How much should you move from a 401(k)/traditional IRA into a Roth this year? Built on 2026 federal tax law. Everything runs in your browser — the figures you enter are never uploaded anywhere.
The same answers the search box above gives, laid out to read through. Everything here describes how this tool works — it is not tax advice.
It answers one question: how much should you move from a traditional 401(k) or IRA into a Roth this year?
It works out the largest conversion that fits inside a tax bracket you choose, then checks that amount against the other thresholds a conversion can trip — Medicare IRMAA surcharges, the ACA subsidy cliff, Social Security taxability, and deduction phase-outs — and reports what the conversion truly costs rather than what your bracket implies.
It then projects the decision forward so you can see the effect on required minimum distributions and on what is left after tax.
Moving money from a pre-tax account (traditional 401(k), traditional IRA) into a Roth account. You pay ordinary income tax on the amount moved this year, and in exchange that money — and all its future growth — is never taxed again, and is never subject to required minimum distributions.
The entire game is rate arbitrage: convert in years when your rate is low, so you never pay a higher rate on that money later. This tool is about finding how much fits in the low-rate space you have this year.
No. Recharacterization of Roth conversions was eliminated by the Tax Cuts and Jobs Act in 2018. Once you convert, it is done, and the tax is owed for that year.
That is why the tool is conservative about thresholds and tells you to leave margin. A conversion sized against an estimate that turns out slightly wrong cannot be trimmed afterwards.
No. Every figure you enter — your age, balances, income, Social Security — stays in your browser. The calculation runs on your own machine. There is no server to send it to, and no code on this page reads a form field and transmits it.
Nothing is saved either: closing the tab discards your entries, which is why there is no “save” button. Use Print / save as PDF if you want a record.
The site does use Google Analytics, which counts visits and records which pages are viewed. That is traffic measurement, and it is separate from the calculator — it cannot see what you type. The privacy policy spells out exactly what it does and does not collect.
Your federal filing status, which sets the bracket widths, the standard deduction, the Social Security thresholds, and the IRMAA thresholds.
The single brackets are roughly half as wide as the joint ones, so the same income is taxed harder. That is why the survivor refinement matters so much for couples — and why you can learn a lot by switching this to Single and seeing what happens to the recommendation.
Married filing separately is treated strictly: the Social Security thresholds drop to $0 (assuming you lived together during the year), so benefits are taxed almost immediately.
Age drives more of this model than anything except income. It determines:
How many years the projection runs. It affects only the long-game section, never the recommended amount.
Common choices are life expectancy, or the point at which you expect assets to pass to heirs. A longer horizon flatters conversion, because tax-free compounding has more time to work — so if you set 40 years, treat the resulting number with suspicion.
The total of every pre-tax retirement dollar you hold: traditional 401(k), 403(b), 457, traditional IRA, rollover IRA, SEP and SIMPLE IRAs. Add yours and your spouse's together if filing jointly.
Do not include Roth accounts, taxable brokerage accounts, HSAs, or pensions you do not control.
This is the pool available to convert, the base that required minimum distributions are calculated from, and the ceiling on the recommendation — the tool will never suggest converting more than you have.
What you already hold in Roth accounts. It does not affect the recommended conversion amount at all — it is only a starting balance for the projection, so the ending Roth figure is realistic.
Enter 0 if you have none. If you are unsure, the recommendation is unchanged either way.
Money held outside retirement accounts: a brokerage account, savings, CDs, money market funds.
This is the single most important number for whether a conversion actually works, because paying the conversion tax from here — rather than from the IRA itself — is what makes the strategy pay off. Every dollar of tax you fund from outside is a dollar that stays inside the Roth compounding tax-free forever.
The projection also collects your RMD proceeds here, and pays IRMAA surcharges from here.
Average yearly investment growth, before inflation — inflation is a separate input, so do not subtract it here. Rough guides: conservative 5-6%, a 60/40 portfolio 7-8%, all-stock 9-10%.
Three things to know:
Because it is the softest number on the page, the long-game section re-runs the comparison at plus and minus two points so you can see how much the conclusion really rests on it.
Everything taxed at ordinary rates before any conversion: wages, self-employment income, pension and annuity payments, taxable interest, short-term capital gains, non-qualified dividends, and any RMD you are already required to take.
Do not include Social Security (it has its own field, because only part of it is taxable) and do not include the conversion itself — that is what the tool is solving for.
This is your starting point on the income ladder. The lower it is, the more room you have to convert cheaply.
The full annual benefit before Medicare premiums are withheld — the gross figure on your SSA statement, not what lands in your bank account. Enter the household total if married.
Enter 0 if you have not claimed yet. Doing so usually widens the recommendation considerably, which is exactly why the years between retiring and claiming are the classic conversion window.
Investment income taxed at the preferential rates — qualified dividends and long-term capital gains — expected for the year. Include mutual fund capital gain distributions.
This matters more than people expect. These dollars sit on top of your ordinary income when taxed, so a conversion pushes them upward. If they were in the 0% capital gains bracket, a conversion can shove them into the 15% one — a real cost that never appears in any bracket table. The tool shows this in the marginal cost chart.
Interest from municipal bonds. It is not subject to federal income tax — but it is not invisible.
It counts toward the calculation that decides how much of your Social Security is taxable, toward the MAGI that sets your Medicare IRMAA tier, and toward the MAGI behind your ACA subsidy. So muni interest can push you over a threshold even though it is never taxed itself.
Leave this at 0 unless you itemize. At 0 the tool uses the standard deduction, which is what the large majority of retirees take.
If you do itemize, enter your total. The tool uses whichever is larger. One consequence worth knowing: taking itemized deductions means you lose the extra age-65 standard deduction, so the tool removes it automatically when your itemized total exceeds the standard one.
Your approximate marginal state rate as a percentage. Enter 0 for Florida, Texas, Tennessee, Nevada, Washington, South Dakota, Wyoming, Alaska and New Hampshire.
This is a flat-rate approximation, not a real state bracket schedule — so treat state figures as indicative. Switch on the State tax detail refinement if your state exempts Social Security or gives retirees an income exclusion.
The state rate you expect to face in the projection years. Set it lower than your current rate if you plan to move somewhere cheaper.
This matters enormously and is often overlooked. States tax you as a resident in the year the conversion happens. Converting a large sum the year before moving from California to Nevada can cost tens of thousands that waiting would have avoided entirely. If the two rates differ, the tool flags it in the notes.
The rate you believe the remaining pre-tax money will eventually face — either by you in later years, or by whoever inherits it.
This is the benchmark the whole verdict is judged against. If the effective rate you pay converting now is below this, the tool says the conversion looks worth doing; if above, it tells you to think twice. So it is worth setting deliberately.
Points to weigh: inherited IRAs must now be emptied within 10 years, often landing in an heir's peak earning years; a surviving spouse files single at roughly half the bracket width; and your own RMDs may push you higher than you are today.
Tick this if your state excludes IRA and 401(k) distributions from taxable income — Illinois, Pennsylvania and Mississippi are the clearest examples.
When ticked, the conversion is removed from the state tax calculation entirely, though your other income is still taxed at the rate you entered.
Tick this if you will pay the conversion tax from money outside the IRA. Leave it unticked if the tax will be withheld from the conversion itself.
This single choice often decides whether converting is worth doing. Paying from outside means the full converted amount lands in the Roth. Withholding from the IRA means less arrives, and the strategy loses most of its advantage.
Under age 59½ it is worse still: the withheld portion counts as an early distribution and carries a 10% penalty on top. The tool warns you when this applies.
If you tick this but your taxable balance cannot cover the plan, the tool says so rather than quietly assuming money that is not there.
The bracket you are willing to fill. The tool finds the largest conversion that keeps your taxable income at or below the top of it, without spilling into the next.
There is no universally right answer. Common approaches: fill the 12% bracket if you have a long low-income window; fill 22% or 24% if large RMDs are coming and you want to drain the pre-tax account faster. Compare them in the "Every bracket, priced out" table, which shows the true effective rate of each choice — often quite different from the bracket number.
How many consecutive years the projection assumes you repeat the recommended conversion. It affects only the long-game section, not this year's recommendation.
The amount is treated as a real, today's-dollars figure and inflated forward each year, mirroring how the bracket thresholds themselves move — so year 5 converts more nominal dollars but fills the same bracket space.
The natural stopping point is the year your RMDs begin, since after that the RMD occupies the low-rate room a conversion would have used.
Used to index the tax brackets, deductions and thresholds forward through the projection, to grow your income and conversion schedule in nominal terms, and to convert the final figure back into today's dollars.
Internally the model deflates each future year back to 2026 dollars and applies the 2026 brackets, which is mathematically equivalent to indexing every threshold forward — and avoids inventing bracket figures for years the IRS has not published.
Note that the Social Security taxability thresholds and the NIIT thresholds are not inflation-indexed in real law, and the tool models that correctly: they bite harder every year.
On by default. When ticked, the recommendation is capped so your MAGI stays below the next IRMAA threshold, even if that means leaving bracket room unused.
Untick it to see what filling the bracket would cost. Sometimes crossing is right — if you are converting aggressively over several years, one year of surcharge can be worth it. The tool shows the surcharge as a dollar figure so you can judge rather than guess.
Eight effects that most conversion calculators ignore. All are off by default, because the main form already covers the common case. Each can change the answer substantially when it applies to you:
If you buy health insurance on the exchange before Medicare starts at 65, a conversion raises the income your subsidy is based on.
The enhanced subsidies expired on 31 December 2025, so for 2026 the hard cliff is back. At 400% of the federal poverty level the premium tax credit does not taper — it goes to zero. One dollar over and the entire credit is gone. For a household of two that line sits at $84,600.
For someone retiring at 62 this is normally a tighter constraint than any tax bracket, and the tool caps the recommendation there when the guard is on. It also shows the smooth erosion below the cliff, since your expected contribution percentage rises with income well before you reach it.
The module switches itself off at 65, when Medicare takes over and IRMAA becomes the relevant threshold.
The full, unsubsidised annual premium of the second-lowest-cost silver plan available to your household in your county — the plan the subsidy formula is pegged to, whether or not you actually buy it.
Find it on healthcare.gov or your state exchange by entering your county and ages. A couple in their early sixties commonly sees $24,000-$36,000 a year. The bigger this number, the more a subsidy is worth and the more costly it is to cross the cliff.
If you ever made non-deductible IRA contributions, you have basis — money already taxed once. That share of every conversion comes out tax free. Enter the figure from Form 8606, line 14.
Two cautions the tool models:
From age 70½ you can send money directly from an IRA to charity. It never enters your AGI at all, and it counts toward your RMD. The 2026 limit is $111,000 per person.
A QCD beats writing a cheque and taking a deduction, because staying out of AGI also lowers the numbers that drive Social Security taxability, IRMAA and — before 65 — your ACA subsidy.
Pairing a QCD with a conversion in the same year is a common move: the QCD makes room under a threshold that the conversion then fills.
Note the age is 70½, not your RMD age of 73 or 75.
When one spouse dies, the survivor files as single from the following year — roughly half the bracket width and half the standard deduction, on close to the same income, plus the single IRMAA thresholds which are half the joint ones.
The survivor also keeps only the larger of the two Social Security benefits, not both.
This is one of the strongest arguments for converting while both spouses are alive, and almost no calculator models it. In testing it added around $104,000 of lifetime tax in one scenario and roughly tripled the benefit of converting early.
Set the year you want to model. The year-by-year table shades the affected rows and shows the filing status.
If you raise the tax money by selling appreciated shares, that sale is itself a taxable event — capital gains stacked on top of the conversion, which can also push you over an IRMAA or ACA threshold.
Enter your cost basis as a share of the account. 100% means no embedded gain (cash or money market); 40% means 60 cents of every dollar sold is a taxable gain.
The tool solves this as a small fixed point, since the extra gain raises the tax, which requires selling more. If you hold anything at a loss, harvesting that first is the cheapest way to fund a conversion.
Two refinements to the flat state rate:
A 3.8% surtax on investment income once modified AGI passes $250,000 (married) or $200,000 (single). Those thresholds are not inflation-indexed, so they catch more people every year.
A conversion is not itself investment income — but it raises the AGI that triggers the surtax on everything else. That is a hidden 3.8% on top of your bracket, and this refinement makes it visible.
The field asks what share of the ordinary income you already entered is investment income. It is a subset, not additional income. Your qualified dividends and long-term gains are counted automatically.
By default benefits grow at your inflation assumption. Set a lower COLA to model benefits losing ground — which pushes more of your other income into higher brackets over time and generally strengthens the case for converting earlier.
The largest conversion that satisfies every constraint at once: it fills your chosen bracket without spilling over, stays under the next IRMAA threshold if that guard is on, stays under the ACA cliff if that guard is on, and never exceeds your actual balance.
The line beneath it tells you which constraint was binding. If it says "capped by Medicare" or "capped by your health insurance subsidy", the bracket was not the limiting factor — a threshold was.
If it says convert nothing, your income already exceeds the bracket you chose. Pick a higher ceiling to see what is available.
Total extra cost divided by the amount converted — the honest number. It includes federal tax, state tax, any IRMAA surcharge triggered, and any ACA subsidy lost.
It is routinely higher than your bracket. A conversion "in the 12% bracket" can genuinely cost 22%, 30% or more once Social Security taxability, deduction phase-outs and capital gain stacking are counted. That gap is the main thing this calculator exists to show you.
A straight comparison of the effective rate you would pay now against the future rate you entered.
Since it hinges entirely on the future rate you supplied, try a couple of values before trusting it.
The centrepiece. The dashed grey line is your statutory tax bracket — the rate most people assume they pay. The blue line is what an extra dollar of conversion actually costs.
The gap between them is everything the bracket table hides: Social Security being dragged into taxable income, the senior deduction phasing out, capital gains pushed from the 0% band into the 15% one, the NIIT, and the ACA subsidy eroding.
Read it as terrain. Convert into the valleys and stop before the walls. The green line marks the recommendation; red dashes mark an IRMAA cliff; a solid orange line marks the ACA cliff.
What you could convert if you filled each bracket in turn, and what each would really cost — broken into federal, state, IRMAA and lost subsidy, with the effective rate alongside.
The highlighted row is the recommendation. Rows marked "already above this bracket" mean your income exceeds that bracket before converting anything.
This is the table to scan when deciding whether to be more aggressive: it often shows the next bracket up costs far less, or far more, than the bracket numbers suggest.
Repeats the recommended conversion for the number of years you set, then coasts to your horizon, and compares that against doing nothing.
The chart shows the pre-tax balance under both paths plus the Roth building up, with the year RMDs begin marked. The table totals lifetime tax, IRMAA, lost subsidy, and the final after-tax value to you and your heirs.
Treat the bottom line as directional, not precise — it rests on assumptions about returns, inflation and future tax law. If the difference is under 2% of the total, the tool says so plainly.
Re-runs the whole comparison at your return assumption plus and minus two percentage points, because that input is the softest number on the page.
If the conversion stays favourable across all three, the conclusion is reasonably robust. If the sign flips, the projection cannot tell you what to do, and the tool says so — decide on the near-term numbers instead: the effective rate you pay this year against the rate you expect later.
Judge the percentage column rather than the dollars, since a higher return inflates every figure.
A side-by-side of your return with and without the conversion, from gross income down to total cost. It shows the taxable share of your Social Security, each deduction separately, taxable income, the bracket reached, federal and state tax, your Medicare MAGI and tier, and your ACA position if that module is on.
Use it to sanity-check the model against your own return or tax software.
Print / save as PDF produces a clean copy with the input form hidden and the results laid out for paper. It is the practical way to take the analysis to a CPA or advisor.
Nothing is saved automatically — reopening the file gives you the defaults again.
Social Security is taxed on a formula, not a rate. The tool computes provisional income — your other income, plus tax-exempt interest, plus half your benefits — and compares it to fixed thresholds: $32,000 and $44,000 married, $25,000 and $34,000 single.
Below the first, none is taxable. Between them, up to 50%. Above the second, up to 85%.
Those thresholds are not inflation-indexed and never have been, so more retirees cross them every year. The tool follows the IRS Publication 915 worksheet exactly.
In the range where benefits are phasing into taxability, each extra dollar of conversion drags up to 85 cents of Social Security into taxable income along with it. Your taxable income rises by $1.85 for every $1 converted.
So a dollar taxed "at 12%" really costs 22.2%, and at 22% it costs 40.7%. This is the torpedo, and it is clearly visible as a hump in the marginal cost chart.
There is a silver lining: once 85% of your benefits are taxable, the effect is exhausted. Further conversions drop back to the plain bracket rate. Sometimes the right move is to push all the way through the torpedo rather than stop inside it.
An income-related surcharge on Medicare Part B and Part D premiums. Two features make it dangerous for conversions:
2026 thresholds start at $109,000 single and $218,000 married. The tool uses these as the basis for future years, but the actual 2028 figures are not published yet — so leave margin.
A threshold where crossing by a single dollar costs you a fixed amount, rather than tapering.
Two appear here. IRMAA raises your Medicare premiums for a whole year — around $2,297 for a couple at the first step. The ACA subsidy cliff at 400% of poverty is far more brutal: the entire premium tax credit disappears, which can be $20,000 or more.
The practical rule is the same for both: convert right up to a cliff, or well past it, but never just over it. Landing $500 over is the worst possible outcome.
Because the thresholds you are aiming at are estimates and your income is not fully known until the year ends.
The 2028 IRMAA brackets have not been published — the tool projects them. A December mutual fund capital gain distribution, a bonus, or a larger-than-expected dividend can push you over after you have already converted. And conversions cannot be undone.
A few thousand dollars of headroom costs you very little; crossing a cliff costs a lot. Converting late in the year, when your income is nearly certain, is the standard defence.
From your RMD age you must withdraw a percentage of your pre-tax balance every year whether you need it or not. Under SECURE 2.0 that age is 73 if you were born 1951-1959, and 75 if born 1960 or later.
The amount is your prior year-end balance divided by an IRS Uniform Lifetime Table factor — 26.5 at 73, 24.6 at 75, falling every year, so the percentage forced out keeps rising.
This is the core argument for converting early. Every dollar converted before RMDs begin permanently shrinks the base those distributions are calculated from. Roth accounts have no RMDs at all.
Note that an RMD cannot itself be converted — it must be taken first.
Two separate things apply at 65. The long-standing extra standard deduction ($2,050 single, $1,650 per spouse married), and the new senior bonus deduction of $6,000 per person created by the OBBBA.
The bonus phases out at 6% of income above $75,000 single / $150,000 married, which creates a hidden marginal rate: inside the phase-out each dollar converted costs your bracket times 1.06, or times 1.12 for a couple where both qualify.
It expires after 2028. If it is materially widening your conversion room, that window is only three years long.
Qualified dividends and long-term gains are taxed after your ordinary income, sitting on top of it. So a conversion — which is ordinary income — pushes them upward through their own brackets.
If your gains were in the 0% capital gains band, a conversion can shove them into the 15% one. Each converted dollar then costs your bracket rate plus 15 cents of newly taxed capital gain. The tool accounts for this and names it in the marginal cost explanation when it is a factor.
The conversion itself carries no 10% early distribution penalty at any age. But two things apply:
The tool flags both when your age is under 59½. It does not model the five-year clocks as a cash-flow constraint — that is a liquidity question, not a tax cost.
There are two, often confused. One governs earnings: your first Roth IRA must have been open five years before earnings come out tax free. The other governs converted principal: each conversion has its own five-year clock before that principal can be withdrawn penalty-free under 59½.
The calculator does not model either as a constraint. It assumes converted money stays invested, which is the normal case. If you might need the money within five years, that is a reason to convert less than the tool suggests.
The tool does not model timing, but the practical answer is late in the year, for one reason: by then your income is nearly known.
Conversions must be completed by 31 December — unlike IRA contributions, there is no grace period into April. Converting in November or December lets you size the amount against actual figures rather than estimates, which matters enormously when you are aiming just below a cliff that cannot be undone.
The counter-argument is that converting early in the year gives more time in the market. If you are nowhere near a threshold, that argument wins.
2026 federal law throughout. Specifically: brackets and standard deduction from IRS Rev. Proc. 2025-32; the OBBBA senior deduction; Social Security taxability per Publication 915; long-term capital gain stacking; the 3.8% net investment income tax; 2026 Medicare IRMAA tiers from the CMS premium announcement; SECURE 2.0 RMD ages; ACA applicable percentages from Rev. Proc. 2025-25 with the 2025 HHS poverty guidelines; and the 2026 QCD limit of $111,000.
Future years are modelled by indexing these forward at your inflation assumption, except where the law does not index — Social Security and NIIT thresholds are correctly left fixed.
Known simplifications, stated plainly:
No. It is a planning model, not advice, and it was not written by a tax professional.
Use it to understand the shape of your decision — where the thresholds sit, what a conversion really costs, which constraint is binding — and then confirm the numbers with a CPA or fee-only advisor before you act. A conversion cannot be reversed, so the cost of being wrong is permanent.
The calculation engine is covered by more than 450 automated checks against hand-computed IRS figures: every bracket boundary for all four filing statuses, the Publication 915 Social Security worksheet, capital gain stacking across band boundaries, deduction phase-outs, IRMAA tiers, the ACA applicable percentage table, and the pro-rata basis rule.
The projection is checked for conservation of dollars — every dollar must be accounted for as a balance or a tax paid — and against a rate-matched case where converting should be close to a wash.
That said, tests confirm the model matches its own intent. Check the line-by-line table against your own return before relying on it.
Only questions about this calculator: what a field means, how a number is computed, how to read a chart, what a threshold does, and what the model does not cover.
It cannot answer questions about your personal tax situation, give advice on what you should do, or discuss topics outside this tool. For those, talk to a CPA.