FREE NEWSLETTER
I’M PONDERING WHETHER to make my biggest transaction in four years—and it might be the trickiest financial decision I’ve ever made. My quandary: Should I take advantage of today’s low tax rates to convert a big chunk of my traditional IRA to a Roth?
This financial navel-gazing was sparked by an article by John Yeigh, one of HumbleDollar’s contributors. As John pointed out, you can now have a much higher annual income and still avoid the top federal tax brackets, thanks to the tax law passed in late 2017 and which took effect in 2018. That means it’s possible for folks to make a large conversion from a traditional IRA to a Roth and pay taxes at a relatively low rate. Remember, you have to pay federal—and possibly state—income taxes on the taxable sum converted to a Roth. To cover the resulting tax bill, you’ll want money set aside in a regular taxable account.
In return for paying those taxes today, you get a handsome reward: Money in a Roth IRA grows tax-free, plus there’s no required minimum distributions once you turn age 70½. Sound appealing? Problem is, the current window of opportunity may slam shut at the end of 2025, when many provisions in 2017’s tax law sunset, and it could close even earlier—depending on what happens in 2020’s election.
To understand what’s at stake, suppose you claim the standard deduction. In 2019, if you’re married filing jointly, you could have total income of up to $345,850 and still remain in the 24% federal tax bracket. Above that level, additional income is taxed at 32%. By contrast, in 2017, before the new tax law took effect, you’d have been taxed at 25% once your total income got above $96,700. This latter figure reflects 2017’s standard deduction, plus two personal exemptions.
What if you’re single? In 2019, you can have total income of up to $172,925 and pay taxes at 24% or less. Contrast that with 2017, when you would have been taxed at 25% on income above $48,350. Meanwhile, if you file taxes as head of household and claim one dependent, you can have total 2019 income of as much as $179,050 and remain in the 24% bracket, versus reaching the 25% tax bracket at $68,250 in total income in 2017.
The bottom line: Unless you already have a huge income, it looks like there’s plenty of room to do a hefty Roth conversion in 2019 and get taxed at 24% or less. That alone, however, isn’t enough to justify a big conversion. Instead, you also need to be reasonably confident you’ll be taxed at a higher rate later on, especially once you reach age 70½ and start taking required minimum distributions from your retirement accounts.
I’m 56½, which means I’m 14 years from that point. If my traditional IRA notches a 4% after-inflation annual return, it’ll balloon 73% over the next 14 years—and I’ll be looking at required minimum distributions of around $95,000, based on the IRS’s uniform lifetime table. (I’m using my portfolio’s potential after-inflation return because, over the 14 years, I presume tax thresholds will continue to rise with inflation.)
Take that $95,000 IRA distribution and add my Social Security benefit, my taxable account’s gains, my wife’s Social Security and her portfolio income, and it looks like we could easily be paying taxes at a 25% marginal federal rate and perhaps more—assuming we return to something that looks like the tax law that applied to 2017 and earlier years.
To be sure, I could reduce the tax hit somewhat by spending maybe a third of my IRA in my 60s. But even then, it looks like my wife and I would still be in the 25% tax bracket once we’re in our 70s—again, assuming we revert to something like the old tax law.
The question: Should I do a big Roth conversion this year and perhaps in the years that follow, paying federal income taxes at 24% or less, so that—14 years from now—we could potentially avoid paying taxes at 25% or more. A big conversion today, and hence less taxable income later, may also result in lower Medicare premiums. What about my Social Security benefit? That, alas, will almost certainly be taxable.
As I weigh all of this, there are so many unknowns. Maybe we won’t revert to something that looks like the old tax law. Maybe we’ll cut federal income taxes and instead adopt a national sales tax, which would make income-tax-free withdrawals from a Roth less valuable. Maybe Congress, in its infinite wisdom, will decide to tax Roth accounts.
I long ago learned that you don’t want to get into the forecasting business—not just when it comes to the financial markets, but also when contemplating future tax rates. At this juncture, all we can say with any certainty is that 2017’s individual income tax rates will return in 2026, unless Congress acts.
So if I don’t want to forecast, how do I decide whether to take advantage of today’s low federal tax rates? As always, the trick is to eschew predictions and focus on managing risk. Because none of us can be sure what will happen to tax rates, my inclination is to hedge my bets and convert maybe $50,000 to a Roth this year and perhaps a similar sum in the years that follow.
But I might convert far more—if the stock market plunges. That’ll allow me to convert a larger percentage of my traditional IRA to a Roth, while still paying the same amount in taxes. My fondest wish: The stock market nosedives, I convert, the market comes roaring back—and the entire gain is tax-free.
Follow Jonathan on Twitter @ClementsMoney and on Facebook. His most recent articles include That’s Enough, Third Rail and Get Happy. Jonathan’s latest books: From Here to Financial Happiness and How to Think About Money.
Want to receive our weekly newsletter? Sign up now. How about our daily alert about the site's latest posts? Join the list.
What about the SECURE ACT working it’s way thru Congress as we speak? It would mandate RMDs on inherited Roth IRAs. The Roth distribution wouldn’t be taxed, but any future earnings on those reinvested funds would be subject to tax.
Is the 1% difference between 24% and 25% really that significant?
My wife died in May and the difference in filing single and joint is breathtaking. I did the numbers and decided to roll $55,000 to my Roth this year. Next year I will go over 12% rate much quicker due to filing widower for the next two years and then single forever. My current portfolio is 30% Roth and 70% IRA.
My favorite analogy for selecting a Roth over traditional….”would you rather be taxed on the seed or the crop”?
Good article. Another unknown is the possibility that the starting RMD age will be raised.
The current proposal is talking about 72 — a little bit of relief, but not a huge amount.
This is definitely a tough question. How are the high state taxes in NY entering into your decision? If someone is thinking about possibly moving out of state in the future that could have a big impact. And medicare tax may also be an issue. Depending upon filing status and current income, a conversion to a Roth could potentially subject someone to the additional medicare tax.
Both the Medicare surtax and state income taxes are indeed important considerations. While we may move out of NY, it’s unlikely we’ll leave the NE — where state income taxes are high — because this is where our kids are.
“But I might convert far more—if the stock market plunges. That’ll allow
me to convert a larger percentage of my traditional IRA to a Roth, while
still paying the same amount in taxes.”
To pay the same amount in taxes (assuming tax rates don’t change), you would be converting the same amount, not “far more”. What you may have meant is merely that your (unchanged) planned conversion amounts would constitute a larger percentage of your IRA.
However, even this seems to fly in the face of your hedging strategy – that since the future is unknowable, you’ll split the difference and convert only a given percentage of your T-IRA. What you seem to be suggesting is that you can, at least to some extent, predict the stock market.
That if it drops a significant amount you’ll expect above average returns going forward. Thus you’ll convert a larger percentage rather than adjust the amount to keep the same percentage.
But isn’t the converse also just as likely? That since we’ve just had a 10 year bull market, returns going forward are likely to be below average? And so you might consider converting a lower fraction of your T-IRA (i.e. less than “splitting the difference”) at the start, because you are converting overpriced securities now?
Simple heuristics (I don’t know what will happen, so I’ll hedge my bets) aren’t so simple after all. The problem with the market impacting this heuristic is that it starts with the implicit assumption that the market is now “neutral” – not underpriced, not overpriced. Is it?
What you said: “What you may have meant is merely that your (unchanged) planned conversion amounts would constitute a larger percentage of your IRA.”
What I said: “That’ll allow me to convert a larger percentage of my traditional IRA to a Roth.”
I would suggest you bother to read what I wrote before you start criticizing.
Meanwhile, if you think future returns are likely to be lower after a 10-year bull market, wouldn’t you expect them to be higher if stocks plunge? I certainly would.
Most of the comments on HD tend to be thoughtful and respectful. Your comments failed on both scores.
I have the “annuity conundrum” to deal with. If I die too soon my conversions may not be the optimal situation for my primary beneficiary. Beyond that, my IRAs will go to a charitable organization so they won’t have to pay any tax at all. As usual, covering every single financial base is a big challenge.
For me, charitable contributions are the biggest reason to hold back on conversions. Both from a legacy perspective and for funding contributions via QCDs. Especially now that it’s become more difficult to itemize charitable deductions.
I totally appreciate your ambivalence Jonathan, there are a LOT of moving parts to this question, and math is only part of it. You’ve made some very good points, and in addition might consider…
1) At some point, either you or your wife will be single, and with Roth’s the tax burden on which ever one of you remains should be somewhat lessened, which leads me to a second consideration–
2) The rules surrounding the Roth account are much simpler than the traditional account when it comes to both withdrawals and inheritance.
There are quite a few less “gotchas” and tax traps with the Roth, both present and future, and that alone is a big consideration for many, including myself.I’d rather have my wife and/or my heirs deal with a Roth account over a traditional account, hands down.
My approach – Revisit in December to see how the portfolio is doing, expenses for kids go in the meantime, and cash position is to cover paying the Roth conversion taxes. If still in doubt, Roth convert halfway between your zero and max, and you’re assured of only being Half Wrong – see: https://humbledollar.com/2019/01/half-wrong/
We were half wrong last year, converting about half my wife’s IRA. The rest is to be converted this year if last year’s tax bite doesn’t dissuade her (it doesn’t me). Of course now what’s left in her Traditional is >20% more than it was before, and of course we’ll convert it and then the market and account value will then drop next year. I suppose we could be half wrong again and convert half of what’s left…
My December conversion last year happened to occur when the market was in a trough – half right so far through simple luck. I know of one very conservative individual who converted some IRA monies about 8 years ago and invested 100% in bonds, mostly municipals – he is frustrated in that the Roth has barely recovered the tax payments. He would be better structured with his Roth most aggressively invested for the long term, IRAs more moderately aggressive, and after-tax money most conservatively invested. Over any ten year period, the stock market increases 95% of the time – You’ll eventually be half right.
On the bit about doing a big Roth conversion when stocks are down, so as to get more conversion buck for the tax-hit bang, is that you don’t actually have to wait for a market decline. Presumably your tax deferred accounts hold some bond fund money, and Uncle Goody is as happy to accept your deferred tax payments on conversions of these fixed income assets as on equities. Then, when stocks eventually do get whacked, just move that Roth bond fund money back into stocks in the same Roth account.
Downside: Yes this is a form of market-timing, but note that your fixed-versus-equiities asset allocation doesn’t need to change during these transactions (especially if you account and adjust for the higher value of after-tax Roth dollars vs. the-taxman-cometh dollars in your deferred accounts). But it’s a harmless form of timing, with one minor exception:
I want my Roth-space filled with asset classes most likely to go zoom when (if) they have their day – emerging, value, etc. (and some S&P 500 too). I certainly don’t want it stuffed with 2% bonds. At least not for long. But if low-return bonds sit in my Roth for a year or two waiting for stocks to take a hit I won’t lose any sleep over that.
Interesting comment. I’ve thought about the same thing. If nothing else, it helps minimize regret: If you convert money earmarked for bonds, you don’t have to worry about stocks declining and then kicking yourself because “I should have waited.”
I’m a big supporter of doing Roth conversions although, because i’m now having to take RMD’s, I’m past the point where converting makes any sense for me. My experience, and of course this is anecdotal, is that my IRA grew and continues to grow faster than what I expected when I retired in 2001.
The chance for faster growth, I think, isn’t as appreciated as much as it should be. In your case, you are only planning on 4% compounded. What happens to your IRA if it turns out to be 5% or more? What happens to your future RMD’s?
Over time, I converted about 47% of my IRA balance at retirement. Today, my Roth only represents 40% of present IRA balance. But, its value is more than double the amount I converted. Each RMD I have to take is 40% lower than it would otherwise be. I’m not paying IRMAA surcharges.
We do live in a state without a state income tax.
Thinking about today’s THINK entry (bottom of HD home page), separate from converting Tradtional IRAs and 401ks to Roths, is it also reasonable to do some tax gain harvesting in taxable accounts so the taxes are paid at today’s historically low rates, especially if taking the perspective that taxes will be higher in the future? Editing to add there’s also the possibility that in future years we will not live in a no income tax state as we do now. My wife is a California girl and us going there is a possibility.
I would be reluctant to realize capital gains in a taxable account, assuming you’re happy to hold those positions for the long haul. Remember, converting to a Roth is a potential benefit both for you and your heirs. But that isn’t the case if you realize gains in a taxable account. You’ll be paying unnecessary taxes if you never spend that money–thanks to the step-up in cost basis upon death.
Thanks, that makes sense. I’m just wary of someday wanting a bunch of the money in the taxable accounts at once – let’s say for a house in the South Bay area of LA. Which of course may never happen, so back to your point…
The whole conversation reinforces that decumulating is going to be a lot harder than accumulating. First world problem for which we’re thankful.
Do note that the SECURE act requires non-spouse beneficiaries to empty IRA accts — INCLUDING ROTHS — in 10 yrs. (Or 5 in the Senate version …)
We had an article about that last month:
https://humbledollar.com/2019/06/new-rules/
When doing Roth conversions to lower age 70+ RMD’s, for many, it would save a bunch on Medicare premiums if after you turn 63 (~2 years before you join Medicare), if you keep your total annual income below $85K (single). Another choice available after you are age 70 on RMD’s is to do the paperwork to do a qualified charitable distribution from your Traditional IRA each year to avoid the taxable RMD. RMD’s are low at age 70, but they climb every year and by age 85, they are a lot of money.