Why I must convert to Roth

[FD1000]

When I retired in 2018, I thought I would make 6% annually, which is all I needed, based on my system. It turned out I made more than 11% annually.
That' a good thing, but our T-IRA grew to 2.1 million. Our Roth is at 900K and joint taxable is at $600K.
In the last 2 years I converted $100K total from both T-IRA to stay below IRMAA of $200K.
Yesterday I checked it again, and I realized that our T-IRA will grow so much in the future that we are going to pay a lot of taxes.
All the numbers will be rounded, no inflation will be used. This is a close guesstimation. The future is unknown.

2024 taxes?
In 2024 I receive $30K SS + $100K conversion + $70K from capital gain = AGI/IRMAA(same for us) = $200K. On that amount we paid taxes=$28K. Fed=$27K + State=$1000 (GA has $65K per person, per year tax free).

Let's see what will happen to T-IRA in the next 25 years. I'm 68 years old, and my wife is 65. My mom died at 91; on my wife's side, they live to 95-103.

I'm going to use 10% growth annually + withdrawal of $100K per year.
I have used this site(https://www.mackenzieinvestments.co...-calculators/investment-withdrawal-calculator)

As you can see, the tax burden will be huge. If I'm gone first, the tax burden on one person will be enormous.

This is what I'm going to do.
Convert $300K annually instead.

In 10 years the TIRA will be depleted. I can really stop any time under $200-300K because by then, till death, the taxes should not be a problem.
What about taxes and Part-B?

Right NOW =Total taxes + Part B
Total taxes = $28K.
Part B (Medicare)? We both have Advantage, so no part D.
right now it's $185 PP times 24 months = $4440
Total taxes + Part B = 28K + 4.4K = $32.5 (rounding)

If we convert $300K,Total taxes + Part B
My software says: Total taxes = $90K.
IRMAA=$400K, it's $480 PP times 24 months = 11520
Rounding up = additional $7K for Part-B
Total taxes + Part B = $97K (90+7)

Total additional taxes + Part-B are $65K (97 - 32.5 = 64.5, rounded to $65K)

What about our joint taxable account? We will use it to pay for the conversion.

We will run out of money in Joint-Taxable after 9 years.

What about RMD?
I'm not going to calculate it because it's negligible. By age 73 the TIRA for both is $1.17 million. Half of that is $600K. Let's assume 4% RMD = $24K.
This means I can convert only $276K ($300K - 24K).

What will I achieve?
* In 10 years I will pay an additional $650K taxes (65 per years times 10). If I convert only $100K, the tax burden will be much worse
* After 10 years, our TIRA+Joint taxable will be gone. Roth will have all the money.
* No taxes, or minimum taxes for life for us after 10 years and much more important for my wife. Later for the kids.
* Part-B goes back down within 2 years after the conversion ends

What do you think? Yes, No , maybe, why?
I know we can use Qualified Charitable Distributions (QCDs) from an IRA, to lower taxes without a deduction.
I don't want to discuss it here.

IRMAA

 

[Route246]

You’ve put a lot of analysis into Roth conversions—out of curiosity, where are you at in terms of your x-FI number using the 4% rule? The strategies around pre-retirement tax optimization definitely feel more urgent when that financial independence buffer is limited.

For me, as my x-FI multiple has grown, I’ve found that my attention to maximizing tax efficiency has relaxed quite a bit. When resources are on a knife’s edge, every tax-saving move feels essential. But once there’s a substantial cushion, I worry less about the tax tail wagging the dog—sometimes optimizing for simplicity, flexibility, or peace of mind brings more value than perfecting every tax angle.

Curious if your perspective on tax strategy shifts once FI is well above the minimum, or if you feel that dialing in tax efficiency always deserves maximum focus regardless of your FI buffer.

 

[sakowitzm]

I'm no good at these kinds of analyses, so take this with several grains of salt. But what I see is that IF your spreadsheet pans out, you would kick off this mortal coil with some $13 MILLION! Isn't that something we all would love? Why all the fretting over how much you pay in taxes? As Route246 said, that sounds like letting the tax tail wag the dog.

Now... that said, I think 10% is far too optimistic to use as a planning metric.

 

[FD1000]

You’ve put a lot of analysis into Roth conversions—out of curiosity, where are you at in terms of your x-FI number using the 4% rule? The strategies around pre-retirement tax optimization definitely feel more urgent when that financial independence buffer is limited.

For me, as my x-FI multiple has grown, I’ve found that my attention to maximizing tax efficiency has relaxed quite a bit. When resources are on a knife’s edge, every tax-saving move feels essential. But once there’s a substantial cushion, I worry less about the tax tail wagging the dog—sometimes optimizing for simplicity, flexibility, or peace of mind brings more value than perfecting every tax angle.

Curious if your perspective on tax strategy shifts once FI is well above the minimum, or if you feel that dialing in tax efficiency always deserves maximum focus regardless of your FI buffer.

FI isn’t part of my analysis, because the only thing that matters here is total return.
Since 2018, I’ve invested about 95% in bond OEFs.
FI, investing style, and timing aren’t the focus of this discussion—they’ll only take it off track.

 

[TheWizard]

What I think: no.
You'll be prepaying considerably higher taxes over the next several years to allow you have low or zero taxes in your later years.

Because of our progressive tax system, this is a very suboptimal approach.

It's neither necessary nor desirable to completely deplete your tax-deferred accounts. Continue to Roth convert a modest amount each year and you'll be fine.

You mentioned kids, so set up your plans to have a PORTION of your tax-deferred money go to those kids upon the first spouse's death. Start the first ten year clock early.

And your IRMAA table is out of date. The circled AGI MFJ amount is $416k or higher for 2025 income. Use TFB site to track this.

2025 2026 2027 Medicare IRMAA Premium MAGI Brackets

You pay higher Medicare Part B and Part D premiums if your income exceeds certain thresholds. Here are the IRMAA income brackets for 2025, 2026, and 2027.

thefinancebuff.com

 

[SecondCor521]

* No taxes, or minimum taxes for life for us after 10 years and much more important for my wife. Later for the kids.

What do you think? Yes, No , maybe, why?

Yes, but I think you're converting too aggressively because of the one bullet point of yours I quoted.

I didn't check your numbers, but they are broadly similar enough to me that I know the general landscape of the tax area you are in.

Comparing your incremental case to your base case, you're paying ~$65K/$200K = 32.5% between taxes and additional IRMAA now and 0% later, when you could probably pay 22% plus some lesser IRMAA (swag of about 2%) both now and later.

10 years of the additional $200K is $2M. 32.5% - 24% * $2M is $170K. and that $170K could compound over the next two decades if you don't pay it in voluntary extra taxes, so it's probably over $300K in real money.

That is a lot to pay for tax simplification.

That being said, you're married. The widow tax trap is a real planning concern for married couples. Still, you could probably find a sweet spot which is less aggressive than your current idea which balances the taxes better. It would take a bit of work and spreadsheeting and making assumptions about life expectancies and kids' future tax rates, but it's not that hard in the grand scheme of things, and once you've got the work done you can revisit it as time goes by.

 

[TheWizard]

By comparison, I'm single with similar per capita wealth and in same IRMAA tier you circled.
My simple goal the past few years has been:
1) stay in 24% marginal tax bracket, avoid 32%
2) avoid getting into next higher IRMAA tier

But I'm 75 now, so I can only do token Roth conversions with those rules, in the $10-20k range last year and probably this year also.

And because I'm 95% stock funds across all my accounts, I could find it challenging to stay in my chosen bracket/tier if the market continues upwards...

 

[Exchme]

Now... that said, I think 10% is far too optimistic to use as a planning metric.

Agreed, plus OP is ignoring inflation, so OP is counting on 10% real growth!

If OP makes Roth Conversions that are optimized for this level of growth, they would be highly non-optimum at lower growth rates. If his system stops working or he declines in health and loses the capacity to implement it, then the returns would revert to their underlying asset types and OP can end up losing money vs. not converting.

I would take a middle ground. Between IRMAA tiers and the OBBBA senior deduction phase-out, moving up the income ladder from where OP is now, the marginal tax rates run 24-28% on conversions up to $500K AGI and then there is a zone where it's worse as the enhanced state and local tax deduction phases out. So converting to the top of the 24% bracket or 2.6X IRMAA tier (at least $416,000 in 2025), whichever is lower for the OP for a few years would be helpful if the growth numbers are anywhere close. If the hyper growth does not pan out, that's still not a disaster as the marginal rates are pretty flat.

 

[TheWizard]

...That being said, you're married. The widow tax trap is a real planning concern for married couples. Still, you could probably find a sweet spot which is less aggressive than your current idea which balances the taxes better. It would take a bit of work and spreadsheeting and making assumptions about life expectancies and kids' future tax rates, but it's not that hard in the grand scheme of things, and once you've got the work done you can revisit it as time goes by.

As I mentioned in #5, there's an easy way to sidestep much of the widow trap with proper planning...

 

[Focus]

My sense is the OP's mind is made up already.

 

[FD1000]

Yes, but I think you're converting too aggressively because of the one bullet point of yours I quoted.

I didn't check your numbers, but they are broadly similar enough to me that I know the general landscape of the tax area you are in.

Comparing your incremental case to your base case, you're paying ~$65K/$200K = 32.5% between taxes and additional IRMAA now and 0% later, when you could probably pay 22% plus some lesser IRMAA (swag of about 2%) both now and later.

10 years of the additional $200K is $2M. 32.5% - 24% * $2M is $170K. and that $170K could compound over the next two decades if you don't pay it in voluntary extra taxes, so it's probably over $300K in real money.

That is a lot to pay for tax simplification.

That being said, you're married. The widow tax trap is a real planning concern for married couples. Still, you could probably find a sweet spot which is less aggressive than your current idea which balances the taxes better. It would take a bit of work and spreadsheeting and making assumptions about life expectancies and kids' future tax rates, but it's not that hard in the grand scheme of things, and once you've got the work done you can revisit it as time goes by.

Thank you for the above.
"you're paying ~$65K/$200K = 32.5%"
The tax software is using 2024 table
On the Fed tax Now 27, after convertion 77 = 50 / 200 = 25%
GA tax now=1K, after 12K = this is the big jump.
You are right, additional tot = 32.5%

I have done many of these calcualtion in 2018.
If I convert just $200K.
The TIRA will go down to $614K by age 90.

Taxes will be Fed=$51.8K + State=$6.4K = $58 rounded
Now, converting 100K = Total tax $28K
Converting 200K = Total tax $58K

The additional $100K costs me $30K more in taxes = 30% more taxes.
The key is of course, that GA doesn't taxe the first $130K (65K per person).

Part-B goes from 185 PP to 370 PP. From $4440 to 8880 = 4440

Additional taxes + Part-B = 30 + 4444 = 34.5K per year.
But, I will pay for 25 years and still will not deplete TIRA. 25 * 34.5 = 862.5 in taxes

 

[Al18]

The industry experts are saying 5.5% growth from the SP500 is expected for the next decade.

 

[jazz4cash]

My takeaway is a tale of two outcomes. We have been laser focused on not running out of money so we tend to be very conservative using high estimates of expenses and low estimates of returns to balance it all out. I use Fidelity’s income planner tool and select the much below average market returns view. We rest easy knowing we have enough under poor market conditions and Roth conversions look silly. It’s more likely we get average returns and then the portfolio balloons and we’ll wish we did the Roth conversions. My solution is to not worry about optimizing, just worry about having enough to pay for everything including the taxes and use QCDs which I don’t fully understand yet.

 

[38Chevy454]

I think doing some Roth conversions is smart, but only up to 22 or maybe 24% max tax rates. Consider the IRMAA penalty, but don't let it drive your choices.

Worse case is you (or heirs) pay the higher tax rates later.

 

[FD1000]

For many years, I worried about whether I had enough. Then I retired—and ended up making more than I expected.
Sure, I could settle for 6–8%, but what if my investments return 10%?
Keep in mind, I don’t rely on stocks or traditional bonds. I use specialized bond OEFs along with trading and timing strategies.
I’ve managed to avoid every decline greater than 1%.
So even if growth averages only 8%, the conversion will still be OK.

My conversion are within the 24% tax bracket; So, instead of $300K conversion, I can do $283K.
We get $100K from SS + capital gains.
Total taxes go up more than the 24% because GA taxes go up from $1K for $200K IRMAA to much more for the rest.

My sense is the OP's mind is made up already.

I didn't.

The wizard: You mentioned kids, so set up your plans to have a PORTION of your tax-deferred money go to those kids upon the first spouse's death. Start the first ten year clock early.

I don't get the above; please explain.

 

[RetiredHappy]

For many years, I worried about whether I had enough. Then I retired—and ended up making more than I expected.
Sure, I could settle for 6–8%, but what if my investments return 10%?
Keep in mind, I don’t rely on stocks or traditional bonds. I use specialized bond OEFs along with trading and timing strategies.
I’ve managed to avoid every decline greater than 1%.
So even if growth averages only 8%, the conversion will still be OK.


I didn't.


I don't get the above; please explain.

The surviving spouse can disclaim part of the IRA and the disclaimed portion go to the children if they are the contingent beneficiaries, thereby the IRA becomes smaller to deal wrt withdrawals and taxes.

I plan to do this if my spouse predeceases me.

 

[FD1000]

You’re right—I forgot that once a spouse passes, the kids can start withdrawing their share.
BTW, taxes will be higher in 10-20 years. That's a guarantee.

More calculation and converting $250K looks good too.

 

[Route246]

FI isn’t part of my analysis, because the only thing that matters here is total return.
Since 2018, I’ve invested about 95% in bond OEFs.
FI, investing style, and timing aren’t the focus of this discussion—they’ll only take it off track.

Your statement is a contradiction, "the only thing that matters is total return" yet you also say "This is a close guesstimation. The future is unknown" which doesn't make sense. It seems contradictory to be focused on total return when everything is just a guess or estimation because the future is unknown. You seem to be leaving yourself no contingencies for unknown events that could happen.

Example: Congress changes their mind and starts taxing Roth accounts. Don't say that will never happen because all it takes is for things to line up and it will happen. This is not an outlier or black swan, either. They have shown a propensity to act in ways that are very surprising. Just think about when they started taxing SS benefits. Nobody ever believed that would be politically feasible yet it happened. Improbable? Yes. Impossible? Absolutely not.

 

[TheWizard]

The industry experts are saying 5.5% growth from the SP500 is expected for the next decade.

Ah, the experts!
Yes indeed!!

 

[FD1000]

Your statement is a contradiction, "the only thing that matters is total return" yet you also say "This is a close guesstimation. The future is unknown" which doesn't make sense. It seems contradictory to be focused on total return when everything is just a guess or estimation because the future is unknown. You seem to be leaving yourself no contingencies for unknown events that could happen.

Example: Congress changes their mind and starts taxing Roth accounts. Don't say that will never happen because all it takes is for things to line up and it will happen. This is not an outlier or black swan, either. They have shown a propensity to act in ways that are very surprising. Just think about when they started taxing SS benefits. Nobody ever believed that would be politically feasible yet it happened. Improbable? Yes. Impossible? Absolutely not.

I don't see a contradiction.
Just because I made a prediction, it doesn't mean it will be true.
The idea is to plan far enough to make decisions now. It served me very well in the past.
At retirement, I asked myself, can I make it on 6% for another 40 years? The answer was yes, and then I did much better.
It's similar now. What if I make 10%?
If I make just 8%, no harm.

Ah, the experts!

I have been documenting the "experts" since 1990 and it's not pretty, and that's why I watch my own indicators.

 

[TheWizard]

...I don't get the above; please explain.

Married couples sometimes get into what they call Widow Trap after the first of them passes.
This is because the surviving spouse, filing Single, can be paying more income tax than they did as a couple.

This often is because of RMDs on tax-deferred accounts when the survivor inherits all of the deceased spouse's accounts.

Let's assume $2M tax-deferred which has an RMD of maybe $80-120k or more, depending on age.

If the deceased spouse had willed $700k of tax-deferred to their offspring instead of 100% to spouse, then surviving spouse's tax burden would be less.

And offspring get partial inheritance sooner with ten years to take entire amount...

 

[TripleLindy]

I went through the Roth conversion exercise earlier this year. I don’t recall the details, but for us, the conclusion was to skip it. Too much tax liability now for the possibility of avoiding taxes much later in life - primarily an advantage for our heirs. I may revisit the issue at some point, but I don’t recall any significant advantage.

 

[pb4uski]

I agree with others that 10% future return is too optimistic (even though I suspect it is a leveraged return). Also, your Roth conversion plan is too aggressive and paying boatloads of tax now to avoid taxes later seems suboptimal to me.

Have you run a scenario of more moderate Roth conversions calibrated to levelize taxable income between now and, say, age 85? I suspect that might be a sweet spot for you.

 

[Route246]

Yes, but I think you're converting too aggressively because of the one bullet point of yours I quoted.

I didn't check your numbers, but they are broadly similar enough to me that I know the general landscape of the tax area you are in.

Comparing your incremental case to your base case, you're paying ~$65K/$200K = 32.5% between taxes and additional IRMAA now and 0% later, when you could probably pay 22% plus some lesser IRMAA (swag of about 2%) both now and later.

10 years of the additional $200K is $2M. 32.5% - 24% * $2M is $170K. and that $170K could compound over the next two decades if you don't pay it in voluntary extra taxes, so it's probably over $300K in real money.

That is a lot to pay for tax simplification.

That being said, you're married. The widow tax trap is a real planning concern for married couples. Still, you could probably find a sweet spot which is less aggressive than your current idea which balances the taxes better. It would take a bit of work and spreadsheeting and making assumptions about life expectancies and kids' future tax rates, but it's not that hard in the grand scheme of things, and once you've got the work done you can revisit it as time goes by.

I’m thinking one layer deeper. Converting aggressively into a Roth assumes that tax laws will remain static and that the benefits of a Roth are untouchable. That assumption is flawed. A Roth is not a covenant; it’s simply a structure, and Congress can change that structure anytime it wants. Another flaw in the OP’s sentiment is the reliance on hindsight—20/20 vision only works backward. As the saying goes, past performance is not an indicator of future returns.

Optimization has its place, but let’s be clear: financial planning built on rigid assumptions is not true optimization—it’s wishful thinking.

 

[TheWizard]

I agree with others that 10% future return is too optimistic (even though I suspect it is a leveraged return). Also, your Roth conversion plan is too aggressive and paying boatloads of tax now to avoid taxes later seems suboptimal to me...

I agree on the suboptimal part, but 10% annual return, nominal not real, is entirely possible if you're 90%+ in stock funds...