Do you Roth much?

[Ron Scott]

I Roth.

Retired, many years to go before RMD tax depression. Currently living primarily on interest and dividends from a 6040 portfolio. When I do need extra $$ I take it from a staunch of employer stock I never got rid of, to diversify. Between that and the coming forced income from RMDs I can’t actually see the need for a big Roth when a small one will suffice for emergencies.

My accountant tells me I do Roths for my daughter, not for me. He enjoys saying this!

But I’m conservative and I’m going to do another one again now to fill up a low bracket.

What’s your philosophy on this, for yourselves?

[flyingaway]

It depends largely on how large is your tax-advantaged portfolio and on the need for ACA subsidies, and on other things at a small scale.

[reeshau]

I live primarily on LTCG's and qualified dividends, so base tax rate is 0%.  I do Roths as part of tax planning, to soak up non-refundable credits that would otherwise go to waste.  (DS is 8)  Tax software is out Nov. 4, so it's just about time to start the next cycle.  My goal is to write a check < $100.

I doubt this will be enough to totally escape RMD's, but it has been about $40k per year.  (this will be year #3)

[2sk22]

The main reason to do Roth conversions is if your RMDs will force you to withdraw far more than you need.

[secondcor521]

Each year, I Roth convert up to close to the maximum marginal rate I expect to pay between now and my 50% life expectancy age.

I do a pro forma return in December using the Case Study Spreadsheet to see where I stand, then adjust from there in late December.

I have started Roth converting some at the start of the year and some more if the market is down somewhat, then top off in December.

I view it mostly as tax rate arbitrage, and I probably save 15% of the amount converted in federal taxes, so I think it's worth it.

With three kids and a legacy mindset, at some point the impact on their tax returns for the 10 years after I die will become more of a practical concern and less of a theoretical one.  Right now, RMDs(*) from inheriting an IRA could push the adult kid beneficiaries into high brackets in their later working years if it doesn't enable them to retire early themselves.

(*) IRS proposed regulation.

[seattlecyclone]

I contributed a bunch to Roth when I worked at a place with a mega backdoor option. Also most of my IRA contributions have been Roth because I've rarely earned little enough to make tax-deferred IRA contributions. Beyond that, in a lower-income year I'll sometimes take advantage of a low bracket to convert some pre-tax savings and lock in a low tax rate.

Regarding concern over RMDs, I've said it before and I'll say it again: the RMD formula is basically designed to get you on a path drawing down toward zero during your remaining life expectancy. If this amount is significantly less than you were planning to withdraw and spend anyway, congratulations! You have won the wealth accumulation game and have saved more money than you have any use for. This goes double if a big chunk of your money is in Roth or taxable accounts that have no RMD requirements to start with. If just the RMD-able assets are more than you wanted to spend, you're at the next level of winning. One nice thing for such winners is that the tax code allows for you to divert up to $100k/year of this surplus wealth to charity tax-free once you turn 70½. This is a complete exclusion from income that doesn't even touch your AGI to get deducted later, so it doesn't affect the taxability of your social security, it doesn't get you on IRMAA, none of that.

[Loren Ver]

DH and I both have little in our Roths, less than 30K each, so we can take money out of the principle in an emergency fund situation, if needed.  This is actually one of our backup plans if things go plaid.

As for converting from our 401ks (403bs, IRA, or similar) to Roth to avoid future RMD.  No, we haven't done anything like that.  We are on the ACA so keeping our income low, like really low, is the goal so we don't generate any additional income.  Also, our 401ks etc don't make up that much of our overall portfolios, about 26% so when RMD come to pass I'm not that concerned about the amount we will be required to take out.  We should be able to just shift   what we take out from non-taxed advantaged accounts to taxed advantaged accounts when the time comes.

Also, just because I have to take money out of a 401k because of a RMD doesn't mean I need to spend it.  It would have to be a LOT to get me out of the 0% tax bracket.  Then I just reinvest the portion I don't want to use that year in a taxable account. 

@seattlecyclone That is some great information about donations!  If it doesn't affect AGI and therefore maybe not MAGI (??) then that is another option.

When your MAGI is generally between 28k-32k there just isn't a lot of conversion space when you also intend to live off the income.

Loren

[Mr. Green]

Roth conversions are the lifeblood of our financial plan over the next 20 years, barring income from unexpected sources. We didn't even have Roth IRAs when we were in our 20s because we didn't know the value of them. Our stash is so heavily weighted in traditional IRAs that we'll have to convert more than we need just to keep RMDs from being insane when we're in our 80s and 90s. Maybe we'll be giving away gobs of money by then.

[Purple_Crayon]

The vast majority of my stash in ROTH. I spent the majority of my working years not making a ton of money, so figured I'd pay the taxes now. However, as mentioned above, this became an issue for health insurance. In my state, you cannot purchase from the ACA exchange until you make over 148% of the FPL -- you can only be placed on Medicaid. So, depending on where you live, being too heavily allocated to ROTH can take away options.

I'm maxing 401 TRAD now until FIRE to build up some flexibility.

[jimmyshutter]

I'm not post-fire but my intentions are to ROTH early until I enter higher tax brackets. I understand the point of "do it for your kids" but recently I've given this a second thought because my son needs the money now so I've been sending him more than I used to. It's a nice feeling to help him so why not enjoy it now rather than wait until I'm almost dead?

[Car Jack]

I retired the middle of this year.  So for the first time, I've taken a deep dive into Medicare and found the dreaded IRMAA.  I made enough 2 years ago to push me 2 clicks into IRMAA and with an appeal to recalculate based on only half the income because I quit, they just (5 months, 2 phone calls and a personal visit) approved.  If you haven't looked into this, for me (MFJ), it doubled my Medicare premiums.  Moving forward, I am Roth converting to keep my total income just under the first IRMAA limit.  To not do that, RMDs would be enormous as almost all of our savings came from 401k's and we very much over saved. 

[flyingaway]

DH and I both have little in our Roths, less than 30K each, so we can take money out of the principle in an emergency fund situation, if needed.  This is actually one of our backup plans if things go plaid.

As for converting from our 401ks (403bs, IRA, or similar) to Roth to avoid future RMD.  No, we haven't done anything like that.  We are on the ACA so keeping our income low, like really low, is the goal so we don't generate any additional income.  Also, our 401ks etc don't make up that much of our overall portfolios, about 26% so when RMD come to pass I'm not that concerned about the amount we will be required to take out.  We should be able to just shift   what we take out from non-taxed advantaged accounts to taxed advantaged accounts when the time comes.

Also, just because I have to take money out of a 401k because of a RMD doesn't mean I need to spend it.  It would have to be a LOT to get me out of the 0% tax bracket.  Then I just reinvest the portion I don't want to use that year in a taxable account. 

@seattlecyclone That is some great information about donations!  If it doesn't affect AGI and therefore maybe not MAGI (??) then that is another option.

When your MAGI is generally between 28k-32k there just isn't a lot of conversion space when you also intend to live off the income.

Loren

Does that mean almost zero cost for ACA?

[achvfi]

Not post-fire and still in accumulating stage. Our not so exact goal is to reach 1/3 in pretax, 1/3 in Roth, 1/3 in Brokerage in investments over time. There is no clear logic to this but idea is to have options in any scenario. Scenarios such as Layoff, early retirement, Sabbatical, inheritance RMDs e.t.c..

For first 10 years of investment journey my contributions were mostly pre-tax, except for annual maxing out the Roth IRA. Remaining savings went into taxable brokerage account and savings accounts.
 
After finding about Mega Back Door Roth, I stopped brokerage contributions and started filling up MBR.

Now the question is when to start adding to Brokerage accounts again as I don't have enough to max MBR and add significantly to brokerage account. Should I stop or reduce MBR at some point? Or just contribute until MBR is available.
I don't know that there is a clear answer on that.

Once I start early retirement, my plan is to fund lifestyle + Roth conversions with pre-tax funds for that I will need savings outside of qualified accounts to pay taxes. If any of you have experience doing this?

I don't know man. I think it is going to be mentally hard to start decumulating. When you have been putting away for so long I am not sure if we have the muscle to start using some of this money.

[Ron Scott]

Once I start early retirement, my plan is to fund lifestyle + Roth conversions with pre-tax funds for that I will need savings outside of qualified accounts to pay taxes. If any of you have experience doing this?

I don’t know how old you are, but if I had 15-20 years before RMD’s, I would not be looking at Roth. That law is about 25 years old and has already undergone changes. Depending on your individual circumstances, future changes could make today’s conversions good or bad. So you have to ask yourself: Do I feel lucky?

Funding the conversion? Taxes are an expense, even if you manufacture your own. Fund the taxes on your Roth like you would your daily bread.

I don't know man. I think it is going to be mentally hard to start decumulating. When you have been putting away for so long I am not sure if we have the muscle to start using some of this money.

Relax. It is quite possible that you will not begin decumulating (seeing a nominal reduction in your invested assets) for many years. You will have time to get used to the idea.

Also, if you are following the general approach taken by many on this forum in which “retirement” doesn’t actually mean retirement, and you go back to work for a while, take part-time work, “side hustles“, etc.—and you do not increase your spending much—the actual decumulation phase may keep being pushed back.

Of course, all this is contingent on the performance of stock and bond markets in the future, which is not predictable so…

[seattlecyclone]

@Ron Scott do you have any plans for the future at all? Seems like it would be very hard to make any with all that pessimism. Better not book any plane tickets, after all you might drop dead before then.

[ender]

We're putting in about the same Roth vs traditional at this point. With about the total into taxable too.

But because our existing balances are about 2/3 traditional, we're never going to catch up with Roth proportionally.

In 2024, we're contributing roughly:

• ~$4k HSA
• $23k 401k pretax
• $13.8k match
• $32.2k megabackdoor 401k
• $7k backdoor roth IRA
• everything else taxable

So for tax advantaged, we're going to end up with roughly $40k of both pretax/Roth each. With a fair bit more than $40k in taxable as well too.

The primary tax change which would hurt our planning would be if megabackdoor contributions become ineligible for withdrawing penalty free. Right now we have over $200k worth of Roth IRA/megabackdoor contributions we can withdraw. By the time I'm FI that number will be considerably higher because each year I'm adding almost $40k.

[Sandi_k]

We are almost all pre-tax, and with almost nothing in brokerage accounts. DH had a SEP-IRA and Roth during his working years (now retired, age 60), and I have a Roth.

We've been converting his SEP-IRA to his Roth since 2019, and the conversions have been paid from my paycheck, as I force over-withholding sufficient to cover his conversions.

We've now converted enough that he'll use the remainder in the SEP for living expenses for 2024 and 2025.

My employer added a Roth 457 mid-year this year, so I switched from the MBR to the Roth 457. I will push big chunks to that account for 2023, 2024, and 2025, and then retire.

Even so, my calcs on retirement estimate that we'll have 75% in pre-tax accounts upon retirement, and only 25% in Roth.  That 25% will be helpful for house projects that will need some big lump sums from our savings going forward. It's unlikely that we'd do much converting after I retire in 2025, simply because of the pension and IRMAA intersection. We'll see where the COLA'd brackets are at that point. I could see bunching conversion years, maybe every third year or something.

And at some point, it's likely that both of us will have inheritances in the six figure range, from parents who own CA homes. Which will give us a big bucket of funds in a brokerage account, and help balance everything out for our old age.

[ca-rn]

I want to Roth some but only have about 130k so far.

My plan is to FIRE sometime in 2025 and access 401k via rule of 55 for some roth conversions.

Have about 700k in 401k (35% of stash) with majority in brokerage account.

My target AA is 70/30-  401k holds bond portion so need a plan to keep 70/30, Roth some while managing ACA cost for 10 years until Medicare.
Rental income (avg 40k+) limits conversion amount to qualify for any ACA assistance (single, no dependents).

Would be ideal if a friend could list me as their domestic partner but not sure how much it would be scrutinized by HR/employer?  There is an affidavit to sign...

[elysianfields]

We just executed a large (~$60k) set of Roth conversions because we have space in 2023 and we likely won't have it in the next three years.  Also, I realized we have a big honking chunk of traditional IRA and TSP (which cannot be converted to Roth while actively working for Uncle Sugar) that have an excellent chance of doubling twice before we hit our MRD age of 75.  The MRDs would vault us into much higher brackets under current law (TCJA expires on December 31, 2025 for personal income taxes).  If one of us passes on before MRDs, filing single would surely push the survivor into the highest bracket under current law.  Rich people problems, I know, and the crystal ball remains murky, so we've followed our current best guess under the circumstances.

[Turtle]

We just executed a large (~$60k) set of Roth conversions because we have space in 2023 and we likely won't have it in the next three years.  Also, I realized we have a big honking chunk of traditional IRA and TSP (which cannot be converted to Roth while actively working for Uncle Sugar) that have an excellent chance of doubling twice before we hit our MRD age of 75.  The MRDs would vault us into much higher brackets under current law (TCJA expires on December 31, 2025 for personal income taxes).  If one of us passes on before MRDs, filing single would surely push the survivor into the highest bracket under current law.  Rich people problems, I know, and the crystal ball remains murky, so we've followed our current best guess under the circumstances.

My late spouse never put anything into Roth.  Thankfully I compensated by utilizing Roth whenever I had an employer who offered it, but the RMD tax bomb will still be there in my 80s.  Initial years between 75-80 won't be completely awful, depending on the market between now and then.  I'm also potentially looking at some charitable contribution from Traditional 401k to help when RMDs hit. Roughly 20% of my 401k/IRA total is Roth, the rest is subject to tax haircut.

The accounts are being split 5 ways for inheritance, so their individual brackets won't be hit as badly as my upper RMD years.  Most of them are likely to be retired or nearly so by that time, so they can use these proceeds to allow their own accounts to continue to grow.

[Mr. Green]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

[ender]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

This doesn't make sense at all.

There's really no reason you'd prefer to convert a smaller number of shares for the same tax obligation (ie dollar value converted).

[Mr. Green]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

This doesn't make sense at all.

There's really no reason you'd prefer to convert a smaller number of shares for the same tax obligation (ie dollar value converted).

This is one of those counterintuitive ideas most folks won't experience but I can assure you it's correct. It's a two fold problem too. Not only is there a very real risk of us running out of tIRA money to convert if we focused on big Roth conversions when the market was down, we can only spend Roth principal so we also limit the amount of money available to us penalty-free prior to age 60. Perhaps if we were older it wouldn't matter but we still have 20 years before early withdrawal penalties go away.

[secondcor521]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

This doesn't make sense at all.

There's really no reason you'd prefer to convert a smaller number of shares for the same tax obligation (ie dollar value converted).

This is one of those counterintuitive ideas most folks won't experience but I can assure you it's correct. It's a two fold problem too. Not only is there a very real risk of us running out of tIRA money to convert if we focused on big Roth conversions when the market was down, we can only spend Roth principal so we also limit the amount of money available to us penalty-free prior to age 60. Perhaps if we were older it wouldn't matter but we still have 20 years before early withdrawal penalties go away.

[Emphasis added.]

You can withdraw and spend Roth conversions tax- and penalty-free prior to age 60 if you wait for each Roth conversion to remain in the Roth for five tax years.  This is the basis of the Roth conversion ladder aka Roth pipeline, something I'm sure you're already familiar with.

So since I think you know that, I'm not sure how that squares with what I bolded in your statement above.  If you entirely emptied your traditional IRA by making Roth conversions while young, you'd still be able to spend all of your Roth contributions and all of the conversions on a 5 year delay.  The only part of your Roth that would have to wait until 59.5 would be any Roth earnings (which would be Roth balance - Roth contributions - Roth conversions).

What am I missing?

[Mr. Green]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

This doesn't make sense at all.

There's really no reason you'd prefer to convert a smaller number of shares for the same tax obligation (ie dollar value converted).

This is one of those counterintuitive ideas most folks won't experience but I can assure you it's correct. It's a two fold problem too. Not only is there a very real risk of us running out of tIRA money to convert if we focused on big Roth conversions when the market was down, we can only spend Roth principal so we also limit the amount of money available to us penalty-free prior to age 60. Perhaps if we were older it wouldn't matter but we still have 20 years before early withdrawal penalties go away.

[Emphasis added.]

You can withdraw and spend Roth conversions tax- and penalty-free prior to age 60 if you wait for each Roth conversion to remain in the Roth for five tax years.  This is the basis of the Roth conversion ladder aka Roth pipeline, something I'm sure you're already familiar with.

So since I think you know that, I'm not sure how that squares with what I bolded in your statement above.  If you entirely emptied your traditional IRA by making Roth conversions while young, you'd still be able to spend all of your Roth contributions and all of the conversions on a 5 year delay.  The only part of your Roth that would have to wait until 59.5 would be any Roth earnings (which would be Roth balance - Roth contributions - Roth conversions).

What am I missing?

By principal I really meant basis, whether contribution or conversion. Poor choice of words.

If we convert every penny of tIRA money to Roth prior to age 55 and spend it as quickly as it's converted the last years of our 50s get a little complicated. Plus we royally fucked up our lifetime tax curve and probably paid way more in taxes than we needed to. That's a possible scenario if we focus on big conversions in down markets over the next 20 years.

I've modeled it both ways and emphasizing Roth conversions in down markets lowers the ceiling of spending prior to age 60, as it should if one thinks about it simply. By emphasizing tax-tree gains in the Roth, one also emphasizes the availability of their money after age 60. That is not a desireable characteristic to me.

If we had FIREd closer to age 60 there'd be no concern. I'd focus on converting in down markets.

[secondcor521]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

This doesn't make sense at all.

There's really no reason you'd prefer to convert a smaller number of shares for the same tax obligation (ie dollar value converted).

This is one of those counterintuitive ideas most folks won't experience but I can assure you it's correct. It's a two fold problem too. Not only is there a very real risk of us running out of tIRA money to convert if we focused on big Roth conversions when the market was down, we can only spend Roth principal so we also limit the amount of money available to us penalty-free prior to age 60. Perhaps if we were older it wouldn't matter but we still have 20 years before early withdrawal penalties go away.

[Emphasis added.]

You can withdraw and spend Roth conversions tax- and penalty-free prior to age 60 if you wait for each Roth conversion to remain in the Roth for five tax years.  This is the basis of the Roth conversion ladder aka Roth pipeline, something I'm sure you're already familiar with.

So since I think you know that, I'm not sure how that squares with what I bolded in your statement above.  If you entirely emptied your traditional IRA by making Roth conversions while young, you'd still be able to spend all of your Roth contributions and all of the conversions on a 5 year delay.  The only part of your Roth that would have to wait until 59.5 would be any Roth earnings (which would be Roth balance - Roth contributions - Roth conversions).

What am I missing?

By principal I really meant basis, whether contribution or conversion. Poor choice of words.

If we convert every penny of tIRA money to Roth prior to age 55 and spend it as quickly as it's converted the last years of our 50s get a little complicated. Plus we royally fucked up our lifetime tax curve and probably paid way more in taxes than we needed to. That's a possible scenario if we focus on big conversions in down markets over the next 20 years.

I've modeled it both ways and emphasizing Roth conversions in down markets lowers the ceiling of spending prior to age 60, as it should if one thinks about it simply. By emphasizing tax-tree gains in the Roth, one also emphasizes the availability of their money after age 60. That is not a desireable characteristic to me.

If we had FIREd closer to age 60 there'd be no concern. I'd focus on converting in down markets.

I agree with you about maybe paying too much in taxes if you convert too much too early.

To me, the dollar amount of Roth conversions, the amount of spending each year, and whether I'm converting at a relative low or high are all mostly independent variables.  it seems you're connecting them to each other somehow.

My process, in a nutshell:  Figure my max marginal rate between now and 85 when I've a decent chance of being deceased.  Then convert a dollar amount up to that marginal rate this year.  If I can convert when the market is at a relative low, then that might get me ahead a smidge on the tax curve.  But since I have decades of Roth conversions ahead and I redo the math every year, it would self correct with slightly smaller conversions for the next two decades.

I know I can't get it perfect, so I just try to get it close (in terms of the marginal rate).

But I have plenty of Roth conversions ahead of me.  I'm older than you.  And I underspend what I could by a lot so I'm mostly playing for my kids' future benefit.

One question, I guess:  How do you model future down markets?  Probably not Monte Carlo.  Do you just assume a 10% correction every X years and a 20% correction every Y years or something like that?

[ender]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

This doesn't make sense at all.

There's really no reason you'd prefer to convert a smaller number of shares for the same tax obligation (ie dollar value converted).

This is one of those counterintuitive ideas most folks won't experience but I can assure you it's correct. It's a two fold problem too. Not only is there a very real risk of us running out of tIRA money to convert if we focused on big Roth conversions when the market was down, we can only spend Roth principal so we also limit the amount of money available to us penalty-free prior to age 60. Perhaps if we were older it wouldn't matter but we still have 20 years before early withdrawal penalties go away.

[Emphasis added.]

You can withdraw and spend Roth conversions tax- and penalty-free prior to age 60 if you wait for each Roth conversion to remain in the Roth for five tax years.  This is the basis of the Roth conversion ladder aka Roth pipeline, something I'm sure you're already familiar with.

So since I think you know that, I'm not sure how that squares with what I bolded in your statement above.  If you entirely emptied your traditional IRA by making Roth conversions while young, you'd still be able to spend all of your Roth contributions and all of the conversions on a 5 year delay.  The only part of your Roth that would have to wait until 59.5 would be any Roth earnings (which would be Roth balance - Roth contributions - Roth conversions).

What am I missing?

By principal I really meant basis, whether contribution or conversion. Poor choice of words.

If we convert every penny of tIRA money to Roth prior to age 55 and spend it as quickly as it's converted the last years of our 50s get a little complicated. Plus we royally fucked up our lifetime tax curve and probably paid way more in taxes than we needed to. That's a possible scenario if we focus on big conversions in down markets over the next 20 years.

I've modeled it both ways and emphasizing Roth conversions in down markets lowers the ceiling of spending prior to age 60, as it should if one thinks about it simply. By emphasizing tax-tree gains in the Roth, one also emphasizes the availability of their money after age 60. That is not a desireable characteristic to me.

If we had FIREd closer to age 60 there'd be no concern. I'd focus on converting in down markets.

It still doesn't make sense though.

If you have 1000 shares in your traditional IRA all at $1k/share (so $1M total), then you will eventually convert 1000 shares if you fully drain your traditional IRA.

The only way the problem you are trying to avoid happens is if you convert, and the market goes down meaningfully and consistently until you turn 60.

Lets say you want $50k in Roth conversions/year roughly. So around 50 shares if you convert them on Jan1, you have 950 shares left in traditional. If it's a year like this year, converting in Dec might mean you convert only 40 shares. So 950 vs 960 shares. Both situations $50k in conversions.

Basically if you end up converting in a "down" time period, it means you'll have more traditional IRA money available to convert later, so it's self correcting.

If you end up running out of Roth conversions you can still withdraw the earnings, you just pay a slight penalty (10% penalty + taxed as income, which likely will be only 10% total). If you're paying more than 10ish percent on the marginal conversions this nets out a wash, possibly even better depending on your spending vs the marginal conversion rates you are paying.

[secondcor521]

I just executed our last batch of Roth today, bringing our total for the year to 60k. As soon as January 1 hits I'll convert some more for 2024 while the market is high. It's always better for us to convert when the market is high because we're young enough that converting while low risks the chance of us exhausting our supply of tIRA funds before age 60.

This doesn't make sense at all.

There's really no reason you'd prefer to convert a smaller number of shares for the same tax obligation (ie dollar value converted).

This is one of those counterintuitive ideas most folks won't experience but I can assure you it's correct. It's a two fold problem too. Not only is there a very real risk of us running out of tIRA money to convert if we focused on big Roth conversions when the market was down, we can only spend Roth principal so we also limit the amount of money available to us penalty-free prior to age 60. Perhaps if we were older it wouldn't matter but we still have 20 years before early withdrawal penalties go away.

[Emphasis added.]

You can withdraw and spend Roth conversions tax- and penalty-free prior to age 60 if you wait for each Roth conversion to remain in the Roth for five tax years.  This is the basis of the Roth conversion ladder aka Roth pipeline, something I'm sure you're already familiar with.

So since I think you know that, I'm not sure how that squares with what I bolded in your statement above.  If you entirely emptied your traditional IRA by making Roth conversions while young, you'd still be able to spend all of your Roth contributions and all of the conversions on a 5 year delay.  The only part of your Roth that would have to wait until 59.5 would be any Roth earnings (which would be Roth balance - Roth contributions - Roth conversions).

What am I missing?

By principal I really meant basis, whether contribution or conversion. Poor choice of words.

If we convert every penny of tIRA money to Roth prior to age 55 and spend it as quickly as it's converted the last years of our 50s get a little complicated. Plus we royally fucked up our lifetime tax curve and probably paid way more in taxes than we needed to. That's a possible scenario if we focus on big conversions in down markets over the next 20 years.

I've modeled it both ways and emphasizing Roth conversions in down markets lowers the ceiling of spending prior to age 60, as it should if one thinks about it simply. By emphasizing tax-tree gains in the Roth, one also emphasizes the availability of their money after age 60. That is not a desireable characteristic to me.

If we had FIREd closer to age 60 there'd be no concern. I'd focus on converting in down markets.

It still doesn't make sense though.

If you have 1000 shares in your traditional IRA all at $1k/share (so $1M total), then you will eventually convert 1000 shares if you fully drain your traditional IRA.

The only way the problem you are trying to avoid happens is if you convert, and the market goes down meaningfully and consistently until you turn 60.

Lets say you want $50k in Roth conversions/year roughly. So around 50 shares if you convert them on Jan1, you have 950 shares left in traditional. If it's a year like this year, converting in Dec might mean you convert only 40 shares. So 950 vs 960 shares. Both situations $50k in conversions.

Basically if you end up converting in a "down" time period, it means you'll have more traditional IRA money available to convert later, so it's self correcting.

If you end up running out of Roth conversions you can still withdraw the earnings, you just pay a slight penalty (10% penalty + taxed as income, which likely will be only 10% total). If you're paying more than 10ish percent on the marginal conversions this nets out a wash, possibly even better depending on your spending vs the marginal conversion rates you are paying.

Well, there is the problem of converting too fast.  Most people don't run into this situation, but @Mr. Green might, and that may be what he means.

To use an extreme example with your numbers, suppose the market is down somewhat like last year and a taxpayer decides that Roth converting when the market is down is a good thing.  And they hear about the TCJA expiring in 2026.  So they convert 500 shares ($500K) in 2024 and 500 shares ($500K) in 2025, completely draining their tIRA in the process.

I'm not going to do the math, but they'd pay a lot more in taxes in that scenario than they would by doing 50 shares a year for 20 years, which is closer to how most people do it.

I agree with you about converting in down markets, though, @ender.  To me, the same number of shares for a lower tax cost seems like it has to be a correct argument.  But I also don't fully understand Mr. Green's optimization process, or his numbers, or his goals, or his total picture.  There is a disconnect somewhere, but I don't know if it's with me or with Mr. Green.  And I know he's looked at it every way from Sunday and is confident in his numbers, so... :shrug:.

I personally think of the issue as one of getting my marginal rate this year up to my maximum marginal rate later, thus reducing or eliminating me ending up in any bracket I don't have to end up in.  My marginal rate now can be converted into an AGI target via the CSS, which converts into a Roth conversion target in dollars.  I try to fill that dollar amount with as many shares as I can each year, so I try to convert when the market is "low".

With my set of numbers, I'll be Roth converting for decades, and I'll have more than I will spend, so I think I have different goals and different financial tasks than Mr. Green.

[Mr. Green]

@secondcor521 It's definitely a unique situation. My goal is to convert as large a dollar figure as possible in the next 15 years, while still being tax efficient. This is what leads to the largest amount of Roth basis available to spend. If our future spending approaches six figures or beyond, paying the 10% penalty and ordinary income taxes on the early withdrawal of Roth gains could be quite salty. Does that make sense? I feel like I'm not doing a very good job of explaining.

I think most people don't encounter this because they convert what they need to support their spending and call it a day. We aim to convert as much as possible such that it allows as much spending as possible before age 60.

My latest model requires Roth conversions of over 100k for seven years starting in 2024 and then drops down to ~50k for seven more years in order to maximize spending. That supports spending 150k from ages 46 to 56, and 200k from ages 57 to 70. Spending slowing tapers off after age 70. This exhausts virtually all after-tax money and Roth basis in the age 59 year, or landing the plane on fumes as we turn 60 and Roth gains and tIRA funds are available penalty free.

If we made our Roth conversions during big dips to maximize future Roth gains, that model fails.

[secondcor521]

@Mr. Green, that does help, and I do see more of what you're saying.

What people sometimes misunderstand - but you clearly get it - is that only the dollar amount of a Roth conversion is spendable before age 59.5; any earnings on the Roth conversion inside the Roth goes into the earnings bucket, which as you properly note is not available prior to 59.5 without taxes and penalties.

There are several differences between your situation and mine (and maybe ender's):

1.  It sounds like your plan calls for converting most of your tIRAs before age 55 or so.  I'm already 54.5 and am nowhere near done.

2.  You have a lifetime spending plan and appear to want to spend most of your assets, maybe preferentially before age 70 or so.  None of those three things are true of me.

3.  You appear to weight somewhat heavily access to spendable funds between now and 60.  I don't.

4.  You retired earlier than I did and have done more optimization work than I did at your age.

5.  I'm guessing you don't have much in taxable, or it's mostly going to pay your taxes on your conversions.  I have some extra taxable, so don't need to worry much about my Roth ladder being big.  Also, I'm closer to 59.5 so I have more flexibility than you do at the moment.

6.  I'm single, and I'm sure that my spending, tIRA, Roth IRA, taxable, and SS situation is different from yours (and everyone else on the planet).

If you did deplete your tIRA by conversions and your Roth basis before 59.5, instead of drawing on the Roth earnings it could make sense to not convert the last few years in your plan and just spend from the tIRA instead.  You'd still have EWP plus income taxes, but you'd be paying the income taxes anyway; that would leave the Roth earnings to be tax free after 59.5.

I sort of wonder what iORP would say about your optimization.  Have you tried it?  If so, how does it's plan compare to yours?  If different, have you examined those differences?  These sorts of optimization problems can be tricky to code right unless you have the theoretical underpinning - I know just enough to know I couldn't do it properly myself, and I think the iORP author does know enough to do it right.

When you say "the model fails" in your last sentence, is that because you run out of tIRA funds to convert, or some other reason?

[ender]

@secondcor521 It's definitely a unique situation. My goal is to convert as large a dollar figure as possible in the next 15 years, while still being tax efficient. This is what leads to the largest amount of Roth basis available to spend. If our future spending approaches six figures or beyond, paying the 10% penalty and ordinary income taxes on the early withdrawal of Roth gains could be quite salty. Does that make sense? I feel like I'm not doing a very good job of explaining.

I think most people don't encounter this because they convert what they need to support their spending and call it a day. We aim to convert as much as possible such that it allows as much spending as possible before age 60.

My latest model requires Roth conversions of over 100k for seven years starting in 2024 and then drops down to ~50k for seven more years in order to maximize spending. That supports spending 150k from ages 46 to 56, and 200k from ages 57 to 70. Spending slowing tapers off after age 70. This exhausts virtually all after-tax money and Roth basis in the age 59 year, or landing the plane on fumes as we turn 60 and Roth gains and tIRA funds are available penalty free.

If we made our Roth conversions during big dips to maximize future Roth gains, that model fails.

Converting $100k a year has a fairly high "penalty" on a good chunk of the money via taxes as it is though. Or do you have a way to drop your federal marginal rate to basically zero currently?  $100k in conversions is solidly 22% marginal even for MFJ. Do you have rental depreciation you are offsetting?

Based on your profile age of 40, it seems you will stop Roth conversions at 55ish?  If so, you are wasting at the very least the standard deduction worth of free conversions and/or income tax free earnings withdrawals until you the year you turn 59.5. Or will you have other taxable income sources then, that you can't withdraw? Even with a 10% penalty from 55-59.5, you'd be paying meaningfully less than your current marginal rate on the conversions if you withdrew Roth earnings (preferably in this situation you'd just directly withdraw leftover traditional money and pay the 10% penalty), especially if you are able to pay 22% less on taxes for conversions themselves throughout the time frame.  Also obviously add state/local income taxes here too.

I'm assuming too you are considering the impact of high(er) roth conversions on ACA subsidies, too, to ensure you are correctly calculating the cost of converting larger amounts vs paying the penalties? For us, quick math shows converting $50k vs $100k results in a bit over $3k hit on ACA subsidies. That's a reasonably high marginal rate addition for higher Roth conversions.

If your healthcare is ACA it probably makes sense to batch your conversions somewhat instead of evenly distributing - instead of $100k/year, doing something more like $150k and then $50k. Basically whatever it would take to keep the lower year out of the 22% federal bracket, by moving all of the converted amount above 22% into the "big" year where you've already lost most of the ACA subsidy anyways, then the next year you can get a higher subsidy.

From what you've said though, your marginal rate currently is pretty high (30%+) on a good chunk of your planned $100k yearly Roth conversions vs other options.




@secondcor521 likely the biggest difference for me personally planning wise is I already have a large amount of roth contributions via megabackdoor and regular Roth IRA available. Almost 1/4 of our total investments are Roth principal via one of those two sources. That will continue to grow as each year I'm contributing around $40k in withdraw-able contributions (the backdoor Roth ones require 5 years; megabackdoor require me to roll it to an IRA).

[seattlecyclone]

@secondcor521 It's definitely a unique situation. My goal is to convert as large a dollar figure as possible in the next 15 years, while still being tax efficient. This is what leads to the largest amount of Roth basis available to spend. If our future spending approaches six figures or beyond, paying the 10% penalty and ordinary income taxes on the early withdrawal of Roth gains could be quite salty. Does that make sense? I feel like I'm not doing a very good job of explaining.

I think most people don't encounter this because they convert what they need to support their spending and call it a day. We aim to convert as much as possible such that it allows as much spending as possible before age 60.

My latest model requires Roth conversions of over 100k for seven years starting in 2024 and then drops down to ~50k for seven more years in order to maximize spending. That supports spending 150k from ages 46 to 56, and 200k from ages 57 to 70. Spending slowing tapers off after age 70. This exhausts virtually all after-tax money and Roth basis in the age 59 year, or landing the plane on fumes as we turn 60 and Roth gains and tIRA funds are available penalty free.

If we made our Roth conversions during big dips to maximize future Roth gains, that model fails.

Converting $100k a year has a fairly high "penalty" on a good chunk of the money via taxes as it is though. Or do you have a way to drop your federal marginal rate to basically zero currently?  $100k in conversions is solidly 22% marginal even for MFJ.

Not necessarily. The 12% bracket goes up to $89,450 taxable income this year. Add back the standard deduction and that allows for $117,150 gross. Sounds like their plan depends very heavily on withdrawal of Roth basis, so I wouldn't be surprised to find that the Roth conversion makes up the vast majority of their gross income.

I tend to believe it's unlikely most of us need to worry about the problem of running out of all our money except for Roth earnings pre-59½, but if you have little to no assets outside of retirement accounts and FIRE relatively early I could see that being a risk.

Like you, I have a pretty good chunk of mega backdoor basis and I also have a portion of our stash in taxable, so my optimization problem is different from Mr. Green's. That's part of what makes the drawdown phase so much more complex than the accumulation phase. We all start with our own individual allocation into the different tax shelters and that affects the math a lot.

[ender]

@secondcor521 It's definitely a unique situation. My goal is to convert as large a dollar figure as possible in the next 15 years, while still being tax efficient. This is what leads to the largest amount of Roth basis available to spend. If our future spending approaches six figures or beyond, paying the 10% penalty and ordinary income taxes on the early withdrawal of Roth gains could be quite salty. Does that make sense? I feel like I'm not doing a very good job of explaining.

I think most people don't encounter this because they convert what they need to support their spending and call it a day. We aim to convert as much as possible such that it allows as much spending as possible before age 60.

My latest model requires Roth conversions of over 100k for seven years starting in 2024 and then drops down to ~50k for seven more years in order to maximize spending. That supports spending 150k from ages 46 to 56, and 200k from ages 57 to 70. Spending slowing tapers off after age 70. This exhausts virtually all after-tax money and Roth basis in the age 59 year, or landing the plane on fumes as we turn 60 and Roth gains and tIRA funds are available penalty free.

If we made our Roth conversions during big dips to maximize future Roth gains, that model fails.

Converting $100k a year has a fairly high "penalty" on a good chunk of the money via taxes as it is though. Or do you have a way to drop your federal marginal rate to basically zero currently?  $100k in conversions is solidly 22% marginal even for MFJ.

Not necessarily. The 12% bracket goes up to $89,450 taxable income this year. Add back the standard deduction and that allows for $117,150 gross. Sounds like their plan depends very heavily on withdrawal of Roth basis, so I wouldn't be surprised to find that the Roth conversion makes up the vast majority of their gross income.

I tend to believe it's unlikely most of us need to worry about the problem of running out of all our money except for Roth earnings pre-59½, but if you have little to no assets outside of retirement accounts and FIRE relatively early I could see that being a risk.

Like you, I have a pretty good chunk of mega backdoor basis and I also have a portion of our stash in taxable, so my optimization problem is different from Mr. Green's. That's part of what makes the drawdown phase so much more complex than the accumulation phase. We all start with our own individual allocation into the different tax shelters and that affects the math a lot.

Even still, at 12% federal - if they use ACA, and live in NC (which seems to have roughly 4.75%), their marginal rate is more than 20% (roughly 22% if you add all those).

That's pretty high if you ask me to be paying for Roth conversions even if they aren't into the 22% federal bracket.

[secondcor521]

@ender, I think people always have to start with their current situation, and then look at their prioritized goals.

I think one of your and my relatively high prioritized goals is to maximize lifetime after-tax spendable dollars.  I think @Mr. Green's higher priority goal is having lots to spend prior to 59.5.  He may be willing to spend more in taxes to achieve that.

Or, like you said, he might have refundable credits.  This would be especially likely if he has kids (which I can't recall offhand or not if he does) - CTC and EITC come to mind.  This creates a larger 0% "phantom bracket" that can sop up ACA additional premium taxes or income taxes.

[ender]

@ender, I think people always have to start with their current situation, and then look at their prioritized goals.

I think one of your and my relatively high prioritized goals is to maximize lifetime after-tax spendable dollars.  I think @Mr. Green's higher priority goal is having lots to spend prior to 59.5.  He may be willing to spend more in taxes to achieve that.

Or, like you said, he might have refundable credits.  This would be especially likely if he has kids (which I can't recall offhand or not if he does) - CTC and EITC come to mind.  This creates a larger 0% "phantom bracket" that can sop up ACA additional premium taxes or income taxes.

Well, CTC is mostly refundable EITC only applies to earned income (so conversions don't count; unless they have earned income EITC = $0) ;-)

[secondcor521]

@ender, I think people always have to start with their current situation, and then look at their prioritized goals.

I think one of your and my relatively high prioritized goals is to maximize lifetime after-tax spendable dollars.  I think @Mr. Green's higher priority goal is having lots to spend prior to 59.5.  He may be willing to spend more in taxes to achieve that.

Or, like you said, he might have refundable credits.  This would be especially likely if he has kids (which I can't recall offhand or not if he does) - CTC and EITC come to mind.  This creates a larger 0% "phantom bracket" that can sop up ACA additional premium taxes or income taxes.

Well, CTC is mostly refundable EITC only applies to earned income (so conversions don't count; unless they have earned income EITC = $0) ;-)

I was guessing Mr. Green might have some side gig income to qualify for EITC.  Of course, $100K Roth conversions would prevent EITC eligiblity pretty definitively.

[ender]

@secondcor521 I guess what I'm getting at is nothing presented in this thread makes his Roth conversion strategy make sense to me.

That doesn't mean there are not possible reasons it can make sense, but from what's in this thread I don't see it.

(also, as a meta comment, I think this conversation is useful for the thread purpose because it effectively is discussion around Roth as a portion of your portfolio vs traditional)

[Mr. Green]

I want to reply to some of the other comments but I'm fried. Had a long day building a custom spice rack for my wife. @secondcor521 I did manage to put our data in iORP and look what it spit out! Six-figure Roth conversions for 8 years and then down to ~50k for six years. Makes me feel pretty damn good about my model. I used the "Reality Retirement Planning" option for spending, though I think our curve would be a bit more bell shaped.

The model ID is "M2312280MqXVuKfN7eU" in case you want to take a look.

Edit: Something seems off on the Social Security calculations though. It has us netting over 100k from SS and that isn't right. I put in the proper PIA amounts we currently expect to receive. Maybe those are future dollars? I always use real dollars because inflation-adjusted is too hard to take into perspective over such a long time horizon. Anyway, I can't stare at the screen anymore to figure it out yet tonight. Maybe this would change the math on the early Roth conversions.

[ender]

I want to reply to some of the other comments but I'm fried. Had a long day building a custom spice rack for my wife. @secondcor521 I did manage to put our data in iORP and look what it spit out! Six-figure Roth conversions for 8 years and then down to ~50k for six years. Makes me feel pretty damn good about my model. I used the "Reality Retirement Planning" option for spending, though I think our curve would be a bit more bell shaped.

The model ID is "M2312280MqXVuKfN7eU" in case you want to take a look.

Edit: Something seems off on the Social Security calculations though. It has us netting over 100k from SS and that isn't right. I put in the proper PIA amounts we currently expect to receive. Maybe those are future dollars? I always use real dollars because inflation-adjusted is too hard to take into perspective over such a long time horizon. Anyway, I can't stare at the screen anymore to figure it out yet tonight. Maybe this would change the math on the early Roth conversions.

Oooooook, looking at the results to that makes what you were saying about your goals earlier a lot more clear - you're realistically aiming to spend almost everything by 70 as it looks like the spending model has you down to roughly $250k in assets ($137k in todays dollars) when you start drawing social security at age 70.


For SS, what it's doing is taking your annual inflation (2%) and then since you delay until 70, it's saying your starting SS is 26.2 * 1.02^30 * 1.24.  It must be assuming you input the normal retirement age SS numbers?

[secondcor521]

Maybe those are future dollars?

Not sure, but probably so.  Most of these models work in nominal dollars, not real.  (Which is I think what ender is saying immediately above.)

Oooooook, looking at the results to that makes what you were saying about your goals earlier a lot more clear - you're realistically aiming to spend almost everything by 70 as it looks like the spending model has you down to roughly $250k in assets ($137k in todays dollars) when you start drawing social security at age 70.

Right.  Not a goal I have, but seems to be what Mr. Green wants to do.  I think he said something about landing the plane on fumes at age 70 a few posts back.

[ender]

Maybe those are future dollars?

Not sure, but probably so.  Most of these models work in nominal dollars, not real.  (Which is I think what ender is saying immediately above.)

Yes, it's calculating the actual dollar amount and using that. But it's also adding a 24% SS benefit increase by delaying to 70 vs taking at 67 which might not match the initial input numbers.

[seattlecyclone]

Back to the original question, I just put in a request to do a ~$40k Roth conversion on the last trading day of the year. I was not employed for the first several months of the year, and now I am, so we're falling in the 12% bracket. That's a pretty nice rate to lock in. I doubt we'll get significantly lower than that in the future, especially in non-employed years when we have ACA phase-outs to deal with.

[2Birds1Stone]

Can someone provide a sanity check?

Is this mostly a problem for folks that have normal to fatFIRE levels of income/spending in retirement? I'm referring to those worried about how to best convert their pre-tax dollars before RMD's kick in or to get them out to maximize lifetime spending/minimize lifetime taxes.

I am in @Mr. Green's camp of worrying about running out of pretax dollars to convert to keep myself eligible for ACA and off of Medicaid for the next ~30 years.

Our situation is a little unique because we are in the leanFIRE camp (spending $42-48k/yr) and have over 50% of our portfolio in after tax dollars. It should be relatively easy to convert the $10-40k/yr of tIRA funds to Roth at nearly 0% taxes. Or am I missing something obvious?

[chasesfish]

This is an interesting and timely discussion.   We FIREd with limited Roth assets, 16% of our overall total.  After much debate and guidance, I elected to make significant conversions over the past two years, around $300,000 spread between 2022 and 2023's tax year.

I'll preface everything I say below with the following:  These are decisions made with imperfect information.  We only know tax laws today, know the general need for more governmental revenue, and have to make the best decisions we can in a world of imperfect information.    As someone closer to fatFIRE, my goal is to minimize our lifetime tax burden, which will be one of our top five largest expenses over the next few decades.

Some reasons we are choosing large conversions:

- We were so heavy in pre-tax dollars, we needed / wanted to build a Roth IRA ladder for security.  Getting this started boosts the contributions available in a couple of years.

- The balances in pre-tax accounts were well over $1mil.  If we didn't convert and they see an 8% return over time, in a few years we could have the good problem that the account sizes run away from us.  With that comes all the current punitive taxes on high income elderly (IRRMA, Taxable Social Security).   I'd rather have runaway balances in the Roth accounts.

- I had some self employment income, when combined with a deferred compensation plan, pushed us mostly out of the world of ACA tax credits.   This will continue in the future, but I have some flexibility on the timing of it.

- In our situation, $50,000 to $80,000 in income carries the highest marginal rate due to the tax credit phase outs, then there's minimal difference between $80,000 and $230,000 in tax rate, which comes out to +/- 22%.   Going over $250,000 brings into play the NII tax surcharge.   Where exactly these high rates hit you will depend on family size / age and your state's insurance premiums.   Those are our numbers at 41, married w/o kids in Florida. 

- The deferred comp plan runs out when we turn 52, at which point I want to be mostly done with Roth conversions to participate in the ACA tax credits again.   The last couple years of this alone will push us out of the most beneficial credits.   Insurance costs also continue to outpace inflation and are priced based on age.   

Where this leaves us is $500,000 or so in additional dollars we want to convert in the next ten years.  That'll leave $200,000 to $400,000 in the tIRA if we need to manufacture income for the ACA and charitable stuff that's allowed straight from the IRA in the future.   

The recent conclusion I've come up with is it's likely better to alternate high/low years in income on those conversions instead of a steady amount each year.   The tax burden is lower if we have a $50,000 AGI year followed by a $200,000 AGI year vs. a consistent $125,000/year.   I'll likely target a low year in 2024 and 2026, but we'll see how that goes throughout the year. 

Each person's situation is different, but I thought it would be worth posting how these conversions have been going for us.

[Mr. Green]

@2Birds1Stone If you're looking to convert just what you need to cover year-to-year spending or create just enough income to keep you off medicaid you're not going to have a problem of running out of money to convert. That's where my thought process was a few years ago. It was a very comforting stage to be in.

What happened to me is that after we were good for a couple years and the money pile got larger, I started to think "what if we wanted to move to a HCOL location later in life?" Or other questions like that. Essentially, "what if I needed access to a large lump sum sometime in the future?" If I kept creating as small an income as possible, I would not be able to do this without potentially undesirable tax consequences. After-tax monies will have capital gains. Pulling from tIRAs early will have a penalty. The answer was an easy one to come to. By converting more than we need now while following some tax efficiency guidelines, we increase the likelihood that we'll have access to a large lump sum tax-fee while also being more tax-efficient across our lifetime.

What we model isn't necessarily what we do but it still informs my decision-making. For instance, my model says that we should have Roth converted at least enough income up to the (suspended) ACA subsidy cliff of ~92k for a family of 3. Even doing that we're still not even close to exhausting all our money later in life. But I chose to stop at 78k because I couldn't make myself pay the 33% marginal tax rate on the next 14k. I didn't need to do even that. We could have left our income at 40k or something like that if we just wanted to stay off medicaid.

Will we ever actually spend 200k annually? I honestly can't fathom that. I literally don't know how I'd do it. I don't know what will happen later. For us, it now feels like the best way to position ourselves for "hey, what if we..." and that's basically all our life is these days!

[keyvaluepair]

Late to this, but yes, both DW and I Roth for the following reasons:

1) SS Tax Torpedo
2) Medicare premium impact of RMD
3) ACA subsidy effects

See: Wade Pfau article for more nitty gritty

[MustacheAndaHalf]

Back at the end of 2019, my Roth IRA was 1/5th of my NW.  I figured it would supplement retirement, if needed.  Not needing it allowed me to take a high degree of risk at the end of March 2020.  My signature has "An experiment" with a real-time replay of those events, and I wound up with a Roth IRA weighing in at 3/5th of a much higher NW.  That's probably a bit excessive.

I've also moved Roth IRA assets into self-directed IRAs (SDIRAs), which allow investments in anything except a few limited categories (like collectibles).  An SDIRA can own real estate, gold, crypto, or shares in a hedge fund.  I'm convinced venture capital investments can beat the S&P 500, so some of my Roth IRA has been invested there (using SDIRAs).
https://www.investopedia.com/terms/s/self-directed-ira.asp

[jim555]

The conundrum, Roth convert $20K because my state has a $20K exclusion over 59.5 and blow out my ACA situation, or not.

[Turtle]

The conundrum, Roth convert $20K because my state has a $20K exclusion over 59.5 and blow out my ACA situation, or not.

Several factors to consider when doing the math:

What is the additional cost for ACA; the state tax rate you'd be saving on that 20K; your current vs future expected tax rate; your RMD expected situation?

Is your state's 20K exclusion a one time thing, or per year?