Why Roth conversions always pay off—if you can hold on long enough

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

marcopolo wrote: Mon Sep 27, 2021 2:55 pm The error (i believe) is in cells G84-G92. Here the source is hard-coded to come from column P instead of O.
Ah, yes, I see what you mean. I think you are correct.

If MDM sees this, I'm sure he/she will be glad for the system test assist.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

dodecahedron wrote: Sun Sep 26, 2021 6:18 pm
McQ wrote: Sun Sep 26, 2021 5:18 pm Next, the reductio. I have no story to tell here; simply the math of the worst imaginable case under the current tax structure. That would be to convert at the top rate of 37%, to avert RMDs that would only have been taxed at … zero. This is a pure reductio. I cannot imagine a real world case that would produce this course of events;
I would suggest you are lacking in imagination. Simply imagine a taxpayer couple with very successful investments (e.g., a family business) throwing off large amount of taxable income such that the TPs project being in the top tax bracket for the rest of their lives (and their children's lives.)

Then imagine that, unfortunately, down the road, an embezzler employee defrauds the family business or the child designated to take over the business turns out to be incompetent, or some environmental disaster destroys the business or some unscrupulous person or charity preys upon the taxpayers with a scam that destroys their portfolio and shrinks their AGI well below original projections.

In my own small town (population 20,000), I have read of several extremely wealthy senior citizens who eventually developed dementia and were taken advantage of by embezzling attorneys who served as fiduciary trustees of large trusts designed to provide for their old age in great luxury and comfort. I am sure this happens elsewhere.
Based on this post, in conjunction with your earlier chain-of-spouses post, I would l like to state for the record that your imagination has greater range than mine :sharebeer
I hope you will continue to exercise it here and on other threads: it adds value for me.
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

marcopolo wrote: Mon Sep 27, 2021 3:33 am
McQ wrote: Sat Sep 25, 2021 1:50 pm You guessed wrong about future tax rates--how bad will it be?

This post examines unfortunately timed Roth conversions, or perhaps, poorly analyzed conversions where future tax rates turned out not to be as expected (or feared). This post is limited to realistic pratfalls. Next post looks at terrible, horrible, no good, very bad Roth conversions that never, ever should have been done (spoiler alert: they still pay off—eventually).

.... Charts deleted for brevity
Prof. McQuarrie,

Thanks you for starting this interesting discussion.
I took the liberty of reproducing your spreadsheet. Your descriptions were quite clear and easy to follow, thanks for the detailed explanations!
I first confirmed that i was getting the same exact results as what you posted for several input scenarios.

I also ran a couple of sanity checks, like setting both tax rates equal, and setting cap gains tax rate (tax drag) to zero.
In this case, as expected, we see that the convert and non-convert cases are identical as expected.

I then took the further liberty of adding two additional changes.

1) Added a parameter that specifies the yield of the taxable account. This separates the unrealized capital gains (not taxed each year) from the annual dividend and cap gains distributions (taxed annually at specified cap gains rate).

2) Added a couple of columns to keep track of the embedded capital gains, and then at each ending year, applied the capital gains rate to the previous 10 years of accumulated unrealized gains. Since we are considering the case where the dollars under consideration are not spent, this models the case where the beneficiaries get step up basis at death of the investor, and then wait the maximum 10 years prior to liquidating the accounts, and incur capital gains taxes of the gains during those 10 years after step up.

I believe this is is a much more realistic scenario than taxing all gains in the taxable account each year.
This changes the "break even" time frame quite dramatically.

For example, at tax rates of 32% at conversion,and 24% at evaluation, your approach shows the Roth Conversion pulling ahead at age 95.
If instead, we use a 2% yield (each year has 10% gain, consisting of 2% taxable distribution, and 8% unrealized gains), what we see is that the break even age becomes 105, that is meaning the investor died at age 95, and the heirs let the accounts grow another 10 years before withdrawing.

The tax drag will eventually make up any difference in tax rates, but i think under more realistic scenarios, the time to do that can be quite significant.
Thank you marcopolo—I really appreciate your efforts. I was hoping my spreadsheet could be reproduced from the image without too much effort, but now you have provided the proof. I am also gratified you did not find any hidden calculation errors. I live in terror of those.

When I do get around to attempting a version with capital gains embedded, my plan is to proceed much as you did (=separate income, track cost basis). My intuition was that the change would impair returns on the conversion strategy, but I did not anticipate breakeven being pushed out so much as you found.

And I will pay particular attention to the case in your later post, where the Roth never recovered. It’s easy to show that outcome when the Roth gets tapped early, as so many posters anticipate doing; but I hadn’t found that in any of the “left undisturbed” scenarios I have looked at.

“Always” is indeed a tough taskmaster … to be continued.
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

cas wrote: Mon Sep 27, 2021 12:24 pm
(To give appropriate credit, McQ said earlier he plans to add these features to his spreadsheet, given some more time. But he is also busy providing these thorough posts to us, which I'm sure is taking lots of time that he could otherwise be using enhancing his spreadsheet.)
Cas- I appreciate that slack you cut me. FWIW, my modus operandi in this and similar multi-post threads is to:
1. Envision what I want to do. In this case, get more feedback from BH to assist in revising the SSRN paper; and also, to motivate that feedback by showing that I responded to the initial round of BH criticism (hence the new spreadsheet).
2. Start writing. Although I identify as a writer, I’m not actually very good at writing; rather, I’m excellent at revising and re-writing. So a few weeks get spent masticating what was supposed to be 2-3 starting posts but which the process turned into six or seven posts. In the meantime, the spreadsheet is being massaged and improved.
3. At last, the thread is launched. Several days worth of posts are in hand and ready to go.
4. But, after a few days, as now on this thread, there are no more posts finished and waiting to be pasted, and I have to write new ones, which is why things slow down.
5. Then there is the need to respond to ongoing comments
6. Not to speak of continuing to push ahead on the two or three other papers I have under way on other topics.

So, to all commenters: please forgive me if the follow up post/ re-geared spreadsheet I’ve promised doesn’t show up until well into October.
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
User avatar
FiveK
Posts: 15691
Joined: Sun Mar 16, 2014 2:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by FiveK »

cas wrote: Mon Sep 27, 2021 4:25 pm
marcopolo wrote: Mon Sep 27, 2021 2:55 pm The error (i believe) is in cells G84-G92. Here the source is hard-coded to come from column P instead of O.
Ah, yes, I see what you mean. I think you are correct.

If MDM sees this, I'm sure he/she will be glad for the system test assist.
As you predicted: thanks to Boglehead's member marcopolo for that one

Got any stock tips while you're at it? :)
cbeck
Posts: 640
Joined: Sun Jun 24, 2012 1:28 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cbeck »

McQ wrote: Sun Sep 26, 2021 2:19 pm
cbeck wrote: Sat Sep 25, 2021 6:12 pm I miss the part about the source of funds to pay the taxes. Does this analysis assume that taxes are paid from the IRA or from other, after-tax funds? If from after-tax funds, then that amounts to a conversion of taxable funds to Roth, which is a guaranteed winner, no?
The first series of posts assumes funds will be paid from the conversion. A second series considers paying funds from outside.

As FiveK indicated in his/her reply, that will make a difference, but (my words now) rather less than many people expect.
I guess I am among those who do expect a greater effect. So, let's assume you are doing a conversion at a 22% marginal fed tax rate and a 7% state tax rate, which assumes you live in one of the 38 states that do tax IRA distributions. So, the government owns 29% of your IRA. By paying your taxes with other, fully taxable funds, you are in effect contributing that 29% of your IRA in otherwise fully taxable funds to the Roth where it will be tax-free forever. That 29% of the whole pre-conversion IRA amounts to 40% of the post-conversion Roth if the taxes come out of the IRA itself. How could it be that increasing your Roth by 40% for free does not have a large effect?

The answer to that also partly depends on how long "forever" is. In my own case I expect my Roth to be feeding my wife forty years from now. For others it could be even longer. The ultimate value of my Roth depends on the longevity of the Roth, not my own longevity.
Last edited by cbeck on Mon Sep 27, 2021 7:42 pm, edited 1 time in total.
2pedals
Posts: 1988
Joined: Wed Dec 31, 2014 11:31 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by 2pedals »

McQ wrote: Fri Sep 24, 2021 12:34 pm
2pedals wrote: Thu Sep 23, 2021 8:19 pm ^^ +2
I just don't know what's going to happen. It doesn't feel too good using discretionary lumpy expenses and at the same time jumping into higher tax brackets and IRMAA payments. With adequate Roth it enables one to balance things out better when lumpy expenses are desired. RMDs are not friendly to single tax brackets for surviving spouse and 10 year distribution limits for single heirs.
I will take up the lumpy expense case in a week or two. Although your point is a staple of Roth discussions ("if I have a Roth, then I can fix the roof / replace the car / help with a down payment, without tapping my TDA and bumping up a bracket"), I'm not sure that will pencil out as expected. To be continued.
If I pay a 22% tax rate now on a Roth conversion in Washington state (no state tax on conversion), have lumpy expenses in a few years, move to Arizona. I could avoid paying Arizona taxes, avoid paying higher marginal tax rates (25%) plus higher IRMAA levels for me and my wife. It wouldn't take a long time to break even. I look at it as winning now because the utility of Roth amounts and tax-deferred traditional amounts are different due to the tax laws.
sc9182
Posts: 2165
Joined: Wed Aug 17, 2016 7:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by sc9182 »

2pedals wrote: Mon Sep 27, 2021 9:43 pm ..
If I pay a 22% tax rate now on a Roth conversion in Washington state (no state tax on conversion), have lumpy expenses in a few years, move to Arizona. I could avoid paying Arizona taxes, avoid paying higher marginal tax rates (25%) plus higher IRMAA levels for me and my wife. It wouldn't take a long time to break even. I look at it as winning now because the utility of Roth amounts and traditional amounts are different due to the tax laws.
Geographic (tax) Arbitrage — yes, cool with this idea - gr8 likelihood of success!

I am afraid - some Roth believers overly prescribe Roth conversions— even when one is in NY/MA/NJ etc areas, pay state tax during conversion., as they age, eventually retire to warmer and no-state tax state such as: FL, TX or Tennessee etc. Effectively, they pre-paid rather high state-taxes during Roth conversions., but retired to no-income-tax state !! Its hard to figure/fix this feature/bug in their Roth-conversion Calculator/tool :-(
User avatar
FiveK
Posts: 15691
Joined: Sun Mar 16, 2014 2:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by FiveK »

2pedals wrote: Mon Sep 27, 2021 9:43 pm If I pay a 22% tax rate now on a Roth conversion...to...avoid paying higher marginal tax rates (25%)...It wouldn't take a long time to break even. I look at it as winning now....
Yes, converting at a lower rate to avoid paying a higher rate later is favorable no later than when the higher rate starts.
User avatar
FiveK
Posts: 15691
Joined: Sun Mar 16, 2014 2:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by FiveK »

sc9182 wrote: Mon Sep 27, 2021 9:53 pm ...they pre-paid rather high state-taxes during Roth conversions., but retired to no-income-tax state
That is the point of this thread: it can be favorable to have done the Roth conversion, even for the quoted situation.

Not "will be favorable", not "will not be favorable", but "might or might not be favorable, depending on various things".
User avatar
FiveK
Posts: 15691
Joined: Sun Mar 16, 2014 2:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by FiveK »

cbeck wrote: Mon Sep 27, 2021 7:39 pm I guess I am among those who do expect a greater effect.
Toggle cell B6 in the MDM spreadsheet (after setting other inputs to match your situation the best you can) to see what the effect might be for you.
The answer to that also partly depends on how long "forever" is. In my own case I expect my Roth to be feeding my wife forty years from now. For others it could be even longer. The ultimate value of my Roth depends on the longevity of the Roth, not my own longevity.
Yes, for spousal heirs who can and do choose to treat the inherited IRA as their own, that is how to view it. Not sure if either the MDM or McQ tools handle all variations of this theme.
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

McQ wrote: Mon Sep 27, 2021 5:51 pm
And I will pay particular attention to the case in your later post [from marcopolo], where the Roth never recovered. It’s easy to show that outcome when the Roth gets tapped early, as so many posters anticipate doing; but I hadn’t found that in any of the “left undisturbed” scenarios I have looked at.

“Always” is indeed a tough taskmaster … to be continued.
It dawned on me yesterday that that there is a whole class of conversions, where the "left undisturbed" scenario will NOT recover. (Under current and, I think, all tax law since 2010 when Roth conversion became legal for people with AGI over $100,000.) (And, yes, I'm staying within the realm of the "RMD effect" Roth conversion scenarios being discussed in this thread and not searching over the whole universe of possible Roth conversion scenarios.)

And that class of "failed" conversions can be explained conceptually, in a few sentences, using general concepts on how taxable vs Roth accounts work.

You (McQ) are very, very close to having the lightbulb go on yourself, I think, so I'll be a bit coy and not say what it is just now.

Hint: Your most recent 2 case studies are just on the tippy edge of being examples. marcopolo's examples are also on the tippy edge. Just tweak one assumption, and they'll be there.

(There is also the risk that once this class of conversion scenarios is revealed, you'll just say "Oh. That is so obvious, it is cheating. I'll just put it in my caveats at the beginning of the paper that it is obvious and that that the rest of the paper won't consider that situation.)
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by Lee_WSP »

McQ wrote: Mon Sep 27, 2021 5:42 pm
dodecahedron wrote: Sun Sep 26, 2021 6:18 pm
McQ wrote: Sun Sep 26, 2021 5:18 pm Next, the reductio. I have no story to tell here; simply the math of the worst imaginable case under the current tax structure. That would be to convert at the top rate of 37%, to avert RMDs that would only have been taxed at … zero. This is a pure reductio. I cannot imagine a real world case that would produce this course of events;
I would suggest you are lacking in imagination. Simply imagine a taxpayer couple with very successful investments (e.g., a family business) throwing off large amount of taxable income such that the TPs project being in the top tax bracket for the rest of their lives (and their children's lives.)

Then imagine that, unfortunately, down the road, an embezzler employee defrauds the family business or the child designated to take over the business turns out to be incompetent, or some environmental disaster destroys the business or some unscrupulous person or charity preys upon the taxpayers with a scam that destroys their portfolio and shrinks their AGI well below original projections.

In my own small town (population 20,000), I have read of several extremely wealthy senior citizens who eventually developed dementia and were taken advantage of by embezzling attorneys who served as fiduciary trustees of large trusts designed to provide for their old age in great luxury and comfort. I am sure this happens elsewhere.
Based on this post, in conjunction with your earlier chain-of-spouses post, I would l like to state for the record that your imagination has greater range than mine :sharebeer
I hope you will continue to exercise it here and on other threads: it adds value for me.
It requires far less than that amount of imagination. It happened to thousands of families in 1929.

Or the family farm stops producing due to changing weather.

I can come up with far more historical examples if you'd like.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

cas wrote: Tue Sep 28, 2021 7:48 am
McQ wrote: Mon Sep 27, 2021 5:51 pm
And I will pay particular attention to the case in your later post [from marcopolo], where the Roth never recovered. It’s easy to show that outcome when the Roth gets tapped early, as so many posters anticipate doing; but I hadn’t found that in any of the “left undisturbed” scenarios I have looked at.

“Always” is indeed a tough taskmaster … to be continued.
It dawned on me yesterday that that there is a whole class of conversions, where the "left undisturbed" scenario will NOT recover. (Under current and, I think, all tax law since 2010 when Roth conversion became legal for people with AGI over $100,000.) (And, yes, I'm staying within the realm of the "RMD effect" Roth conversion scenarios being discussed in this thread and not searching over the whole universe of possible Roth conversion scenarios.)

And that class of "failed" conversions can be explained conceptually, in a few sentences, using general concepts on how taxable vs Roth accounts work.

You (McQ) are very, very close to having the lightbulb go on yourself, I think, so I'll be a bit coy and not say what it is just now.

Hint: Your most recent 2 case studies are just on the tippy edge of being examples. marcopolo's examples are also on the tippy edge. Just tweak one assumption, and they'll be there.

(There is also the risk that once this class of conversion scenarios is revealed, you'll just say "Oh. That is so obvious, it is cheating. I'll just put it in my caveats at the beginning of the paper that it is obvious and that that the rest of the paper won't consider that situation.)
Cool, I’ll play (I am not above dealing the “tease” card myself, as below).

My guess at your realization: Roth conversions always pay out, iff one of you lives long enough. In my coup de grace example (convert at 37%, RMDs at 0%), breakeven was at age 122, and I was at some pains to show that that age was not ridiculous, in terms of the heirs’ 10 years plus having a younger spouse with good longevity be the second-to-die.

But if the second to die was only three years younger, and died at 95, that Roth conversion will not breakeven before it had to be wound up, when the heirs’ ten years are over (95 + 3 + 10 = 108). Once wound up, the Roth funds are placed in a taxable account, the taxable account from reinvested RMDs is stepped up in basis, and the two are now equivalent, which means there is no longer tax drag, and the Roth never catches up.

If marcopolo’s result checks out (delayed breakeven with a step up in basis, and not just lower return), then there will be a larger amount of cases where death comes too soon and the conversion fails. Poor Bob and Barb, converting at 22% to save tax at 12%: if the same age, and both die at 85, heirs will have to distribute before breakeven at 96 (and that was with capital gains taken every year).

True, this is easily handled with a footnote to the subtitle: “hold on long enough entails one of you staying alive long enough.” But I hadn’t put two and two together myself, so you get your second acknowledgement in that footnote.

Your post is also timely because at this point in thread I’m about to switch gears from “always” to “not soon and not by much.” Across the thread, and its dozen or two posts, I’ll be striving to live up to F. Scott Fitzgerald’s famous dictum in this treatment of Roth conversions.

PS: marcopolo, you can’t get a 2% dividend on anything but a value fund today. Why not assume the Russell 2000 Growth ETF from Vanguard, with annual dividend yield of 0.18%--should help your case / support cas’ point.
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
User avatar
Lee_WSP
Posts: 10346
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by Lee_WSP »

Before I comment, can I get clarification on something?

Is the driver of Roth conversions eventually pulling ahead the tax drag from either or both RMD's and dividends from a taxable account?

If so, you can come up with a scenario where someone converts nearly all of a TDA into Roth and RMD's + dividends never touch the LTCG rate of 0%, thus allowing nearly unlimited tax free growth up until the taxable account becomes large enough to start spitting off dividends that are larger than the LTCG 0% bracket.

This can continue onto the next generation if the beneficiary is disabled and draws no income. Or is retired or otherwise draws no income.

That second scenario can theoretically go on forever without much of a stretch in imagination where the decedent is 90-100 and the beneficiary is 65-80.

Either case, both the 0% LTCG bracket ad infinitum and the drastic over convert scenario are the two pole ends that are talked about a lot, but not in actual fact.

If not, exactly what drives the Roth catchup?
Last edited by Lee_WSP on Tue Sep 28, 2021 4:27 pm, edited 1 time in total.
User avatar
LadyGeek
Site Admin
Posts: 95466
Joined: Sat Dec 20, 2008 4:34 pm
Location: Philadelphia
Contact:

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by LadyGeek »

McQ wrote: Tue Sep 28, 2021 4:08 pm My guess at your realization: Roth conversions always pay out, iff one of you lives long enough.
I assume "iff" is the academic shorthand for "If and only if", i.e. "if and only if one of you lives long enough."
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
marcopolo
Posts: 8411
Joined: Sat Dec 03, 2016 9:22 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by marcopolo »

Lee_WSP wrote: Tue Sep 28, 2021 4:26 pm Before I comment, can I get clarification on something?

Is the driver of Roth conversions eventually pulling ahead the tax drag from either or both RMD's and dividends from a taxable account?

If so, you can come up with a scenario where someone converts nearly all of a TDA into Roth and RMD's + dividends never touch the LTCG rate of 0%, thus allowing nearly unlimited tax free growth up until the taxable account becomes large enough to start spitting off dividends that are larger than the LTCG 0% bracket.

This can continue onto the next generation if the beneficiary is disabled and draws no income. Or is retired or otherwise draws no income.

That second scenario can theoretically go on forever without much of a stretch in imagination where the decedent is 90-100 and the beneficiary is 65-80.

Either case, both the 0% LTCG bracket ad infinitum and the drastic over convert scenario are the two pole ends that are talked about a lot, but not in actual fact.

If not, exactly what drives the Roth catchup?
From playing around with several scenarios, it looks to me like the tax drag is the main driver, and it is dominated by the assumption that ALL gains in the taxable account (housing RMDs) are taxed each year at 15% LTCG rate. As someone pointed out above, this was true even for the cases where the withdrawal rate was set to 12%, and even 0%. So, in this iteration of the model, there is no such thing as a 0% LTCG.

I honestly do not see the point (other than as an academic exercise) in modeling this issue in such a way. Who pays taxes on all gains each year in real life? Especially when the assumptions also say none of the money is spent, and is passed to heirs. Why pay taxes each year on all gains when step up basis is coming soon?
Once in a while you get shown the light, in the strangest of places if you look at it right.
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

McQ wrote: Tue Sep 28, 2021 4:08 pm
cas wrote: Tue Sep 28, 2021 7:48 am
McQ wrote: Mon Sep 27, 2021 5:51 pm
And I will pay particular attention to the case in your later post [from marcopolo], where the Roth never recovered. It’s easy to show that outcome when the Roth gets tapped early, as so many posters anticipate doing; but I hadn’t found that in any of the “left undisturbed” scenarios I have looked at.
It dawned on me yesterday that that there is a whole class of conversions, where the "left undisturbed" scenario will NOT recover.

My guess at your realization: Roth conversions always pay out, iff one of you lives long enough.
No. The opposite point.

I'm pointing to a whole class of plausible Roth conversions scenarios that would NEVER pay out ... a refutation of your "Roth ALWAYS pays out."

(Edit: where "NEVER" means "never as shown by either McQ's or MDM's spreadsheets." Real life is more a case of "never say never." )

Hint #2: the question I asked up above:
cas wrote: Sun Sep 26, 2021 6:23 pm
Your hypothetical invokes some scenario where RMDs end up taxed at 0%. (OK. Fair enough. One plausible example that is frequently discussed in bogleheads threads is a scenario where high long term care medical expenses lead to a high medical deduction.)

But then you are leaving the taxation of realized capital gains and dividends at 15%.

Is that intentional?
(Hint #2.5. Under the current tax code, in order to have a 0% nominal tax rate on RMDs (ordinary income), but a 15% tax rate on qualified dividends and long term cap gains (QD/LTCG), you would need a huge taxable account "off camera" kicking off huge amounts of QD/LTCG year after year.

e.g. Here's a tax visualization tool showing such a starting scenario. (With the limitation that using standard deduction is hard coded in to the tool. You could have much more ordinary income in the 0% bracket if you had a huge itemized deduction, e.g. big medical long term care expenses.)

Did you really mean to invoke a new multi-million dollar account "off-camera"?

If not, then should the tax rate on the QD/LTCG in your scenario really be 15%?

If it isn't 15%, what happens to your scenario results?)
Last edited by cas on Tue Sep 28, 2021 10:54 pm, edited 1 time in total.
marcopolo
Posts: 8411
Joined: Sat Dec 03, 2016 9:22 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by marcopolo »

cas wrote: Tue Sep 28, 2021 4:56 pm
McQ wrote: Tue Sep 28, 2021 4:08 pm
cas wrote: Tue Sep 28, 2021 7:48 am
McQ wrote: Mon Sep 27, 2021 5:51 pm
And I will pay particular attention to the case in your later post [from marcopolo], where the Roth never recovered. It’s easy to show that outcome when the Roth gets tapped early, as so many posters anticipate doing; but I hadn’t found that in any of the “left undisturbed” scenarios I have looked at.
It dawned on me yesterday that that there is a whole class of conversions, where the "left undisturbed" scenario will NOT recover.

My guess at your realization: Roth conversions always pay out, iff one of you lives long enough.
No. The opposite point.

I'm pointing to a whole class of plausible Roth conversions scenarios that would NEVER pay out ... a refutation of your "Roth ALWAYS pays out."

Hint #2: the question I asked up above:
cas wrote: Sun Sep 26, 2021 6:23 pm
Your hypothetical invokes some scenario where RMDs end up taxed at 0%. (OK. Fair enough. One plausible example that is frequently discussed in bogleheads threads is a scenario where high long term care medical expenses lead to a high medical deduction.)

But then you are leaving the taxation of realized capital gains and dividends at 15%.

Is that intentional?
(Hint #2.5. Under the current tax code, in order to have a 0% nominal tax rate on RMDs (ordinary income), but a 15% tax rate on qualified dividends and long term cap gains (QD/LTCG), you would need a huge taxable account "off camera" kicking off huge amounts of QD/LTCG year after year.

e.g. Here's a tax visualization tool showing such a starting scenario. (With the limitation that using standard deduction is hard coded in to the tool. You could have much more ordinary income in the 0% bracket if you had a huge itemized deduction, e.g. big medical long term care expenses.)

Did you really mean to invoke a new multi-million dollar account "off-camera"?

If not, then should the tax rate on the QD/LTCG in your scenario really be 15%?

If it isn't 15%, what happens to your scenario results?)
It might not take much "off camera" dollars.
Under the assumption of the model, all gains in taxable are realized each year. That will eventually make the LTCG quite large each year.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
FiveK
Posts: 15691
Joined: Sun Mar 16, 2014 2:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by FiveK »

cas wrote: Tue Sep 28, 2021 7:48 am It dawned on me yesterday that that there is a whole class of conversions, where the "left undisturbed" scenario will NOT recover. (Under current and, I think, all tax law since 2010 when Roth conversion became legal for people with AGI over $100,000.) (And, yes, I'm staying within the realm of the "RMD effect" Roth conversion scenarios being discussed in this thread and not searching over the whole universe of possible Roth conversion scenarios.)
Is the scenario below getting close?

Code: Select all

+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
| Original owner     |           |       | Non-spouse heir  |     |   |   |                                          |            |
+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
| RMD Start age      | 72        |       | Death age        | 85  |   |   |                                          |            |
| Roth conv. age     | 71        |       | Basis step       | 1   |   |   |                                          |            |
| Roth conv. amount  | $100,000  |       | RMD delay        | 0   |   |   |                                          |            |
| 1st yr tax rate    | 22%       |       | RMD span         | 10  |   |   |                                          |            |
| Pay tax w/ cash    | 1         |       | EOY RMD?         | 1   |   |   |                                          |            |
| Basis fract.       | 100%      |       |                  |     |   |   |                                          |            |
| EOY conversion?    | 1         |       |                  |     |   |   |                                          |            |
| EOY RMD?           | 1         |       |                  |     |   |   |                                          |            |
|                    |           |       |                  |     |   |   |                                          | Roth Gain  |
|                    | Original  | Heir  |                  |     |   |   | Just before owner's death                | -$22,205   |
| Annual tax rate    | 15.00%    | 15%   |                  |     |   |   | Just after inheritance                   | -$22,205   |
| Stocks             | 100%      | 100%  |                  |     |   |   | After 10 years                           | -$47,939   |
| Bonds              | 0%        | 0%    |                  |     |   |   | After 10 years (if unspent)              | -$47,939   |
| Turnover           | 10%       | 10%   |                  |     |   |   | Owner's age at which conversion wins     | >125       |
| Cap gain           | 6%        | 6%    |                  |     |   |   | Years after inheritance for Roth to win  | Never      |
| Dividend           | 2%        | 2%    |                  |     |   |   |                                          |            |
| Interest           | 5%        | 5%    |                  |     |   |   |                                          |            |
| Tax-ad gain        | 8%        | 8%    |                  |     |   |   |                                          |            |
| CG tax             | 0.0%      | 0%    |                  |     |   |   |                                          |            |
| Div. tax           | 0.0%      | 0%    |                  |     |   |   |                                          |            |
+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
marcopolo
Posts: 8411
Joined: Sat Dec 03, 2016 9:22 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by marcopolo »

FiveK wrote: Tue Sep 28, 2021 5:23 pm
cas wrote: Tue Sep 28, 2021 7:48 am It dawned on me yesterday that that there is a whole class of conversions, where the "left undisturbed" scenario will NOT recover. (Under current and, I think, all tax law since 2010 when Roth conversion became legal for people with AGI over $100,000.) (And, yes, I'm staying within the realm of the "RMD effect" Roth conversion scenarios being discussed in this thread and not searching over the whole universe of possible Roth conversion scenarios.)
Is the scenario below getting close?

Code: Select all

+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
| Original owner     |           |       | Non-spouse heir  |     |   |   |                                          |            |
+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
| RMD Start age      | 72        |       | Death age        | 85  |   |   |                                          |            |
| Roth conv. age     | 71        |       | Basis step       | 1   |   |   |                                          |            |
| Roth conv. amount  | $100,000  |       | RMD delay        | 0   |   |   |                                          |            |
| 1st yr tax rate    | 22%       |       | RMD span         | 10  |   |   |                                          |            |
| Pay tax w/ cash    | 1         |       | EOY RMD?         | 1   |   |   |                                          |            |
| Basis fract.       | 100%      |       |                  |     |   |   |                                          |            |
| EOY conversion?    | 1         |       |                  |     |   |   |                                          |            |
| EOY RMD?           | 1         |       |                  |     |   |   |                                          |            |
|                    |           |       |                  |     |   |   |                                          | Roth Gain  |
|                    | Original  | Heir  |                  |     |   |   | Just before owner's death                | -$22,205   |
| Annual tax rate    | 15.00%    | 15%   |                  |     |   |   | Just after inheritance                   | -$22,205   |
| Stocks             | 100%      | 100%  |                  |     |   |   | After 10 years                           | -$47,939   |
| Bonds              | 0%        | 0%    |                  |     |   |   | After 10 years (if unspent)              | -$47,939   |
| Turnover           | 10%       | 10%   |                  |     |   |   | Owner's age at which conversion wins     | >125       |
| Cap gain           | 6%        | 6%    |                  |     |   |   | Years after inheritance for Roth to win  | Never      |
| Dividend           | 2%        | 2%    |                  |     |   |   |                                          |            |
| Interest           | 5%        | 5%    |                  |     |   |   |                                          |            |
| Tax-ad gain        | 8%        | 8%    |                  |     |   |   |                                          |            |
| CG tax             | 0.0%      | 0%    |                  |     |   |   |                                          |            |
| Div. tax           | 0.0%      | 0%    |                  |     |   |   |                                          |            |
+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
As soon as you reduce the turnover to something reasonable, the tax drag becomes quite small, and there are a myriad of scenarios in which the conversion will never payoff.
Once in a while you get shown the light, in the strangest of places if you look at it right.
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

FiveK wrote: Tue Sep 28, 2021 5:23 pm Is the scenario below getting close?

Code: Select all

+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
| Original owner     |           |       | Non-spouse heir  |     |   |   |                                          |            |
+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
| RMD Start age      | 72        |       | Death age        | 85  |   |   |                                          |            |
| Roth conv. age     | 71        |       | Basis step       | 1   |   |   |                                          |            |
| Roth conv. amount  | $100,000  |       | RMD delay        | 0   |   |   |                                          |            |
| 1st yr tax rate    | 22%       |       | RMD span         | 10  |   |   |                                          |            |
| Pay tax w/ cash    | 1         |       | EOY RMD?         | 1   |   |   |                                          |            |
| Basis fract.       | 100%      |       |                  |     |   |   |                                          |            |
| EOY conversion?    | 1         |       |                  |     |   |   |                                          |            |
| EOY RMD?           | 1         |       |                  |     |   |   |                                          |            |
|                    |           |       |                  |     |   |   |                                          | Roth Gain  |
|                    | Original  | Heir  |                  |     |   |   | Just before owner's death                | -$22,205   |
| Annual tax rate    | 15.00%    | 15%   |                  |     |   |   | Just after inheritance                   | -$22,205   |
| Stocks             | 100%      | 100%  |                  |     |   |   | After 10 years                           | -$47,939   |
| Bonds              | 0%        | 0%    |                  |     |   |   | After 10 years (if unspent)              | -$47,939   |
| Turnover           | 10%       | 10%   |                  |     |   |   | Owner's age at which conversion wins     | >125       |
| Cap gain           | 6%        | 6%    |                  |     |   |   | Years after inheritance for Roth to win  | Never      |
| Dividend           | 2%        | 2%    |                  |     |   |   |                                          |            |
| Interest           | 5%        | 5%    |                  |     |   |   |                                          |            |
| Tax-ad gain        | 8%        | 8%    |                  |     |   |   |                                          |            |
| CG tax             | 0.0%      | 0%    |                  |     |   |   |                                          |            |
| Div. tax           | 0.0%      | 0%    |                  |     |   |   |                                          |            |
+────────────────────+───────────+───────+──────────────────+─────+───+───+──────────────────────────────────────────+────────────+
Yep.
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

marcopolo wrote: Tue Sep 28, 2021 5:29 pm
As soon as you reduce the turnover to something reasonable, the tax drag becomes quite small, and there are a myriad of scenarios in which the conversion will never payoff.
True, but that isn't the key assumption that leads to a whole class of scenarios where the Roth NEVER succeeds.
Last edited by cas on Tue Sep 28, 2021 6:48 pm, edited 1 time in total.
Roothy
Posts: 145
Joined: Wed Aug 26, 2015 9:20 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by Roothy »

Can someone answer the question posed above: if Roth conversions "always" (eventually) pay out, why does it ever make sense to use traditional retirement accounts?
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

marcopolo wrote: Tue Sep 28, 2021 5:20 pm It might not take much "off camera" dollars.
Under the assumption of the model, all gains in taxable are realized each year. That will eventually make the LTCG quite large each year.
This "off camera" account would need to kick off a minimum of $80K(ish) every year, starting at age 72 in the scenario. So, true, that is "only" a $800K account at 10% annual return and 100% annual turnover. (As opposed to a multi-million dollar account.)

But, I was trying to give the scenario-people a modicum of credit that they would have learned not to shoot themselves annually in their tax foot if they actually had accumulated an $800K+ "off camera" taxable account during the years when they were in the 22%+ ordinary income bracket / 15%+ QD/LTCG bracket.

Anyway ... the "off camera" account size/tax-behavior is a distraction from the point at hand. Nobody should let it distract them.
User avatar
FiveK
Posts: 15691
Joined: Sun Mar 16, 2014 2:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by FiveK »

Roothy wrote: Tue Sep 28, 2021 6:58 pm Can someone answer the question posed above: if Roth conversions "always" (eventually) pay out, why does it ever make sense to use traditional retirement accounts?
The primary answer: conversions don't always pay out.

A secondary answer: contributions are different than conversions.

In general, "it depends..." on many situation-specific conditions.
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

cas wrote: Tue Sep 28, 2021 7:48 am I'm pointing to a whole class of plausible Roth conversions scenarios that would NEVER pay out ... a refutation of your "Roth ALWAYS pays out."
It occurs to me I need to qualify that "NEVER" statement.

By "NEVER," I meant "never as shown by either McQ's or MDM's spreadsheets."

Both spreadsheets make the simplifying assumption that the tax rate is allowed to be changed only once during the life of the original owner - and only immediately after the conversion year. This would seem to be a perfectly adequate assumption for teasing out general behaviors of the "RMD effect."

However, I will fully allow that real life and the real tax code are more complicated. One should probably never say never (or always).

Among other possible complications, excellent compounding over decades in the "on-camera" accounts could cause annual "on-camera" distributions to start being taxed across multiple tax brackets, which might cause tax drag to start accelerating more rapidly than the spreadsheets show.
wrongfunds
Posts: 3187
Joined: Tue Dec 21, 2010 2:55 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by wrongfunds »

What does "on-camera" account mean?
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

wrongfunds wrote: Wed Sep 29, 2021 11:21 am What does "on-camera" account mean?
Short answer:

The "on camera" accounts are the accounts that we can see in the screenshots of the spreadsheet in McQ's posts. (The "camera" is looking at the spreadsheet, and McQ's posts show the picture it takes.)

Long answer:
The scenario is that some 71 year old person has 100K worth of assets in a tIRA that he predicts will never be needed for his needs in his lifetime. The fate of these assets (no Roth conversion vs Roth conversion) is tracked before our eyes through the years in the spreadsheet rows. Since we can see the numbers change through the years, I called these the "on camera" accounts/assets.

However, we know that the owner must also have another set of assets/income-streams that he is living on. We know nothing about the nature of these assets/income-streams except that they are affecting the tax rate on the accounts that we can see in the spreadsheet. Since we know that they must exist, but we have no visibility into them, I called them "off camera."

Example:

In McQ's first scenario in this thread...

In the option where no Roth conversion was done, the age-72 RMD from the 100K tIRA is $3650. But the spreadsheet shows that that $3650 is being taxed at 24%.

If that $3650 RMD was the *only* income that the owner received that year, obviously the RMD wouldn't be taxed at 24%.

So we know that there have to be other accounts/income streams out there (invisible to us) that are filling up the owner's 0%-> 22% brackets.

The $3650 RMD (whose source is visible to us "on camera") is sitting on top of that opaque stack of income (whose source(s) are invisible to us "off camera"). Otherwise the $3650 RMD wouldn't be taxed at 24%.
L84SUPR
Posts: 343
Joined: Sat Feb 01, 2014 12:28 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by L84SUPR »

cas wrote: Wed Sep 29, 2021 12:15 pm
wrongfunds wrote: Wed Sep 29, 2021 11:21 am What does "on-camera" account mean?
Short answer:

The "on camera" accounts are the accounts that we can see in the screenshots of the spreadsheet in McQ's posts. (The "camera" is looking at the spreadsheet, and McQ's posts show the picture it takes.)

Long answer:
The scenario is that some 71 year old person has 100K worth of assets in a tIRA that he predicts will never be needed for his needs in his lifetime. The fate of these assets (no Roth conversion vs Roth conversion) is tracked before our eyes through the years in the spreadsheet rows. Since we can see the numbers change through the years, I called these the "on camera" accounts/assets.
Is there a thread discussing the case of middle class Californians (mfj), or other states with similar taxes. I have done some sensitivity analysis with RPM and would like to compare to other discussions. I found the case of the affluent Californians but we are not quite at that level.

More specifically I looked at filling the 12+9.3 percent bracket from 63 to 70 and as a separate increment filling the 22+9.3 at age 60-62 during the last years working.

If you know of such a discussion I would appreciate a link.
1/3rd VTWAX, 1/3rd Wellington, 1/3rd G Fund | All models are wrong. Some are useful.
cas
Posts: 2245
Joined: Wed Apr 26, 2017 8:41 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by cas »

All of this "on camera" and "off camera" distinction is a complete distraction to the real discussion, which is on the "RMD effect." So I apologize for that.

But ... McQ seemed to be heading towards pinning the whole trajectory of his upcoming discussion on the fact that his spreadsheet could NOT find a counter-example to his "Why Roth conversions always pay off—if you can hold on long enough" thread title.

But he was supporting this "I can't find a counter-example, even though I've tried really, really hard" assertion with scenarios (which he stipulated were using current tax law) that are not impossible for quite affluent people, but were really kind of odd. For example ... a scenario where ordinary income is taxed in the 0% nominal bracket, but qualified dividend/long term capital gain income is taxed at 15%.

Meanwhile, he was missing an almost identical income scenario that is talked about all the time on bogleheads: ordinary income taxed at 0% (or at least at 12% or below) and QD/LTCG income taxed at 0%.

And that common scenario (combined with doing the spreadsheet's Roth conversion at a higher rate) DOES provide an example where his spreadsheet shows that a Roth conversion NEVER pays off, no matter how old Methuselah gets.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

LadyGeek wrote: Tue Sep 28, 2021 4:27 pm
McQ wrote: Tue Sep 28, 2021 4:08 pm My guess at your realization: Roth conversions always pay out, iff one of you lives long enough.
I assume "iff" is the academic shorthand for "If and only if", i.e. "if and only if one of you lives long enough."
Yes; but I had assumed it was an abbreviation in widespread use
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
User avatar
dodecahedron
Posts: 6563
Joined: Tue Nov 12, 2013 11:28 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by dodecahedron »

McQ wrote: Wed Sep 29, 2021 5:20 pm
LadyGeek wrote: Tue Sep 28, 2021 4:27 pm
McQ wrote: Tue Sep 28, 2021 4:08 pm My guess at your realization: Roth conversions always pay out, iff one of you lives long enough.
I assume "iff" is the academic shorthand for "If and only if", i.e. "if and only if one of you lives long enough."
Yes; but I had assumed it was an abbreviation in widespread use
Widely used in theoretical math and related academic disciplines (logic, etc.) Not really used much outside academia.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

cas wrote: Wed Sep 29, 2021 12:52 pm All of this "on camera" and "off camera" distinction is a complete distraction to the real discussion, which is on the "RMD effect." So I apologize for that.

But ... McQ seemed to be heading towards pinning the whole trajectory of his upcoming discussion on the fact that his spreadsheet could NOT find a counter-example to his "Why Roth conversions always pay off—if you can hold on long enough" thread title.

But he was supporting this "I can't find a counter-example, even though I've tried really, really hard" assertion with scenarios (which he stipulated were using current tax law) that are not impossible for quite affluent people, but were really kind of odd. For example ... a scenario where ordinary income is taxed in the 0% nominal bracket, but qualified dividend/long term capital gain income is taxed at 15%.

Meanwhile, he was missing an almost identical income scenario that is talked about all the time on bogleheads: ordinary income taxed at 0% (or at least at 12% or below) and QD/LTCG income taxed at 0%.

And that common scenario (combined with doing the spreadsheet's Roth conversion at a higher rate) DOES provide an example where his spreadsheet shows that a Roth conversion NEVER pays off, no matter how old Methuselah gets.
Actually, Cas, I think the on-camera, off-camera distinction is quite helpful here. I liked your explanation of it upthread; it shows the level of insight I’ve come to expect.

To begin: why is there a new spreadsheet, when the SSRN paper spreadsheet could easily have been patched up to fix any problems?

Answer: the SSRN spreadsheet came to seem to me to be too complex, precisely because it kept everything “on camera.” It had to show the $90,000 social security payments, and the $4,000,000 TDA, that together just barely nudged Rob and Sue into the 24% bracket in the 2027 time frame. It also had to allow for income spread across tax bracket boundaries, varying rates of capital gains tax depending on age of liquidation, yada yada. Plus it opened up all kinds of unproductive discussions (what if I have pension income too? What if I have a large pre-existing taxable account?); unproductive, that is, from my standpoint of seeking the foundation of any conversion benefits.

For the new spreadsheet, that’s all off camera. I just assume the assets and income necessary to have the indicated tax rate, 24% in the base case. The ordinary income that puts you in that bracket puts you at least in the 15% cap gains bracket, if not the 18.8% bracket; and if capital gains are embedded rather than taken annually, and have to be taken while alive, then the rate will quickly reach 23.8%, unless Congress, ahem, chooses to hike the current top LTCG rate still further.

How could a scenario have a 0% tax rate on RMDs while the LTCG rate stays 15%? In my imagining, Congress passed a law exempting RMDs from taxation but stacking them at the bottom for computing tax on other income, so that other income is taxed as if RMDs had filled the bottom brackets. Just like Congress did for international income. Again, the extreme cases are theoretical demonstrations. Most of the thread will be concerned with the middle brackets where most BH reside. Confession: I don't have much to offer BH who can arrange to be in the 0% bracket; I'll never be there in my life (actually, not ever even in the 12% bracket), hence tend to slight it.

Next, and acknowledging Lee_WSJ (and FiveK), if future tax rates are constant or move lower, all the tax drag comes from the taxable account where RMDs were re-invested. If the tax on the taxable account is or becomes zero, there is no (more) tax drag and no chance for the conversion to catch up and breakeven when future rates drop. Conversely, if there is any tax drag at all, then eventually the conversion will prevail, but maybe not in human time, pace dodecahedron’s endless chain of spouses (see in the paper my development of Sydney Homer’s example of the Roman soldier investing a drachma).

Bottom line: there are simple spreadsheets and comprehensive spreadsheets, but I lack the wit to construct one that is both.

Oh wait: even this thread’s ss was labeled “very detailed” in an earlier comment …
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

Roothy wrote: Tue Sep 28, 2021 6:58 pm Can someone answer the question posed above: if Roth conversions "always" (eventually) pay out, why does it ever make sense to use traditional retirement accounts?
Why you would NOT contribute only to a Roth 401(k) throughout your career

Because then you would have given up decades of tax deduction at 22%, 24%, or 32%, or more, with the result that you did not have enough TDA to generate RMDs to fill up the 10% or the 12% brackets when retired, and could not harvest the benefits of deducting at 32% and distributing at 12%. Pity.

On the other hand, we can imagine someone aged 60+, near the end of their working life, doing the math and realizing they are on track to have more than ample TDA funds, and switching at that point to a Roth 401(k). Somewhat iffy, if they are still in the 32% bracket, and giving up that level of deduction; but almost certainly viable in the 22% or 24% bracket. (Viable meaning "pay off while still alive even if you die young")
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
Beehave
Posts: 1166
Joined: Mon Jun 19, 2017 12:46 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by Beehave »

Watty wrote: Fri Sep 24, 2021 8:47 am
Why Roth conversions always pay off—if you can hold on long enough
I did not follow the details but you need to be careful about using the word "always" since one exception is enough to disprove your primis.

For example a single person might do a Roth conversion when they are in the higher single tax brackets, but they could then get married and then be in the lower joint tax brackets.

Another is that someone might have a year when they have exceptionally high income and are in a higher than normal tax bracket, like when they sell a business. Doing a Roth conversion that year would likely not pay off either.

For either person the Roth conversion would not pay off so your argument would be shown to not be correct.

You obviously put a lot of thought and work into this you might want to rephrase that "always" claim since it sounds too extreme and bit like clickbait.
+1 There are additional cases in which 401K could beat Roth:

. A . Under current tax laws - - if you use the 401K to pay high LTC or healthcare bills, they may be tax deductible. If you've converted all to Roth you'd miss out.

. B . Under conceivable future tax laws, it would be possible for a VAT to replace in income taxes in part or in whole, making prior Roth conversions less valuable or possibly unfortunate mistakes.

When it comes to questions of finance, educated tilts are fine, but for my money, I prefer diversified holdings over all-in "always-type" surefire bets.
Roth is fine and probably the best alternative. But not definitely.
Topic Author
McQ
Posts: 1414
Joined: Fri Jun 18, 2021 12:21 am
Location: California

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by McQ »

Part II

At this point in the thread a second series of posts begins. The first series had the key word “always.” The second series picks up on the sub-subtitle to the thread: “but not by very much and seldom very soon.” And the theme of today’s post might be glossed as, “yes, it is theoretically possible to receive a speedy and lucrative outcome from a Roth conversion, but it may not be likely in your case, if you convert in the middle brackets.”

All the caveats laid out in the initial post continue to apply, of course: if you can convert at 0%, 10%, 12% today, to save tax of 37% or 39.6% later—go for it! Huge benefit from converting in that situation. However, such halcyon circumstances will not be the focus of this post or the next few.

The stipulation here:
-you might convert at 22% today to save 25% later, post-TCJA; or,
-that you convert at 22% today, in hopes to save 28% later (because the post-TCJA tax brackets turn what would have been a 24% rate, in the next bracket up, back into the pre-TCJA rate of 28%); or,
-you convert at 22% today to save both 28% and IRMAA charges later; or,
-just maybe, the RMDs to be reduced would have fallen into the upper part of the social security tax torpedo, carrying a marginal rate of almost 41% (46% post-TCJA).

These are all plausible “best cases” for a relatively early and relatively lucrative payoff from a Roth conversion made in the middle brackets. But I’ll also consider a constant 22% rate as a benchmark. I showed in the first series that a conversion under constant rates pays off eventually; here in the second series the focus will be on how to calibrate that payoff, to compare how much better off one would be under one of the “best plausible cases” just introduced.

At first glance, these conditions should still apply to a large number of BH; I’ve certainly heard, on a variety of threads, that whether or not to convert at 22% is a real issue up for debate. But you need to ask yourself: is your 22% conversion really going to avert RMDs at 28% or more? Or are you cruising to be Bob and Barb, from two posts back, converting at 22% today to save … 12% on your RMDs later?

Now you see the stakes.

Straight talk about future tax brackets

How could you tell whether you were hurtling down the track to pay only 12% on your RMDs, or instead, are at a real risk of paying 28% or more? The one would discourage a Roth conversion in the 22% bracket, the other might strongly encourage one.

Green eyeshade time: the AGI ceiling for the 22% bracket today is $200,450 if 65 or over (IRS stated bracket plus standard deductions). The bottom / top of the first IRMAA bracket is $176,000 / $222,000 AGI. Note how the 22%/24% threshold divides the first IRMAA bracket almost exactly in half.
And last: the floor of the 22% bracket in AGI terms is $108,750 (I’ll ignore the inconvenient detail that the 22% bracket is less than $100,000 wide. Bear with me as I continue to enter $100,000 as the conversion amount.)

How to figure marginal rates with IRMAA

The full IRMAA charge kicks in on the first dollar of excess income; hence, the dollar of income that takes you from $176,000 to $176,001 has a marginal rate in the thousands of percent. Savvy taxpayers who stray into IRMAA territory know to voluntarily accelerate income up to the top of the IRMAA bracket into which they strayed. For the first three IRMAA brackets, the quotient [IRMAA dollars / IRMAA bracket span] will equal 4%, plus or minus (see SSRN paper appendix). So that’s one way to estimate the “combined” marginal rate: income tax rate + 4% IRMAA rate.

But there’s a problem with the first IRMAA bracket; as noted above, it straddles the 22% / 24% income tax threshold. The combined income tax plus IRMAA marginal rate changes at $200,451. Here is a table to organize the numbers, assuming taxpayers over 65 filing as married joint here in 2021.

Code: Select all

Income range -----> combined tax rate
$108,750 - $176,000 22%
$176,001 - $204,850 30% (22% + [IRMAA dollars / 0.5 IRMAA bracket span], or 30% in all, at the 22% ceiling)
$204,851 - $222,000 32%, declining toward 28% as IRMAA ceiling is approached
$222,001 - $330,000 28% at the top of IRMAA bracket #2 and bracket #3, higher if brackets are not tapped all the way.
These 2021 rates are TCJA rates; if the planner projects the expiration of TCJA, then the corresponding values are

Code: Select all

22% ----->  25%
30% ----->  33%
32/28% --> 36%/32%
28% ----->  32%
To sum up, there is indeed a range of income likely to be taxed, post-TCJA, at combined rates around 32%. To convert at 22% to save 32% certainly sounds attractive. But are you, for lack of a better word, eligible for this favorable outcome? Put another way, is it legitimate for you, with your very particular income and assets, to fear tax rates on your projected RMDs at rates on the order of 32%? Or have you been the victim of fearmongering, of the sort commonly seen in financial journalism, promotional missives from financial advisors, and (sometimes) social media?

Are you eligible / should you be afraid?

All the threshold values just given are for 2021. Tax brackets and IRMAA brackets adjust for inflation each year; as in the spreadsheet, I will assume an inflation rate of 3%.

Next, let’s assume that you, the prospective Roth converter, are in your early 60s, or maybe, mid-50s. Your RMDs are not going to begin for 10, 15, or 20 years yet. Whether you will be eligible for that very favorable Roth conversion outcome, where you convert today at 22% to avert subsequent RMDs taxed at scary high rates like 32%, is a function of what those tax brackets will be 10, 15, and 20 years in the future.

To keep it simple, let’s focus on $176,000, the 1st IRMAA threshold, and $204,850, the 24% (or maybe 28% after 2025) bracket floor. Here are the inflation-adjusted values, rounded to the nearest $250.

Code: Select all

Years out	IRMAA #1	22% ceiling
10	       $236,500 	$270,000 
15	       $274,250 	$313,000 
20	       $318,000 	$362,750 
For discussion purposes, I’ll take the middle time frame of 15 years and the 22%/24% boundary. For you to have a lock on the juicy Roth conversion outcomes discussed above, you would need an AGI, in 2036, RMDs included, of $313,000.

This, from a pair of taxpayers who had been earning somewhere around $110,000 near the peak of their working life …? Really?

Okay, okay—you are actually a retired individual, formerly earning many hundreds of thousands of dollars per year, now with complete control of how much income you take each year, able to (temporarily) depress your income before conversion to $108,750—or even zero. Please forgive me for not focusing on your special case, since you are already able to obtain very lucrative Roth conversion outcomes, simply by depressing your pre-conversion income to zero.

Returning to the main case, the taxpayer couple whose ongoing, recurring income is low enough to have a substantial amount remaining in the 22% bracket, available for conversion: how might they get to an RMD-included income of $313,000, fifteen years out?

-at the low end of the 22% bracket, we cannot give them the maximum social security benefit; their income of $108,750, split equally, puts each at 38% of the maximum income subject to social security tax. Fine; we’ll give them 50% of the max payment at full retirement age in 2021, and project that forward at the inflation rate. I make their collective 2036 SS payment to be about $49,500 (=$3113 monthly max today, x 12 x 2, inflated at 3% for 15 years = $98,939, x 50%).

-the rest of the $313,000 , or $263,500, comes from RMDs, I suppose. That indicates a TDA balance at age 72 of just over $7.2 million dollars. Tell me, how did a couple with peak wages of $108,750, after decades of working and saving, manage to accumulate $7.2 million by their early 70s?

Oh, that’s right—it did not all come from RMDs. They had pension income twice their social security income, an extra $100,000 in 2036. Which means they only have to project $4.5 million in their TDAs to aspire to those juicy Roth conversion outcomes.

How did they get to $4.5 million? Of course :oops: Emerging market small cap value funds, combined with frontier market disruptive innovators and deft movement in and out of 3X leveraged funds.

Please forgive the gentle sarcasm: there actually will be some investors whose wages were modest but whose investing acumen was large, such that they will have $4 to 5 million in their TDAs by age 72, despite never earning much more than $100K as a couple. And I’m happy to accept that more of them will be found here at BH.org than anywhere else on the web. My point: if your personal financial projections do not show such a multi-million dollar TDA balance, then it is unlikely that your conversion today at 22% is going to receive the juicy Roth conversion outcomes to be shown subsequently.

And if you can’t project even $1 million dollars in TDA balance at age 72, in the absence of a conversion; and you don’t have a sweet pension deal, of the kind private sector employees don’t get anymore; then you are at risk of doing a Bob-and-Barb: converting at 22% to save tax at 12%. And that isn’t very pretty.

I go on like this because I find fearmongering in support of Roth conversions to be rampant. Please don’t convert at 22% if you haven’t run the numbers, as described here, to confirm your future projected tax situation. You can compute an inflation index for the desired number of years to project tax and IRMAA brackets; can project your social security based on your wage history; and you can project your TDA balance using an assumed investment return; plus, you can vary at least the latter to do sensitivity testing.

Before converting at 22% today, prove to your satisfaction that you will indeed have age 72 income, including RMDs, greater than the inflation-indexed value of $108,750 today.

A later post will show the spreadsheet for juicy outcomes versus a more humdrum scenario (convert at 22% to avoid tax on RMDs at 22%). To do that it helps to have a metric, which is the subject of the next post.

Finally, in this second series, I will be looking at younger ages, evaluating outcomes at the perhaps more believable ages of 85 and 90, rather than an age of 100+, as in the first series.

A Roth conversion that has not paid off significantly by age 90 will probably be disappointing in the eyes of many.
You can take the academic out of the classroom by retirement, but you can't ever take the classroom out of his tone, style, and manner of approach.
sc9182
Posts: 2165
Joined: Wed Aug 17, 2016 7:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by sc9182 »

McQ wrote: Wed Sep 29, 2021 6:00 pm Part II

At this point in the thread a second series of posts begins. The first series had the key word “always.” The second series picks up on the sub-subtitle to the thread: “but not by very much and seldom very soon.” And the theme of today’s post might be glossed as, “yes, it is theoretically possible to receive a speedy and lucrative outcome from a Roth conversion, but it may not be likely in your case, if you convert in the middle brackets.”

All the caveats laid out in the initial post continue to apply, of course: if you can convert at 0%, 10%, 12% today, to save tax of 37% or 39.6% later—go for it! Huge benefit from converting in that situation. However, such halcyon circumstances will not be the focus of this post or the next few.

The stipulation here:
-you might convert at 22% today to save 25% later, post-TCJA; or,
-that you convert at 22% today, in hopes to save 28% later (because the post-TCJA tax brackets turn what would have been a 24% rate, in the next bracket up, back into the pre-TCJA rate of 28%); or,
-you convert at 22% today to save both 28% and IRMAA charges later; or,
-just maybe, the RMDs to be reduced would have fallen into the upper part of the social security tax torpedo, carrying a marginal rate of almost 41% (46% post-TCJA).

These are all plausible “best cases” for a relatively early and relatively lucrative payoff from a Roth conversion made in the middle brackets. But I’ll also consider a constant 22% rate as a benchmark. I showed in the first series that a conversion under constant rates pays off eventually; here in the second series the focus will be on how to calibrate that payoff, to compare how much better off one would be under one of the “best plausible cases” just introduced.

At first glance, these conditions should still apply to a large number of BH; I’ve certainly heard, on a variety of threads, that whether or not to convert at 22% is a real issue up for debate. But you need to ask yourself: is your 22% conversion really going to avert RMDs at 28% or more? Or are you cruising to be Bob and Barb, from two posts back, converting at 22% today to save … 12% on your RMDs later?

Now you see the stakes.

Straight talk about future tax brackets

How could you tell whether you were hurtling down the track to pay only 12% on your RMDs, or instead, are at a real risk of paying 28% or more? The one would discourage a Roth conversion in the 22% bracket, the other might strongly encourage one.

Green eyeshade time: the AGI ceiling for the 22% bracket today is $200,450 if 65 or over (IRS stated bracket plus standard deductions). The bottom / top of the first IRMAA bracket is $176,000 / $222,000 AGI. Note how the 22%/24% threshold divides the first IRMAA bracket almost exactly in half.
And last: the floor of the 22% bracket in AGI terms is $108,750 (I’ll ignore the inconvenient detail that the 22% bracket is less than $100,000 wide. Bear with me as I continue to enter $100,000 as the conversion amount.)

How to figure marginal rates with IRMAA

The full IRMAA charge kicks in on the first dollar of excess income; hence, the dollar of income that takes you from $176,000 to $176,001 has a marginal rate in the thousands of percent. Savvy taxpayers who stray into IRMAA territory know to voluntarily accelerate income up to the top of the IRMAA bracket into which they strayed. For the first three IRMAA brackets, the quotient [IRMAA dollars / IRMAA bracket span] will equal 4%, plus or minus (see SSRN paper appendix). So that’s one way to estimate the “combined” marginal rate: income tax rate + 4% IRMAA rate.

But there’s a problem with the first IRMAA bracket; as noted above, it straddles the 22% / 24% income tax threshold. The combined income tax plus IRMAA marginal rate changes at $200,451. Here is a table to organize the numbers, assuming taxpayers over 65 filing as married joint here in 2021.

Code: Select all

Income range -----> combined tax rate
$108,750 - $176,000 22%
$176,001 - $204,850 30% (22% + [IRMAA dollars / 0.5 IRMAA bracket span], or 30% in all, at the 22% ceiling)
$204,851 - $222,000 32%, declining toward 28% as IRMAA ceiling is approached
$222,001 - $330,000 28% at the top of IRMAA bracket #2 and bracket #3, higher if brackets are not tapped all the way.
These 2021 rates are TCJA rates; if the planner projects the expiration of TCJA, then the corresponding values are

Code: Select all

22% ----->  25%
30% ----->  33%
32/28% --> 36%/32%
28% ----->  32%
To sum up, there is indeed a range of income likely to be taxed, post-TCJA, at combined rates around 32%. To convert at 22% to save 32% certainly sounds attractive. But are you, for lack of a better word, eligible for this favorable outcome? Put another way, is it legitimate for you, with your very particular income and assets, to fear tax rates on your projected RMDs at rates on the order of 32%? Or have you been the victim of fearmongering, of the sort commonly seen in financial journalism, promotional missives from financial advisors, and (sometimes) social media?

Are you eligible / should you be afraid?

All the threshold values just given are for 2021. Tax brackets and IRMAA brackets adjust for inflation each year; as in the spreadsheet, I will assume an inflation rate of 3%.

Next, let’s assume that you, the prospective Roth converter, are in your early 60s, or maybe, mid-50s. Your RMDs are not going to begin for 10, 15, or 20 years yet. Whether you will be eligible for that very favorable Roth conversion outcome, where you convert today at 22% to avert subsequent RMDs taxed at scary high rates like 32%, is a function of what those tax brackets will be 10, 15, and 20 years in the future.

To keep it simple, let’s focus on $176,000, the 1st IRMAA threshold, and $204,850, the 24% (or maybe 28% after 2025) bracket floor. Here are the inflation-adjusted values, rounded to the nearest $250.

Code: Select all

Years out	IRMAA #1	22% ceiling
10	       $236,500 	$270,000 
15	       $274,250 	$313,000 
20	       $318,000 	$362,750 
For discussion purposes, I’ll take the middle time frame of 15 years and the 22%/24% boundary. For you to have a lock on the juicy Roth conversion outcomes discussed above, you would need an AGI, in 2036, RMDs included, of $313,000.

This, from a pair of taxpayers who had been earning somewhere around $110,000 near the peak of their working life …? Really?

Okay, okay—you are actually a retired individual, formerly earning many hundreds of thousands of dollars per year, now with complete control of how much income you take each year, able to (temporarily) depress your income before conversion to $108,750—or even zero. Please forgive me for not focusing on your special case, since you are already able to obtain very lucrative Roth conversion outcomes, simply by depressing your pre-conversion income to zero.

Returning to the main case, the taxpayer couple whose ongoing, recurring income is low enough to have a substantial amount remaining in the 22% bracket, available for conversion: how might they get to an RMD-included income of $313,000, fifteen years out?

-at the low end of the 22% bracket, we cannot give them the maximum social security benefit; their income of $108,750, split equally, puts each at 38% of the maximum income subject to social security tax. Fine; we’ll give them 50% of the max payment at full retirement age in 2021, and project that forward at the inflation rate. I make their collective 2036 SS payment to be about $49,500 (=$3113 monthly max today, x 12 x 2, inflated at 3% for 15 years = $98,939, x 50%).

-the rest of the $313,000 , or $263,500, comes from RMDs, I suppose. That indicates a TDA balance at age 72 of just over $7.2 million dollars. Tell me, how did a couple with peak wages of $108,750, after decades of working and saving, manage to accumulate $7.2 million by their early 70s?

Oh, that’s right—it did not all come from RMDs. They had pension income twice their social security income, an extra $100,000 in 2036. Which means they only have to project $4.5 million in their TDAs to aspire to those juicy Roth conversion outcomes.

How did they get to $4.5 million? Of course :oops: Emerging market small cap value funds, combined with frontier market disruptive innovators and deft movement in and out of 3X leveraged funds.

Please forgive the gentle sarcasm: there actually will be some investors whose wages were modest but whose investing acumen was large, such that they will have $4 to 5 million in their TDAs by age 72, despite never earning much more than $100K as a couple. And I’m happy to accept that more of them will be found here at BH.org than anywhere else on the web. My point: if your personal financial projections do not show such a multi-million dollar TDA balance, then it is unlikely that your conversion today at 22% is going to receive the juicy Roth conversion outcomes to be shown subsequently.

And if you can’t project even $1 million dollars in TDA balance at age 72, in the absence of a conversion; and you don’t have a sweet pension deal, of the kind private sector employees don’t get anymore; then you are at risk of doing a Bob-and-Barb: converting at 22% to save tax at 12%. And that isn’t very pretty.

I go on like this because I find fearmongering in support of Roth conversions to be rampant. Please don’t convert at 22% if you haven’t run the numbers, as described here, to confirm your future projected tax situation. You can compute an inflation index for the desired number of years to project tax and IRMAA brackets; can project your social security based on your wage history; and you can project your TDA balance using an assumed investment return; plus, you can vary at least the latter to do sensitivity testing.

Before converting at 22% today, prove to your satisfaction that you will indeed have age 72 income, including RMDs, greater than the inflation-indexed value of $108,750 today.

A later post will show the spreadsheet for juicy outcomes versus a more humdrum scenario (convert at 22% to avoid tax on RMDs at 22%). To do that it helps to have a metric, which is the subject of the next post.

Finally, in this second series, I will be looking at younger ages, evaluating outcomes at the perhaps more believable ages of 85 and 90, rather than an age of 100+, as in the first series.

A Roth conversion that has not paid off significantly by age 90 will probably be disappointing in the eyes of many.
Thanks Prof. McQ

There are sufficient number of folks on BH - who get these points - and there are a few swear by it that their situation befits sure-shot Roth-conversion payout (to beneficiaries).

Unfortunately, current set of tooling is still figuring out intricacies of Roth conversions — and thanks to you and other BH’er who are cross-checking math against a few other tools/calculators. This conversion knowledge is continue to evolve., thanks to you for hashing out Roth conversions scenarios very well !!

Much appreciate it !
2pedals
Posts: 1988
Joined: Wed Dec 31, 2014 11:31 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by 2pedals »

^ Great comments above

I would like to add some bogleheads who are married in the early retirement plan to pass all or most assets including a significant pension to the surviving spouse and are trying to get conversions done while they are still able to use married tax tables. This can mitigate the risk of some higher possible tax rates for several years if one spouse is deceased early in retirement. On top of that, they may have single heirs and high-income brackets.
RetiredAL
Posts: 3513
Joined: Tue Jun 06, 2017 12:09 am
Location: SF Bay Area

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by RetiredAL »

2pedals wrote: Wed Sep 29, 2021 9:46 pm ^ Great comments above

I would like to add some bogleheads who are married in the early retirement plan to pass all or most assets including a significant pension to the surviving spouse and are trying to get conversions done while they are still able to use married tax tables. This can mitigate the risk of some higher possible tax rates for several years if one spouse is deceased early in retirement. On top of that, they may have single heirs and high-income brackets.
I am in that group, concerned about both the Single Tax Bracket impacts AND IRMAA up-charges.

When my Mom passed in Feb 2013, the income loss was relatively minor, but the next year Dad got a largely increased tax bill as he was thrust much higher, percent of his income-wise, into the 24% tax bracket. And the year after that, by IRMAA.

This has driven me to model how DW or I, as the last person standing, will be taxed in the aggregate. Unfortunately, there is no simple cut-n-dried solution to the question of convert or not, or convert how much.

I take comfort in understanding that even if I pick a poorer conversion path for converting part of my differed I plan on converting, the result is likely to be no worse than not having converted.
sc9182
Posts: 2165
Joined: Wed Aug 17, 2016 7:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by sc9182 »

RetiredAL wrote: Wed Sep 29, 2021 10:24 pm
2pedals wrote: Wed Sep 29, 2021 9:46 pm ^ Great comments above

I would like to add some bogleheads who are married in the early retirement plan to pass all or most assets including a significant pension to the surviving spouse and are trying to get conversions done while they are still able to use married tax tables. This can mitigate the risk of some higher possible tax rates for several years if one spouse is deceased early in retirement. On top of that, they may have single heirs and high-income brackets.
I am in that group, concerned about both the Single Tax Bracket impacts AND IRMAA up-charges.

When my Mom passed in Feb 2013, the income loss was relatively minor, but the next year Dad got a largely increased tax bill as he was thrust much higher, percent of his income-wise, into the 24% tax bracket. And the year after that, by IRMAA.

This has driven me to model how DW or I, as the last person standing, will be taxed in the aggregate. Unfortunately, there is no simple cut-n-dried solution to the question of convert or not, or convert how much.

I take comfort in understanding that even if I pick a poorer conversion path for converting part of my differed I plan on converting, the result is likely to be no worse than not having converted.
Sorry for Mom’s loss.

Has your Pops received any life-insurance? Or did he get any Step-Up basis on Brokerge/LTCG assets ? Has his expenses and/or medical costs have decreased somewhat or they also remain steady (Vs. Pre-2013). Over the last few years — HSAs (triple/Quadruple tax advantaged) come to fruition, most likely your Pops generation didn’t have access to (Also doubt Pre-2012 IRS clarity — how much 100k+ Roth conversion opportunity your Pops under MFJ had .. just saying)

Yes, you want to reduce taxes paid, but at the end of the day have to look at Net-Monies in hand (and/or Net Average monies/person)

What are likelyhood of an asteroid striking human, twice - so how do you prepare for such doomsday !? Does your family has tradition of Pops significantly out-living Moms (or marrying much older wife/spouse). Most of the life expectancy tables say/show — female spouses die later (sorry don’t share us one-off/two-off evidence, we are talking generalization here)

Yes. Planning is important - but fear not. Many a planning opportunities exist while MFJ, and/or when Widow(oed) — such as Geographic Tax arbitrage, possible insurance, possible higher SS of the two, and/or possible beneficial tax-filing status changes down the line. Or worse yet, if large nursing-home etc medical expenses come due - then Trad IRA monies may come in lot handy than Roths (even overcoming MFJ or single IRMAA 1st or 2nd tier tax impact) !!
Last edited by sc9182 on Wed Sep 29, 2021 11:10 pm, edited 5 times in total.
marcopolo
Posts: 8411
Joined: Sat Dec 03, 2016 9:22 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by marcopolo »

RetiredAL wrote: Wed Sep 29, 2021 10:24 pm
2pedals wrote: Wed Sep 29, 2021 9:46 pm ^ Great comments above

I would like to add some bogleheads who are married in the early retirement plan to pass all or most assets including a significant pension to the surviving spouse and are trying to get conversions done while they are still able to use married tax tables. This can mitigate the risk of some higher possible tax rates for several years if one spouse is deceased early in retirement. On top of that, they may have single heirs and high-income brackets.
I am in that group, concerned about both the Single Tax Bracket impacts AND IRMAA up-charges.

When my Mom passed in Feb 2013, the income loss was relatively minor, but the next year Dad got a largely increased tax bill as he was thrust much higher, percent of his income-wise, into the 24% tax bracket. And the year after that, by IRMAA.

This has driven me to model how DW or I, as the last person standing, will be taxed in the aggregate. Unfortunately, there is no simple cut-n-dried solution to the question of convert or not, or convert how much.

I take comfort in understanding that even if I pick a poorer conversion path for converting part of my differed I plan on converting, the result is likely to be no worse than not having converted.
This is only true if the portfolio continues to grow.
Not to take this thread off-topic, but this analysis is mostly for wealthy people, or those with large protected income, who will not need their portfolio to live on, and are mostly concerned about maximizing after tax inheritance to heirs.

If one has to live off their portfolio, and the goal is to ensure they have sufficient funds to support their spending needs, they should at least consider the assymetric nature of the risks of converting vs. not converting.

Not converting carries the risk that if the portfolio does well, one ends up paying some additional taxes, and have less money to spend, but they probably don't need the extra money to support their spending because markets have done so well.

Converting carries the risk that the person has less money to support their spending when markets perform poorly, just when they need the extra money the most, simply to support their spending.

One needs to decide which risk is most important to them to mitigate.

On a forum where people are always worried about poor future market performance (See all the "2% SWR is too risky", or "Valuations and low bond yields doom future return" type threads), I am surprised there is so little consideration about how such concerns might inform the decision to convert or not.
Once in a while you get shown the light, in the strangest of places if you look at it right.
Big Dog
Posts: 4588
Joined: Mon Sep 07, 2015 4:12 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by Big Dog »

better get those conversions in process, as they may not be around much longer.

Note, not trying to discuss pending legislation, but just alerting members to current activity which they should follow on their own.... Hopefully this PSA doesn't violate ToS.

https://www.marketwatch.com/story/congr ... 1632861718
User avatar
FiveK
Posts: 15691
Joined: Sun Mar 16, 2014 2:43 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by FiveK »

Big Dog wrote: Wed Sep 29, 2021 11:02 pm better get those conversions in process, as they may not be around much longer.

Note, not trying to discuss pending legislation, but just alerting members to current activity which they should follow on their own.... Hopefully this PSA doesn't violate ToS.

https://www.marketwatch.com/story/congr ... 1632861718
Note that the conversions being discussed in this thread would be unaffected because that article discusses a proposal that "...prohibits after-tax IRA contributions from being converted to Roth regardless of income level...." We can return to our regular thread discussion already in progress.
RetiredAL
Posts: 3513
Joined: Tue Jun 06, 2017 12:09 am
Location: SF Bay Area

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by RetiredAL »

sc9182 wrote: Wed Sep 29, 2021 10:44 pm
RetiredAL wrote: Wed Sep 29, 2021 10:24 pm
2pedals wrote: Wed Sep 29, 2021 9:46 pm ^ Great comments above

I would like to add some bogleheads who are married in the early retirement plan to pass all or most assets including a significant pension to the surviving spouse and are trying to get conversions done while they are still able to use married tax tables. This can mitigate the risk of some higher possible tax rates for several years if one spouse is deceased early in retirement. On top of that, they may have single heirs and high-income brackets.
I am in that group, concerned about both the Single Tax Bracket impacts AND IRMAA up-charges.

When my Mom passed in Feb 2013, the income loss was relatively minor, but the next year Dad got a largely increased tax bill as he was thrust much higher, percent of his income-wise, into the 24% tax bracket. And the year after that, by IRMAA.

This has driven me to model how DW or I, as the last person standing, will be taxed in the aggregate. Unfortunately, there is no simple cut-n-dried solution to the question of convert or not, or convert how much.

I take comfort in understanding that even if I pick a poorer conversion path for converting part of my differed I plan on converting, the result is likely to be no worse than not having converted.
Sorry for Mom’s loss.

Has your Pops received any life-insurance? Or did he get any Step-Up basis on Brokerge/LTCG assets ? Has his expenses and/or medical costs have decreased somewhat or they also remain steady (Vs. Pre-2013). Over the last few years — HSAs (triple/Quadruple tax advantaged) come to fruition, most likely your Pops generation didn’t have access to (Also doubt Pre-2012 IRS clarity — how much 100k+ Roth conversion opportunity your Pops under MFJ had .. just saying)

Yes, you want to reduce taxes paid, but at the end of the day have to look at Net-Monies in hand (and/or Net Average monies/person)

Yes. Planning is important - but fear not. Many a planning opportunities exist while MFJ, and/or when Widow(oed) — such as Geographic Tax arbitrage, possible insurance, possible higher SS of the two, and/or possible beneficial tax-filing status changes down the line. Or worse yet, if large nursing-home etc medical expenses come due - then Trad IRA monies may come in lot handy than Roths (even overcoming IRMAA 1st or 2nd tier tax impact) !!
Dad will be 97 in Nov. He is now very frail and in Assisted Living.

Yes, there was step-up re-valuation for his home and a vacation home. He still has both. The step-up to the investment accounts was just somewhat. They had a large amount invested with and advisor, the turnover was significant, so no really long term gains existed to get the step-up. A separate non-managed account had some pretty good gains but those are still held today. All these assets are held in a Revocable Trust. He is a Calif resident. His RMD (His and Mom's IRA's combined) was slightly north of $20K last year. I manage all of his affairs.

Dad's out-of-pocket expenses, other than the increase tax bills, has been pretty steady since Mom passed and similar to before she passed. He was career Navy and retired 0-4 with thirty years. That retirement plus SS plus RMD means he has pretty substantial income, driving significantly higher tax bill and IRMAA up-charges. Having removed the advisor, his holdings today are stagnant, so the turnover cap-gain income has dramatically dropped, getting him down to IRMAA-2. When Mom was alive, they were at IRMAA-Base.

His ALF charges are mostly being paid by his LTCI. He has about 2 year of LTCI coverage left. Him running out of money is extremely unlikely.

Yes, having IRA monies is important. If he survives longer that his LTCI coverage, the IRA $ will be the next monies spent.

Although I am in the convert camp, it's conversion in moderation. In my case, <12% has/will-be converted.

I am lucky to be able to see and actually practice the estate planning tools that most only get to talk about or plan for. My sister and I don't desire to receive more differed tax monies from his estate, so the emphasis is towards LTCG and step-up basis. If it turns out either the differed IRA or the LTCG needs to be tapped to pay his care bills, his care tax deduction will largely nullify his tax bills.
dknightd
Posts: 3715
Joined: Wed Mar 07, 2018 10:57 am

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by dknightd »

McQ wrote: Wed Sep 29, 2021 6:00 pm Part II

Before converting at 22% today, prove to your satisfaction that you will indeed have age 72 income, including RMDs, greater than the inflation-indexed value of $108,750 today.
I'm in this boat. Trying to figure out what my RMD will be, and what tax rates might be. Since I do not know either of those with any certainly, for now I'm not converting at 22% this year. I might change my mind next year, or a few years from now . . .

If it looks likely that RMD will put me in 22% bracket, I might convert at 22% to avoid 25%, we'll see.

My thinking is, if we find ourselves not spending RMD, we'll obviously have to take it anyway, but, we might use some of it to pay taxes on a Roth conversion. This would reduce next years RMD. It would also reduce "tax drag" on after tax account. But that is years away. I'll probably keep making this decision on a year by year basis.
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 ;)
wrongfunds
Posts: 3187
Joined: Tue Dec 21, 2010 2:55 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by wrongfunds »

How does step-up work on a joint/wros/tod type of accounts or home titled in both the names? I thought step up only comes in to picture when both owners pass away.
RetiredAL
Posts: 3513
Joined: Tue Jun 06, 2017 12:09 am
Location: SF Bay Area

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by RetiredAL »

wrongfunds wrote: Thu Sep 30, 2021 8:38 am How does step-up work on a joint/wros/tod type of accounts or home titled in both the names? I thought step up only comes in to picture when both owners pass away.
Wrongfunds -- It depends on if the state is "community property" or not.

Calif is a community property state, so there is full step-up on the passing of one. (simple explanation)

No direct experience, but as I understand it, in a non-community property state, a "joint titled asset" is stepped up 50% on the passing of one.
wrongfunds
Posts: 3187
Joined: Tue Dec 21, 2010 2:55 pm

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by wrongfunds »

RetiredAL wrote: Thu Sep 30, 2021 9:26 am
wrongfunds wrote: Thu Sep 30, 2021 8:38 am How does step-up work on a joint/wros/tod type of accounts or home titled in both the names? I thought step up only comes in to picture when both owners pass away.
Wrongfunds -- It depends on if the state is "community property" or not.

Calif is a community property state, so there is full step-up on the passing of one. (simple explanation)

No direct experience, but as I understand it, in a non-community property state, a "joint titled asset" is stepped up 50% on the passing of one.
Is Massachusetts "community property"? Also does it matter if we are talking about married couple?
RetiredAL
Posts: 3513
Joined: Tue Jun 06, 2017 12:09 am
Location: SF Bay Area

Re: Why Roth conversions always pay off—if you can hold on long enough

Post by RetiredAL »

My thanks to McQ and several other posters who have put a lot of thought and time into this conversion question.
Post Reply