FREE NEWSLETTER
I’M NOT IN THE HABIT of celebrating half-birthdays, but my next one has me thinking. In a few days, I’ll turn age 59½.
That, of course, is the age at which you can tap your retirement accounts without paying the 10% early withdrawal penalty. Though I don’t currently need to pull spending money from my retirement accounts, I like the feeling that I can now do so penalty-free.
Even without that 10% penalty, however, there’s still the small issue of income taxes. The good news: More than a fifth of my investment portfolio is in tax-free Roth accounts, with another tenth in a regular taxable account. The bad news: The other two-thirds are in a traditional IRA. Every dollar coming out of that traditional IRA will be dunned at ordinary income-tax rates.
That prompted me to do a quick, back-of-the-envelope calculation. If I put off all traditional IRA withdrawals until they’re required at age 72, and I assume modest investment gains, and I add in Social Security benefits and other income, I’ll likely find myself near the top of the 24% federal income-tax bracket when I’m in my 70s. A reliable estimate? Given all the variables involved, I consider it more of a rough guess.
Indeed, based on current tax law, there’s a possibility I could end up paying a 32% marginal rate—and there’s a decent chance tax laws will be different a dozen years from now, especially with parts of 2017’s tax law scheduled to sunset at year-end 2025. On top of that, I’ll most likely have to pay higher premiums for Medicare Part B and Part D, thanks to so-called IRMAA surcharges. For me, those surcharges could amount to an extra tax equal to 2% or 3% of income.
Meanwhile, this year, it looks like I’ll land in the lower part of the 24% marginal federal tax bracket. The upshot: I figure there’s some incentive to shrink my traditional IRA over the next few years, so there’s less risk I’ll end up paying higher taxes—and heftier Medicare premiums—later on. To be sure, by opting to draw down my traditional IRA before I’m compelled to, I’ll be inflicting large tax bills on myself. That’s hardly a pleasant prospect.
Still, it strikes me as the rational thing to do. To that end, now that I’m turning age 59½, I could start pulling money penalty-free from my traditional IRA, and continue to do so throughout my 60s. But given that I don’t need the spending money right now, it makes more sense to convert chunks of my traditional IRA to a Roth each year, where the money will then grow tax-free. Keep in mind that you don’t have to be age 59½ to do a Roth conversion. Indeed, I’ve done a few over the years, including a big one in 2010, when I converted my nondeductible IRA.
With the Roth conversions I’m currently planning, my goal is to make the most of the 24% income-tax bracket, but try mightily to avoid generating so much extra income that some of it gets dunned at 32%. The incentive to shrink my traditional IRA is especially great this year and in the three years that follow.
Why? There are two reasons. First, in 2026, I turn age 63, meaning I’ll be two years from claiming Medicare. At that point, any extra income I generate has the potential to boost my Medicare premiums, because the IRMAA surcharges are based on your tax return from two years earlier. Second, in 2026, parts of today’s tax law sunset and, at that juncture, it may take far less income to land in a high tax bracket.
My plan: I’ll convert $60,000 now. Later in the year, when I have a better handle on my 2022 income and how close I am to the top of the 24% tax bracket, I might convert another $10,000 or so. Why not just wait until late 2022 and do a big conversion then? I don’t know whether the stock market will be depressed later in the year—but I know it’s depressed right now. I like the idea that the $60,000 I convert might bounce back with the broad stock market, and that those gains would be tax-free.
Jonathan Clements is the founder and editor of HumbleDollar. Follow him on Twitter @ClementsMoney and on Facebook, and check out his earlier articles.
Want to receive our weekly newsletter? Sign up now. How about our daily alert about the site's latest posts? Join the list.
I started looking at this many years ago, especially with the Trump tax cuts and the market that did so well during his presidency. I started with a spreadsheet that got more and more complicated each year as I plan an upper end 30 year retirement. The result is that I am making Roth conversions to fill in my highest tax bracket while staying under 182,000 where penalties kick in for Medicare. Even assuming now 4.1% inflation for my particular situation, the RMD’s are a killer if you don’t average out distributions over a longer time period. The divider in the Uniform Lifetime Table for RMD’s is wicked in the end. It also doesn’t help that most likely you or your spouse end up in the single tax bracket in the latter part of retirement.
In your shoes, Jonathan, I think my inclination would be to analyze it a bit differently. We’re talking about 8 extra points of tax on some amount of money, correct (i.e. 32% marginal rate vs 24%)? So, to me, the question would not be the marginal rates, but what happens to my effective rate (i.e. total tax divided by total taxable income). If it moves just a point or two, I think I’d stick with continued tax-deferred compounding, as one would need to impute a compounded value of the investment growth lost to the early payment of taxes to determine the economic result (or, just keep compounding, capture that growth, and not worry about a small increase in the effective tax rate, given the income you’ll be enjoying in retirement).
Several years ago, I also was facing this dilemma. Two things happened:
First, I read a commenter in Humble Dollar who said the IRA/Roth IRA rules were set up so that the end results (taxes paid, investment balances, etc.) would be about the same no matter which direction you went. I found that hard to believe because I assumed the tax-free accumulation in a Roth IRA would create much better results over the long haul.
Second, in order to prove my assumption correct, I went to my trusted old friend, Excel. I developed a spreadsheet, which then went through multiple upgrades as I kept adding more and more parameters for tax rates, growth rates, and income rates for different times in the future. I went out 20+ years and assumed I died at age 85. At that time, my children inherited my IRA and/or Roth IRA. Then, for the next ten years, they had options on how they emptied the account(s). To be honest, there are many more parameters I could have added, but did not.
The result: It did not make much difference; just like the commenter in Humble Dollar said. The conversion of IRA monies to a Roth IRA did look better, but the impact over 30+ years seemed almost insignificant. In fact, I believe some of the parameters I did not bother to incorporate (IRMMA, NIIT) would have narrowed the difference even more.
Regardless, in 2020 and 2021, I converted IRA monies to my Roth. My old friend Excel did help me determine how much to convert in order to stay in the 24% tax bracket and out of the 32% bracket. Writing those checks for quarterly estimated taxes made me miserable, and that misery added to my feeling of “Is it worth it?”
I do not plan on converting any IRA money to my Roth this year, UNLESS the stock market drops so significantly that I can move a lot of more shares for the same tax dollars.
It does not make much difference as long as one assumes consistent tax rates. If rates go up in the future, then Roth becomes the clear leader.
As one of my parameters, I played with different tax rates, especially for the day when my children would inherit an IRA. Of course, no one knows if I went high enough ! ! !
I agree with you that the Roth conversion did come out ahead in terms of dollars. My problem was the angst of having to pay large quarterly taxes and the cash flow issues they caused, in comparison to the surprisingly small dollar gains way down the road.
Currently going through this very thought process. One item to remember for others who may as well is the net investment income tax (NIIT). Depending on your deductions, it’s possible your conversion will cause you to hit the NIIT threshold, which is based on MAGI, before you hit the ceiling of your target tax bracket, which is based on taxable income, which would increase the marginal tax rate on the conversion by 3.8%.
Yes the IRA distribution (whether or not converted to a Roth) raises the MAGI as it is ordinary income, but no the IRA distribution is not subject to the NIIT. It would only be actual investment income that would be subject to the NIIT to the extent tha the 250K threshold is breached.
You have a problem all of us would love to have. The only way to accurately assess whether to pay those income taxes at 72 or when one is 59 is to calculate the present value of your tax bill(federal and state) if you convert to a Roth when 72 versus converting to the Roth now.
Love these threads and commentary. Based on my studies and analysis, there have typically been two scenarios where the Roth conversion did not work.
the benefit for Roth is to be able to control distributions, have tax free withdrawals, and to gift tax free assets to heirs if legacy is s concern.
My philosophy has always been a tax delayed is a tax not paid. I understand the NPV calculation of the tax rate arbitrage, but a lot can happen between now and then. If I turn out to be in higher tax bracket in my seventies than I am now, I will celebrate.
Does it make any sense for a high earner in top tax bracket to do this with possible taxes rising in future? I have maybe another 12-20 years before I will stop working.
You should check out this article, which lists some of the key questions to ask yourself:
https://humbledollar.com/2020/05/to-roth-or-not/
What until you are approaching 62 and realize you can apply for social security. Yes, I know, it’s not FRA. That’s another “milestone” birthday. That will be me this time next year. For me hitting 62 was a sigh of relief. No, I didn’t quit my job, but I knew there was that plan B if I needed it. Work is so much less stressful when I don’t have to worry about “what’ll I do if I lose my job.”
Your concern at 62 may not be income, but what I find is that retirees overarching concern is healthcare. 65 is a significant milestone to offset rising healthcare cost between 62-65.
Here is a very tactical process for annual decisions about the complex brew of factors discussed in the article and comments. The idea is to understand the impact of tax brackets on these decisions. Create a small spreadsheet, starting with all of the tax brackets (this year) and account for the “add-ons” such as the 3.8 percent Net Investment Income Tax, etc. Then estimate income, capital gains & losses, taxfree-muni income. I separate Social Security because taxes are a function of income/AGI. Then calculate the maximum amounts that you can utilize within each tax-bracket. Example: Brackets of 84K, 178k, 340k, 432k, 648k; your taxable income is about 100k. You can “use” up to 78k within your current tax bracket (marginal rate), 162k in the next bracket. This tells you to withdraw no more than 78k from an IRA at one marginal rate and any more at the next, etc.
Rather than convert a fixed dollar amount to a Roth each year, I carefully look at how much, maximum, I can convert in each tax bracket and then decide. Looking to withdraw funds in later years? Withdraw up to $X in a low bracket and then some from a Roth, if needed. The Roth withdrawl is saving marginal higher tax if it were from an IRA. That is what I mean by a “tactical” process that helps me optimize annually.
My personal spreadsheet now includes my state income tax brackets and deductions, such as out of state munis, plus the effect on Medicare premiums. Medicare uses “MAGI” (add back muni income) greater than 176(182?)k from 2 YEARS AGO, so I project the tax effects for a few years in the future. A huge lump sum Roth conversions puts you in progressively higher tax brackets this year and higher premiums in two years. The 3.8% Medicare NIC kicks in at 200k. Lots of different “bracket” effects for the spreadsheet to keep track of. But easy to do “what if” scenarios to consider different alternative.
Keep the spreadsheet for next year, as rates, brackets, etc. change annualy. Easy to maintain once you have it setup for your situation.
The calculus of comparing the regular IRA vs Roth options depends quite a bit on your estimates – mainly how long you will live and how you think the market will perform in the future. A flat market that persists for a few years can prevent a Roth conversion from making enough of a comeback to justify it.
Also, most of the comments I’ve read talk about the beneficial impacts of moving between the 12% and 22% or 24% range. The promise of benefit is far less likely when you talk about moving between the rates for low six-figure incomes – why pay 32% now to avoid 35% later, for example? (And to counter that, the risk of rates reverting in 2026 become a lot more significant.)
(The upside of the hard Roth conversion question is that if you guess wrong, it may wind up not hurting you all that much.)
Keep in mind that IRA capital gains are taxed at ordinary rates, i.e., taxes on gains strictly due to inflation. Roth IRAs are not taxed on inflation. So if you only keep up with inflation in future gains, the ROTH has zero real return and the IRA has a negative real return, approximately negative your marginal tax rate at withdrawal. Ouch!
Great. Sounds like something we just discussed on our last podcast. Haha. This is the heart and soul of retirement planning at its finest.
Also, listen to our latest podcast with Jonathan Clements https://podcasts.apple.com/us/podcast/the-9innings-podcast/id1558127474?i=1000567699273
Wonderful article Jonathan! Many of us struggle with this conundrum.
I’m in my late 60’s and have not done any Roth conversions per guidance from my tax advisor, but feel intimidated by what I read all the time that I should be doing so.
Do you have any good links to Roth conversion tools on the web that can help one get a broad picture of their situation? When my RMDs kick in, I get a little nervous that I will be in the same boat as you.
I use an internet-based comprehensive financial planner called MaxiFi (maxifiplanner.com) which can give you very precise modeling of the effect of all of these factors on any scenario or Roth conversion plan you want to test, including the effect of different possible future tax rates, inflation rates, withdrawal rates, Roth conversion amounts, income, lifespan, etc. It is fee-based, $109 for the first year (you don’t have to continue). (I have no connection to this company other than as a customer, but please feel free to delete this comment if a commercial mention is taboo.) I have been making bracket-filling Roth conversions the last few years while working.
I suspect there are internet calculators available, but I haven’t tried any of them. My suggestion — and I know it’s tricky — is to estimate your total household income as of age 72, including required retirement account distributions, Social Security (which can be partly taxable) and anything else, and then compare it to your expected total income this year. If it looks like your income will be a whole lot higher at 72, a conversion today might make sense — and thus it might be worth another conversation with your tax advisor.
There may be a small twist to the thinking about IRMAA Medicare surcharges depending on taxable income. Assuming you will still have a sufficient self employment (SE) income from your single member LLC at age 65 your qualified medical insurance, including Medicare and related supplement premiums, should continue to be deductible above the adjusted gross income (AGI) line on your federal 1040. Thus, you will still be out the cash for the higher premiums but the premium deduction would wash against taxable income limited by your net SE income. Such medical premiums are typically not deductible for SE tax purposes. It sounds like the IRMAA income threshold may become your targeted taxable income when you become age 63, thus your great comment about converting deferred retirement income to a Roth prior to age 63. When you reach age 72 and have to start RMDs you may be able to fine tune your taxable income to stay below the IRMAA limit by making strategic contributions after year end to a traditional IRA or solo 401(k) account as the 2019 SIMPLE act eliminated the prior age restriction on IRA contributions. The new SIMPLE 2.0 bills that both the house and senate have passed in 2022 have proposed raising the required beginning date for RMDs to age 75 over a long phase in period. It appears likely to me the final tax act may pass in 2022 and may impact your tax planning.
There is one more wrinkle to consider. Perhaps you are purchasing Affordable Health Care insurance because you are not yet eligible for Medicare. If your income is low enough, you qualify for premium subsidies. Withdrawals you take from your Roth IRA do not count toward your income when determining eligibility for premium subsidies.
Very timely as I have been going through this tax optimization process for the last few years. Since I’m self employed and I have lumpy income, I’ve found that I should spread my IRA conversions towards the end of the year so I can better estimate my tax rate. The IRRMA costs are real but are hard to avoid. I did appeal my initial IRRMA decision to SSA and won a lower rate.
Jonathan – Another variable for many is marriage status. If filing MFJ, one day a spouse will pass and the surviving spouse goes into the (much higher) single tax brackets. There are many variables when a spouse passes, but don’t forget to think about the surviving spouse’s income tax burden. And, in general, the prospect of higher income taxes make Roth conversions more attractive.
I’ve been warning family and friends of likely taxes in our 70’s and 80’s due to IRA RMD’s on top of SS benefits and IRRMA penalties. Thanks for the explanation and suggestions. I’ve been fortunate to have sheltered some taxable income with rental real estate (depreciation) and NOL’s from investments in business start ups with my son. It’s a serious concern to which more “savers” need to pay attention.
You said it, not me Jon: it makes more sense to convert chunks of my traditional IRA to a Roth each year, where the money will then grow tax-free.
Good luck & Best wishes….
I too have 59 1/2 on my calendar, but for inservice Ira rollover process to start. Perhaps you can talk about that, pros and cons?
I’m thinking it will get some ( bulk) of my 403b into an IRA and allow for more options, and charitable giving options when we’re at RMD phase. In the meantime lower costs and consolidation seems helpful for planning.
I am subject to the RMD rules in 2022 on my traditional IRA and have also been thinking about an in-service distribution from my 401(k). I am still an active employee and not a 5% owner of my employer so the RMD requirement at age 72+ does not start until I end my employment with my current employer under ERISA and my 401(k) plan rules. I have currently decided to not volunteer for additional current tax by moving funds from my 401(k) that are not subject to a current RMD to my IRA where they are subject to the RMD rules. The costs from my decision are the higher 401(k) fees and selection limitations which seem to be minimal problems compared with the tax cost. Some 401(k) plans will allow a rollover from a IRA to a 401(k) which effectively delays the RMD on the IRA investments but my plan does not allow such a transfer. Thus, I will need to take my 2022 IRA RMD or pay the 50% tax penalty of the IRA RMD amount, ouch, for failure to do so. The proposed SIMPLE 2.0 bills currently in congress includes a reduction in this penalty percentage which would be nice.
The inservice rollover is a great way to lower your investment costs if your 403b has high fees and minimal options. As you mention, other investment options also become available including Roth conversions.
If you’re allowed to do an in-service rollover, that seems like a smart strategy, especially if it’ll lower your investment costs.
I got hit with IRMAA based on income when I was 62 because I applied when I was 64. Perhaps it depends on one’s birth month. I was born in June.
How can you get IRMAA premiums at 62 before Medicare or are you collecting disability?
Applied a couple months before his 65th birthday when he was still 64?The IRMA premium is based on your income 2 years prior. So if you are eligible in 2022, the IRMA is based on 2020 income.
Hope your stress level goes down, Richard. Always enjoy your articles.