FREE NEWSLETTER
A FEW DAYS AGO, I RAN the numbers on our likely 2023 taxes, and reached two conclusions: We have a small refund coming—and we should find a way to pay more taxes.
How can both be true? I project that our 2023 taxable income will be well below $190,750, which is the top of the 22% tax bracket for those married filing jointly. Getting taxed at 22% strikes me as a good deal, given the likelihood that we’ll be taxed at an even higher rate later in retirement. There are a few things we could do by Dec. 31 to make sure this opportunity doesn’t go to waste.
For instance, we could convert some holdings from a traditional IRA to a Roth IRA, paying tax at 22% on the sum converted. That marginal rate is probably low compared to what we’ll pay when we draw down our traditional retirement accounts later in retirement. By the time we make those withdrawals, we hope our accounts will have grown, plus income tax rates may have gone up.
Today’s historically low tax brackets are set to sunset at year-end 2025. The costs of the pandemic, foreign wars and the government’s growing debt burden create further pressure to raise taxes. Anything can happen, of course, but it seems highly likely that our future tax burden will be higher than our present one.
We might even consider Roth conversions that would push us into the next bracket, which goes up to $364,200 for married couples. Even at that high amount, the effective marginal rate would only be 24%. While the net investment income tax (NIIT) is triggered for couples at $250,000 in modified adjusted gross income, distributions from IRAs and 401(k) accounts are exempt from this additional 3.8% tax.
Still, converting such a large sum would mean a big tax bill. Moreover, there are reasons to keep a healthy sum in traditional retirement accounts, rather than moving too much into Roth accounts. We also don’t have children, so we don’t have the legacy considerations that make Roth conversions appealing to many. But while converting at a 24% rate is debatable, converting enough to get to the top of the 22% bracket seems like a no-brainer for us, improving our tax diversification across Roth and traditional retirement accounts.
Another way we might take advantage of our low bracket is to realize some capital gains by selling appreciated investments held in our regular taxable account. We have some tax losses that we harvested last year and which are available to offset gains.
What about generating additional gains beyond the size of our realized losses? Our taxable income is too high to realize gains at the zero percent rate. On the other hand, even if we realized enough gains to get our income all the way to the top of the 24% income tax bracket, the tax on those gains would be just 15%, with the 3.8% NIIT layered on top of that 15%.
We have a few holdings in mind to sell. These are perfectly fine funds and stocks, but selective selling could reduce our large number of holdings, lower the average expense ratio of our portfolio and increase its tax efficiency. Then again, if we just sit tight, when the first of us dies, these holdings will get an adjustment in cost basis that makes the embedded gains go away.
This will be especially significant if we’re still residents of a community property state. In other words, we have a choice between paying some tax today to improve our portfolio or avoiding that tax payment and letting nature do the job for us—though hopefully not until the far distant future.
Our default would probably be to reduce some of the holdings mentioned above by the amount we can cover with carried-over losses, for a net zero capital gains tax liability for 2023. An argument could be made for preserving the carried-over losses for use in future years, when the capital gains tax rate is perhaps higher than 15%. After all, if basic income tax brackets are likely to go up, why not capital gains tax brackets?
We have to be cognizant of the NIIT in realizing capital gains. Unlike Roth conversions, dividends and capital gains are subject to the NIIT and its additional 3.8% tax. If we did both, while the Roth conversion amount itself wouldn’t be subject to the NIIT, it would count toward the $250,000 threshold beyond which dividends and capital gains would get hit with the NIIT.
Next year, we’ll have another, perhaps bigger bite at this tax apple. I project our taxable income in 2024 will be even lower, which would give us more room to do a Roth conversion or take capital gains at a relatively low rate. We’ll also still be more than three years away from Medicare, which means not having to worry about triggering higher Medicare premiums.
As I write this, we haven’t decided. I expect we’ll realize capital gains up to the point where these gains are offset by the losses harvested last year. I also expect we’ll do enough Roth conversion to get us near the top of the 22% tax bracket. Beyond that, well, we still have a few weeks to decide.
Michael Perry is a former career Army officer and external affairs executive for a Fortune 100 company. In addition to personal finance and investing, his interests include reading, traveling, being outdoors, strength training and coaching, and cocktails. 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.
Nice article and rationale. I’ve been chipping away at my IRA account balances by doing Roth conversions each year. I use the IRS draft versions of the relevant tax forms to update a spreadsheet that allows me to do pro forma tax calculations regarding various conversion scenarios. Just finished this year’s calculations. This, in conjunction with a retirement planning spreadsheet that I also update annually, gives me peace of mind. Don’t like “surprises” at tax time, nor down the road 🙂
If you already have enough and are charitably inclined, why pay any additional taxes when you can utilize qualified charitable distributions (QCD)? One can start as early as 70.5, give up to $100,000 per year, have it count towards some or all of one’s RMD (when required to take an RMD) and–best of all–there are no tax consequences! It doesn’t even show up as income. This is such a great gift from the IRS, that I worry it will go away one day. And, it can be a wonderful way to give meaningful amounts to charities that align with one’s values. There are so many smaller (and often all volunteer) worthy nonprofits where a QCD could make a world of difference. I have struggled making the transition from accumulating savings to using it, but I have found giving via QCDs to be a great option and very rewarding.
Thanks Rick, I just talked to my IRA manager about this and am starting a couple QCDs next year.
He had to call the IRA company to determine what they required and we both learned how it works with qualified, tax-exempt 501(C)3 organizations. My guy thanked me, as he has others that use the same company.
I just wonder why tax preparers don’t suggest doing this to others!
I’m single, 74 and stopped working full time 23 years ago. When I retired, I had some very large unrealized capital gains and almost no unrealized capital losses. The tax rates were much higher then, so I spread out taking capital gains over several years. Later, when my income was lower, I converted enough of my traditional IRA to a Roth to reach the top of my tax bracket.
Now, I have mostly spent down my taxable account but have more in my IRAs than I will likely ever spend. Since my only close relative probably won’t need any inheritance from me, most of my estate will go to charities. I’m pondering how much is prudent for me to donate now using QCDs vs. waiting until I die.
I have realized that I converted too much from my traditional IRA to my Roth. Since the charities are the major beneficiaries of my IRAs, they won’t pay the tax anyway.
Gary, Another newly provided approach for charitable giving from IRAs is a Chairtable Gift Annuity where the donor retains the annuity payments until death at which point the money goes to the charity. Kathleen Rehl provided excellent guidance here:
https://humbledollar.com/2023/02/better-than-cake/
We’re also not concerned about leaving money to heirs, and also have a bunch of accumulated capital gains. It will still make sense to realize these and pay the tax before taking distributions from retirement accounts. This might be by realizing some each year as you have done. Or, we might decide that we want to buy a house and have little choice but to realize lots of gains to do so. This is one argument for taking more gains now within the thresholds discussed rather than doing more Roth conversions.
I’m filling my 12% bracket with Roth conversions, careful not to go into the 15% capital gains bracket; otherwise my marginal rate would be 27%.
Thanks for the comment Randy. I don’t think your marginal rate would become 27%. As I understand, CGs would suddenly cost you 15% instead of zero, but I think the marginal rate on every dollar of ordinary income would still be 12% or 22%.
Still, avoiding that 15% is significant. It makes me think of one of my first articles on HD.
https://humbledollar.com/2021/10/getting-to-zero/
Sad but true, Michael: at least in my case, $100 in extra Roth conversions costs $12 in ordinary income tax PLUS $15 in qualified dividend/capital gains tax (according to the QD/CG worksheet).
Well thought out article, very informative. I’m starting to realize I may be the only person on HD not trying to avoid IRMAA and even believes paying premiums based on income is fair.
Thanks Dick.
I think IRMAA would be easier for many to swallow if it were more gradual and knowable in advance.
Yes, sir, we make more, we pay more, what the hek?
However, in 2022 my first tax preparer over stated my MAGI by….$51K! So, IRMAA would have been astronomical so, I paid another preparer $2500 to redo the return to drop the earnings to where they should have been. We don’t have to just throw the IRS money. haha.
Your article dovetails nicely with Jonathan’s post of “Better and Better” this week.
Old way of thinking: “Each year you should strive to pay the absolute minimum amount of income tax, regardless of future consequences, whatever that takes.”
New way of thinking: “As you approach retirement, look for ways of strategically managing your tax bill, in an attempt to minimize the overall tax burden that you (and possibly your heirs) will be subject to throughout your retirement.”
Well done!
Thanks for the comment. I suppose Jonathan’s old way of thinking was still an improvement on my even older way, which was to get to the end of the year and let the tax chips fall where they may. Nowadays am definitely in the new way, or at least trying to be.
I agree with your new way of thinking for tax planning comment. I would also add considering the time value of money into such tax planning is an additional factor I consider.
I think the appropriate discount rate for deciding to realize future taxes sooner for individuals is unique to each taxpayer based on their appropriate marginal interest expense if they have debt and/or the manner they are taxed on the type of income sources they expect to have in addition to tax rate difference considerations already noted in the article. Both are moving targets and my current plans are to be debt free by 2025 and I will take distributions in 2024 and 2025 in excess of my RMDs to achieve that goal.
2024 and 2025 are shaping up to be particularity good years of certainty for taking tax actions in my thinking as I do not expect the provisions of the tax cuts and jobs act (TCJA) that expire at the end of 2025 to change in material unexpected respects. I do expect beginning in 2026 that the sunset of the TCJA combined with the interest on federal debt will create a greater need for more federal tax collections and tax law changes.
I will hopefully survive and be in my mid age 80’s around 2034 when the payment of currently scheduled social security benefits will need to be reduced absent law changes. I do not expect congress will do anything until the last minute in regards to changing social security benefits or their funding. While I am planning for an approximate 20% cut in our own social security benefits in 2034 the impact of changes in social security benefits for those a decade or more younger than me appears to be another critical factor that younger taxpayers needs to consider for long term tax and cash planning.
I am also considering a purchase of qualified longevity annuity contracts (QLAC) to start at age 85 to be funded with an investment from our t-IRAs and I expect we will make those decisions after we reach our age 75 with decision criteria based on our then health and the revised tax laws after TCJA provisions expire.
Your 22% or 24% tax-rates on Roth converting with zero NIIT or IRMAA looks like a winner to me, especially as in a couple years your IRMAA-free window definitely closes and these tax rates have a good chance of disappearing. This also reduces future RMDs when you indicate potentially being in a higher tax bracket anyway.
Thanks John. Still not quite settled on taking more capital gains or doing Roth conversion but definitely won’t let the low bracket go to waste.
From what you have written about yourself, I am about your age, apparently in a similar station of life, and with similar financial concerns. I have seemingly gone through the exact same thought processes as yourself.
For the last couple years, I have opted to convert more Traditional to Roth even though it has pushed us into the 24% bracket, along with the NIIT tax. Of course, as I get to 63, I must reduce my conversions back down to avoid Dear Aunt IRMAA from medicare (and her two-year look back) when I hit 65.
But, a lot also has to do with out-year calculations when I hit 75 and will be even more likely have to deal with IRMAA, as those 401k and Traditional IRA RMDs start rolling in.
And, as you suggest, at the end of 2025, something is likely to change with taxes (of course no one knows for sure; “they” could discover how cold fusion works and all our problems would be solved, but probably not).
Although I am not a betting person, it seems to me that, politics aside, at some point, as taxes coming in pay a larger and larger amount of interest on the debt, taxes will need to go up.
So, taking my best SWAG, I am paying now instead of later (of course, they might even decide to tax Roths, but, you do the best your can).
Thanks for the comments Joe. While it’s hard to say what might happen in 2025, it’s hard to imagine our taxes not going up sometime between now and when we start RMDs. Although I suppose people might have said that 20 years ago too.
Nice article Michael. I have an “estimated tax” spreadsheet that I use to track yearly finances and tax implications. I got serious about it in 2017 when I started consulting and became responsible for estimated payments and SE tax. I haven’t added a NIIT calculation so far, but you’ve inspired me to add that to the mix. As you clearly demonstrate, there are lots of confusing connections in various parts of the tax code, that are difficult to think about individually. State taxes can also play a part indecision making.
Thanks Rick. Yes NIIT is definitely worth adding I think, partially because it isn’t just marginal tax on income over a threshold, but a cliff that applies to all investment income once the threshold is crossed.
And great point on state tax, which is also a factor for us. We don’t have any now, but can’t say that will be true in a few years.
The tax system is very complex. If you have a large stream of dividends taxed at 0% and 15%, and your are subject to IRMAA and NIIT, you have to model your income in a spreadsheet and see what the impact of various changes might be.
You should know that taking more ordinary income may push your qualified dividends and capital gains up a bracket, because that income rides on top of your ordinary income. You might also push them into the NIIT area. As a single retiree, I used to like to fill up the 10% and 12% brackets, and then have the rest of my income as qualified dividends taxed at 15%. However, with favorable Treasury bill rates, I’ve modified that goal, and I’m now filling up the 22% bracket too. With the standard deduction, single retirees can have up to about $115K in ordinary income without hitting the 24% bracket. The rest you can fill with qualified dividends taxed at 15% up to $200K and 18.3% up to $492K.
There is an interesting trick with NIIT. You can adjust your income for the impact of state taxes when computing your NIIT. There is no form or worksheet – the instructions say you can use any reasonable method. I consulted with a friend who is a CPA, and showed him what I was going to do, and he said yes, that’s what everybody does.
Thanks Ormode. I didn’t know that about state taxes, probably since we have none. Will have to keep that in mind for future years when we may.
Taxes tax the brain, don’t they Michael? I hope the right path emerges from the tangle of choices. Thanks for good article on the topic.
Nice one 🙂 They do, but it’s worth the effort I think.
Are there any IRMAA considerations?
There could be if closer to 65, but there aren’t any for us this year or next.