Roth Conversions at 15% or 25% Tax Bracket?

Felix Mulier

Recycles dryer sheets
Joined
Feb 4, 2013
Messages
99
I just spent the last week coming up to speed on the Retiree Portfolio Model from BigFoot48 on Bogleheads. I think it's a great tool (but my head hurts!):cool:.

My original goal was to see if we should stay under the ACA Cliff for the next few years while we convert our tIRAs and 401K to Roth IRAs. The answer, for us, turns out to be "no" - it's better if we convert more sooner than later.

Given that, I ran 2 more scenarios; 1 maxing out (i.e. 95% of) the 15% tax bracket, the other at 25%. I know I have to look into AMT and Medicare Part B Premium increases for Option #2, but here are some Pros and Cons for the 2 scenarios.

I'd appreciate any questions and/or comments as to which option you'd choose if you were me, and also if there are other things than those I already mentioned that I should investigate further. Thanks in advance!

Option #1 - Max out 15% Tax bracket
Pros
  1. Total taxes will be 9% less than those we'd pay in Option #2 (and who wants to pay more taxes, right?):cool:
  2. According to RPM, we would never have to "dip into" our Roth Conversion account. It would just continue to grow. (But we have no kids, so - hmmm!)
  3. Year-end Total Portfolio balances outpace Option #2's balances for the first 19 years (by as much as 8.5%)
  4. The final balance would be ever so slightly higher than Option #2 (<1%)
Cons

  1. We would pay income taxes for the next 16 years (Roth conversions would take that long), but then no taxes :dance:
  2. Our Social Security (at age 70) would be taxable for the first 7 years
Option #2 - Max out 25% Tax bracket

Pros

  1. We would pay (BIG) income taxes for the next 8 years (half as long as option #1), but then no taxes :dance::dance:
  2. Our Social Security would never get taxed
Cons

  1. While we'd only pay taxes for the next 8 years, we'd pay 9% more in Total Taxes than we would pay in Option #1
  2. The Total Portfolio takes a BIG hit in the first 8 years and doesn't catch up to Option #1's portfolio for 19 years :(
  3. The final balance would be ever so slightly lower than Option #1 (<1% if we live long enough!)
  4. We'd start using up Roth $'s 8 years sooner than Option #1
All things being equal, it may come down to how well we can sleep at night with Option #2's whopping tax bills!
 
Last edited:
With Option 2 you say your Social Security would never get taxed. I assume that means your bracket then would be lower than 25%. The general rule is don't Roth convert to max out the 25% bracket now if you expect your bracket later to be lower than 25%.
 
You are making my head hurt thinking about this!!

Obviously a first world problem.

In our case, we cannot lower the RMD enough to make a real difference, so we are going for simplicity. DW will use the 15% bracket to move her lower IRA to Roth, We will then draw off of mine, and when the time comes we will probably be in the 25% bracket, with RMD's (or what ever there is at that time).

While I would love to minimze the tax dollars we spend, i can't worry about every last dollar.
 
i-orp will let you run your scenarios, and it would probably tell you it was "optimal" to convert more sooner. But because the optimization treats the length of the model as fixed, many users felt that they'd rather not pay a bunch of taxes early 'for sure', and take their chances that they might regret living so long, lol! Anyway, the i-orp model now allows limitation of Roth conversions, which gives a sub-optimal result, but one many people are more comfortable with.

I tried to reconcile the results from RPM and i-orp. I walked away with the same game plan from each, but couldn't get the numbers to align very closely. But if I were you, I'd run i-orp with your various scenarios and see how that looks.

EDIT: I notice that Pat has added an ORP comparison tool in RPM now! That's nice. My chat with Pat and work to reconcile was in the spring of 2015, and in the fall of 2015, the comparison tool was added to RPM. What I like about RPM is you can trace formulas (if you have the time, inclination and skill), whereas with i-orp, many of the calculations are going to be opaque with no way to override (specifically the tax calculations).
 
Last edited:
You are making my head hurt thinking about this!!
....
While I would love to minimze the tax dollars we spend, i can't worry about every last dollar.

Depending upon my mood at the time, I'll run various scenarios on i-orp, see that they're really not very different in terms of overall results, and then throw up my hands and say "it doesn't really matter".

We won't talk about how my head feels after a day when I'm in the mood to tinker with every possibility, lol!
 
..................................... The general rule is don't Roth convert to max out the 25% bracket now if you expect your bracket later to be lower than 25%.

I think the rule is don't convert at 25% if you expect your bracket later to be
lower than 25% in the absence of conversion.
 
To do your analysis, you've obviously had to make assumptions about the growth of your portfolio. My question in these cases is always: "What if my portfolio performs much worse than I've assumed?" (Obviously, you'll have no problems if your portfolio performs better than you'd planned for). If our portfolio performs much worse than anticipated:
1) I'll need every dollar to assure we get to the finish line, AND scarce funds in the future will have even higher marginal utility to us than they do today. In this case, I would very much regret giving Uncle Sam some of my money early.
2) My marginal tax rates in the future will be lower than anticipated (because my income will be lower than anticipated). This, again, is a reason not to pre-pay taxes.

For these reasons, I always tend to be conservative with these pre-payment of tax questions. So, I would choose the Option 1 in the OP. I'm not trying to maximize the number of dollars I will get if things go exactly "according to plan." Instead, I'm trying to maximize my expected marginal utility across a wide range of possible (but presently unknowable) investment outcomes. If we are surprised by outstanding investment results and we have to pay more taxes--that's a great problem to have! If we get the exact returns I'd expected, then we are covered. But if our investment results are poor, then underpaying taxes now and keeping that money as a badly needed cushion to get us through tough times will be really important.
 
Last edited:
I think it was you, samclem, that pointed out that nursing home expenses are tax deducible, so if in the last years, one may find their tax bracket to be lower than anticipated, in spite of RMDs.
 
My original goal was to see if we should stay under the ACA Cliff for the next few years

People are still thinking that ACA will still be around?
 
I'll have to look into RPM.

I built my own model when I first retired and decided to convert to the top of the 15% tax bracket. While i-orp and even my own model suggest that I might be better off converting into the 25% tax bracket, I just can't bring myself to do it, particularly since once I go over the top of the 15% tax bracket my marginal rate is 30% (more ordinary income taxed at 15%, and pushed up capital gains taxed at 15%).

Who knows what the new administrations tax reform will be, but if rates for my income level go down then I may be happy that I limited my Roth conversions. Time will tell.

Like you say, a first world problem.
 
I wouldn't make any long-term decisions now based on future tax rates assumptions & the ACA.
 
I've been on the fence re: Roth conversions for 5 years, as most calculators seemed to show the lifetime tax liability as a wash, including i-ORP. Converting to the 15% limit seemed like a no-brainer, but DW's income plus our div/STCG income puts us at the 15% limit so there's been no room for Roth conversions. OTOH, our current tax liability is crazy low, and I realize we're going to get clobbered when we hit 70 and RMD and Soc Sec kick in.

For whatever reason when I run full i-ORP today it shows if I convert to the 25% bracket, I can avoid almost $300K in lifetime taxes and enjoy a minor increase in spending. So I guess I'm in the 'makes my head hurt camp' since today's runs seem quite a bit different than earlier runs, and no two answers have ever been the same in the past :confused:

Since i-ORP is a black box, I finally got frustrated enough with myself for sitting on the fence re: Roth conversions for so long, that I've asked VG to develop a comprehensive withdrawal plan. So I am really looking forward to seeing those results and the accompanying discussion. [Turns out it's free with Flagship status, something my rep never mentioned before, even though I'd asked several times in past years :mad:]

There's no absolute answer, but I'd like to go with the odds based on assumptions I'm comfortable with.

I may look at RPM too, but as I recall I did look at it a while ago, and it's a black box too. At least VG is a known quantity.
 
Last edited:
[snip] I've asked VG to develop a comprehensive withdrawal plan. So I am really looking forward to seeing those results and the accompanying discussion. [Turns out it's free with Flagship status, something my rep never mentioned before, even though I'd asked several times in past years :mad:]

Hey, that's good to know! Did you initiate that with a phone call or something online?

I may want to take advantage of that, also.

Thanks!
 
Thanks for your responses. We've decided to go with option #1 and stay within the 15% tax bracket. I still plan to run ORP, but am not expecting any surprises there.

samclem said:
If we are surprised by outstanding investment results and we have to pay more taxes--that's a great problem to have! If we get the exact returns I'd expected, then we are covered. But if our investment results are poor, then underpaying taxes now and keeping that money as a badly needed cushion to get us through tough times will be really important.
Excellent point. I actually graphed out the Total Portfolio balances for the 2 options and the gaping chasm between them in the early years gave me considerable pause!

GrayHare said:
The general rule is don't Roth convert to max out the 25% bracket now if you expect your bracket later to be lower than 25%.
kaneohe said:
don't convert at 25% if you expect your bracket later to be lower than 25% in the absence of conversion.
This is a great guideline. Without conversions, we'd be in the 15% bracket.

OTOH, our current tax liability is crazy low, and I realize we're going to get clobbered when we hit 70 and RMD and Soc Sec kick in...I may look at RPM too, but as I recall I did look at it a while ago, and it's a black box too.

I highly recommend trying RPM. If you are familiar with Excel, it is not nearly as black a box as I find ORP to be. I printed out the 4 page "Full Case" analysis for the options I ran. (Got out my trusty scissors and tape and put them together!)

Then, I took the 1st year and said "These are my expenses. Now, where is the money coming from to cover them?" Went through that for every 5 years to get a good feel for it. Repeated for each option. It was a valuable exercise.

I could see exactly where e.g. money had to be moved from my 401K to cover living expenses and tax payments; where RMDs kicked in and for how long; where SS would get taxed - you get the idea.

Granted, it took a week to get comfortable with the tool and to stop feeling as if I were drinking out of a fire hose, but now that I understand it, I feel really good about being able to do "what-if" scenarios from here on out. Also, as real life happens, I'll know how to make adjustments. If I'm gonna get clobbered, I rather see it coming!

P.S. Glad to help with the RPM learning curve if I can.
 
I'll have to look into RPM.

I built my own model when I first retired and decided to convert to the top of the 15% tax bracket. While i-orp and even my own model suggest that I might be better off converting into the 25% tax bracket, I just can't bring myself to do it, particularly since once I go over the top of the 15% tax bracket my marginal rate is 30% (more ordinary income taxed at 15%, and pushed up capital gains taxed at 15%).

Who knows what the new administrations tax reform will be, but if rates for my income level go down then I may be happy that I limited my Roth conversions. Time will tell.

Like you say, a first world problem.

There's a limit to that 30% marginal rate. At some point you've pushed all of your cap gains into being taxed, at which point the marginal rate drops back to 25%. If you haven't already looked at this, you may want to consider how long you get taxed at 30%.
 
Yes, I realize that...my marginal tax rate will be 30% until all QDI and LTCG are pushed into the 25% tax bracket (our until my ordinary income equals the top of the $75,300 top of the 15% tax bracket) and then any additional Roth conversions would be at 25%.

I would almost have to quadruple my Roth conversions to get back to that 25% incremental rate as my qualified income is 93% of my taxable income.
 
A couple of thoughts from my own calculations:

a) You probably don't want to go all the way down to nothing in your tIRAs/401k. Just get the balance down far enough that RMDs stay mostly within the 0% to 10% tax bracket for you. There's no good reason to pay 15%/25% now when you can pay 0% to 10% later.

b) Your optimum may be converting partially into the 25% bracket. No reason you have to max it out all the time. However, I suspect that since you can fully convert using the 15% bracket only, there is no need to enter the 25% bracket at all.

c) 16 years of Roth conversions is a long time. Do you have a taxable account balance or other income that can support you and pay Roth conversion taxes during that entire time? Generally it is not beneficial to Roth convert if you are using tIRA withdrawals to pay for it.

d) Converting after age 70.5 means RMDs must be withdrawn first and cannot be Roth converted. Ideally your conversions would be complete before then, since keeping RMDs from pushing you into the 25%+ tax bracket at age 70.5 is your target goal.

e) Consider adjusting tIRA balances by some average tax factor when calculating net worth. You know Roth IRAs are more valuable dollar for dollar than tIRAs, but looking only at pre-tax balances makes Roth conversions look crazy. Your goal is to withdraw each dollar from tIRA/401k at the lowest tax rate you can, now or in the future. If that means you withdraw tIRA money that you don't currently need, then it gets Roth converted instead of spent.

f) I'd double check the ACA subsidy calculations. In my case one year it was worth it to me to skip a $2500 education tax credit only if I Roth converted $40k or more (calculating how much we could spend yearly during retirement). ACA subsidies can be worth a lot more than $2500. I would think the subsidy would win over Roth conversions for many. For 2017 I would have been considering getting the subsidy, but we will probably be in an off-Marketplace plan that has a better network for us.

g) All things being equal (yearly spending is the same for both options), since both options end up with only Roth balances, I'd pick the one with the largest balance at the end. Or the one that allows the highest yearly spending level while the ending balances are equal.
 
Hey, that's good to know! Did you initiate that with a phone call or something online?

I may want to take advantage of that, also.

Thanks!
Sent a message to my designated rep, and he told me I was entitled to a free consultation with one of their CFPs. Wish I knew why he didn't tell me that the times I'd asked before over the past few years? That said, shame on me for not looking into it further myself...
 
iORP told me to be much more aggressive with conversions than I could bring myself to do and because of our personal circumstances I wish I had done more. In year 1 I converted all my tIRA since it had a very large basis (the following year I rolled over my 401k so had a large IRA with zero basis. I also converted some of DW's IRA since about 15% of the total of her IRA was after-tax. The next 2 years I concentrated on converting DW's IRA but again nowhere near as much as iORP recommended.

Then we made the decision to split our time evenly between the UK and the US starting in 2016, and when I looked at how the UK tax system worked I discovered that each person was taxed individually (wasn't like that when we lived there 25 years earlier). Because of my pensions it meant that I was going to be in the 20% bracket and move into the 40% bracket at 70 when I started taking SS. This means that RMD's will be taxed, for me, at 40% so this last 3 years I have been converting up to the limit of the 25% bracket.
 
Midpack: I hold a seven-figure amount of Vanguard funds, but on the Schwab platform. I know this is just a hair off-topic (apologies to moderator), but does anyone know if I am or could become flagship even though these funds are custodied at Schwab?
 
My original plan, as commented above, was to convert to the top of the 15% bracket. Now I am having second thoughts about paying an extra $6k in taxes.

At 70, we will be in the 25% bracket. And we will pay taxes on 85% of our SS. Short of a major market downturn that does not come back. So, maybe a good strategy is to minimize taxes for the next 9 years?

At this point, it appears any long term taxes I "save" will be for the benefit of my DS, since we would likely never touch the Roth IRA.

Likely I will compromise and convert less than planned, with the objective still to get the DW's IRA fully converted before RMD's. Like I said above, this is more for simplicity, since only one account will be subject to RMD's.
 
Midpack: I hold a seven-figure amount of Vanguard funds, but on the Schwab platform. I know this is just a hair off-topic (apologies to moderator), but does anyone know if I am or could become flagship even though these funds are custodied at Schwab?

Doubtful.
Eligibility is based on total household assets held at Vanguard, with a minimum $1 million to qualify for Vanguard Flagship Services®. We determine eligibility by aggregating assets of all qualifying accounts held by you and your immediate family members who reside at the same address.
Assets that qualify
Vanguard mutual funds.
Vanguard ETFs®.
Certain annuities through Vanguard.
The Vanguard 529 Plan.
Certain small-business accounts.
 
My original plan, as commented above, was to convert to the top of the 15% bracket. Now I am having second thoughts about paying an extra $6k in taxes.

At 70, we will be in the 25% bracket. And we will pay taxes on 85% of our SS. Short of a major market downturn that does not come back. So, maybe a good strategy is to minimize taxes for the next 9 years?

At this point, it appears any long term taxes I "save" will be for the benefit of my DS, since we would likely never touch the Roth IRA.

Likely I will compromise and convert less than planned, with the objective still to get the DW's IRA fully converted before RMD's. Like I said above, this is more for simplicity, since only one account will be subject to RMD's.

Are you saying to not even convert to the top of 15%? I would take full advantage of 15% now over 25% later, especially since once converted you can grow it tax free, whereas you've had 9 more years of growth before paying 25%.

If you're saying not to go above 15%, I'd probably agree. It's a guess whether taxes will stay the same in the future, or which direction they'll go for the various income types. Plus, as others have noted, the first part above 15% is usually at 30% rather than 25%.
 
I'm converting to the top of the 15% tax bracket and paying a hair over 2% of the amount I convert in tax.

How's that you ask... I have about $70k of QDI & LTCG that end up not being taxed at all, $8k of ordinary income (interest, unqualified dividends, pension) and $26k of Roth conversion (which is also ordinary income). $20k of itemized deductions and $8k of exemptions reduce the total of $34k of ordinary income to $6k, which is taxed at 10% so a tax of $600 on $26k of Roth conversions... sweet. (Based on MFJ).

Unfortunately, state income taxes increase the 2% to 9%, but still....hard to beat that. I'd be crazy not to convert.
 
Last edited:
The main benefit of a Roth conversion is that all the money in the Roth is yours, but in a tIRA 15%-25% is claimed by the IRS so only 75%-85% is yours.
Which means that if you pay the tax on the conversion from the amount converted, there is no net benefit -- assuming the same tax rate. But if you pay the tax from outside funds, the net effect is that you effectively add money to the Roth.

But if your tax rate later will be more than your tax rate now, it is good to convert now. For example, if you convert now at Joint rate but you think that a withdrawal in the future might be at Individual rate, like if your spouse dies before then. Or if the eventual RMD's will push you into a higher bracket.

But a lot of this tax planning this year may well be moot, if/when (and how much of) Trump's new proposed tax schedule comes to pass.
https://www.kitces.com/blog/end-of-year-2016-tax-planning-for-2017-trump-income-tax-reform/

Likewise all the planning on optimizing income for ACA subsidy. That is surely going to change bigly, no matter what. Would have had to change no matter how the election turned out.
 
Back
Top Bottom