Reasons to Consider a Roth Conversion (and Reasons Not To)

Take a hypothetical couple, both 64, both retired the previous spring, sitting on about $1.6 million in one old 401(k) and reasonably pleased with a taxable income that year of roughly $40,000. On the face of it they have done everything right. They are also about to spend the next eight years being taxed on almost nothing, and the twenty after that being taxed on a great deal.

That gap is the whole argument. Not "Roth good, traditional bad" (that framing has done more damage than help), but the specific observation that most people's lifetime tax rate is not flat, and there are usually a handful of years where it dips well below where it started and well below where it is going to end up. A Roth conversion is the tool for moving income into those years on purpose.

What follows is when that is worth doing, and when it very much is not.

What a conversion is, in one paragraph

You move money from a pre-tax account (a traditional IRA, a SEP, an old 401(k)) into a Roth IRA. The amount you move gets added to your ordinary income that year and taxed at your marginal rate. In exchange, everything in the Roth grows tax-free and comes out tax-free, with no required distributions during your lifetime. There is no dollar limit and no income limit. The only real question is whether the rate you pay today is lower than the rate you would have paid later. If you want to put numbers on your own situation, the Roth conversion calculator does the arithmetic.

Reason one: the window between the last paycheck and the first RMD

This is the big one, and it is the situation the couple above were in without knowing it.

Between the day you stop working and the day required minimum distributions begin at 73, there is often a stretch of years where your taxable income is the lowest it has been since your twenties and the lowest it will ever be again. You are living off taxable savings. You have not claimed Social Security yet, or you have claimed a reduced amount. Nothing is forcing money out of your retirement accounts.

That window is usually somewhere between five and ten years long, and it is the single best opportunity most people get to move pre-tax money at a low rate. Convert enough each year to fill up the bracket you are in and then stop. Do it again next year. By the time RMDs start, the balance they are calculated on is materially smaller, and a meaningful share of your retirement savings has moved permanently out of the taxable column.

The mistake people make here is not converting too much. It is not realizing the window exists, letting it close, and then discovering at 73 that their required distributions have pushed them back into a bracket they thought they had left behind for good.

Reason two: a year your income fell off a cliff

Same logic, different cause. A sabbatical. A layoff. A business that had a genuinely bad year. A move to part-time work. Any year where your income lands well below your normal is a year where converting is cheaper than usual, and those years tend to feel like the worst possible time to voluntarily generate a tax bill.

I understand the instinct. But the rate is the rate, and a year at 12% is a year at 12% regardless of why you are there. Business owners in particular should be looking at this. If your income swings by a factor of two or three across a cycle, the down years are not just something to survive. They are the cheapest tax years you will get, and the broader tax planning picture for an owner-managed business should have conversions in it somewhere.

Reason three: you can already see the RMD problem coming

Required minimum distributions are not optional and they are not sized to your needs. They are sized to your balance. If you arrive at 73 with a large pre-tax balance and modest spending, you will be forced to take money you do not want, at rates you did not choose, every year for the rest of your life.

The knock-on effects are the part people underestimate. Higher taxable income raises the share of your Social Security that is taxable. It can push you over an IRMAA threshold and raise your Medicare premiums. It can drag capital gains into a higher bracket. One forced distribution quietly makes three other things more expensive.

If you run your projected balance forward and the required distribution at 75 looks larger than the income you actually plan to spend, that is a signal. Converting during the window is how you shrink the base the calculation runs on. Our RMD planner will show you what the unconverted version looks like, which is usually the number that settles the argument.

Reason four: the market just fell

This is the tactical one, and it is genuinely useful. The tax on a conversion is assessed on the value of what you move, on the day you move it. If your account is down twenty percent, converting the same number of shares costs twenty percent less in tax, and the recovery, when it comes, happens inside the Roth where it is never taxed again.

You are not timing the market here. You are not changing what you own. The shares go across in kind and stay invested throughout. You are simply choosing to pay the toll on a temporarily lower valuation, which is the closest thing to a free option the tax code offers.

The practical version: keep a rough figure in mind for how much you would be willing to convert this year, and if a real drawdown arrives, do a chunk of it then rather than waiting for December. You can always do the rest later.

Reason five: what your heirs will actually inherit

Non-spouse beneficiaries who inherit a traditional IRA generally have to empty it within ten years. There is no stretching it across a lifetime any more. So the money comes out over a decade, taxed as ordinary income, landing on top of whatever your children are already earning, which for a beneficiary in their fifties is frequently their own peak earning years.

An inherited Roth still has to be emptied in ten years, but none of it is taxable. If your marginal rate today is lower than your children's will be when they inherit, converting is a straightforward transfer of tax liability from a high-rate payer to a low-rate one, and it happens to be you doing the paying.

The caveat is that this turns on their tax situation, not yours, and people are often wrong about it. If your beneficiary is a charity, converting is actively counterproductive: charities pay no income tax on an inherited IRA, so you would be volunteering to pay tax nobody was ever going to pay. It is worth a conversation before assuming. Some of the same logic runs through what happens on the receiving end of an inheritance, which is worth reading if this is the reason you are considering it.

The reasons not to, which are real

Conversions have a fan club, and the enthusiasm outruns the arithmetic more often than it should. The cases against:

You will be in a lower bracket later. This is the ordinary case for a high earner heading toward a modest retirement, and it is exactly what tax deferral was built for. Converting at 35% to avoid 22% is not clever. It is just paying more.

You would have to pay the tax out of the IRA. Paying from outside money is most of what makes conversions work: it effectively lets you shelter more, because the full amount goes across. Take the tax out of the account instead and you have shrunk the balance and paid tax on it. Worse, if you are under 59½, the withheld portion counts as a distribution rather than a conversion and picks up a 10% penalty.

You give to charity from your IRA. Qualified charitable distributions let you move IRA money to charity with no income tax to anyone, ever. Converting first and giving later throws that away entirely.

You are close to a threshold that matters. IRMAA tiers, the ACA subsidy cliff if you are on a marketplace plan before 65, the point where Social Security taxation steps up. These are cliffs, not slopes. A conversion that lands one dollar past one of them can cost more than the tax on the conversion itself.

You are relying on the 2026 argument. A great deal of conversion content still leans on the idea that tax rates were about to rise when the 2017 cuts expired. They did not. Those rates were made permanent. If someone is still pitching conversions on a deadline, they are working from an old script, and the case now has to stand on your own bracket trajectory instead.

If you decide to do it

Three things matter more than people expect.

Size it, do not just do it. Work out where the top of your current bracket is, subtract your expected income, and convert into the gap. The first dollars are cheap and the last dollars are not, and there is a point on that curve where it stops being worth it. That point is the answer.

Wait until you know your year. Conversions cannot be undone: recharacterization went away in 2018. Converting in January means guessing at a bonus, a business result, or a capital gain that has not happened yet. November and December are when you actually know. The exception is a market drawdown, which is worth acting on when it arrives.

Check for other pre-tax IRA money first. The pro-rata rule aggregates every traditional, SEP, and SIMPLE IRA you own and taxes any conversion proportionally across all of them. You cannot select the after-tax dollars and convert only those. This is the single most common way a backdoor Roth goes wrong.

None of this is complicated in isolation. What makes it hard is that the bracket, the IRMAA tier, the Social Security taxation, the capital gains rate and the subsidy cliff all move together when you convert, and getting one right by getting another wrong is the usual outcome of doing it alone. Run the numbers first: the calculator will tell you whether this is even close, and if it is close, that is the point at which it is worth an hour with someone who does it regularly.