By Dr. Jim Dahle, WCI Founder
There is a lot of angst out there on the interwebs about IRMAA. So, I decided to look into it a little more deeply to see if I was missing something. Nope, I wasn't. What a silly thing to worry about. I can't believe people spend any time at all on this subject while doing their financial planning. Let me explain.
What Is IRMAA?
IRMAA stands for Income Related Monthly Adjustment Amount. Note that it is not IRMMA (just like HIPAA is not HIPPA). IRMAA applies to Medicare premiums. It is a method to means-test Medicare. Means-testing means that those with “means” (i.e. the wealthy) pay more than those without the “means” for the same product or service.
How Medicare Works
“Remind me again how Medicare works,” you say.
Don't feel bad. Even most doctors whose income primarily comes from Medicare don't actually understand how Medicare works. Here's a brief reminder of what you need to know about Medicare to understand IRMAA.
Medicare Part A is hospital insurance. You do not pay a Medicare Part A premium. It has a $1,600 per year deductible. Basically, if you go to the hospital on Medicare, you pay $1,600 [2023] and Medicare pays the rest. Congratulations! Thank you for working and paying into Medicare for so many years. This is a big part of what you get for paying 2.9% (3.8% for many of us) of everything you earned for decades into Medicare.
Medicare Part B is doctor insurance. It covers doctor visits (including specialists), ambulance services, vaccines, home health, and medical supplies. It does not cover hospitalizations, and it does not cover medications. The Medicare Part B deductible for 2023 is $226. Believe it or not, that's actually lower than it was in 2022. The “standard premium” is also lower for 2023 at $164.90 per month.
Medicare Part C is generally called Medicare Advantage. There are lots of different Medicare Advantage plans that cover all kinds of different things for different prices. Typically, people buy either Medicare Advantage or traditional Medicare parts A, B, and D. For today's discussion, we can ignore Medicare Part C.
Medicare Part D is drug insurance. It covers the cost of your prescriptions. Annual deductibles vary, but they can be no more than $505 in 2023. Unlike Parts A and B, Part D also has co-pays with a maximum out-of-pocket total of $7,400 in 2023. The “standard premium” for 2023 is $31.50 per month (also a decrease from 2022).
Medigap policies are sold by private companies, and they cover co-pays and deductibles that Medicare doesn't normally cover. It comes with an additional premium averaging $155 per month in 2023.
Note that all of these premiums, deductibles, and co-pays are PER PERSON, not per family or per couple.
More information here:
Healthcare in Retirement Could Cost You $500,000; Here’s How to Plan for It
Health Insurance in Early Retirement
Yes, Medicare Costs Money
This is an important point that lots of people don't understand. You hear it all the time.
“I can't retire, I need the medical insurance.”
“I just need to get to 65 when I can get Medicare.”
“What do people do for health insurance when they FIRE?”
As you can see, Medicare isn't free. It might be a lot cheaper than buying insurance on your own, but it's still going to cost a couple at least $400 a month (plus deductibles and co-pays). Note that you can pay Part B, Part C (Medicare Advantage), and Part D premiums using a Health Savings Account (HSA), but you typically can't pay health insurance premiums (or Medigap premiums) with HSA dollars.
Where Does IRMAA Come In?
Now that you understand how Medicare works, you can understand IRMAA. IRMAA simply means that rich people pay higher premiums than poor people for Medicare Parts B and D. I can hear you grumbling now, given that the WCI audience is high-income professionals. But you shouldn't grumble very loudly. You have to be pretty rich for your premiums to go up at all, and even then, they don't go up very much. Let me show you using the charts for 2023:
Let me decipher this for you. Let's say you've done very well for yourself. You and your spouse are now retired, and you have an Adjusted Gross Income (AGI) of $200,000.
“Oh no!” you say. “We're making more than $194,000; we're going to get IRMAA'd!”
Yes, you are. You're going to get IRMAA'd. So, what does that mean? That means you are going to pay the following additional monthly amounts:
- Spouse #1 Part B: $65.90
- Spouse #2 Part B: $65.90
- Spouse #1 Part D: $12.20
- Spouse #2 Part D: $12.20
- Total: $156.20 per month
- Total annual amount: $1,874.40
Now ask yourself,
“How much financial planning am I willing to do in order to avoid this cost?”
This is less than 1% of your AGI (and we all know that a lot of your spending in retirement is not even included in your AGI—such as basis; Roth money; 15% of Social Security income; and borrowing against life insurance policies, your portfolio, or your house.) For me, the answer to this question is “pretty much nothing.”
Even at the top end of IRMAA, it really doesn't look too bad. We're talking about a single person with an AGI of $500,000+ or a married couple with an AGI of $750,000+.
- Spouse #1 Part B: $395.60
- Spouse #2 Part B: $395.60
- Spouse #1 Part D: $76.40
- Spouse #2 Part D: $76.40
- Total: $944 per month
- Total annual amount: $11,328
Again, that is no more than 1.5% of your AGI. This isn't exactly going to break the bank. Does it feel unfair because you paid in so much more to Medicare than anyone else and yet you are getting so much less out of it? Sure. That's why it's called Medicare TAX. You've been paying more to drive on the roads, to have a functioning 911 system, and to cover the national debt for decades. Why would it be any different now? But this little tax is just not a large enough amount to be worth revamping your financial life. Maybe you can figure out a way to get your AGI just below $750,000. Now you saved $1,000 in annual taxes. Big whoop. Not going to move the needle.
More information here:
How to Optimize Your Money in Retirement
Healthcare Is Expensive
I've been self-employed for more than a decade and have been buying my own health insurance that entire time. These days, we buy health insurance for our employees' families, too. We know exactly what health insurance costs. I always find it fascinating how inexpensive people think healthcare is. For some reason, they think it should resemble their cell phone bill or even their Netflix bill more than it does their grocery bill or even a small mortgage. That's not the case, folks. Healthcare spending makes up 20% of our nation's Gross Domestic Product (GDP). On average, Americans should expect about 20% of their monthly spending to be on healthcare between premiums, deductibles, and co-pays. It'll be a higher percentage for those with lower incomes and a lower percentage for those with higher incomes. But the point is that it will always be expensive because it is expensive stuff and because we consume a lot of it.
Those who pay for their own healthcare without any sort of government or employer subsidization know that it is typical to spend more than $1,000 per month on it (and not hard at all to spend $2,000 per month on it). Some families even spend $3,000 per month. When we're talking about your cost for it going up by $156 a month if you have taxable retirement income 50% higher than the average American household, it's hard to feel too much compassion. Even for those so well off that they have a seven-figure AGI in retirement, Medicare is still cheaper than going out and buying your own insurance.
More information here:
A New Way of Doing Business (and Saving Tons of Money) in My Retirement
The Counterargument
The counterargument is simply that IRMAA is one of the few places in the tax code where making just a few dollars more can cost you far more than you actually earned in additional income. Most tax rules phase in and phase out gradually, but not this. So, if your AGI is very close to these income levels ($97,000, $123,000, $153,000, $183,000, $500,000, or the MFJ equivalents) why not make a little change to avoid paying extra if you can? I don't see this as a particularly strong counterargument, but $1,900 is $1,900 and I've probably done more for less at some point in the past.
IRMAA is a surprise to many well-to-do retirees. While it is fun to gripe about it, don't spend much time and effort trying to avoid it with complicated financial planning. Doing so won't save you much money.
If you need extra help with planning for retirement or have
questions about the best way to save your money in tax-protected accounts, hire a WCI-vetted professional to help you figure it out.
What do you think? Am I downplaying IRMAA too much? Should people worry about it more? Comment below!
Just a caution…I typically “fill my bracket” by rolling regular IRA funds into a Roth every year. I didn’t consider IRMMA when doing this. The bracket breaks for the IRS are not the same as those for IRMMA. The last $5000 I rolled cost my wife and I about an extra $2000 in premiums. That coupled with a 22% federal and 7% state plus the Obamatax supplement yielded an over 70% tax rate.
You meant 70% tax off the “$5000” – correct !?
No wonder IRMAA brackets are called cliffs – be mindful of those .. especially if you AGI is hanging nearby those cliffs !!
If you Roth-converted only $2500 (instead of $5000) — you prolly could have lost full 100% of it to taxes too ! Durn Cliffs again !!
IRMAA is based on your Modified Adjusted Gross Income (MAGI) from two years ago. This “delayed “ taxation throws off lots of tax calculations, and tax software— so, be mindful if one is specifically hanging close to those cliffs.
Only 7 % of folks pay IRMAA – and, there are life changing circumstances which you possibly can “apply” or work with to remove IRMAA. Just don’t simply lay-down and pay the surcharge- at least give the “request “ a try ..
A majority of the times – IRMAA could be a “self inflicted “ problem – due to excessive/overly -greedy Roth conversions ; so, “converters” – be extra careful with these brackets/cliffs 🙂 (especially in conjunction with unexpected year end Dividends or surprise interest/1099 – which now could push your AGI over one of those Brackets/Cliffs)
If you have a life changing event such as retirement that lowers your income you can reduce or eliminate your IRMAA surcharge.
https://www.ssa.gov/medicare/lower-irmaa
Great tip, thanks for sharing.
(I inadvertently posted this as a reply rather than a comment when this topic first hit the web (WCI))
Yep, it was a bit of a wake up when I crossed that age 65 threshold. My healthcare had been covered by Tricare up to that point-which for me had no annual premium cost. I was still working 50% time and my younger spouse was at 50% time as well. Even at reduced hours, with “unearned” income and military retirement we crossed the next to highest threshold and therefore I was required to pay a surcharge for parts A+B. My status for Tricare changed to essentially a Medicare supplement policy which also covered medication (therefore no need for part D).
A couple of points: Medicare considers you income from 2 years prior to determine your surcharge (which you can appeal), and as long as you are still working in an IC status, the insurance (Medicare) premiums are a business expense.
Interesting that you’re one of the few whose health care insurance costs went up when you went on to Medicare. My parents had a similar experience (state pension/health care until 65).
Via email:
Yep, it was a bit of a wake up when I crossed that age 65 threshold. My healthcare had been covered by Tricare up to that point-which for me had no annual premium cost. I was still working 50% time and my younger spouse was at 50% time as well. Even at reduced hours, with “unearned” income and military retirement we crossed the next to highest threshold and therefore I was required to pay a surcharge for parts A+B. My status for Tricare changed to essentially a Medicare supplement policy which also covered medication (therefore no need for part D).
A couple of points: Medicare considers you income from 2 years prior to determine your surcharge (which you can appeal), and as long as you are still working in an IC status, the insurance (Medicare) premiums are a business expense.
But if you get hit by IRMAA because of cap gains or something not continuous, doesn’t the penalty for too much earnings go down automatically in 2 years?
Yea. There’s a two year delay in IRMAA. Should have mentioned that in the article.
My financial advisor said IRMAA premiums were ‘eligible’ to be revised lower in succeeding years (when income fluctuates downward), but it was not automatic.
That is, a taxpayer has to initiate a request.
That’s right.
For medium- to long-range tax planning, IRMAA adds about 5% to one’s expected marginal tax rate for taxpayers in the range of the first through third IRMAA tiers (MAGI ~$97-183k single or $194-366k married, ranges which will include many doctors). That can easily be enough to tip some “traditional versus Roth” decisions in favor of Roth.
Except to get the Roth you will likely need to do conversions, which again, will increase IRMAA tax.
Not if you contribute to Roth accounts directly. And Roth conversions before age 63 do not trigger IRMAA.
Here’s an example that may blow your mind: single doc, 30 years old, top (37%) tax bracket, independent contractor with $66k solo 401k limit, no pension, expects to retire at 65 with close to maximum (~$43k/year) SS benefits. Going 100% pre-tax with the solo 401k is the obvious choice, right? Or is it?
When you account for the effects of SS taxation phase-in, IRMAA, and the mathematical benefit to Roth when contributing the maximum, you will find it only makes sense for this doc to enter retirement with about $800k in pre-tax. His/her “lower brackets” are 0%, 18.5%, and 22.2%, which together span only about $30k/year of total pre-tax withdrawals (~$30k / 4% = ~$800k). Above that, there is a big 40.7% bump for SS taxation phase-in (49.95% if s/he has any qualified dividends), then the rate drops down again, but IRMAA starts up at ~$66k. Mathematically, Roth is the better choice for all savings that will come in above that $30k/year. How much pre-tax savings does it take to get to $800k by age 65 with a 4% real return? If saving equal proportions each year it’s about $11k pre-tax and $55k Roth. Or, the doc could go 100% Roth until age 55 then switch to 100% pre-tax for the last 10 working years. This assumes no years of Roth conversion before retirement, but those are limited to about ~$100k/year after age 63 without paying the IRMAA surcharge.
You left an important variable out of the equation, how much this doc earning at least $600K is saving. If that doc is saving 20% and working to age 65, there’s no way the doc is going to be playing in the lowest bracket.
=FV(5%,35,-120000) = $10,838,436.88 in today’s dollars
Even if it was all invested super tax-efficiently, it’s going to be hard to have less than $300K a year in income. Even paying max IRMAA isn’t a big deal at that point. Talking about breakpoints at $66K is kind of silly when 4% is more than $400K a year.
But hey, there are lots of WCIers for whom Roth is already the right choice and this could possibly make it a little more right. But as far as how many it flips from tax-deferred being right to tax-free being right, I’m a little more skeptical.
A couple things.
First, in the personal finance enthusiast community, we’re used to seeing 20, 30, 40%++ savings rates. In reality, a tiny fraction of Americans save this much. If my example doc saved just $66k/year in a solo 401k mostly as Roth, plus a backdoor Roth IRA, and nothing in taxable, s/he would have about (using your 5% real return):
FV(5%, 35,-66000-6500)*4% + $43500 [SS] – $4500 [income tax on the small pre-tax withdrawals] = ~$301k/year after-tax income in retirement
Sure, that’s less than the $600k gross income during working years, but it’s not directly comparable – you have to subtract income and payroll taxes, retirement savings, mortgage on a surely expensive house that will be paid off by retirement, child-related expenses that are gone by retirement, etc. My family grosses about $500k in a VHCOL area, and after subtracting taxes, mortgage, savings, and child expenses [daycare etc.] we live on about $85k/year. $300k free and clear would be a MASSIVE increase in lifestyle. So, I reject the premise a doc earning $600k/year would NEED to enter retirement with a $10M savings to maintain a similar lifestyle as while working. But I acknowledge a certain percentage of your readers with that income will save that much anyway.
Second, let’s go with your numbers and say this doc saves ~$50k/year in taxable on top of a solo 401k and backdoor Roth, to hit a 20% overall savings rate. After 35 years that would be worth about:
FV(5%-2%*23.8%,35,-50000) = ~$4M [in addition to ~$6.5M in retirement accounts]
Dividends from that portfolio would be about $80k/year, which fill up all the lower brackets and create a complex tax curve, with 27%, 22%, 24%, and 27.8% brackets with IRMAA spikes sprinkled in. When contributing the maximum, over a 35 year time horizon, traditional savings @ 37% is just about equal to Roth (within a couple percent) when the withdrawal tax rate is in the mid-20%’s. It’s far from obvious that 100% pre-tax is the way to go. IRMAA is just one factor here, but I would argue it’s a significant one.
I think your main point is that people with a taxable income of under $100K in retirement should definitely think about IRMAA when doing financial planning and I agree with that. Anything more specific, well, we’d have to get into specific numbers and actually run them. Hard to make rules of thumb on this topic.
Don’t forget that LTCGs and qualified dividends stack on top of ordinary income when it comes to filling brackets.
The ranges of greatest IRMAA impact are about $100-200k (single) or $200-400k (married) adjusted gross income, which includes pre-tax withdrawals, dividends/cap gains, and whatever percentage of Social Security is taxable based on the phase-in calculation. Those ranges will include many if not most doctors. Being in the IRMAA range doesn’t automatically mean Roth is the right answer – IRMAA in that range adds about 5% to one’s marginal tax rate on average – but it should be included in the calculation and can make a significant difference. In my example, IRMAA made the difference between pre-tax clearly winning, and it being a toss-up between pre-tax and Roth. It makes less of a difference for married taxpayers, because the thresholds are doubled and married taxpayers tend to have less than twice the income of single taxpayers (in my experience).
Yes, long-term capital gains and qualified dividends stack “on top of” ordinary income. When you are combining phase-in of Social Security, LTCG/qualified dividends, and IRMAA, throw the traditional tax brackets (standard deduction, 10%, 12%, 22%, etc.) right out the window, and use some tax software to generate plots of your actual marginal tax rate vs. additional income. In my example above with $43.5k SS and $80k QD, the first dollar of tIRA withdrawals is taxed at 27%: 12% on the dollar itself, and another 15% because it pushes up a dollar of QD from the 0% to 15% brackets. It’s basically impossible for this taxpayer to pay less than ~27-28% on any pre-tax withdrawals.
That’s not really true. The ranges are actually very small because IRMAA is dumb with its cliffs. If you look at MFJ, the range is $193-195K. Then NOTHING happens from $195K to $245K. Then there is another range from $245K to $247K. Then NOTHING from $247K to $305K. And so on.
So there is zero benefit or penalty for most people to make little adjustments. If your MAGI is $275K, you can go up or down in income by $25K with no change whatsoever.
So it’s pretty impossible to write a rule of thumb about this other than, hey, take a look at the MAGI cliffs AND the top of your next tax bracket when doing Roth conversions. And maybe, if your accumulation tax rate and your retirement tax rate look like they’re going to be very close, maybe favor Roth a little more because you might get burned by IRMAA.
Otherwise, you just have to run the numbers and even then, so much of your taxable income can be out of your control that you still might get hit.
But my point is that even if you do get hit, it’s not that big of a deal. So you go from $428 to $527 a month. $99 more. $1200 a year. For someone with $400K in taxable income in retirement. This is a non-issue and not something one should spend a lot of brain power on.
I agree using tax software is the best way to run these calculations, should you feel the desire to do so.
I might seem huge to you (“5% higher tax rate on this conversion due to IRMAA!”) but when you calculate the actual dollars, it’s not a big deal.
You’re right that IRMAA is a series of small tax cliffs, rather than brackets. I’m not sure where your $2k-wide ranges come from, though – even if you are $1 (or $0.01 if it rounds you up to the next $1) over the MAGI threshold, you pay the entire surcharge for that tier. If you are close to an IRMAA spike and go over it with a small amount of income, the marginal rate on that extra income can be large: trigger a $1,500 IRMAA surcharge with $3,000 of income, and the marginal rate is 50% on top of the base tax rate. If the income doesn’t cross tiers and is small enough to stay within a tier, the marginal rate is zero. But for anything but very short-range planning, it’s not practical to maneuver around the spikes like that. There’s too much uncertainty with investment performance and other variables. That’s why I usually recommend taking the average marginal rate, which is around 5%, and adding it to one’s predicted future marginal rate if they expect to be in the densest IRMAA range of the first through third tiers. That 5% may or may not tip the balance for making traditional versus Roth decisions; if it’s already a close call, it might. If someone a priori expects to be fourth or higher tiers, or well under the first tier, then IRMAA can be ignored as part of one’s tax rate.
Heck, the IRMAA spikes are difficult to plan around even in a single year, because the two-year delay requires estimating the next two years of inflation to get the income limits that you’ll actually be measured against. Because inflation numbers come out around November, folks who Roth-convert in December only have to estimate one year.
If I gave the impression I thought 5% is an enormous increase in tax rate, I apologize – it’s not. Not too small to ignore, not too big to dominate. I do recommend folding it into tax planning calculations as one of many factors, then run the numbers and see if it changes the decision. I’d also agree that a likely 5-figure IRMAA cost won’t make a meaningful difference in the financial life of most of your readers, who will have multi-6-figure incomes in retirement. But your readers also usually care about squeezing out that last few percent from their investments and tax planning, so many will want to tax-plan around IRMAA.
yea, the truth is they’re $1 cliffs but the closer you get to them the more attention you have to pay to income/deductions.
Easy for the WCI Author to say “No big deal…don’t sweat it.”
IRMAA has a cliff feature, and $1 error in planning income this year can cost $1,000s in 2025. That’s a big deal to me.
Moreover, it is a penalty whose cost and actual cliff aren’t known. For instance, the IRMAA penalty for 2025 will be based on 2023 income, and won’t be known until 4Q24. Good luck planning around that.
Even more reason not to worry about it.