Roth conversions and life expectancy

[SecondCor521]

For the past few years, I've tried to Roth convert up to the level where my current marginal rate was approximately equal to my projected marginal rate at age 75.

I'm thinking about increasing that age 75 target to age 82 based on the SSA longevity tables indicating that age 82 is the age at which half of the men my age have died.

A recurring thought is that I may not be paying those tax rates at age 82, either because (a) my assets do not grow as much as I am projecting, (b) tax rates will probably change up or down, and most particularly (c) I might be dead. (Less likely in my particular case but I know @pb4uski has mentioned a fourth: (d) a person might move to a lower tax state.)

I could envision a scenario where I convert at, say, a total marginal rate of 36% now to avoid something higher than that in my 80s; then I die at 72 and my kids' average total marginal rate might only be 30%.

I tried incorporating the SSA longevity table into my RMD spreadsheet, but I'm not sure if I believe the answers. At age 90, I figure about a 45% top marginal rate and about a 21% chance I'll still be kicking then, which only results in 9.36% if I multiply those two factors together. If I do that for every year between now and then, I get a weighted rate of only 15% or so, which seems way too low.

How do y'all account for the fact that you might die before having to pay those higher rates?

 

[RunningBum]

You've got two goals: optimizing tax rates for yourself, and optimizing them for your kids. The methods to meet those goals conflict: convert more now, or convert less. The basic answer is that you're going to have to prioritize which goal to meet.

I don't think I buy multiplying your future top tax rate with the chance of being around then. If you multiple 5 apples by 3 oranges, you get 15, but 15 of what? Is it 15 useful things? It's not to say that the likelihood of reaching an old age and what your tax rate would be then, but I don't think there's a convenient way to combine them and get a useful and meaningful single number. Maybe I'm wrong about that. If I am, then you have to factor in just how high your kids' tax rates will be if they inherit, and what if one or all of them passes before you do?

I've decided to calculate Roth conversions solely for my benefit, comforted by the fact that my son will have some benefit to getting a Roth rather than a tIRA, even if it might be more tax efficient for him to pay the tax.

I'm hedging my bets by gifting him money each year in the form of appreciated funds that he can cash in at a 0% CG rate. He can and has made good use of that money now (buying a nice house in a nice neighborhood) which we both view as better than him living in an apartment that he really dislikes and eventually inheriting far more money than he needs. Is this really related to my Roth conversions strategy? I don't know. It does seem like making some decisions for my benefit and others for his is a nice compromise.

In general I think the best strategy is to make as certain as you can that you won't be a financial burden to your kids. If you feel secure in that, then consider what the best way is to help your kids financially. In that respect you don't make a single financial choice in a vacuum; instead, you consider all options.

 

[FI_RElater]

I can't see any reason for conversions at 36%...

For the past few years, I've tried to Roth convert up to the level where my current marginal rate was approximately equal to my projected marginal rate at age 75.

I'm thinking about increasing that age 75 target to age 82 based on the SSA longevity tables indicating that age 82 is the age at which half of the men my age have died.

A recurring thought is that I may not be paying those tax rates at age 82, either because (a) my assets do not grow as much as I am projecting, (b) tax rates will probably change up or down, and most particularly (c) I might be dead. (Less likely in my particular case but I know @pb4uski has mentioned a fourth: (d) a person might move to a lower tax state.)

I could envision a scenario where I convert at, say, a total marginal rate of 36% now to avoid something higher than that in my 80s; then I die at 72 and my kids' average total marginal rate might only be 30%.

I tried incorporating the SSA longevity table into my RMD spreadsheet, but I'm not sure if I believe the answers. At age 90, I figure about a 45% top marginal rate and about a 21% chance I'll still be kicking then, which only results in 9.36% if I multiply those two factors together. If I do that for every year between now and then, I get a weighted rate of only 15% or so, which seems way too low.

How do y'all account for the fact that you might die before having to pay those higher rates?

I certainly can't see ANY reason for conversions at 36%

(Cali residents at 24% fed and state at 10% are the most likely to be in that situation... but once retired I can't see, except those with truely large portfolios, being beyond that... so no need to pay for the tax earlier... just let the portfolio grow)

there's really no effective tax arbitrage at that level... did you calculate just how LARGE your portfolio/pension/SS would have to be to get to that level?? and even getting beyond the first IRMAA tier, while it costs you higher Medicare for both, doesn't add that much relative to income (we know we'll be in first IRMAA tier, but because we have a good mix of tax-deferred, Roth, and taxable it's probable that we won't get hit into the second tier)

 

[SecondCor521]

I certainly can't see ANY reason for conversions at 36%

(Cali residents at 24% fed and state at 10% are the most likely to be in that situation... but once retired I can't see, except those with truely large portfolios, being beyond that... so no need to pay for the tax earlier... just let the portfolio grow)

there's really no effective tax arbitrage at that level... did you calculate just how LARGE your portfolio/pension/SS would have to be to get to that level?? and even getting beyond the first IRMAA tier, while it costs you higher Medicare for both, doesn't add that much relative to income (we know we'll be in first IRMAA tier, but because we have a good mix of tax-deferred, Roth, and taxable it's probable that we won't get hit into the second tier)

I try to avoid mentioning anything about assets or income too specifically for privacy reasons, but the 36% mentioned is a combination of federal income tax, IRMAA, and state income tax (6%) and would only be around age 82. And that's assuming a healthy investment growth rate, modest inflation, and unchanging tax laws. And all of that is, of course, subject to change.

I've mentioned it before, but I think the rate of increase in IRMAA premiums is a particularly squirelly one to try to predict. I just put in the most average estimates for everything I can find and try to get the spreadsheet right then see what it says.

 

[SecondCor521]

You've got two goals: optimizing tax rates for yourself, and optimizing them for your kids. The methods to meet those goals conflict: convert more now, or convert less. The basic answer is that you're going to have to prioritize which goal to meet.

I don't think I buy multiplying your future top tax rate with the chance of being around then. If you multiple 5 apples by 3 oranges, you get 15, but 15 of what? Is it 15 useful things? It's not to say that the likelihood of reaching an old age and what your tax rate would be then, but I don't think there's a convenient way to combine them and get a useful and meaningful single number. Maybe I'm wrong about that. If I am, then you have to factor in just how high your kids' tax rates will be if they inherit, and what if one or all of them passes before you do?

I've decided to calculate Roth conversions solely for my benefit, comforted by the fact that my son will have some benefit to getting a Roth rather than a tIRA, even if it might be more tax efficient for him to pay the tax.

I'm hedging my bets by gifting him money each year in the form of appreciated funds that he can cash in at a 0% CG rate. He can and has made good use of that money now (buying a nice house in a nice neighborhood) which we both view as better than him living in an apartment that he really dislikes and eventually inheriting far more money than he needs. Is this really related to my Roth conversions strategy? I don't know. It does seem like making some decisions for my benefit and others for his is a nice compromise.

In general I think the best strategy is to make as certain as you can that you won't be a financial burden to your kids. If you feel secure in that, then consider what the best way is to help your kids financially. In that respect you don't make a single financial choice in a vacuum; instead, you consider all options.

I'm trying to take a balanced approach: Consider my tax rate in my 80's, but only to the extent that I might be alive, because if I'm not it doesn't matter.

I'd like to consider my kids' tax rates as well, but trying to guess what those will be in 20 years is nearly impossible. Like most parents, I simply assume they will outlive me, which is a reasonable guess.

Regarding your second paragraph, yeah, I was struggling when I wrote that "15%" in the OP. I don't even know what it means. But it was my effort at trying to weight those future tax rates less because the chances of me living longer gets less and less.

Regarding your last paragraph, I'm not rationally worried about running out or being a burden. I do try to make decisions holistically and taking the probabilities and my kids' situation in mind, but I can also see the rationale for just doing what's best for me and letting the chips fall where they may with my kids.

 

[TheWizard]

"For the past several years, I've Roth converted to get my AGI up approximately to where it will be after starting RMDs at age 72."

That's my replacement of the OP first sentence.

Point being: I don't really know what my marginal Federal tax rate will be over the next decade or two so I focus on AGI, which is easier and more relevant to my IRMAA tier as well.

And for example, if my current 24% bracket reverts back to 28% AND that nasty SALT limit is removed, then I'll hardly see any difference.

So, having started RMDs this year, I'm finished doing Roth conversions. If that secure 2.0 act got voted into effect earlier, then I'd likely be Roth converting for another year or so.

My expected lifespan doesn't factor into this thinking at all, given that I'm still in decent health...

 

[donheff]

It is a complicated analysis. In our case, we will never tap the Roths so the issue is the kids' brackets and the effect on AGI of the accelerated RMDs based on the likely size of the inherited IRAs.

 

[CRLLS]

It is hard enough to guestimate our own future tax rates that I don't consider my kid's future tax rates. Both of them will be difficult to plan for, as one is a teacher and the other is currently military. Both have spouses who's work is covered by SS. We do Roth conversions for us and do not consider them in our planning decisions. What is left when we pass they will get. Period. It may seem a bit callous. In the meantime, we do what we need to do. We taught them to be self-sufficient and they have been so for oh-so-many years now. We hope that they continue to be so. When we pass, it will likely be a spend-that-dough event for them.

 

[SnowballCamper]

How do y'all account for the fact that you might die before having to pay those higher rates?

In my situation, being under 59yo, withdraw more now in order to level income compared to what I expect it to be in the future (when RMDs kick in). The withdraw part is to avoid the higher tax rates I expect in the future. But I spend the money in order to account for the fact that I might die.

I think the goal of optimizing taxes isn't particularly well defined. Minimizing taxes isn't particularly useful because creating a situation with zero, or very low income, nets zero taxes. The prudent goal is to maximize income, over some defined time period, and this is what the I-orp calculator does.

For my own situation (generous pension, mix of modest TSP and Roth balances) I've decided the prudent goal is to level income over the future years. As a result I recently started a SEPP plan for the TSP.

Good Luck!

 

[FI_RElater]

I try to avoid mentioning anything about assets or income too specifically for privacy reasons, but the 36% mentioned is a combination of federal income tax, IRMAA, and state income tax (6%) and would only be around age 82. And that's assuming a healthy investment growth rate, modest inflation, and unchanging tax laws. And all of that is, of course, subject to change.

I've mentioned it before, but I think the rate of increase in IRMAA premiums is a particularly squirelly one to try to predict. I just put in the most average estimates for everything I can find and try to get the spreadsheet right then see what it says.

2ndCor

As I mentioned for those in Cali, more likely to get into that case. For us... not on the coasts (No Cali or NY/NJ etc), when we were in the thirty-x federal...plus state... all we could do is put into non-deductible IRA; one of us was HCE and limited to 10% into 401k(I'd maxed mine) and both of us were triple digit income, and when that's added to high bonuses and ISO's you can see the tax burden. It didn't make sense to do any conversions then... we waited until retirement before starting... and I just finished the last of the pro-rata stuff from all that... and will then push my last 401k into my IRA next year.
Have yet to convert spouse's ... we'll probably convert that but only pull from my regular IRA to supplement as needed... unlikely to need to do more conversions.

As a result of the HCE etc we've got a fair mix of deferred, Roth, and taxable so we can try to pull the maximum while still keeping in lower brackets now in retirement. We expect to keep a fair bit in deferred as we expect somewhere that there's gonna be some higher medical sometime in the future where we won't have to pay the tax...just the higher bills that are necessitated.

 

[flyingaway]

I'm hedging my bets by gifting him money each year in the form of appreciated funds that he can cash in at a 0% CG rate. He can and has made good use of that money now (buying a nice house in a nice neighborhood) which we both view as better than him living in an apartment that he really dislikes and eventually inheriting far more money than he needs. Is this really related to my Roth conversions strategy? I don't know. It does seem like making some decisions for my benefit and others for his is a nice compromise.

Hi, RunningBum,

I am interested in knowing how did you gift your son money in the form of appreciated funds?

For example, if my son and I both have a Fidelity investment accounts, should we ask a Fidelity representative to transfer a few shares of a fund from my account to his account?

 

[SecondCor521]

Hi, RunningBum,

I am interested in knowing how did you gift your son money in the form of appreciated funds?

For example, if my son and I both have a Fidelity investment accounts, should we ask a Fidelity representative to transfer a few shares of a fund from my account to his account?

I'm not RunningBum, but I give appreciated mutual fund shares to my kids by doing DTC transfers from my Vanguard taxable to their Schwab taxable accounts.

I just go online, fill out a form, and submit it. A day or two later, the shares disappear from my account. A few days later, they show up in my kids' account.

Slick and easy, although this is my first year doing it and I haven't yet confirmed the arrival of the cost basis information in their accounts. I've been told it can take a while to show up.

Note that the capital gains rules for gifted shares are a bit unusual if you gift shares at an FMV below cost basis. OTOH, if you gift shares where basis < FMV, then the recipient gets your basis and acquisition date.

Finally, beware the kiddie tax if your kids sell while still young.

 

[RunningBum]

Hi, RunningBum,

I am interested in knowing how did you gift your son money in the form of appreciated funds?

For example, if my son and I both have a Fidelity investment accounts, should we ask a Fidelity representative to transfer a few shares of a fund from my account to his account?

Yes, everything SecondCor said above. My son and I have VG accounts and they have a form I fill out to transfer the shares. The shares show up in his account, with the cost basis I had. Since I keep it to yearly gift exemption amount, this is not a taxable transaction. When he sells, of course, it is.

I assume Fido has the same type of form, and their rep should be able to point you to it if you can't find it.

 

[flyingaway]

I'm not RunningBum, but I give appreciated mutual fund shares to my kids by doing DTC transfers from my Vanguard taxable to their Schwab taxable accounts.

I just go online, fill out a form, and submit it. A day or two later, the shares disappear from my account. A few days later, they show up in my kids' account.

Slick and easy, although this is my first year doing it and I haven't yet confirmed the arrival of the cost basis information in their accounts. I've been told it can take a while to show up.

Note that the capital gains rules for gifted shares are a bit unusual if you gift shares at an FMV below cost basis. OTOH, if you gift shares where basis < FMV, then the recipient gets your basis and acquisition date.

Finally, beware the kiddie tax if your kids sell while still young.

Thank you for your explanation.

 

[flyingaway]

Yes, everything SecondCor said above. My son and I have VG accounts and they have a form I fill out to transfer the shares. The shares show up in his account, with the cost basis I had. Since I keep it to yearly gift exemption amount, this is not a taxable transaction. When he sells, of course, it is.

I assume Fido has the same type of form, and their rep should be able to point you to it if you can't find it.

Thank you. That seems to be doable.

 

[Midpack]

How do y'all account for the fact that you might die before having to pay those higher rates?

Instead of converting to the top of the 24% bracket (the most advantageous if I live as long as my parents), I’m converting to the top of the 22% bracket. Gets most of the Roth bang for the buck versus no conversions. But it all comes down to what your assumptions are, largely future tax rates. What I’m expecting may be completely different than someone else…

 

[RetireBy90]

I expect trying to minimize the tax hit is shooting at a moving target, except you can't just lead and shoot as too many variables for me to compute and if I get a handle on one part the rules change or I find another variable that pops up.

We have a mix close to 1/2 TIRA and 1/2 Roth. I've only done 3 years of conversions and decided to slow down to a lower amount going forward. Having to write big quarterly checks and monthly Medicare just gives too much grief for the perceived gain.

We have sufficient income to pay bills and live a little without touching IRA accounts and will have more when I start SS. Therefore I'll convert about $60K/yr and take a small RMD from beneficiary IRA through 2025 and see what changes happen at that point. Our plan is to leave the Roth to sons and TIRA to charity. RMDs from TIRA will also go to QCDs.

I know this doesn't answer your question of how to figure optimal conversions based on taxes but perhaps there are other ways to minimize taxes by giving the funds in question to charity and feds get 0. Very effective in minimizing taxes.

 

[Montecfo]

2Corr,

You have identified key risks associated with Roth conversions. There are many variables that you must plan for, accurately, over a long period of time. And as you point out, a number of key variables are unknowns (e.g., longevity, market returns, future tax rates, kids' tax rates.)

It is a tactic not suited for optimization for that reason, in my opinion.

You are making an investment in tax rate arbitrage. Like any investment, you want a margin of safety.

Accordingly I think Roth conversions are best pursued when the case for doing so is strongest. This to me is when you are setting off 5-10 percentage points of tax rate differential, not 0-2.

That enhances the likelihood of a positive payoff.

 

[ERhoosier]

"I expect trying to minimize the tax hit is shooting at a moving target, except you can't just lead and shoot as too many variables for me to compute and if I get a handle on one part the rules change or I find another variable that pops up."


RetireBy90 is 100% correct. I have not tried to engage in Roth conversion 'arbitrage' for that reason. Roth conversions offer NO tax benefit to someone dying prior to RMD age and with an estate below Fed tax threshold. I may well fall in that category considering my family record of longevity (or lack thereof )

 

[RunningBum]

Roth conversions offer NO tax benefit to someone dying prior to RMD age and with an estate below Fed tax threshold.

I'm 60. If I wanted to do something today like buy a 2nd home for $500K without going through the mortgage process without job income, and not wanting a huge tax hit this year, I could tap my Roth for that money and not change my income at all. Big tax benefit since I converted most of that money at a rate much lower than what most of a one-time $500K withdrawal would be taxed at.

I may well fall in that category considering my family record of longevity (or lack thereof )

Then maybe you shouldn't have deferred any income at all. You reduced your taxes, but put a far greater amount out of your reach in a 401K or tIRA you think you may well never use.

 

[pb4uski]

....Roth conversions offer NO tax benefit to someone dying prior to RMD age and with an estate below Fed tax threshold. ...

How are you figuring that? It is very situational. We convert about $80k a year and pay about 11.5% in tax on those conversions.

DS is in the 12% bracket so not much savngs there but DD and DSIL are in a much higher tax bracket so would pay more that 11.5% in tax than I paid.

So I guess the benefit is for them and less for me, but it is likely that either DW or I will live to our early 90s so I think we will benefit.

 

[sengsational]

I-orp's approach is to maximize after tax spendable funds, which I think is the most logical goal. I've run scenarios that pay less tax, but generate less money to spend.

As to the life expectancy issue, if it's a couple, and one of you might go sooner, that puts more pressure on early conversations since the single tax rates are quite a bit higher.

 

[imnontrad]

I do not think I will ever be able to spend al that is in my pre-tax retirement account let alone my Roth. But I know that heirs of my pre-tax retirement account would have to empty it in 10 years and pay income taxes on the money (except that portion that will be donated to non-profits). So I vies my conversions of pre-tax to Roth is my gift to them since they will get the Roth tax free.

 

[Janet from FIRE Planet]

I enjoyed readng this recent article on Roth Conversions by Andy Panko. I belong to his Facebook retirement group mentioned at the end of the article. I think he has good advice and sound opinions. https://rethinking65.com/2022/04/07...ICIQDtvdeMXPy-dwGUmQqfkLMpxf6PnDRDQxkkUzsCHgU

 

[cb2008]

FIDO share transfers

Yes, everything SecondCor said above. My son and I have VG accounts and they have a form I fill out to transfer the shares. The shares show up in his account, with the cost basis I had. Since I keep it to yearly gift exemption amount, this is not a taxable transaction. When he sells, of course, it is.

I assume Fido has the same type of form, and their rep should be able to point you to it if you can't find it.

We convinced all three kids to open FIDO brokerage accounts with the $100 opening bonus they offered last fall. When we want to transfer appreciated shares we just call FIDO and advise them of the # of shares of which stock positions and the kids account numbers. The cost basis shows up immediately on their accounts. After seeing those crazy year end capital gains last year, this will be the only way we “gift” money to the kids. We also opened a Fidelity Charitable Giving account for the same reason.