.
A third is asset location. This can be changed tax-free inside tax advantaged accounts. Cash would not usually be in a Roth account, and significant spending from one location may cause rebalancing in other locations.
A fourth overall issue is secondary taxes like IRMAA, Net Investment Income Tax, ACA Partial Tax Credit, taxation of Social Security benefits, and other credits and deductions related to MAGI. Distributions from some locations may be MAGI tax free.
A fifth issue is that spending is not flat over time, but may rise with travel, new houses/cars/boats, or medical like Long Term Care. However there may be methods like distributions from tax deferred accounts being mostly tax-free based on itemized deductions. Spending on charities can be tax-free.

I probably left off some more interrelated issues, and my attempt to color code may not have helped.
Well, you pay a little bit, we're a little bit tough. | You pay very much, very much tough. | You pay a too much, we're too much a tough. | How much you pay? ... Well, then we're plenty tough. - Marx
sc9182
Posts: 2301
Joined: Wed Aug 17, 2016 7:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by sc9182 »

One poster mentioned an example where: there may be couple in 30s inheriting $2M tIRA - and causing a tax-bomb thus to those heirs.

If some parents been creating $2M tIRA in/by their 60s at the time of their demise -- likley, they have:
A) decent chunk of brokerage amounts (step-up basis ?)
B) they should have decent chunk of term-life - to be paid-out to heirs mostly tax-free (or other lousy life policies - but we all know to avoid those)
C) decent chunk of house (step-up basis?)
plus $2M tIRA (10-year rule)

If A, B, C -- are ALL tax-free to heirs -- don't you think their overall/aggregate taxed inheritance rate be at decent level - instead of thinking exorbitant (only looking thru the limited lense of taxes ONLY an inherited tIRA)

To provide a counter-example, at/by the age of 59 -- say the said parents disposed off their "home", sold off all their "LTCG" brokerage/assets paying about 18% LTCG incl NIIT (they did not have term-life policy)., and converted all that $2M tIRA into Roths (using the funds/cash from house/brokerage sales to pay for Taxes - leaving any remaining into bank account).

Would the heirs be better off inheriting Roth (with its 10-year rule/limit) + left-over peanut sized cash? If so - would it be financially beneficial to them (highly doubt it - then again, you see/do the math yourself)

Also - if (likely chances are that most folks inherit in late 50s to 60s or possibly into 70s) -- likely most of the heirs are already retired, or about to retire - and if one $2M tIRA lands in their lap -- may be CONSIDER early-retire or accelerate retirement, and enjoy a life little (besides, Charitable angle other posters mentioned). A large inherited tIRA is NOT that much different than a salary-deferral some of us folks have enjoyed/put-to-use. And many Salary-deferral folks use those moneys to retire timely/early, and bridge towards delayed-larger-double SS. And those folks deferring salaries ain't all that dull in their tax-planning - are they ? For tIRA benefit (over Salary deferral) - it provides creditor protection(s), large/lumpy Medical costs deductions, stretch IRA (for disabled children), ability-to/selectively Roth-convert, very long run-way (joint lifetime + 10-years), brokerage-bonuses ( :happy ) - among many more benefits. Love tIRA

** The only plausible complaint I heard of large tIRA to heirs is that - their Daughter-in-Laws complaining at funeral/at-tax-season that half-the-moneys went to paying taxes and their tax-rate shot up thru the roof (Among another complained lobbed -- PLEASE do not conflate this complaint with tax-bomb oft-mentioned here by a couple of posters) - so the inheritance was practically useless. If we have such brood/next-gen ., we would rather pass it further along to grand-kids generation., and/or increase our charitable donations/bequests - than leaving "tIRA" to foul-mouthing/bad-intentioned one daughter-in-law .. just saying ..
Last edited by sc9182 on Sat Jul 06, 2024 11:22 am, edited 8 times in total.
rs9876lg
Posts: 31
Joined: Wed Jul 03, 2024 8:48 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by rs9876lg »

A great reply by VGG! I will add couple more parameters to it. Some are evaluated retroactively and the cliff location is known only afterwards e.g. IRRMA brackets and two year look back period. The second and probably the most important point is all the factors such as expected gain of return on each asset class or the inflation or your continued earnings for next multiple decades are subject to random changes and are completely unpredictable. YES we can assign certain probability and compute optimal strategy but never forget that reality and assumptions can and will differ
rs9876lg
Posts: 31
Joined: Wed Jul 03, 2024 8:48 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by rs9876lg »

By the way I always thought that those references to “prank” or proverbial daughter-in-law complaining at funeral were jokes but reading few serious replies in this forum I have one thing to say “Really???”
Cah
Posts: 243
Joined: Fri Jun 30, 2023 10:39 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by Cah »

syc wrote: Fri Jun 21, 2024 7:12 pm In this post:

viewtopic.php?t=434084

I read, among the replies, the following, from two different respondents:

"In general withdraw first from a taxable brokerage account during the first years of retirement . . . ."

versus

" . . . there's common themes to tap the Tax-Deferred accounts first (reduces future RMDs and associated "tax time-bomb"), then Taxable accounts, then Tax-Free Roth last . . . ."

To be fair, both respondents qualified their remarks with statements to the effect that there is no right answer, that it depends on individual circumstances. I respect that. But I'd love to hear some discussion of the pros and cons of these seemingly opposite philosophies. I struggle with them myself, and see merit in both. (I'll leave aside the Roths for now, since there seems to be a relatively firm consensus that for most folks, Roth should be tapped last, and if able, not at all.)

Thanks.
My best advise, try to live off the interest of taxable stuff unless you need to withdraw from the principle to meet needs or fund a major purchase. Let the tax differed ride as long as possible. Forget about Roth conversions they only hasten the tax and by all means spend your money . You didn't work hard and save to pass away the richest person in the graveyard.
dknightd
Posts: 3916
Joined: Wed Mar 07, 2018 10:57 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by dknightd »

snic wrote: Mon Jun 24, 2024 1:00 pm
syc wrote: Fri Jun 21, 2024 9:08 pm I've heard it said that the worst thing to inherit is a traditional IRA. True?
No, it's a variable annuity.

And actually, there are plenty of worse things to inherit, including land in a distant, chaotic, war-torn country where the inheritor does not speak the local language. (Ask me how I know...)
The worst thing to inherit are bills
Retired 2019. So far, so good. I want to wake up every morning. But I want to die in my sleep. Just another conundrum. I think the solution might be afternoon naps ;)
GAAP
Posts: 2784
Joined: Fri Apr 08, 2016 12:41 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by GAAP »

A potential consideration for some would be the presence of a State Estate Tax. Some of those are indexed to inflation, some are not, Washington is indexed to a no-longer-published statistic (and therefore effectively not indexed). In Washington (probably other states also), the estate includes all of the funds in an IRA, whether Traditional or Roth. That means Roth conversions can reduce the estate size and thus the estate tax -- or even eliminate that estate tax.

Life Insurance proceeds are also considered part of the estate in Washington. Large policies lead to large estate tax bills. "Tax-free" isn't always that free.

The likely tax brackets of the heirs may play into a decision to do Roth conversions. If the retiree's withdrawals and the heirs withdrawals will both be taxed at x%, then it is a wash. If one is taxed at x%, and the other is taxed at x+y%, then the choice may come down to which party has the lower rate. With multiple heirs, the decision may come down to the least common denominator -- which may not be optimal for all.

It is surprisingly easy to build a $2M+ combination of TIRA, 401(k), and Roth IRA without any taxable brokerage amounts. All that is required is the will to save (pay yourself first). It will get easier in the future, since the contribution limits are quite a bit higher than they used to be. People without the benefit of employer-provided DC plans can do the same with a bit more willingness to save after-tax. A strategy that works for the first person will probably not be a good choice for the second -- and vice-versa.
“Adapt what is useful, reject what is useless, and add what is specifically your own.” ― Bruce Lee
User avatar
Hacksawdave
Posts: 1054
Joined: Tue Feb 14, 2023 4:44 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by Hacksawdave »

Each citizen/resident has their civic duty to pay the required amount of tax based upon the imposed income, property, and usage taxation liabilities. How the amount is derived is based upon the personal choices made by each individual in the course of their plans, accumulations and execution. Some will pay lower, some will pay higher, and some can even choose to voluntarily make extra payments if they wish.

There are some that will decide making as much fully taxable ordinary income as is humanly possible is the way to go (I know a couple that do). Others will use holding of long-term assets for capital gains rates, some will choose a tax-exempt route, and some might decide on being a landlord with rental properties that generate large amount of property taxes is the way to go.

I chose a hybrid plan many years ago because it would work the best for me as I play by the rules. The 'rulebook' for federal taxation known as 'Title 26' is around 7,000 pages long. How much time and energy one wants to expend to learn the rules will differ.
Topic Author
syc
Posts: 142
Joined: Sat Feb 17, 2024 1:11 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by syc »

Cah wrote: Sat Jul 06, 2024 7:41 am
My best advise, try to live off the interest of taxable stuff unless you need to withdraw from the principle to meet needs or fund a major purchase. Let the tax differed ride as long as possible. Forget about Roth conversions they only hasten the tax and by all means spend your money . You didn't work hard and save to pass away the richest person in the graveyard.
That's an interesting perspective. Could you expand on the rationale for letting tax-deferred ride as long as possible, and your unfavorable view of Roth conversions? Not that I have strong feelings opposite (that's what I'm trying to figure out by starting this thread). It's just different from what I often see--Roth conversions seem very popular. Thanks.
User avatar
ruralavalon
Posts: 26841
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: sequence of withdrawals during retirement--differing philosophies

Post by ruralavalon »

I suggest letting tax-advantaged (both tax-deferred and Roth) compound as long as possible, and do Roth conversions during the lower-tax period after retirement before Required Minimum Distributions (RMDs) begin.

Defer taking Social Security benefits too, if in good health.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

Cah wrote: Sat Jul 06, 2024 7:41 am
syc wrote: Fri Jun 21, 2024 7:12 pm In this post:

viewtopic.php?t=434084

I read, among the replies, the following, from two different respondents:

"In general withdraw first from a taxable brokerage account during the first years of retirement . . . ."

versus

" . . . there's common themes to tap the Tax-Deferred accounts first (reduces future RMDs and associated "tax time-bomb"), then Taxable accounts, then Tax-Free Roth last . . . ."

To be fair, both respondents qualified their remarks with statements to the effect that there is no right answer, that it depends on individual circumstances. I respect that. But I'd love to hear some discussion of the pros and cons of these seemingly opposite philosophies. I struggle with them myself, and see merit in both. (I'll leave aside the Roths for now, since there seems to be a relatively firm consensus that for most folks, Roth should be tapped last, and if able, not at all.)

Thanks.
My best advise, try to live off the interest of taxable stuff unless you need to withdraw from the principle to meet needs or fund a major purchase. Let the tax differed ride as long as possible. Forget about Roth conversions they only hasten the tax and by all means spend your money . You didn't work hard and save to pass away the richest person in the graveyard.
"My best advise, try to live off the interest of taxable stuff unless you need to withdraw from the principle to meet needs or fund a major purchase. Let the tax differed ride as long as possible."
If you model your numbers and goals and this stratagey comes out ahead by all means this can be a good move.
On the other hand if you have not modeled your own numbers and goals this could be a really poor approach.
Which method/tool was used to model your approach?
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

syc wrote: Sun Jul 07, 2024 9:09 pm
Cah wrote: Sat Jul 06, 2024 7:41 am
My best advise, try to live off the interest of taxable stuff unless you need to withdraw from the principle to meet needs or fund a major purchase. Let the tax differed ride as long as possible. Forget about Roth conversions they only hasten the tax and by all means spend your money . You didn't work hard and save to pass away the richest person in the graveyard.
That's an interesting perspective. Could you expand on the rationale for letting tax-deferred ride as long as possible, and your unfavorable view of Roth conversions? Not that I have strong feelings opposite (that's what I'm trying to figure out by starting this thread). It's just different from what I often see--Roth conversions seem very popular. Thanks.
This was posted eartlier in this thread as a very good overview of the question you presented.....
https://www.kitces.com/blog/tax-efficie ... ing-needs/

In order to fully answer this question for yourself you need to run your speciifc numbers with your specific goals and review those models.
There is no 'rule of thumb' that works well.
Potential tools to model these include RPM and/or Pralana. I am sure there are others I am not familair with.
rockstar
Posts: 7241
Joined: Mon Feb 03, 2020 5:51 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by rockstar »

Trying to get my head around taxable first. With taxable, I’ll pay near zero taxes. The biggest drag would be dividends, but you would live off of those anyway.

Why wouldn’t you withdraw from tax deferred plus dividends, so that RMDs don’t build up? You might end up with higher taxes due to RMDs. And at some point you’re going to collect social security, so you have a limited window to do Roth conversions at reasonable marginal tax rates.

Help me out. I have to be missing something.

Does any software exist, where I can model this versus using Excel or Google Sheets?
User avatar
ruralavalon
Posts: 26841
Joined: Sat Feb 02, 2008 9:29 am
Location: Illinois

Re: sequence of withdrawals during retirement--differing philosophies

Post by ruralavalon »

rockstar wrote: Mon Jul 08, 2024 10:19 am Trying to get my head around taxable first. With taxable, I’ll pay near zero taxes. The biggest drag would be dividends, but you would live off of those anyway.

Why wouldn’t you withdraw from tax deferred plus dividends, so that RMDs don’t build up? You might end up with higher taxes due to RMDs. And at some point you’re going to collect social security, so you have a limited window to do Roth conversions at reasonable marginal tax rates.

. . . . .
This where individual modelling pays off, to see if the amount you currently have in tax-deferred accounts means that monster Required Minimum Distributions (RMDs) are likely in your future, or not.

In our case my rollover IRA was large (about 80% of total portfolio), but not so large as to predict enormous RMDs during our likely lifetimes. We both had serious chronic health issues. I did do some Roth conversions.

rockstar wrote: Mon Jul 08, 2024 10:19 amDoes any software exist, where I can model this versus using Excel or Google Sheets?
I don't know the best for modelling this, I have forgotten how I did this. Probably just did a back-of-the-envelope estimate.
Last edited by ruralavalon on Mon Jul 08, 2024 10:59 am, edited 1 time in total.
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link: Bogleheads® investment philosophy
rockstar
Posts: 7241
Joined: Mon Feb 03, 2020 5:51 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by rockstar »

ruralavalon wrote: Mon Jul 08, 2024 10:46 am
rockstar wrote: Mon Jul 08, 2024 10:19 am Trying to get my head around taxable first. With taxable, I’ll pay near zero taxes. The biggest drag would be dividends, but you would live off of those anyway.

Why wouldn’t you withdraw from tax deferred plus dividends, so that RMDs don’t build up? You might end up with higher taxes due to RMDs. And at some point you’re going to collect social security, so you have a
. . . . .
This where individual modelling pays off, to see if the amount you currently have in tax-deferred accounts means that monster Required Minimum Distributions (RMDs) are likely in your future, or not.

In our case my rollover IRA was large (about 80% of total portfolio), but not so large as to predict enormous RMDs during our likely lifetimes. We both had serious chronic health issues. I did do some Roth conversions.

rockstar wrote: Mon Jul 08, 2024 10:19 amDoes any software exist, where I can model this versus using Excel or Google Sheets?
I don't know the best for modelling this, I have forgotten how I did this.
If my tax deferred doubles three more times between now and 75, I can expect a RMD tax bomb. It really depends on the market.
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

rockstar wrote: Mon Jul 08, 2024 10:19 am Trying to get my head around taxable first. With taxable, I’ll pay near zero taxes. The biggest drag would be dividends, but you would live off of those anyway.

Why wouldn’t you withdraw from tax deferred plus dividends, so that RMDs don’t build up? You might end up with higher taxes due to RMDs. And at some point you’re going to collect social security, so you have a limited window to do Roth conversions at reasonable marginal tax rates.

Help me out. I have to be missing something.

Does any software exist, where I can model this versus using Excel or Google Sheets?
In order to fully answer this question for yourself you need to run your speciifc numbers with your specific goals and review those models.
There is no 'rule of thumb' that works well.
Potential tools to model these include RPM and/or Pralana. I am sure there are others I am not familair with.
User avatar
Hacksawdave
Posts: 1054
Joined: Tue Feb 14, 2023 4:44 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by Hacksawdave »

rockstar wrote: Mon Jul 08, 2024 10:56 am
ruralavalon wrote: Mon Jul 08, 2024 10:46 am
rockstar wrote: Mon Jul 08, 2024 10:19 am Trying to get my head around taxable first. With taxable, I’ll pay near zero taxes. The biggest drag would be dividends, but you would live off of those anyway.

Why wouldn’t you withdraw from tax deferred plus dividends, so that RMDs don’t build up? You might end up with higher taxes due to RMDs. And at some point you’re going to collect social security, so you have a
. . . . .
This where individual modelling pays off, to see if the amount you currently have in tax-deferred accounts means that monster Required Minimum Distributions (RMDs) are likely in your future, or not.

In our case my rollover IRA was large (about 80% of total portfolio), but not so large as to predict enormous RMDs during our likely lifetimes. We both had serious chronic health issues. I did do some Roth conversions.

rockstar wrote: Mon Jul 08, 2024 10:19 amDoes any software exist, where I can model this versus using Excel or Google Sheets?
I don't know the best for modelling this, I have forgotten how I did this.
If my tax deferred doubles three more times between now and 75, I can expect a RMD tax bomb. It really depends on the market.
If I use the rule of 72 and an optimistic 5% based upon the asset classes in the 401k with no withdrawal, my first RMD would be $122K. If I continue doing what I am, the RMD would be in the $60k range. A double-double would definitely nail me.
Cah
Posts: 243
Joined: Fri Jun 30, 2023 10:39 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by Cah »

rockstar wrote: Mon Jul 08, 2024 10:19 am Trying to get my head around taxable first. With taxable, I’ll pay near zero taxes. The biggest drag would be dividends, but you would live off of those anyway.

Why wouldn’t you withdraw from tax deferred plus dividends, so that RMDs don’t build up? You might end up with higher taxes due to RMDs. And at some point you’re going to collect social security, so you have a limited window to do Roth conversions at reasonable marginal tax rates.

Help me out. I have to be missing something.

Does any software exist, where I can model this versus using Excel or Google Sheets?
The entire amount you withdraw from tax deferred is taxed vs only the interest or cap gains on taxable. Probably doesn't make much difference on small withdrawals but a major purchase like a car or trip may cost you a bundle in tax.
rockstar
Posts: 7241
Joined: Mon Feb 03, 2020 5:51 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by rockstar »

Cah wrote: Mon Jul 08, 2024 11:56 am
rockstar wrote: Mon Jul 08, 2024 10:19 am Trying to get my head around taxable first. With taxable, I’ll pay near zero taxes. The biggest drag would be dividends, but you would live off of those anyway.

Why wouldn’t you withdraw from tax deferred plus dividends, so that RMDs don’t build up? You might end up with higher taxes due to RMDs. And at some point you’re going to collect social security, so you have a limited window to do Roth conversions at reasonable marginal tax rates.

Help me out. I have to be missing something.

Does any software exist, where I can model this versus using Excel or Google Sheets?
The entire amount you withdraw from tax deferred is taxed vs only the interest or cap gains on taxable. Probably doesn't make much difference on small withdrawals but a major purchase like a car or trip may cost you a bundle in tax.
Capital gain taxes are far lower than ordinary income tax rates that I would pay out of tax deferred.
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

smitcat wrote: Mon Jul 08, 2024 9:22 am
syc wrote: Sun Jul 07, 2024 9:09 pm
Cah wrote: Sat Jul 06, 2024 7:41 am
My best advise, try to live off the interest of taxable stuff unless you need to withdraw from the principle to meet needs or fund a major purchase. Let the tax differed ride as long as possible. Forget about Roth conversions they only hasten the tax and by all means spend your money . You didn't work hard and save to pass away the richest person in the graveyard.
That's an interesting perspective. Could you expand on the rationale for letting tax-deferred ride as long as possible, and your unfavorable view of Roth conversions? Not that I have strong feelings opposite (that's what I'm trying to figure out by starting this thread). It's just different from what I often see--Roth conversions seem very popular. Thanks.
This was posted eartlier in this thread as a very good overview of the question you presented.....
https://www.kitces.com/blog/tax-efficie ... ing-needs/
Yes, that's a good article.
In order to fully answer this question for yourself you need to run your speciifc numbers with your specific goals and review those models.
There is no 'rule of thumb' that works well.
Potential tools to model these include RPM and/or Pralana. I am sure there are others I am not familair with.
Maybe it's a question of how one defines "rule of thumb". Aiming "this year" for a marginal tax rate that will consequently remain more or less constant in future years (as that article recommends) seems like a rule of thumb. It does, however, require "running your specific numbers" so one could say that goes beyond "rule of thumb". A rose by any other name....

As the Kitces article says:
Kitces' article wrote:Of course, [the target marginal tax rate] will depend on the retiree’s overall income and wealth levels. For some, [that marginal rate] may be to “just” fill up the 15% ordinary income tax bracket, and try to avoid any rates that are 25% or higher. For those with more significant retirement accumulations, the sheer size of the family balance sheet may make it impossible to avoid being in at least the 25% bracket, and the goal will just be to stay there and not drift up into the 28% or 33% brackets. For those with very significant wealth, any tax bracket that’s less than the top 39.6% bracket may be appealing to fill with partial Roth conversions!

Nonetheless, the strategy for the tax-efficient spend-down of a retirement portfolio remains the same – to allow for maximal tax-preferenced growth in retirement accounts by spending from taxable brokerage accounts first, but not letting low tax brackets “go to waste” by filling them with partial Roth conversions along the way!
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

FiveK wrote: Mon Jul 08, 2024 1:52 pm
smitcat wrote: Mon Jul 08, 2024 9:22 am
syc wrote: Sun Jul 07, 2024 9:09 pm
Cah wrote: Sat Jul 06, 2024 7:41 am
My best advise, try to live off the interest of taxable stuff unless you need to withdraw from the principle to meet needs or fund a major purchase. Let the tax differed ride as long as possible. Forget about Roth conversions they only hasten the tax and by all means spend your money . You didn't work hard and save to pass away the richest person in the graveyard.
That's an interesting perspective. Could you expand on the rationale for letting tax-deferred ride as long as possible, and your unfavorable view of Roth conversions? Not that I have strong feelings opposite (that's what I'm trying to figure out by starting this thread). It's just different from what I often see--Roth conversions seem very popular. Thanks.
This was posted eartlier in this thread as a very good overview of the question you presented.....
https://www.kitces.com/blog/tax-efficie ... ing-needs/
Yes, that's a good article.
In order to fully answer this question for yourself you need to run your speciifc numbers with your specific goals and review those models.
There is no 'rule of thumb' that works well.
Potential tools to model these include RPM and/or Pralana. I am sure there are others I am not familair with.
Maybe it's a question of how one defines "rule of thumb". Aiming "this year" for a marginal tax rate that will consequently remain more or less constant in future years (as that article recommends) seems like a rule of thumb. It does, however, require "running your specific numbers" so one could say that goes beyond "rule of thumb". A rose by any other name....

As the Kitces article says:
Kitces' article wrote:Of course, [the target marginal tax rate] will depend on the retiree’s overall income and wealth levels. For some, [that marginal rate] may be to “just” fill up the 15% ordinary income tax bracket, and try to avoid any rates that are 25% or higher. For those with more significant retirement accumulations, the sheer size of the family balance sheet may make it impossible to avoid being in at least the 25% bracket, and the goal will just be to stay there and not drift up into the 28% or 33% brackets. For those with very significant wealth, any tax bracket that’s less than the top 39.6% bracket may be appealing to fill with partial Roth conversions!

Nonetheless, the strategy for the tax-efficient spend-down of a retirement portfolio remains the same – to allow for maximal tax-preferenced growth in retirement accounts by spending from taxable brokerage accounts first, but not letting low tax brackets “go to waste” by filling them with partial Roth conversions along the way!
"Maybe it's a question of how one defines "rule of thumb". Aiming "this year" for a marginal tax rate that will consequently remain more or less constant in future years (as that article recommends) seems like a rule of thumb. It does, however, require "running your specific numbers" so one could say that goes beyond "rule of thumb". A rose by any other name...."

Generally a good rule but is not easy to apply for folks with many moving parts over many years.
If in doubt best to utilize a tool which can help optimize both drawdown and possible Roth conversions at the same time while allowing you to model any of your potential variables as well as all of your goals.
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

smitcat wrote: Mon Jul 08, 2024 2:03 pm
FiveK wrote: Mon Jul 08, 2024 1:52 pm Maybe it's a question of how one defines "rule of thumb". Aiming "this year" for a marginal tax rate that will consequently remain more or less constant in future years (as that article recommends) seems like a rule of thumb. It does, however, require "running your specific numbers" so one could say that goes beyond "rule of thumb". A rose by any other name....
Generally a good rule but is not easy to apply for folks with many moving parts over many years.
Maybe, maybe not. The more moving parts, the more uncertainty about how they will actually move.
If in doubt best to utilize a tool which can help optimize both drawdown and possible Roth conversions at the same time while allowing you to model any of your potential variables as well as all of your goals.
Yes, optimization under uncertainty is a worthy goal. Even if one could assign all the probabilities (market returns, life expectancy, heirs' situations, etc.) correctly, in the end it still comes down to "what marginal tax rate should I target this year?" And then repeat the process in each subsequent year.

Some people enjoy creating complex models and are capable of doing so correctly. Hats off to them. A much larger number of people either aren't interested in doing so, or may do so incorrectly (e.g., assuming higher returns in a Roth account when evaluating Roth conversions).

I suspect that, given the facts of an individual case, you and I would reach similar conclusions. :)
User avatar
LilyFleur
Posts: 3670
Joined: Fri Mar 02, 2018 9:36 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by LilyFleur »

livesoft wrote: Sat Jun 22, 2024 4:32 pm
delamer wrote: Sat Jun 22, 2024 4:12 pmA professional married couple in their 30’s could easily have an income of $300,000+ without having built up sufficient assets to be wealthy enough to donate large amounts to charity. If one of them inherits a $2 million Traditional IRA that has to be withdrawn over 10 years, that’a a significant tax burden
... unless they used the withdrawn inherited Traditional IRA money over those 10 years to donate to charity. Basically it would be almost no tax burden since one can donate up to 50% of AGI as cash (as the inherited IRA money would be cash) and add that to itemized deductions.
I have a young adult child with serious illness. My charity--at considerable personal sacrifice--goes to them. When it works to file taxes as head of household (when their income is below the ceiling allowed), I do get a bit of a tax break, as my usual bracket is single. But my support of them bumps up my MAGI considerably, which will mean higher IRMAA down the road. Additionally, a trust for them would be best funded by a Roth, as my t-IRA would have to be emptied in 10 years, and trust income is heavily taxed. I'm trying to find the balance between supporting them and adding to my Roth.

And it's further complicated by my other child who is successful and will likely be in a high tax bracket as the years go by.

I live in a high-tax state, and long-term capital gains are taxed as regular income for state taxes. My taxable account is only 16% of my portfolio, and to date, my Roth is 16% of the portfolio. So, taxes are rather inevitable. My current strategy is to save the taxable account for my heirs at the stepped-up cost basis, as well as the Roth.
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

FiveK wrote: Mon Jul 08, 2024 3:09 pm
smitcat wrote: Mon Jul 08, 2024 2:03 pm
FiveK wrote: Mon Jul 08, 2024 1:52 pm Maybe it's a question of how one defines "rule of thumb". Aiming "this year" for a marginal tax rate that will consequently remain more or less constant in future years (as that article recommends) seems like a rule of thumb. It does, however, require "running your specific numbers" so one could say that goes beyond "rule of thumb". A rose by any other name....
Generally a good rule but is not easy to apply for folks with many moving parts over many years.
Maybe, maybe not. The more moving parts, the more uncertainty about how they will actually move.
If in doubt best to utilize a tool which can help optimize both drawdown and possible Roth conversions at the same time while allowing you to model any of your potential variables as well as all of your goals.
Yes, optimization under uncertainty is a worthy goal. Even if one could assign all the probabilities (market returns, life expectancy, heirs' situations, etc.) correctly, in the end it still comes down to "what marginal tax rate should I target this year?" And then repeat the process in each subsequent year.

Some people enjoy creating complex models and are capable of doing so correctly. Hats off to them. A much larger number of people either aren't interested in doing so, or may do so incorrectly (e.g., assuming higher returns in a Roth account when evaluating Roth conversions).

I suspect that, given the facts of an individual case, you and I would reach similar conclusions. :)
This rule of thumb is one that we would follow....
"maximize inflation adjusted spendable dollars (after tax) over the entire portfolio and over the entire lifetime of that portfolio including heirs"
Similar to the other 'rule of thumb' it has little value as it is not actionable as is without a bunch of descriptions, instructions and calculations.
I have found that any simple rule of thumb that says something like ...."If you have $XXX dollars in tax defferred then it would be best to do YYY"
to be incorrect some of the time.

"A much larger number of people either aren't interested in doing so, or may do so incorrectly (e.g., assuming higher returns in a Roth account when evaluating Roth conversions)."
Or perhaps testing for taxes works best when someones AA is the same during retirement.

"Even if one could assign all the probabilities (market returns, life expectancy, heirs' situations, etc.) correctly,"
I would test each one of these that concern you - see what the results are to the total spendable dollars.

"in the end it still comes down to "what marginal tax rate should I target this year?" And then repeat the process in each subsequent year."
Yes - in order to get a valid comparison you would need to add up each year of the portfolios lifetime.
cognovimus
Posts: 156
Joined: Sat Aug 10, 2019 1:07 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by cognovimus »

snic wrote: Mon Jun 24, 2024 1:00 pm
syc wrote: Fri Jun 21, 2024 9:08 pm I've heard it said that the worst thing to inherit is a traditional IRA. True?
No, it's a variable annuity.

And actually, there are plenty of worse things to inherit, including land in a distant, chaotic, war-torn country where the inheritor does not speak the local language. (Ask me how I know...)
Could you please explain why a variable annuity is a bad thing to inherit?
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

cognovimus wrote: Tue Jul 09, 2024 6:43 pm
snic wrote: Mon Jun 24, 2024 1:00 pm
syc wrote: Fri Jun 21, 2024 9:08 pm I've heard it said that the worst thing to inherit is a traditional IRA. True?
No, it's a variable annuity.

And actually, there are plenty of worse things to inherit, including land in a distant, chaotic, war-torn country where the inheritor does not speak the local language. (Ask me how I know...)
Could you please explain why a variable annuity is a bad thing to inherit?
No basis step-up, all gains are ordinary income, and the investment vehicle itself is usually overpriced and not a good deal for anyone except the salesperson who sold it. See Variable annuity - Bogleheads.
snic
Posts: 898
Joined: Wed Jun 28, 2023 11:37 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by snic »

FiveK wrote: Tue Jul 09, 2024 7:02 pm
cognovimus wrote: Tue Jul 09, 2024 6:43 pm
snic wrote: Mon Jun 24, 2024 1:00 pm
syc wrote: Fri Jun 21, 2024 9:08 pm I've heard it said that the worst thing to inherit is a traditional IRA. True?
No, it's a variable annuity.

And actually, there are plenty of worse things to inherit, including land in a distant, chaotic, war-torn country where the inheritor does not speak the local language. (Ask me how I know...)
Could you please explain why a variable annuity is a bad thing to inherit?
No basis step-up, all gains are ordinary income, and the investment vehicle itself is usually overpriced and not a good deal for anyone except the salesperson who sold it. See Variable annuity - Bogleheads.
And unless the annuity has a "stretch" provision, the inheritor has to either take distributions over a maximum of five years (which could bump up their marginal tax rate in those years) or annuitize (which is usually not optimal for a younger person who would rather invest the money than receive income).
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

smitcat wrote: Tue Jul 09, 2024 7:40 am This rule of thumb is one that we would follow....
"maximize inflation adjusted spendable dollars (after tax) over the entire portfolio and over the entire lifetime of that portfolio including heirs"
That looks more like an objective function, and no disagreement with using it as such.
I have found that any simple rule of thumb that says something like ...."If you have $XXX dollars in tax defferred then it would be best to do YYY"
to be incorrect some of the time.
Agreed - more information would be needed.
"in the end it still comes down to "what marginal tax rate should I target this year?" And then repeat the process in each subsequent year."
Yes - in order to get a valid comparison you would need to add up each year of the portfolios lifetime.
This is where the "aim for constant marginal rates" as an optimization rule of thumb comes in. In addition to this being the recommendation in the Kitces article mentioned a few posts back, it makes intuitive sense: paying a lower rate now to avoid a higher rate later should be good, while paying a higher rate now to avoid a much lower rate later will probably be bad. Keeping things ~constant is likely to be close to, if not at, the optimum.

Similar to bunching charitable contributions in some years to allow a higher itemized deduction amount, it's theoretically possible that one might incur higher income in some years to reduce IRMAA charges in other years. For anyone subject to IRMAA, however, that's likely getting into roundoff error when compared with overall net worth.

Using a complex model to estimate those future marginal rates may or may not be useful, depending on the taxpayers' circumstances and their comfort with financial modeling.

If anyone is aware of a common situation in which "aim for a constant marginal rate" provides a significantly sub-optimal result, it would be good to publicize that. Otherwise, as Kitces and others have suggested, that seems reasonable advice for most.
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

FiveK wrote: Tue Jul 09, 2024 7:59 pm
smitcat wrote: Tue Jul 09, 2024 7:40 am This rule of thumb is one that we would follow....
"maximize inflation adjusted spendable dollars (after tax) over the entire portfolio and over the entire lifetime of that portfolio including heirs"
That looks more like an objective function, and no disagreement with using it as such.
I have found that any simple rule of thumb that says something like ...."If you have $XXX dollars in tax defferred then it would be best to do YYY"
to be incorrect some of the time.
Agreed - more information would be needed.
"in the end it still comes down to "what marginal tax rate should I target this year?" And then repeat the process in each subsequent year."
Yes - in order to get a valid comparison you would need to add up each year of the portfolios lifetime.
This is where the "aim for constant marginal rates" as an optimization rule of thumb comes in. In addition to this being the recommendation in the Kitces article mentioned a few posts back, it makes intuitive sense: paying a lower rate now to avoid a higher rate later should be good, while paying a higher rate now to avoid a much lower rate later will probably be bad. Keeping things ~constant is likely to be close to, if not at, the optimum.

Similar to bunching charitable contributions in some years to allow a higher itemized deduction amount, it's theoretically possible that one might incur higher income in some years to reduce IRMAA charges in other years. For anyone subject to IRMAA, however, that's likely getting into roundoff error when compared with overall net worth.

Using a complex model to estimate those future marginal rates may or may not be useful, depending on the taxpayers' circumstances and their comfort with financial modeling.

If anyone is aware of a common situation in which "aim for a constant marginal rate" provides a significantly sub-optimal result, it would be good to publicize that. Otherwise, as Kitces and others have suggested, that seems reasonable advice for most.
"it makes intuitive sense: paying a lower rate now to avoid a higher rate later should be good, while paying a higher rate now to avoid a much lower rate later will probably be bad. Keeping things ~constant is likely to be close to, if not at, the optimum."
I do not recommend this asa I cannot pove that it works for all folks that it applies to without adding a host of things to check and caveats about how difficult that might be.
Thsi makes sense but I have not been able to prove it out since the calculations required to adjust all of the portfolio thruout its entre life are quite complex. Just a few are:
- unequal and difficult calculations for collecting all taxes (Fed/State/IRMMA/NIt) and their interactions
- checking all taxes each year not sampling one year here or there
- different aged spouses with differing SS amounts and election age choices
- "trading' tax defferred for Roths which will never be spent (and heirs adjustment after)
- trading lowered taxble accounts for lower tax drag for many years
- rising % of equity within the AA

"If anyone is aware of a common situation in which "aim for a constant marginal rate" provides a significantly sub-optimal result, it would be good to publicize that."
It is an interesting area that we have no intent to pursue since we are not interested in an academic solution to a general topic. If someone wanted to run some tests comparing the different approaches we could supply a set of inputs which may or may not have surprising results.

"Otherwise, as Kitces and others have suggested, that seems reasonable advice for most."
I agree - for most this will likely work out well. As long as they are able to calculate the real comparable tax rates so they have an good indicator for 'spendable' dollars over time. Kitces articles assume that all or most of the accounts are spent down in one lifetime along with a host of other assumptions which may or may not apply to readers.
When folks posting have more complicated , flush, and larger taxable accounts which they have no intentions of spending in their lifetime I suggest the calculators.

Interestingly viewing a goal such as 'smoothing taxes' would appear to be an objective function to us - measuring an indicator rather than the end goal.
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

smitcat wrote: Wed Jul 10, 2024 7:46 am Interestingly viewing a goal such as 'smoothing taxes' would appear to be an objective function to us - measuring an indicator rather than the end goal.
In mathematical optimization, the objective function is the goal. E.g., it could be "maximize after-tax cash flow over some time period".

To achieve that goal, one has independent variables to manipulate, and constraints to stay within. A common set of independent variables for the purpose of this topic is "annual withdrawals/conversions from the various taxable, traditional, and Roth accounts". One could add "when to start taking SS and deferrable pensions" and probably other variables.

Looking at an optimization model's series of Roth conversion amounts isn't as helpful as understanding why the model decided those amounts were optimal. Often (usually? almost always?) they will be amounts that, together with that year's unavoidable income, put one at some tax law boundary (brackets, IRMAA tiers, etc.). That's why some tools allow you to use those boundaries as independent variables, with the Roth conversion amounts then becoming dependent variables.

And that brings us to the "rule of thumb" (vs. running a full non-linear constrained optimization model) suggested by others: assume the complex model results will be at the same tax law boundary each year, and that go to that boundary "this year" based on your expectation for where you will be after all SS, pension, and RMDs are present.

As with any rule of thumb, its utility depends on the extent to which it is "close enough". While there are articles (Kitces, etc.) that show a benefit for "smoothing tax rates" vs. depleting taxable, traditional, and Roth accounts in some fixed order, I don't recall seeing any that show smoothing tax rates is a bad idea.
This makes sense but I have not been able to prove it out since the calculations required to adjust all of the portfolio thruout its entre life are quite complex. Just a few are:
- unequal and difficult calculations for collecting all taxes (Fed/State/IRMMA/NIt) and their interactions
The personal finance toolbox, for example, handles all those (at least federal, and probably close enough for most states). But some people don't like spreadsheets, so that also eliminates tools such as the Retiree Portfolio Model

- checking all taxes each year not sampling one year here or there
Unless one has a very unusual situation (e.g., owner of 20 rental properties and planning to sell 2 or 3 every other year), by the time RMDs come along things should be stable
- different aged spouses with differing SS amounts and election age choices
Fortunately, as OSS's alternative claiming strategy chart usually shows, getting close enough is good enough
- "trading' tax defferred for Roths which will never be spent (and heirs adjustment after)
- trading lowered taxble accounts for lower tax drag for many years
- rising % of equity within the AA
I think those are all covered in other comments.

"If anyone is aware of a common situation in which "aim for a constant marginal rate" provides a significantly sub-optimal result, it would be good to publicize that."
It is an interesting area that we have no intent to pursue since we are not interested in an academic solution to a general topic. If someone wanted to run some tests comparing the different approaches we could supply a set of inputs which may or may not have surprising results.
My main intent in forum conversations such as this is to inform what will go into a wiki article on Retirement income planning.

"Otherwise, as Kitces and others have suggested, that seems reasonable advice for most."
I agree - for most this will likely work out well. As long as they are able to calculate the real comparable tax rates so they have an good indicator for 'spendable' dollars over time. Kitces articles assume that all or most of the accounts are spent down in one lifetime along with a host of other assumptions which may or may not apply to readers.
When folks posting have more complicated , flush, and larger taxable accounts which they have no intentions of spending in their lifetime I suggest the calculators.
Again, I think we agree more than disagree, recognizing that "complicated" is a subjective term that different people will evaluate differently. :)
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

FiveK wrote: Wed Jul 10, 2024 9:46 am
smitcat wrote: Wed Jul 10, 2024 7:46 am Interestingly viewing a goal such as 'smoothing taxes' would appear to be an objective function to us - measuring an indicator rather than the end goal.
In mathematical optimization, the objective function is the goal. E.g., it could be "maximize after-tax cash flow over some time period".

To achieve that goal, one has independent variables to manipulate, and constraints to stay within. A common set of independent variables for the purpose of this topic is "annual withdrawals/conversions from the various taxable, traditional, and Roth accounts". One could add "when to start taking SS and deferrable pensions" and probably other variables.

Looking at an optimization model's series of Roth conversion amounts isn't as helpful as understanding why the model decided those amounts were optimal. Often (usually? almost always?) they will be amounts that, together with that year's unavoidable income, put one at some tax law boundary (brackets, IRMAA tiers, etc.). That's why some tools allow you to use those boundaries as independent variables, with the Roth conversion amounts then becoming dependent variables.

And that brings us to the "rule of thumb" (vs. running a full non-linear constrained optimization model) suggested by others: assume the complex model results will be at the same tax law boundary each year, and that go to that boundary "this year" based on your expectation for where you will be after all SS, pension, and RMDs are present.

As with any rule of thumb, its utility depends on the extent to which it is "close enough". While there are articles (Kitces, etc.) that show a benefit for "smoothing tax rates" vs. depleting taxable, traditional, and Roth accounts in some fixed order, I don't recall seeing any that show smoothing tax rates is a bad idea.
This makes sense but I have not been able to prove it out since the calculations required to adjust all of the portfolio thruout its entre life are quite complex. Just a few are:
- unequal and difficult calculations for collecting all taxes (Fed/State/IRMMA/NIt) and their interactions
The personal finance toolbox, for example, handles all those (at least federal, and probably close enough for most states). But some people don't like spreadsheets, so that also eliminates tools such as the Retiree Portfolio Model

- checking all taxes each year not sampling one year here or there
Unless one has a very unusual situation (e.g., owner of 20 rental properties and planning to sell 2 or 3 every other year), by the time RMDs come along things should be stable
- different aged spouses with differing SS amounts and election age choices
Fortunately, as OSS's alternative claiming strategy chart usually shows, getting close enough is good enough
- "trading' tax defferred for Roths which will never be spent (and heirs adjustment after)
- trading lowered taxble accounts for lower tax drag for many years
- rising % of equity within the AA
I think those are all covered in other comments.

"If anyone is aware of a common situation in which "aim for a constant marginal rate" provides a significantly sub-optimal result, it would be good to publicize that."
It is an interesting area that we have no intent to pursue since we are not interested in an academic solution to a general topic. If someone wanted to run some tests comparing the different approaches we could supply a set of inputs which may or may not have surprising results.
My main intent in forum conversations such as this is to inform what will go into a wiki article on Retirement income planning.

"Otherwise, as Kitces and others have suggested, that seems reasonable advice for most."
I agree - for most this will likely work out well. As long as they are able to calculate the real comparable tax rates so they have an good indicator for 'spendable' dollars over time. Kitces articles assume that all or most of the accounts are spent down in one lifetime along with a host of other assumptions which may or may not apply to readers.
When folks posting have more complicated , flush, and larger taxable accounts which they have no intentions of spending in their lifetime I suggest the calculators.
Again, I think we agree more than disagree, recognizing that "complicated" is a subjective term that different people will evaluate differently. :)
"I don't recall seeing any that show smoothing tax rates is a bad idea."
"If anyone is aware of a common situation in which "aim for a constant marginal rate" provides a significantly sub-optimal result, it would be good to publicize that."
It is an interesting area that we have no intent to pursue since we are not interested in an academic solution to a general topic. If someone wanted to run some tests comparing the different approaches we could supply a set of inputs which may or may not have surprising results.
My main intent in forum conversations such as this is to inform what will go into a wiki article on Retirement income planning.

"There are some articles on this subject that are also typically addressed by CFP's - here is one from Vanguard:
The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
Link...
https://institutional.vanguard.com/insi ... sions.html
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

smitcat wrote: Wed Jul 10, 2024 9:54 am "There are some articles on this subject that are also typically addressed by CFP's - here is one from Vanguard:
The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
Link...
https://institutional.vanguard.com/insi ... sions.html
Yes, that's exactly the issue mentioned in "Traditional plus taxable" vs. Roth.
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

FiveK wrote: Wed Jul 10, 2024 9:57 am
smitcat wrote: Wed Jul 10, 2024 9:54 am "There are some articles on this subject that are also typically addressed by CFP's - here is one from Vanguard:
The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
Link...
https://institutional.vanguard.com/insi ... sions.html
Yes, that's exactly the issue mentioned in "Traditional plus taxable" vs. Roth.
I do not see bypassing the marginal rates to see what happens over the life of the accounts?
I do not see any simple tools which utilize margnal rates to achieve an answer.
Which tool does that in the toolbox?

Where does it not this in the Wiki?
"The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

smitcat wrote: Wed Jul 10, 2024 10:08 am
FiveK wrote: Wed Jul 10, 2024 9:57 am
smitcat wrote: Wed Jul 10, 2024 9:54 am "There are some articles on this subject that are also typically addressed by CFP's - here is one from Vanguard:
The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
Link...
https://institutional.vanguard.com/insi ... sions.html
Yes, that's exactly the issue mentioned in "Traditional plus taxable" vs. Roth.
I do not see bypassing the marginal rates to see what happens over the life of the accounts?
"Bypassing"???
I do not see any simple tools which utilize margnal rates to achieve an answer.
Which tool does that in the toolbox?
E.g., see Worth pushing through the Social Security hump and/or IRMAA cliffs?
Where does it not this in the Wiki?
"The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
In the "Traditional plus taxable" vs. Roth section:
In these situations, tax drag in the taxable account may make Roth preferable even if the withdrawal tax rate is somewhat below the contribution tax rate. Examples of "somewhat" are shown in a table below for situations a) and b), and there are two spreadsheets that will calculate "somewhat" for specific circumstances in those situations:
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

FiveK wrote: Wed Jul 10, 2024 10:33 am
smitcat wrote: Wed Jul 10, 2024 10:08 am
FiveK wrote: Wed Jul 10, 2024 9:57 am
smitcat wrote: Wed Jul 10, 2024 9:54 am "There are some articles on this subject that are also typically addressed by CFP's - here is one from Vanguard:
The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
Link...
https://institutional.vanguard.com/insi ... sions.html
Yes, that's exactly the issue mentioned in "Traditional plus taxable" vs. Roth.
I do not see bypassing the marginal rates to see what happens over the life of the accounts?
"Bypassing"???
I do not see any simple tools which utilize margnal rates to achieve an answer.
Which tool does that in the toolbox?
E.g., see Worth pushing through the Social Security hump and/or IRMAA cliffs?
Where does it not this in the Wiki?
"The traditional wisdom has been that higher future tax rates make conversion more desirable and lower ones make it less so. Vanguard researchers describe a break-even tax rate (BETR) that yields a more accurate view of what future tax rate would make an investor indifferent to a conversion."
In the "Traditional plus taxable" vs. Roth section:
In these situations, tax drag in the taxable account may make Roth preferable even if the withdrawal tax rate is somewhat below the contribution tax rate. Examples of "somewhat" are shown in a table below for situations a) and b), and there are two spreadsheets that will calculate "somewhat" for specific circumstances in those situations:
So you do agree that making Roth conversions at higher tax rates can make sense?
Do you agree with the time related chart in the BETR article?

With these tools how many 'steps' (calculators/spreadsheets/entries) are required to adjust for the comparisons to take place?
Comparisons being drawdown strategies and/or Roth conversions?
rs9876lg
Posts: 31
Joined: Wed Jul 03, 2024 8:48 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by rs9876lg »

Can somebody explain why ending up in next higher bracket is a big deal? For example if extra 20k in income takes you out of 22 bracket and into 25 bracket. What is the material impact of that on your income and on your wealth? Assume $4M in tax advantage account say $200k base income. How does the impact compare with the daily fluctuations of your wealth?
kd2008
Posts: 1327
Joined: Sun Feb 15, 2009 5:19 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by kd2008 »

rs9876lg wrote: Wed Jul 10, 2024 11:52 am Can somebody explain why ending up in next higher bracket is a big deal? For example if extra 20k in income takes you out of 22 bracket and into 25 bracket. What is the material impact of that on your income and on your wealth? Assume $4M in tax advantage account say $200k base income. How does the impact compare with the daily fluctuations of your wealth?
It is not a big deal. The obsessive folks on this forum may have a differing view.

For most folks without a pension, Roth conversions do not move the needle much as their bracket drops in retirement usually - it may pick back up upon RMDs but the fraction of income subject to them is usually tiny enough to not warrant messing around. Plus the future tax brackets are a guessing game at most.

From Vanguard PAS initial consultation to my current Planvision advisor - all showed me with various assumptions how little Roth conversions impacted anything in our situation. The survivor amongst the two of us would be in 22% marginal bracket - same as current contribution (we do pre-tax to bring us down to 12%).

If you have large pensions, SS x 2 plus legacy heir intentions , yes Roth conversions can help but you are working with money that has zero marginal utility to you. YMMV.
User avatar
FiveK
Posts: 16356
Joined: Sun Mar 16, 2014 2:43 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by FiveK »

smitcat wrote: Wed Jul 10, 2024 10:40 am So you do agree that making Roth conversions at higher tax rates can make sense?
Yes. It's math.
Do you agree with the time related chart in the BETR article?
Don't know which chart that is, but it should reflect the same thing either of the two spreadsheets linked in the wiki show, given the same inputs.
With these tools how many 'steps' (calculators/spreadsheets/entries) are required to adjust for the comparisons to take place?
Comparisons being drawdown strategies and/or Roth conversions?
  1. Pick an age, e.g., late 70s, early 80s. Using whatever you expect for real returns, withdrawal rates, SS and pension payouts, life situation, and any other "unavoidable" income, estimate ordinary and QD+LTCG income.
  2. Calculate future marginal rate using current year's tax law
  3. Arrange this year's income so this year's marginal rate is approximately the same as the rate calculated in step #2. Hedge higher or lower depending on what you want to hedge against.
  4. Repeat next year with updated information on all those step #1 inputs
As rs9876lg and kd2008 most recently observed, getting in the ballpark - especially considering the uncertainty of how that ballpark will shift over time - is likely good enough to make no significant difference in one's quality of life.

For those who want want to do the full "optimization under uncertainty" approach, there would be many more steps in which parameters are adjusted and new calculations made. Neither approach is inherently "better", but one or the other may be much more appealing to different population segments.
Spending some time initially to do multiple "what if...?"s to get a better understanding of what drives one's personal situation, then doing simpler calculations with that/those driver(s) fixed, is another option.
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

FiveK wrote: Wed Jul 10, 2024 1:49 pm
smitcat wrote: Wed Jul 10, 2024 10:40 am So you do agree that making Roth conversions at higher tax rates can make sense?
Yes. It's math.
Do you agree with the time related chart in the BETR article?
Don't know which chart that is, but it should reflect the same thing either of the two spreadsheets linked in the wiki show, given the same inputs.
With these tools how many 'steps' (calculators/spreadsheets/entries) are required to adjust for the comparisons to take place?
Comparisons being drawdown strategies and/or Roth conversions?
  1. Pick an age, e.g., late 70s, early 80s. Using whatever you expect for real returns, withdrawal rates, SS and pension payouts, life situation, and any other "unavoidable" income, estimate ordinary and QD+LTCG income.
  2. Calculate future marginal rate using current year's tax law
  3. Arrange this year's income so this year's marginal rate is approximately the same as the rate calculated in step #2. Hedge higher or lower depending on what you want to hedge against.
  4. Repeat next year with updated information on all those step #1 inputs
As rs9876lg and kd2008 most recently observed, getting in the ballpark - especially considering the uncertainty of how that ballpark will shift over time - is likely good enough to make no significant difference in one's quality of life.

For those who want want to do the full "optimization under uncertainty" approach, there would be many more steps in which parameters are adjusted and new calculations made. Neither approach is inherently "better", but one or the other may be much more appealing to different population segments.
Spending some time initially to do multiple "what if...?"s to get a better understanding of what drives one's personal situation, then doing simpler calculations with that/those driver(s) fixed, is another option.
"Don't know which chart that is, but it should reflect the same thing either of the two spreadsheets linked in the wiki show, given the same inputs."
The simple chart that displays the amount of time after conversions vs making conversions in higher rates early a better decision.
Otherwise said - the chart which shows conversion tax rate vs time based on after conversion tax rates.
Please note that the chart does not take into consideration those effects that occurr from any inheritance from those accounts.
It shows that a significantly lower tax rate after conversions can make sence if a large enough time frame transpires after conversions.

"For those who want want to do the full "optimization under uncertainty" approach, there would be many more steps in which parameters are adjusted and new calculations made. Neither approach is inherently "better", but one or the other may be much more appealing to different population segments."
I agree - the detailed calculators can also do this with less effort in our experience.

"Spending some time initially to do multiple "what if...?"s to get a better understanding of what drives one's personal situation, then doing simpler calculations with that/those driver(s) fixed, is another option."
Once you have any of the detailed calculators set up running 'what ifs' are easy and very informative since you can investigate any of the ouputs that interest you.
Running multiple scenarios in Pralana allowed us to see optimized outputs for many of these relationships as I had posted in the past. Last time we ran 16 versions plus a baseline it took about 1-1/2 hours including the time to 'chart' their relative outputs on a simple graph for use later.

"As rs9876lg and kd2008 most recently observed, getting in the ballpark - especially considering the uncertainty of how that ballpark will shift over time - is likely good enough to make no significant difference in one's quality of life."
I do not know if you read what analysis they did or their comments.... but I would not rely on those inputs for guidance.
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

rs9876lg wrote: Wed Jul 10, 2024 11:52 am Can somebody explain why ending up in next higher bracket is a big deal? For example if extra 20k in income takes you out of 22 bracket and into 25 bracket. What is the material impact of that on your income and on your wealth? Assume $4M in tax advantage account say $200k base income. How does the impact compare with the daily fluctuations of your wealth?
It is impossible to answer your question without many more inputs.
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

kd2008 wrote: Wed Jul 10, 2024 12:47 pm
rs9876lg wrote: Wed Jul 10, 2024 11:52 am Can somebody explain why ending up in next higher bracket is a big deal? For example if extra 20k in income takes you out of 22 bracket and into 25 bracket. What is the material impact of that on your income and on your wealth? Assume $4M in tax advantage account say $200k base income. How does the impact compare with the daily fluctuations of your wealth?
It is not a big deal. The obsessive folks on this forum may have a differing view.

For most folks without a pension, Roth conversions do not move the needle much as their bracket drops in retirement usually - it may pick back up upon RMDs but the fraction of income subject to them is usually tiny enough to not warrant messing around. Plus the future tax brackets are a guessing game at most.

From Vanguard PAS initial consultation to my current Planvision advisor - all showed me with various assumptions how little Roth conversions impacted anything in our situation. The survivor amongst the two of us would be in 22% marginal bracket - same as current contribution (we do pre-tax to bring us down to 12%).

If you have large pensions, SS x 2 plus legacy heir intentions , yes Roth conversions can help but you are working with money that has zero marginal utility to you. YMMV.
"From Vanguard PAS initial consultation to my current Planvision advisor - all showed me with various assumptions how little Roth conversions impacted anything in our situation. The survivor amongst the two of us would be in 22% marginal bracket - same as current contribution (we do pre-tax to bring us down to 12%)."

It sounds like you ran all of your numbers and goals and that Roth conversions do not make much of a difference for you given those parameters.
Of course you do not have enough information on the other posters numbers or goals to assess his situation.
Jack FFR1846
Posts: 18950
Joined: Tue Dec 31, 2013 6:05 am
Location: 26 miles, 385 yards west of Copley Square

Re: sequence of withdrawals during retirement--differing philosophies

Post by Jack FFR1846 »

The answer is that it depends on your situation. Where's your savings? What will your SS be? Are you in danger of getting pushed up in tax bracket plus getting IRMAA penalties if you don't reduce your pre-tax account balances?

For someone who barely has 25 times spending saved, you likely won't have these future SS+RMD+Dividend+interest problems pushing your tax bracket and IRMAA. They are fine to take from a taxable account, especially with a low AGI because their cap gains could be zero.

For DW and me, we saved almost all in 401k's and now that I'm on Medicare, we see where that was a mistake. We're "fixing" that mistake with Roth conversions. Today, we have 60.86 times spending saved. With no spending from pre-tax accounts, when we're both taking SS and have to take RMDs, that alone, ignoring all else will be $246,653 in "income" assuming zero growth in our accounts and social security between now and then. IRMAA territory. 28% tax territory (assuming 2026 tax brackets and rates).

So after one year of retirement, all of our spending money has come from hoarded cash. DW did receive some inheritance so going forward, that's going to increase taxable. I just put $25k cash into BRK/b to get it out of interest generation and into non-dividend investment. With Roth conversions just under the IRMAA limit, that will be reduced to $179,450. 25% bracket and not even close to IRMAA. So that's what we're doing and why.
Bogle: Smart Beta is stupid
kd2008
Posts: 1327
Joined: Sun Feb 15, 2009 5:19 pm

Re: sequence of withdrawals during retirement--differing philosophies

Post by kd2008 »

Jack FFR1846 wrote: Wed Jul 10, 2024 2:24 pm The answer is that it depends on your situation. Where's your savings? What will your SS be? Are you in danger of getting pushed up in tax bracket plus getting IRMAA penalties if you don't reduce your pre-tax account balances?

For someone who barely has 25 times spending saved, you likely won't have these future SS+RMD+Dividend+interest problems pushing your tax bracket and IRMAA. They are fine to take from a taxable account, especially with a low AGI because their cap gains could be zero.
Not true in our case. But you explain your unique situation below.
For DW and me, we saved almost all in 401k's and now that I'm on Medicare, we see where that was a mistake. We're "fixing" that mistake with Roth conversions. Today, we have 60.86 times spending saved. With no spending from pre-tax accounts, when we're both taking SS and have to take RMDs, that alone, ignoring all else will be $246,653 in "income" assuming zero growth in our accounts and social security between now and then. IRMAA territory. 28% tax territory (assuming 2026 tax brackets and rates).
[
We are well over 60x, but don't have the "issues" you described.

How much is in 28% tax bracket? What was the bracket going in? Is really making any difference beyond the pearl clutching reaction of the tax bracket you are in? Does it have any marginal utility? (I suspect no, but your assessment may vary).
rs9876lg
Posts: 31
Joined: Wed Jul 03, 2024 8:48 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by rs9876lg »

I thought that was precisely the question I asked. If you have to take $275k in RMD what is the real and material impact? If you do the actual calculations where does it rank with say a drop of your net worth on a RBD? Why would you agonize about misfortune of crossing into 28 bracket? Yes, do your own calculation but you will realize that it will be insignificant in grand scheme of things.
Smitcat keeps on insisting on running the numbers but rarely do I see the actual results expressed in terms of basis point of the net assets.
Every evening before going to bed I pray to God to put me in higher tax bracket 😀
smitcat
Posts: 13737
Joined: Mon Nov 07, 2016 9:51 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by smitcat »

rs9876lg wrote: Wed Jul 10, 2024 6:29 pm I thought that was precisely the question I asked. If you have to take $275k in RMD what is the real and material impact? If you do the actual calculations where does it rank with say a drop of your net worth on a RBD? Why would you agonize about misfortune of crossing into 28 bracket? Yes, do your own calculation but you will realize that it will be insignificant in grand scheme of things.
Smitcat keeps on insisting on running the numbers but rarely do I see the actual results expressed in terms of basis point of the net assets.
Every evening before going to bed I pray to God to put me in higher tax bracket 😀
You have been on this site for one week now and all of your posts were made concerning just one subject. In some posts you indicate that this may be of a concern to you and in others you appear to indicate that they may not be a concern and/or insignificant. There are no methods I am aware of for letting anyone know what might be the results without using their own numbers and their own goals for their future.
If you were serious when you posted that $2 Million and/or maybe a 10% difference in after tax dollars would not be a significant change then you likely will have no reason(s) to continue pursuing this area. Then again I do not believe it was not clear if those future numbers were with or without an inflation calculation.
rs9876lg
Posts: 31
Joined: Wed Jul 03, 2024 8:48 am

Re: sequence of withdrawals during retirement--differing philosophies

Post by rs9876lg »

The thing that lot of us are not getting who slogged middle class life but followed third in taxes third in savings and lived on third of the income for close to 40 years as working professionals (but middle income earners) are going to afford much higher lifestyle than they are used to. Just not having ongoing mortgage and kids college tuition made us rich! Realistically if taken at 70 both SS will cover almost current living standard if the couple was earning in 300k. If they have stashed $4m towards retirement in traditional tax deferred account by all measures they need have no worry in retirement unless they want triple their spending! They can double their spending with only little bit of thinking.

This is real and we are facing it. You can run the numbers.
I accept the fact that we over saved but that was because we never got into big life style creep. I suspect quite a few BH are or will be in similar situation.
Post Reply