The RMD Planning Window Most Solo RIAs Never Open

Executive Summary

Ask an advisor about required minimum distributions and most describe a task that runs itself. The custodian’s system flags the account the year the client turns 73, calculates the number, and the withdrawal happens without anyone touching a keyboard. Congress has spent the last few years making that even more true. The starting age moved from 72 to 73. The penalty for missing one dropped from half the shortfall to a quarter. Fix it inside two years and the number drops again, to a tenth. Every year the IRS nudges the charitable giving number up for inflation. RMDs have gotten quieter, not louder, and most advisors have read that as one less thing to manage.

That’s backward. The RMD itself was never the planning problem, it’s a calculation, and calculations are exactly the kind of thing that runs fine on autopilot. The planning problem sits years earlier, in the bracket a client retires into and never gets revisited before the withdrawals start, and in a $111,000 charitable tool that most advisors haven’t mentioned since the number was $100,000 flat. The quieter the withdrawal gets, the easier it is to miss that the real decision already happened, and already closed.

Diane runs a solo practice outside Sacramento, about ninety households, mostly retirees. Walt had been with her twelve years when he turned 73 this spring. His RMD calculated correctly, hit his account on schedule, and generated a 1099-R with no errors on it. Nothing about the withdrawal itself went wrong.

What went wrong happened three years earlier, the year Walt retired and had the lowest taxable income of his life, before Social Security started and before the RMDs began. Nobody ran a conversion that year. Now his RMD lands on top of Social Security and a small pension, in a bracket that was avoidable, and it’s going to land there every year for the rest of his life.

The RMD Age Moved To 73, And It’s Still Rising

The IRS is direct about the current rule: account holders generally have to start taking withdrawals from an IRA, SIMPLE IRA, SEP IRA, or workplace retirement plan when they reach age 73. That’s already a change from the age 72 rule most advisors trained under, and it isn’t done moving. T. Rowe Price’s own analysis of the SECURE 2.0 rules lays out the next step: the same law that pushed the age to 73 in 2023 schedules a further increase to 75 by 2033.

That sounds like relief and it partly is. A later start gives an account more years of tax deferred growth before the withdrawals begin. What it doesn’t do is move the planning deadline, because the planning deadline was never age 73 to start with. It’s whatever year a client’s income drops, usually the retirement year, before Social Security and before the RMDs stack on top of it. Pushing the withdrawal age back gives a client more years in that low bracket if somebody uses them, and the same number of years to waste if nobody does. Walt’s window was the single year between retiring and turning 73. He had it. Nobody opened it.

The Penalty Got Smaller, Which Made The Mistake Easier To Ignore

The IRS lays out the current penalty structure directly: an account holder who doesn’t take the full RMD owes an excise tax of 25% on the shortfall. Correct it within two years, file Form 5329, and the rate drops to 10%. That’s down from the flat 50% penalty that used to apply to every missed dollar, and advisors talk about the change like it lowered the stakes on RMD planning generally.

It didn’t. It lowered the stakes on one specific failure mode, forgetting to take the withdrawal or miscalculating the amount, which was never the expensive mistake to begin with. Most custodians calculate the number correctly and most clients take it on schedule, Walt included. The excise tax rule protects against a clerical error. It has nothing to say about a client who took a perfectly calculated, perfectly timed distribution into a bracket that a one-time conversation three years earlier would have avoided. A smaller penalty on a rare mistake doesn’t make the common one any cheaper.

The Charitable Giving Number Just Moved Again

The IRS confirmed the 2026 qualified charitable distribution limit in its most recent retirement plan bulletin: the amount an IRA owner can send directly to charity and exclude from taxable income rises to $111,000 for 2026, up from $108,000 in 2025. The number is indexed for inflation now under SECURE 2.0, which means it moves every year, and it started life as a flat $100,000 that held for over a decade before indexing began.

A QCD sent straight from the IRA to a qualified charity counts toward the RMD and never touches the client’s taxable income, which for a charitably inclined client is close to a free option on the entire withdrawal. The catch is that it has to happen before the RMD is otherwise satisfied and it has to go directly from custodian to charity, not through the client’s hands first. Most advisors know QCDs exist. Fewer bring the specific, current-year dollar figure into the year-end conversation, and fewer still are checking whether last year’s number is still this year’s number before they do.

Advisors Are Aging Into The Same Calendar

This isn’t only a client-facing problem. We pulled our own database of registered investment adviser representatives and looked at age directly. Of 24,055 advisors on file, 1,286 are estimated at 65 or older, and 189 of those are 70 or older. That’s roughly one in nineteen advisors in our own dataset sitting inside or approaching the same window their own clients are walking through, and it’s likely an undercount, since our age estimation method caps out at 72 rather than tracking higher.

An advisor running the RMD conversation for ninety households while ignoring it on his own accounts is telling on himself. The same reactive pattern that costs a client a bracket costs an owner a valuation, and it’s the identical mechanism we’ve written about on the succession side, the piece of a transition that doesn’t compress no matter how late you start it is the piece that has to happen years before the deadline everyone’s watching, not on it.

What A Proactive RMD Review Actually Looks Like

Saying “plan the bracket early” is easy. Advisors ask what that means on a calendar, and it’s a fair question, because most of what gets written about RMDs stops at the principle.

It means running a tax projection the year a client retires, while the number is still useful. By the year they turn 73 the decision is already made. It means treating the QCD limit as a number to reconfirm every December instead of a fact learned once and never revisited. It means a client’s beneficiary designations and account titling get checked well before the first distribution, since an inherited account carries its own distribution rules that are easier to sort out with time than under a deadline. None of this is complicated work. It’s work that has to happen on a schedule that doesn’t wait for the client to ask.

Building A Three-Year RMD Readiness Review

We call it the three-year review, because that’s roughly the useful runway once retirement is in sight and the timeline stops being theoretical.

1. Run a bracket projection the year a client retires or the year income first drops, and flag every year afterward where conversions are cheap, before Social Security and RMDs stack on top of the picture.
2. Model whether a QCD can offset some or all of a projected RMD for a charitably inclined client, using the current year’s indexed limit, not last year’s number carried forward out of habit.
3. Confirm beneficiary designations and account titling well ahead of the first distribution year, so an inherited account isn’t the thing that surfaces a gap during a review nobody scheduled for it.

Three steps, not ten, because the useful window is short and most of what fills a longer checklist doesn’t change the outcome. The bracket work is the one that actually moves money. The other two protect it.

Walt’s RMD will keep calculating correctly every year from here, on autopilot, exactly like it’s supposed to. What won’t come back is the bracket he retired into for that one year, the one nobody used. Diane knows it now. It’s not something a custodian’s system was ever going to flag for her.

This kind of gap between what runs automatically and what still needs a human paying attention shows up across retirement income decisions generally. We’ve covered the same pattern in what the safe withdrawal rate number actually measures and in the Social Security claiming advice most clients already know. Our retirement planning coverage on this site walks through the rest of it.

Our goal is to make sure the planning window closes on purpose, not because nobody was watching the calendar.