Deferral calculation on AGI, Taxable or marginal?

[Kings over Queens]

My wife and I deferred the bulk of our retirement money between 1995 and 2015. If I wanted to calculate or gauge the effectiveness of roth conversions, should I be comparing them against AGI, taxable income (line 43 pre 2017) or marginal at the time?

Put another way, assume we deferred $7,800 in 1997, and our AGI then was $58,212, and our taxable was $34,043, how do I compare that deferral when considering a roth conversation into today's (or tomorrows) 22% bracket?

There are significant differences between AGI and Taxable Income for those years.

 

[Sunset]

I just calculate what my RMD income with current IRA amounts + SS (age 70) + normal income for the year, will be at age 73 and see what marginal tax bracket I'll hit.

It's simplistic, but the IRA will grow and the tax deductions and brackets will grow, so using current values is close enough as anything else just introduces more possible errors.

If my expected marginal tax bracket is too high, meaning higher than currently, then any conversion now will save some tax dollars.

 

[Kings over Queens]

If my expected marginal tax bracket is too high, meaning higher than currently, then any conversion now will save some tax dollars.

I'm trying to set an historical benchmark.

Example, "I deferred 10,000 at X%, should I convert now at Y% or wait until potential Z% bracket." I need to solve X.

Is X based on AGI, taxable, or marginal?

 

[Koolau]

I did my conversions without calculations. Good on you for trying to figure this out. Be sure to post your final results for the young'uns. I could have used that 15 years ago. Aloha

 

[Exchme]

You shouldn't be making this decision comparing to possible past actions rates you might have experienced had you taken the income way back when, all you can affect now is the future. (Besides, if the money had been sitting it taxable for 30+ years, you would have lost ~10% of it to tax drag by now.) Your options now are to withdraw the tax deferred on schedule in RMDs/heirs withdrawing it within 10 years of passing, or you can level out our income by doing some Roth Conversions now, before RMD time, or you could make charitable contributions via QCDs.

There are calculator programs that will do the optimization for you, though if the program is thorough enough to help, it is complex enough to take some time to find your way around, set up and understand.

 

[Gumby]

Back when I was working in NYC and making the big bucks, we were at one point in the 39.5% federal marginal bracket and we did everything we could to avoid those taxes. Now we have the majority of our money in tax deferred accounts and need to choose whether to do Roth conversions before RMDs hit. I use today's numbers in my account with today's tax rates and assume a real rate of return going forward, because the relevant brackets are indexed to inflation. After running my models several times, I am convinced that I will turn out roughly the same whether I convert now or wait for RMDs. I will be doing some Roth conversions to build up my Roths just so we can have a sum of money we can take in an emergency without blowing up our tax situation, but I'm not obsessing about it.

I certainly saved on taxes by stuffing our tIRA, 401k, 403b and 457 accounts back then, but that is in the past, cannot be undone and has no bearing on what I should do going forward.

 

[Kings over Queens]

You shouldn't be making this decision comparing to possible past actions rates you might have experienced had you taken the income way back when, all you can affect now is the future.

Disagree, and here is why.

Deferring income is supposed to be done, or works best conceptually, when you defer at a higher tax rate so you can take it at a lower tax rate, right? If that is true, isn't it prudent to know the tax rate when deferred?

Assuming for a second that this money was deferred at 15%, doing a conversion at 22% or 24% today wouldn't be optimal. Granted, there are other factors, and the end result is that I might just have to take my lumps.

Hindsight is always 20/20, I get that.

 

[Koolau]

Back when I was working in NYC and making the big bucks, we were at one point in the 39.5% federal marginal bracket and we did everything we could to avoid those taxes. Now we have the majority of our money in tax deferred accounts and need to choose whether to do Roth conversions before RMDs hit. I use today's numbers in my account with today's tax rates and assume a real rate of return going forward, because the relevant brackets are indexed to inflation. After running my models several times, I am convinced that I will turn out roughly the same whether I convert now or wait for RMDs. I will be doing some Roth conversions to build up my Roths just so we can have a sum of money we can take in an emergency without blowing up our tax situation, but I'm not obsessing about it.

I certainly saved on taxes by stuffing our tIRA, 401k, 403b and 457 accounts back then, but that is in the past, cannot be undone and has no bearing on what I should do going forward.

Right. Regrets (I have a few) don't help. Learning from (my) mistakes can help. YMMV

 

[TheWizard]

Disagree, and here is why.

Deferring income is supposed to be done, or works best conceptually, when you defer at a higher tax rate so you can take it at a lower tax rate, right? If that is true, isn't it prudent to know the tax rate when deferred?

Assuming for a second that this money was deferred at 15%, doing a conversion at 22% or 24% today wouldn't be optimal. Granted, there are other factors, and the end result is that I might just have to take my lumps.

Hindsight is always 20/20, I get that.

There's something called Water under the Bridge.
What you said is true: better to defer money at higher tax rates and take it out at lower tax rates.

But table that thought.
Focus instead on this year and the next 5+ years ..

 

[Koolau]

There's something called Water under the Bridge.
What you said is true: better to defer money at higher tax rates and take it out at lower tax rates.

But table that thought.
Focus instead on this year and the next 5+ years ..

Exactly! What makes sense going forward? Do that and forget the past.

 

[Gumby]

....
What you said is true: better to defer money at higher tax rates and take it out at lower tax rates.
.....

Indeed and I can give you an example of when I put that idea into practice. In the twilight of my legal career, I left private practice and went into public service, which resulted in an immediate 77% pay cut. Our marginal tax rate also plunged in conjunction with that. At that point, I calculated that we would be in the 25% bracket after retirement and reasoned that it made no sense to defer at 25% if I was just going to turn around and take it back out at that rate. So we contributed to our 403b and 457 accounts in an amount sufficient to get ourselves from the 28% bracket down to the 25% bracket, but no more. (The relative rates changed after the 2017 TCJA but the reasoning did not).

Whether my reasoning then was sound or not is irrelevant because, as you note, it's water under the bridge - a sunk cost. The only thing that matters now is how to most tax efficiently get that money out of our tax deferred accounts, and that decision can only be made looking forward.

 

[Kings over Queens]

I called me CPA, he's never had anyone ask the question, and also, never had anyone admit to having every tax return they've ever filed. Guys, on some subjects, I'm basically "rain man" and the only way for me to make a decision looking forward is to solve the echo in my head by looking backward.

Anyway, he gave me the answer I needed, which is to calculate my effective tax rate for the years of deferral.

Appreciate all of the thoughtful replies and sharing of information.

 

[Koolau]

Indeed and I can give you an example of when I put that idea into practice. In the twilight of my legal career, I left private practice and went into public service, which resulted in an immediate 77% pay cut. Our marginal tax rate also plunged in conjunction with that. At that point, I calculated that we would be in the 25% bracket after retirement and reasoned that it made no sense to defer at 25% if I was just going to turn around and take it back out at that rate. So we contributed to our 403b and 457 accounts only in an amount sufficient to get ourselves from the 28% bracket down to the 25% bracket. (The relative rates changed after the 2017 TCJA but the reasoning did not).

Whether my reasoning then was sound or not is irrelevant because, as you note, it's water under the bridge - a sunk cost. The only thing that matters now is how to most tax efficiently get that money out of our tax deferred accounts, and that decision can only be made looking forward.

I always went for the company match and then started thinking about whether it made sense to defer any more. I'm sure I made mistakes as I'm STILL trying to lower my 401(k) total to lower RMDs. Too soon old, too late smart but great problems to have in the great scheme of things.

 

[N02L84ER]

I would also focus on what makes sense going forward, but if you wanted to calculate what the effective tax was when you deferred the IRA/401k money you would take the Taxable Income, add back in the IRA/401k contribution, calculate the tax on that sum, subtract the income tax that was calculated the year of the contribution, and then divide that tax amount by the amount of your contribution. This ignores any matching money on 401k contributions. This is a lot of work for no actionable gain.

 

[Gumby]

In all my working years, I never had a company 401k match, so I never had to choose whether to contribute a non tax optimal amount just to collect the match.

 

[SecondCor521]

I called me CPA, he's never had anyone ask the question, and also, never had anyone admit to having every tax return they've ever filed. Guys, on some subjects, I'm basically "rain man" and the only way for me to make a decision looking forward is to solve the echo in my head by looking backward.

Anyway, he gave me the answer I needed, which is to calculate my effective tax rate for the years of deferral.

Appreciate all of the thoughtful replies and sharing of information.

Unfortunately your CPA is wrong (or you misunderstood him, or I'm misunderstanding you).

The correct answer is to go back to those old tax returns, add the $10K of income (or whatever amount you're looking at Roth converting now), and see what the effective tax rate would have been on that $10K.

This might correspond to the marginal rate bracket you were in at the time, but it can be complicated by things that phase in and phase out, such as ACA subsidies, child tax credits, education credits, Social Security taxation, Pease limitations, charitable AGI limitations, AMT, NIIT, IRMAA and a thousand other things. (*)

As I mentioned on the other thread, though, you've essentially already confirmed that converting now is a win compared to recognizing the income back then. There is still the question of whether you can make it a bigger win by converting or withdrawing at an even lower rate in the future. This bigger win idea does require guessing about the future, which may be what you want to avoid doing by taking the sure win now. You might also not need the bigger win because your finances now might be "good enough". But even so, the possibility of a bigger win still exists.

(*) Some of these wouldn't apply to your working years back then, either because they didn't exist or wouldn't apply; I'm just mentioning them as typical examples.

 

[Out-to-Lunch]

I called me CPA, he's never had anyone ask the question, and also, never had anyone admit to having every tax return they've ever filed. Guys, on some subjects, I'm basically "rain man" and the only way for me to make a decision looking forward is to solve the echo in my head by looking backward.

Anyway, he gave me the answer I needed, which is to calculate my effective tax rate for the years of deferral.

Appreciate all of the thoughtful replies and sharing of information.

FWIW, Bogleheads disagree with your CPA: Traditional versus Roth - Bogleheads

 

[Exchme]

Disagree, and here is why.

Deferring income is supposed to be done, or works best conceptually, when you defer at a higher tax rate so you can take it at a lower tax rate, right? If that is true, isn't it prudent to know the tax rate when deferred?

Assuming for a second that this money was deferred at 15%, doing a conversion at 22% or 24% today wouldn't be optimal. Granted, there are other factors, and the end result is that I might just have to take my lumps.

Hindsight is always 20/20, I get that.

Looking back is as productive as kicking yourself for not buying Apple stock way back when. It's just not relevant to the choices in front of you now.

 

[Koolau]

In all my working years, I never had a company 401k match, so I never had to choose whether to stay in a higher bracket to collect the match.

Bummer!

Megacorp "promised" us 50% match up to 6% of salary though they often gave us 100% (in good years.) Unfortunately, much later, (after the golden handcuffs were cinched up good and tight) they did away with the match for several years during the "lean" times. Instead, they gave us (in lieu of salary) stock options which (of course) were out of the money when they vested (thanks Megacorp.)

Good news is that all those older stocks in the 401(k) I accumulated (and didn't sell) are now worth a small fortune. It's funny how things w*rk out.

 

[Kings over Queens]

The correct answer is to go back to those old tax returns, add the $10K of income (or whatever amount you're looking at Roth converting now), and see what the effective tax rate would have been on that $10K.

Ultimately we arrived at the same conclusion. I just finished the spreadsheet that shows me the effective rate of taxes paid as well as paid as a percentage of AGI. Now I'll add back in the deferral to see how those change, and it gives me one (of many) factors to consider going forward.

 

[SecondCor521]

Ultimately we arrived at the same conclusion. I just finished the spreadsheet that shows me the effective rate of taxes paid as well as paid as a percentage of AGI. Now I'll add back in the deferral to see how those change, and it gives me one (of many) factors to consider going forward.

Since tax laws have changed quite a bit between then and now, I doubt you'll be able to do so with any precision unless your tax situation was extremely simple. Which is unlikely given the marginal rates you mentioned.

Also, water under the bridge, as many others have been trying to point out.

 

[oldtimer]

Mike Piper gave an excellent presentation at this year's Bogleheads Conference. Both the "Roth Conversion: a Deep Dive, with Mike Piper" presentation & his slides are accessible at 2024 Conference - The John C. Bogle Center for Financial Literacy

 

[Kings over Queens]

Mike Piper gave an excellent presentation at this year's Bogleheads Conference. Both the "Roth Conversion: a Deep Dive, with Mike Piper" presentation & his slides are accessible at 2024 Conference - The John C. Bogle Center for Financial Literacy

Thanks @oldtimer just seeing this now and checked out his slides. really appreciate this.

 

[TheWizard]

Thanks @oldtimer just seeing this now and checked out his slides. really appreciate this.

One good point that Mike makes is often ignored.
And that is: retirees with zero earned income cannot legally contribute to a Roth IRA.
But they can EFFECTIVELY do the same thing with a Roth conversion.

Example: $10,000 in a tIRA for me is subject to 24% + 5% tax when withdrawn, netting $7100.
But if I Roth convert $10,000 and pay the $2900 tax from excess income in my checking account, then I've EFFECTIVELY made a $2900 Roth contribution for that year...

 

[pb4uski]

I'm trying to set an historical benchmark.

Example, "I deferred 10,000 at X%, should I convert now at Y% or wait until potential Z% bracket." I need to solve X.

Is X based on AGI, taxable, or marginal?

What you deferred at, or the effective rate of taxes avoided on deferrals is interesting but not really relevant to the Roth conversion decision. Typically, taxes paid on a Roth conversion will be a lower effective rate than the effective rate on the deferral.

However, the relevant comparison for the decision is the effective rate on the conversion compared to the effective rate on RMDs if you don't do the conversion. If the former is lower than the latter, then conversion is beneficial.

For me as an example, most of my referrals were where my federal rate was 28% and 6% state, so 34% in total. The Roth conversions since I've retired are generally 9-11% federal and 0% state since we moved to FL and now TX in retirement. So on those conversions we got a 23-25% benefit.

Once I start RMDs I expect them to be a combination of 12% and 22% brackets, so ~18% blended. So I convert to the top of the 12% tax bracket but not into the 22% bracket as there is no sense to pay 22% now vs paying 22% later. So, I save 7-9% on conversions (18% vs 9-11%).