What the WSJ’s recent Roth conversion piece gets right — and what it leaves for households with substantial retirement balances to figure out on their own.
https://www.wsj.com/personal-finance/taxes/for-retirees-in-their-60s-the-move-that-adds-years-to-a-nest-egg-199e467e (behind a paywall)
OK so there’s this window. Between roughly age 60 and 65 — after you’ve stopped working (probably), before Medicare kicks in (definitely), and long before required minimum distributions arrive at 73 (feels far away, isn’t) — you exist in a kind of financial airlock. Your taxable income is often the lowest it’s been since your first job out of college.
This is either a problem or an opportunity, depending on how you look at it.
The problem framing: less money coming in, closer to running out, scary.
The opportunity framing: less money coming in means a lower marginal tax bracket, which means you can move money from a traditional IRA (taxed when you eventually withdraw it) into a Roth IRA (never taxed again, ever) at a discount that will not come around a second time.
Per the Wall Street Journal piece linked above, retirees are figuring this out en masse. Fidelity reports Roth conversions grew 45% in the twelve months ending July 2024. About 300,000 Americans in their 60s converted $14 billion to Roth accounts in 2020, up from 137,000 who converted $4.4 billion just three years earlier. Baylor’s William Reichenstein estimates that well-timed conversions during the early retirement window can extend a portfolio’s life by up to seven years.
The trend is real. The math is real. So let’s think about what’s actually happening.
You spent your career being told to put money in a 401(k) or a traditional IRA. You did that. You got a tax deduction each year, and you were told the money would grow tax-deferred for decades and that when you eventually withdrew it, you’d be in a lower tax bracket than during your working years.
That last part turned out to be — let’s say — not necessarily accurate.
For many retirees, particularly those with meaningful account balances, retirement tax rates are not obviously lower than working rates. Social Security is largely taxable. Pensions are taxable. Investment income is taxable. And starting at age 73, the IRS requires you to take minimum distributions from your traditional IRA whether you need the money or not, taxed as ordinary income on top of everything else you’re already reporting.
The marginal rates you were “saving” on decades ago — 22%, 24%, maybe 32% — turn out to be roughly the same rates you pay on withdrawal. Sometimes higher.
So the Roth conversion move works like this: in a year when your income is low, you take some money out of your traditional IRA, you pay tax on it at your current low bracket, and the money then goes into a Roth where it grows tax-free forever, doesn’t generate RMDs during your lifetime, and eventually passes to your heirs tax-free.
The mechanics are simple. The math is compelling. The window is real.
The Journal illustrates it with a 65-year-old couple: $2 million in an IRA, $300,000 in the bank. They live off the bank account and convert $400,000 per year to Roth for three years. Three years later, they have $1.1 million in Roth and $1.1 million still in traditional — versus $2.4 million in traditional under the do-nothing scenario. Total estimated lifetime tax savings: $334,433.
Which is real money.
But.
There is a thing the article mentions in passing that turns out to be the actual center of the story for a specific subset of retirees.
Near the end of the piece, a retired CFO named Bill Simonis is quoted explaining why he’s been aggressively converting: he’s “especially eager to stockpile Roth savings for the surviving spouse, since single filers get hit with higher tax rates at lower income levels.”
Bill Simonis — or, more likely, whoever is advising him — has intuited what we’ve come to call the widow’s tax cliff.
When one spouse dies, the surviving spouse files Single. The Single-filer brackets are roughly half as wide as the Married Filing Jointly brackets. The standard deduction is halved. The same required distributions that were taxed at a moderate rate under MFJ are now taxed at substantially higher rates.
Most articles on Roth conversions mention this problem in a sentence, if at all. Most retirement projections silently assume MFJ filing forever, ignoring the near-certainty that one spouse will outlive the other and spend years or decades filing Single on the same accumulated balance.
For most married couples with substantial retirement accounts, the widow phase produces more total lifetime tax than the entire joint-life phase that preceded it.
Bill Simonis has figured this out. Most people haven’t.
Which brings us to the other thing the article doesn’t quite say.
The Journal’s example uses a $2 million IRA and shows a three-year conversion sequence that moves roughly half the account into Roth. That works because the account is small enough — and the taxable-side reserves are large enough — that $400,000 per year of conversions makes meaningful headway on the balance.
For an IRA of $3 million or $5 million, the same conversion pace does not catch up. Converting $400,000 per year for three years is $1.2 million — less than 25% of a $5 million account. The other $3.8 million is still there. Still growing. Still going to trigger mandatory distributions at 73. Still going to compress through Bill Simonis’s very-real widow’s tax cliff. Still going to hit the couple’s children under the SECURE Act’s 10-year drawdown rule during their peak earning years.
For households in this range, basic Roth conversions during the airlock window are a starting point, not a solution. The account compounds faster than a three- or five-year sequence of modest conversions can drain it. What actually solves the problem is a structured, multi-year architecture combining accelerated Roth conversion sequencing with tax-aware partnership investments, real estate-based conversion strategies, and careful basis step-up planning. The full architecture is worth its own conversation.
We’ve written a longer working analytical memo on this version of the problem — the IRA tax-trap that hits households with $3M+ traditional IRAs across the full family lifecycle, including the widow phase that Bill Simonis correctly worries about but that the Journal doesn’t quantify. If you’re in that range and want to see the analysis, request the memo.
The airlock window is real. It’s just, for some households, the beginning of a much longer conversation.
