Roth Conversions Trial Balloon

[Midpack]

We often see posts encouraging Roth conversions to fill to the 12% bracket, and that seems a no brainer. If you convert to 12% and find yourself forced into the 22% bracket from 70 on, you may regret it. Is there more universal advice?

*** Knowing everyone has to assess all the other factors that go into the Roth conversion decision. ***

After the analysis I’ve been though, and though people have probably already said it (and I missed it) - I’m thinking a slightly better starting point might be:

Convert to the fill whatever bracket you’ll be in from age 70 on if you hadn’t done any conversions.

IOW if you do NO conversions and that will put you in the 22% (24%, 32%...) bracket from age 70 on due to Soc Sec, RMD, distributions, cap gains and other income, you’ll probably reduce taxes by converting up to the 22% (24%, 32%...) income bracket every year before age 70. If tax rates don’t change, your nest egg final amount will be about the same - you’ll pay less in federal taxes overall, but you’ll forego the returns on the money used to pay taxes on conversions. However, there are other worthwhile $ benefits to widows, heirs, etc. And if future tax rates take more (as I expect long term), you’ll save even more on federal taxes and increase your nest egg final amount some. State taxes you’ll have to consider separately.

2. Convert Too Much

Roth conversions can be a powerful strategy, but blindly converting can end up causing more harm than good.

A common strategy used to avoid paying more in taxes than you may have to is called “bracket filling.” You determine how much room you have until you hit the next tax bracket, and then convert just enough to “fill up” your current bracket.

For example, if you are married filing jointly and have taxable income of $100,000, then you have about $68,400 of room before jumping from the 22% Federal tax bracket to the 24% bracket. It’s worth noting that certain state tax rates can be impacted as well and the rules vary state to state.

Converting any more than what you have left to fill up your current tax bracket would likely cause you unnecessary taxes. You can work with your tax preparer to find exactly how much room you have and how much to convert.

3. Convert Too Little

On the other end of the spectrum, many people are too conservative when calculating how much to convert and end up missing out on valuable room in their tax brackets.

To be clear, going into the next bracket is probably not what you want to do. However, nobody knows exactly what bonus they’ll get, or how many dividends they will receive, so guessing the exact dollar amount is impossible. Since we no longer have the luxury of undoing a Roth conversion, it’s more important than ever to take extra care when running the numbers.

https://www.betterment.com/resources/common-roth-conversion-mistakes/

 

[GrayHare]

Midpack, I like your approach as a starting point. It nicely sums up the main decision point.

As for the quote from Betterment, "You can work with your tax preparer to find exactly how much room you have and how much to convert." Exactly? No, I don't think so since it is impossible to get exact income numbers in advance, much less predict the all-too-common corrected 1099 that arrives in March, or even later.

 

[OldShooter]

... and there is the little matter of predicting what income tax rates will be far into the future.

To me, the Roth/IRA decision is really a sort of tax rate arbitrage. Yes there are some detail considerations but the main thing is that if tax rates are the same now and in the future, conversion is a don't-care. Only if future rates will be higher is the conversion worthwhile and then the advantage may be quite marginal.

 

[Midpack]

As for the quote from Betterment, "You can work with your tax preparer to find exactly how much room you have and how much to convert." Exactly? No, I don't think so since it is impossible to get exact income numbers in advance, much less predict the all-too-common corrected 1099 that arrives in March, or even later.

Exact no, but you can get pretty close by leaving some amount to convert on or just before 12/31? All my income, interest, dividends etc. fall 10 days or more before 12/31 every year so I know almost exactly how much “room” I have. And I have until mid Jan to do the last estimated taxes too.

 

[GrayHare]

Exact no, but you can get pretty close by leaving some amount to convert on or just before 12/31? All my income, interest, dividends etc. fall 10 days or more before 12/31 every year so I know almost exactly how much “room” I have. And I have until mid Jan to do the last estimated taxes too.

Sure, it's possible, and depends on the diversity of your investments. For example, partnerships and certain other entities do not supply their K-1 forms until March or April.

 

[CRLLS]

I have been thinking along the same lines. I just cannot pull the trigger.

There are some things that may or may not have some impact on that decision. IRMMA for one for those already on Medicare. Another is exactly how far into the next bracket are you before thinking about converting? For instance, how far away from your next bracket are you at currently and is this a 1 time deal or a multi-year plan? If say 50K away, then converting to the top of the next bracket may be paying the higher tax on that 50K where it may not be necessary.

It is a mental game for sure. I think it is a good plan if only to prepare for a MFJ couple's eventuality of being widowed and filing as single.

 

[Midpack]

I have been thinking along the same lines. I just cannot pull the trigger.

There are some things that may or may not have some impact on that decision. IRMMA for one for those already on Medicare. Another is exactly how far into the next bracket are you before thinking about converting? For instance, how far away from your next bracket are you at currently and is this a 1 time deal or a multi-year plan? If say 50K away, then converting to the top of the next bracket may be paying the higher tax on that 50K where it may not be necessary.

It is a mental game for sure. I think it is a good plan if only to prepare for a MFJ couple's eventuality of being widowed and filing as single.

That’s what the preface in red in the OP was about. I’m not suggesting there’s a simple answer, just a better place to start evaluating from.

 

[Fleur58]

Don't you think the tax advantages of Roth conversions are misleading if they do not advise you to consider the amount of state taxes when making the decision? Like in Maryland, even though I am in 12% bracket, I must add on another 7.75% in state/local taxes. It does make converting less attractive because it is only 2.25% savings over a 22% bracket.
Or I am wrong?

 

[ERD50]

I used to be mad at myself if I converted "too much", with "too much" defined as not staying in a lower bracket than what I expect in the future.

But now I'm thinking that converting at the same bracket is a "don't care" state, so no worries. And if the tax rates I'm looking at do go up for us (probably more likely than going down at these income levels), it will be an advantage, and it would likely be an advantage or a wash to our heirs.

That was the crux of your post, did I get that right?

If so, I can take the Alfred E. Neuman approach: "What, me worry?" Just convert, and then some for good measure.

-ERD50

 

[Midpack]

Don't you think the tax advantages of Roth conversions are misleading if they do not advise you to consider the amount of state taxes when making the decision? Like in Maryland, even though I am in 12% bracket, I must add on another 7.75% in state/local taxes. It does make converting less attractive because it is only 2.25% savings over a 22% bracket.
Or I am wrong?

Like Federal, you’re going to pay state taxes on TIRA distributions whether you take it now (5-7 years for us) in Roth conversions or later as RMDs (or more) for the rest of your life.

And like federal, I can’t imagine state tax rates going down long term, only the same or up - but just a guess.

Only exception if you plan to move to a lower income tax state later, but then other fees (maybe unrelated to Roth’s) may be higher. Lower income tax states aren’t always lower state taxes, property, sales, etc.

 

[SecondCor521]

I'm strongly considering the following similar but slightly different decision rule:

Convert enough this year so that my total marginal rate this year would be about equal to my predicted total marginal rate at age 70.5 if I were to continue conversions of similar magnitude between now and then.

Total marginal rate means I consider all of my marginal tax rates now and then. Now includes federal income tax, state income tax, ACA PTC reduction, and FAFSA EFC increase. Then includes federal income tax, state income tax, and IRMAA.

I'm 50. The part about continuing conversions of similar magnitude means: If I need to convert $200K (for example) between now and 70.5 to keep my tax bracket the same, then I'd rather do $10K each year for 20 years rather than $200K this year.

One thing I don't know is how my traditional IRA investments will perform between now and then. Reevaluating every year (which is a kind of obvious thing to do anyway) allows me to incorporate that additional data. The "continuing level conversions" may vary somewhat based on this kind of new data, but they'll probably be in the ballpark and allow me to dial things in reasonably well over the 20 year period I have.

The "continuing level conversions" also points out and helps address the situation where a large conversion today both lowers future tax rates and increases current tax rates. IOW, modeling a conversion by changing the conversion amount obviously changes the current marginal rate but also changes the future tax rate.

I do know that there can be "tax humps" in the code (like the temporary 15% + 12% = 27% rate). I don't really have a good idea about how to handle those in my analysis. What I currently do is put all my stuff into a tax program in December and model out different Roth conversion amounts and try to find where the various flat spots and cliffs happen to be for me, so I hope to find a "well behaved" portion of the slope where the incremental Roth conversion amount pushes me up to where my 70.5 rate would be.

That's my current thought, anyway. I may try to get more sophisticated later, but I think what I have above is "good enough" given my situation.

 

[Clone]

This is how I have organized some of my thoughts on Roth conversions:

I look at what rates we have locked in while MFJ and what the incremental rates would be with my wife filing as a single person. With rate reversion and without rate reversion. In each of those situations, I also look at how much headroom there is for the RMDs or other conversions. In my case, the RMDs will most likely not push us into a higher tax bracket.

There are a lot of moving parts to try find an exact optimum answer. When I look at the incremental rates, there is certainly a penalty for filing single IF the rates revert. But at a 3% premium, it is not exactly a back breaker. The possibility of relocating to a lower income tax rate state is also a consideration.One of the things to do this year (for us) is to investigate what the savings would be to establish residency in a lower tax profile state. We will be within several weeks of flipping the 26 week minimum from one state to another. My gut feel says that moving from a middle of the road tax environment to a slightly better environment will not make a huge difference. A change to a zero tax state is probably not on the table.

We have a few years until I will take SS at age 70, so we have room in the brackets to do Roth conversions. Mentally, it is a challenge to make some big conversions and pay taxes. But I know it is the right thing to do. We may hold back this year, pending the possibility of establishing residency in a better tax state, but it is time to pay some of the deferred taxes.

 

[RunningBum]

I'm strongly considering the following similar but slightly different decision rule:

Convert enough this year so that my total marginal rate this year would be about equal to my predicted total marginal rate at age 70.5 if I were to continue conversions of similar magnitude between now and then.

This is how I look at it too.

I do know that there can be "tax humps" in the code (like the temporary 15% + 12% = 27% rate).

And the SS tax hump, IRMA, etc.

There are a lot of moving parts to try find an exact optimum answer.

There sure are.

So my thought is to compare the marginal rates, and if there is doubt, convert anyway. I feel like tax rates are more likely to go up than go down. I may not try to include those humps, but knowing they might be there make me more likely to convert.

If I were married, the chance of one of us dying and leaving the other with the single filing status would make me even more likely to convert.

Simply put, I see the taxes due on deferred income as a somewhat unknown. If I can convert at a reasonable rate, I'm going to do that. Better to take the known hit now than take the unknown hit that might be higher in the future.

For my own situation, looking to 2020, I'm likely to have enough income from sale of a townhouse my son has been using to take me out of an ACA subsidy, so I may convert up to the top of 24%, which will take out most of my tIRA. 24% isn't much higher than 22%, and it probably leaves me with a higher ACA subsidy in future years, and leaves me free to take SS anytime after 65 if the market drops without worry about conversions. I'm just trying to confirm I'm not being too eager to take out that deferred tax liability that looms over me.

 

[Midpack]

Mentally, it is a challenge to make some big conversions and pay taxes. But I know it is the right thing to do. We may hold back this year, pending the possibility of establishing residency in a better tax state, but it is time to pay some of the deferred taxes.

I think most here would agree, I do. I could have started conversions years ago, and didn’t because a) your mental challenge point and b) relying on simple Roth conversion calculators that said it would be “a wash” - though largely true, there’s a lot more to it than a simple ending balance calculator. Vanguard and others were no help. Until I could see the precise math for myself with all the moving parts laid out, I was stuck. All that’s in the thread I started a while ago.

It’s taken me over a month of looking at it from as many angles as I can comprehend, but I started converting and intend to continue up to the 22% bracket for the next 5-7 years, when DW and I will both be 70. No one can predict all possible scenarios, we can only go with what we consider most likely. IMO odds are we‘ll be better positioned over our retirement years, with some lesser chance we’ll do no better/no worse - acceptable outcomes. Harming our position by converting as planned seems far less likely. That’s my story and I’m sticking to it...

And I can stop converting any year in the next 5-7 if that looks better.

 

[RetireBy90]

I’m currently in 22% marginal and will convert to fill 22% and into 24% bracket starting this year. I expect to stay in 22% bracket from now on and like all I don’t know where fed income will go from here. However you normally pay a risk premium (30 year treasury yields more than 10 years) so it makes sense time to pay extra 2% to remove the term risk not to mention all the other reasons.

 

[pb4uski]

I'm not sure if IRMAA is a good reason to constrain... if we convert to the top of the 22% tax bracket in 2020 that is taxable income of $171,050 and assuming the standard deduction for 65 year olds of $27,400 that is income/MAGI of $198,450.

That income level would put us in the first IRMAA tier and cost us an additional $54.10/month for Part B and $12.40/month... or $1,596 for the two of for the year... not a big deal in the whole scheme of things.

I agree with Midpack's premise of converting to the top of the tax bracket that you expect to be in once all your income streams have started.... mainly because of the factor that if one of you passes the surviving spouse will get thrown into a higher tax bracket.

Another important issue in our case is state income taxes. We tentatively plan to move from an income tax state to a no income tax state in 2020 when we turn 65 and will only convert to the top of the 12% bracket until we are no longer subject to ~4.5% state income tax, then 22%.

If Congress doesn't act to further extend the 2017 TCJA rate decreases for 2026 and beyond we might even go into the 24% tax bracket in 2025.

 

[SecondCor521]

@pb4uski, I treat IRMAA as an additional, parallel tax. For me it's the equivalent of an additional 1.88% on top of the 24% federal rate. So not a big deal, but as long as I can model it reasonably, it's worth adding in to the analysis.

The bummer part about IRMAA, if I understand it correctly, is that the five tiers are quite wide, so if you enter the next tier even by $1, you get to pay the entirety of the next IRMAA tier premium.

 

[Gumby]

I'm strongly considering the following similar but slightly different decision rule:

Convert enough this year so that my total marginal rate this year would be about equal to my predicted total marginal rate at age 70.5 if I were to continue conversions of similar magnitude between now and then.

Total marginal rate means I consider all of my marginal tax rates now and then. Now includes federal income tax, state income tax, ACA PTC reduction, and FAFSA EFC increase. Then includes federal income tax, state income tax, and IRMAA.

If you are reaching up into the 24% bracket, don't forget the NIIT limit - $250k for MFJ -- which imposes a 3.8% surcharge on investment income.

 

[RunningBum]

If you are reaching up into the 24% bracket, don't forget the NIIT limit - $250k for MFJ -- which imposes a 3.8% surcharge on investment income.

I forget, do Roth conversions count toward this limit?

 

[pb4uski]

Yes, that was his whole point.... that Roth conversions can increase your MAGI to a point where this additional 3.8% tax on NII kicks in.

In our case most of our income is pension income (pension, tIRA withdrawals, Roth conversions) and our investment income (interest, dividends, capital gains) are not too significant... so our surcharge would be less than $1k.

But those with bigger taxable portfolios or income producing real estate should be wary of this tax in deciding on Roth conversions.

 

[SecondCor521]

Yes, that was his whole point.... that Roth conversions can increase your MAGI to a point where this additional 3.8% tax on NII kicks in.

In our case most of our income is pension income (pension, tIRA withdrawals, Roth conversions) and our investment income (interest, dividends, capital gains) are not too significant... so our surcharge would be less than $1k.

But those with bigger taxable portfolios or income producing real estate should be wary of this tax in deciding on Roth conversions.

Not my point, but it's true anyway. The general rule, of course, is to consider all applicable "AGI taxes". For me right now that includes a "FAFSA EFC" "tax" which will go away in a few years hopefully.

In my case, I'm just looking at RMDs + 85% of SS - standard deduction. At age 70 my investment income should be rather low if not zero, so I'm personally not concerned with NIIT.

 

[RunningBum]

Yeah, I guess so. I got thrown off by thinking about the last time I had to worry about NIIT, when I took a large cap gain. But the final touch was deciding how much to convert since the gain was already in the books. Too late to retract the question, I suppose.

 

[pb4uski]

Not my point, but it's true anyway. ...

I was referring to Gumby's post #18, not yours.

 

[Gumby]

I forget, do Roth conversions count toward this limit?

Roth conversions count toward your MAGI to determine whether you have reached the threshold of $250K. However, the NIIT is not applied to the conversion itself. See https://www.kitces.com/blog/how-ira...3-8-medicare-surtax-on-net-investment-income/
and
https://www.irs.gov/newsroom/questions-and-answers-on-the-net-investment-income-tax

 

[kaneohe]

I forget, do Roth conversions count toward this limit?

The NIIT is a 3.8% tax on the lower of net investment income or the amount AGI > threshold whichever is less............so if AGI (for MFJ) is 251K but investment income is 50K, NIIT is based on 1K, the amount by which AGI exceeds the 250K threshold. Roth conversions are not investment income but they are a component of AGI so can increase NIIT.

https://www.irs.gov/newsroom/questions-and-answers-on-the-net-investment-income-tax

I guess AGI above should be MAGI but for me the MAGI is AGI.