[ER-02] The Roth Conversion Ladder

Turning “Locked” Pre-Tax Retirement Money into a Penalty-Free Early-Retirement Paycheck

Pillar: Early Retirement Mechanics · Applies to: Anyone retiring before 59½ with significant traditional 401(k)/IRA balances Last verified: August 2026 · Refresh cadence: Annual (bracket/limit figures via FN-02) + event-driven Related: FN-02 Key Numbers · HC-01 ACA Bridge · ER-03 SEPP 72(t) · ER-04 Rule of 55 · ER-05 Accessing Money Early · TX-01 0% Capital Gains · TX-02 Roth Conversion Strategy (post-65 focus)

Not advice. Conversions are irrevocable and interact with health subsidies, financial aid, and state taxes. Model your specific numbers before executing.

---

TL;DR

• The ladder solves early retirement’s central plumbing problem: most FIRE wealth sits in pre-tax accounts that charge a 10% penalty before 59½. A conversion moves money to Roth (paying ordinary income tax now); after a 5-tax-year seasoning period, that converted principal comes out tax- and penalty-free at any age.
• Convert one “rung” every year and, starting in year 6, you harvest a rung annually — a self-renewing pipeline. The catch: you must fund the first five years from somewhere else (taxable brokerage, cash, Roth contribution basis). Under-planning the bridge fuel is the #1 ladder failure.
• Sizing each rung is a tax-geometry problem, not a savings problem. Conversions stack against the standard deduction, the 10/12% brackets, the 0% capital-gains bracket, and — before 65 — the ACA subsidy phase-out from HC-01. Post-2025, the ACA side of that tradeoff got materially more expensive.
• OBBBA (2025) made the TCJA brackets permanent, which changed the rhetoric but not the math: the case for converting is no longer “beat the 2026 sunset,” it’s “convert in the years your personal rate is unusually low” — which, for early retirees, is exactly the bridge years.

---

1. The Problem the Ladder Solves

A typical FIRE portfolio is lopsided: years of maxed 401(k)s produce a large pre-tax balance and a smaller taxable account. Before 59½, traditional IRA/401(k) withdrawals generally incur ordinary income tax plus a 10% early-distribution penalty.

The ladder exploits an asymmetry in the rules:

• Converting traditional → Roth is allowed at any age, in any amount, with no penalty — you just pay ordinary income tax on the converted amount in that year.
• Withdrawing converted principal from the Roth is penalty-free once that conversion has aged five tax years (or once you’re 59½, whichever comes first).

So conversions made during low-income retirement years become spendable money on a five-year delay — usually taxed at 0–12% on the way through, versus the 22–32%+ deduction you took when contributing. That spread is the whole game.

2. The Two Five-Year Clocks (Get This Right First)

The single most-confused topic in Roth planning. There are two unrelated rules:

	Clock A: Conversion seasoning	Clock B: Qualified earnings
What it governs	Penalty-free withdrawal of converted principal before 59½	Tax-free withdrawal of earnings
How many clocks	One per conversion — each year’s rung has its own	One per lifetime — starts with your first-ever Roth contribution or conversion
Starts	January 1 of the conversion’s tax year (a Dec 20, 2026 conversion is deemed Jan 1, 2026)	Jan 1 of the year of your first Roth funding
Satisfied	After 5 tax years — the Dec 2026 conversion is spendable Jan 1, 2031 (~4 years, 1 month of real time)	After 5 years and age 59½ (or death/disability)
Becomes moot	At 59½ (no penalty applies regardless)	Never — but it’s usually long-satisfied by then

Withdrawal ordering rules (automatic, not elective — IRS ordering under §408A):

1. Contributions first — out anytime, tax- and penalty-free.
2. Conversions, oldest first (and within each conversion, the taxable-at-conversion portion before any non-taxable portion).
3. Earnings last.

This ordering is why the ladder works: your withdrawals draw down seasoned rungs before ever touching earnings.

Pro-move on timing: the January 1 back-dating means December conversions season in ~49 months, not 60. Convert late in the year — which conveniently is also when you know your income precisely (HC-01 §5).

3. Building the Ladder: The Pipeline

Retire at 45, needing $60k/year, first conversion December 2026:

Year	Convert (rung in)	Live on	Withdraw (rung out)
2026	$60k	Taxable/cash	—
2027	$62k	Taxable/cash	—
2028	$64k	Taxable/cash	—
2029	$66k	Taxable/cash	—
2030	$68k	Taxable/cash	—
2031	$70k	2026 rung	$60k ✅ seasoned
2032	$72k	2027 rung	$62k ✅
…	continue until ~age 55	…	…

Design notes:

• Bridge fuel: the plan needs ~5 years of spending (here ~$300k+) in penalty-free-accessible form before day one: taxable brokerage (only the gains are taxed), cash, existing Roth contribution basis, I-bonds, even a 457(b) — governmental 457(b)s have no early-withdrawal penalty at any age and are the most underrated bridge asset in the FIRE toolkit (ER-05).
• Stop converting ~5 years before 59½ (here ~54–55): rungs converted later would season after the penalty stops mattering anyway. Whether you keep converting for pure tax-bracket reasons is a separate, post-ladder question (TX-02).
• Inflate the rungs. Each rung should be sized to spending five years out, not today’s.
• Rollover first. Most 401(k)s can’t convert directly on your schedule — roll to a traditional IRA after separation, then convert annually. (Weigh the loss of ERISA creditor protection and any Rule-of-55 access before rolling — ER-04.)

4. Sizing Each Rung: The Stacking Framework

A conversion is ordinary income poured into your tax return from the bottom up. Size it against these lines, in order (2026 figures — live numbers in FN-02):

Layer 1 — The zero bracket. Standard deduction: $32,200 MFJ / $16,100 single. With no other ordinary income, conversions up to this line are taxed at 0% — the closest thing to free money in the code.

Layer 2 — The 10/12% brackets. An MFJ couple with no other income can convert ~$133,000 total ($32,200 + $100,800) at a blended federal rate of roughly 10%. That’s the classic “fat rung” ceiling.

What a rung actually costs. Ladder years are usually the lowest-income years of a lifetime, so the effective rate on a conversion is far below the bracket it reaches into. For an MFJ couple under 65 whose conversion is their only ordinary income:

Rung size	Taxable income	Federal tax	Effective rate	Marginal rate on the next dollar
$32,200	$0	$0	0.0%	10%
$60,000	$27,800	$2,840	4.7%	12%
$133,000	$100,800	$11,600	8.7%	22%
$150,000	$117,800	$15,340	10.2%	22%

The $133,000 rung — the whole 10/12% bracket, filled — costs 8.7%. Against the 24%+ deduction most of that money was contributed at, the spread is the entire point of the strategy, and it is wider than the bracket labels suggest. Note the last column: the marginal rate jumps to 22% the moment you pass $133,000, which is why that figure is the natural rung ceiling and why Layers 3–5 below are about ceilings that bind lower than this one.

Layer 3 — The 0% capital-gains collision. Ordinary income (conversions) stacks under long-term gains. Every conversion dollar pushes a gain dollar upward; once total taxable income passes $98,900 MFJ, gains that were at 0% become 15%. In that overlap zone a conversion dollar effectively costs 12% + 15% = 27%. If you’re also harvesting gains (TX-01), decide which strategy owns the year — they compete for the same bracket space.

Layer 4 — The ACA phase-out (pre-65 retirees). Every conversion dollar is MAGI. From HC-01: within 100–400% FPL you lose roughly 10–15¢ of premium tax credit per dollar, you can fall through CSR steps at 150/200/250% FPL, and $1 past 400% FPL vaporizes the entire credit — with unlimited repayment of advance credits now that the caps are gone. Practical resolutions:

• Pick a lane each year: alternate “subsidy years” (small rung, e.g., to the 199%-FPL line) with “conversion years” (fat rung, accept lean or no subsidy). Alternating usually beats splitting the difference annually because the CSR/cliff losses are step functions, not slopes.
• December sizing: convert to a pre-committed line after fund distributions post, with a $1–2k buffer under any cliff. Conversions cannot be recharacterized.

Layer 5 — Situational ceilings.

• College-bound kids: FAFSA uses prior-prior-year income — a conversion when your child is a high-school sophomore lands in their freshman aid calculation. Ladder parents should map conversion-heavy years around the aid base years (roughly ages 16 through 20 per kid).
• Age 63+: conversions enter the IRMAA two-year lookback (HC-06).
• Age 65+ (2025–2028): the OBBBA senior deduction phases out above $75k/$150k MAGI at 6 cents per dollar, which lifts your effective marginal rate by roughly 1.3–2.9 points inside the band (not by 6 — the 6% is the phase-out rate, not the surtax). A post-ladder concern, sized in TX-02 §5.
• Claiming Social Security: conversions drag more of the benefit into taxation via provisional income (SS-03) — a reason ladder-era conversions (pre-SS) are cheaper than post-SS ones.

5. Execution Mechanics (Where Real Money Gets Lost)

• Deadline is December 31, not tax day. A “2026 conversion” must settle in 2026.
• Never withhold taxes from the conversion itself if under 59½. Withheld amounts don’t reach the Roth — they’re treated as a distribution, incurring the 10% penalty and shrinking the rung. Pay via estimated payments or by cranking withholding on other income (withholding is treated as paid evenly all year — a Q4 lever).
• Estimated taxes: a December rung usually triggers a Q4 estimated payment (due Jan 15) unless you’re inside a safe harbor (100%/110% of prior-year tax). Ladder years with tiny prior-year liability have tiny safe harbors — cheap to satisfy.
• Convert in-kind during drawdowns: moving depressed shares converts more future recovery per tax dollar. (This replaced the old “recharacterize if it drops” play, which died with recharacterization in 2018.)
• Pro-rata & Form 8606: if your traditional IRAs contain any non-deductible basis, every conversion is proportionally taxable across all your traditional/SEP/SIMPLE IRAs (aggregated). Keep every 8606 forever; it’s also your conversion log for the 5-year clocks.
• Track rungs explicitly. A one-line-per-conversion log (year, amount, taxable portion, unlock date) prevents the classic error of spending an unseasoned rung.

6. Ladder vs. the Alternatives

	Roth ladder	SEPP / 72(t) ER-03	Rule of 55 ER-04
Age works from	Any	Any	Year you turn 55 (50 for public safety)
Lead time	5 years of bridge fuel	Immediate income	Immediate
Flexibility	High — resize/skip rungs yearly	Very low — locked schedule; busting it back-charges penalties + interest	Medium — plan must allow partial withdrawals
Tax character	You choose the year’s income	Fixed ordinary income	Ordinary income
ACA compatibility	Excellent (you steer MAGI)	Poor (income is forced)	Poor–medium
Best for	Retirees ≤ ~52 with taxable-side runway	No bridge fuel, need income now	55+ leaving an employer with a cooperative 401(k)

When the ladder is the wrong tool: you’re already 54+ (Rule of 55 or simple waiting dominates); your taxable account alone reaches 59½ (just harvest 0% gains and skip the complexity); pension or rental income already fills the low brackets; or your pre-tax balance is so large that RMD-era bracket projections argue for converting aggressively regardless of ACA cost — that’s a TX-02 optimization wearing an ER-02 costume.

7. Worked Scenarios

A. The classic ladder couple. Ages 45/44, $70k spend, portfolio $1.8M: $900k pre-tax, $600k taxable (basis $420k), $300k Roth (contributions $180k). Bridge fuel = taxable + Roth basis = ample. They alternate lanes: odd years convert $64k (lands ~199% FPL with their ~$12k of dividends and harvested gains → 87% AV Silver); even years convert $110k (subsidy shrinks, bracket space used). Rungs unlock from year 6; Roth contribution basis absorbs any shortfall. Federal tax on ladder income averages ~8–10%; the money went in at 24%+.

B. The short-runway single, 52. Only needs penalty-free income for 7½ years. Converts three fat rungs (52, 53, 54) sized to the top of the 12% bracket, spends taxable until the first unlocks at 57, and lets 59½ moot the rest. Fewer, larger rungs beat a full ladder here — and with no ACA-subsidized years wasted, she times COBRA + a Bronze HSA plan around the fat-rung years.

C. The withholding mistake. Age 48, converts $80k and lets the custodian withhold 20% ($16k). Only $64k reaches the Roth; the $16k is an early distribution → $1,600 penalty, plus a permanently smaller rung. Correct play: convert the full $80k, pay the tax from cash via a January estimated payment. (A 60-day rollover of $16k from other cash could have cured it — if caught in time.)

---

💡 Pro-Tips

1. Convert in December, spend in January thinking: late-year conversions season ~11 months faster and are sized with perfect income information.
2. The first rung is the best rung: your earliest retirement years are usually your lowest-income years ever — don’t leave the 0% and 10% brackets empty while “deciding.”
3. A governmental 457(b) can replace years of bridge fuel — check for one before designing anything.
4. Roth contribution basis is your shock absorber: it’s accessible anytime, so a thin bridge year borrows from it rather than busting an unseasoned rung.
5. Log every conversion like a commit: year, amount, taxable portion, unlock date, Form 8606 filed. Future-you audits past-you.
6. Coordinate with gain harvesting annually — one strategy owns each year’s bracket space; they don’t share well.
7. Converting depressed assets in-kind is the only “discount window” the IRS still offers.
8. Moving states? Sequence the ladder around the move — the same rung can cost 0% or 10%+ in state tax depending on which side of the moving truck it lands.

⚠️ Common Pitfalls

1. No bridge fuel — a ladder without five years of side funding is a diagram, not a plan.
2. Withholding from the conversion under 59½ (penalty + shrunken rung).
3. Confusing the two 5-year rules — or assuming one old Roth account seasons new conversions. Each rung has its own clock.
4. Spending an unseasoned rung — ordering rules pull oldest conversions first, but only seasoned ones escape the penalty before 59½.
5. Blowing the ACA cliff with a habitual December conversion — post-2025, the clawback is uncapped (HC-01).
6. Forgetting the pro-rata rule when non-deductible IRA basis exists — the taxable amount surprises people in April.
7. Converting into a child’s FAFSA base year.
8. Missing Dec 31 — a January 3rd “2026 conversion” is a 2027 conversion, and the whole ladder shifts a year.
9. Converting so much that RMD-era planning was the real problem — if you’re filling the 24% bracket “for the ladder,” you’ve changed strategies without noticing.
10. Treating rate permanence as rate certainty — OBBBA removed the scheduled sunset, not Congress’s ability to legislate. Convert on your bracket math, not predictions.

✅ Actionable Checklist

Design phase (1–2 years before retiring)

• Inventory: pre-tax balances, taxable basis, Roth contribution basis, any 457(b)
• Confirm ≥5 years of bridge fuel; identify the gap-filler if short
• Map ACA lane strategy by year (HC-01 lines for your household size)
• Map kids’ FAFSA base years; block out conversion-heavy years around them
• Plan the 401(k) → IRA rollover timing (vs. Rule-of-55 access you’d forfeit)

Each conversion year

• October: project year-end income incl. fund distributions
• Pick the year’s ceiling: deduction line / 12% top / FPL line / cliff-minus-buffer
• December: convert (in-kind if markets are down), zero withholding
• January: Q4 estimated payment or confirm safe harbor
• Log the rung (amount, taxable portion, unlock date); file Form 8606

Each withdrawal year (year 6+)

• Verify the rung being tapped is seasoned (Jan-1 back-dating)
• Withdraw ≤ cumulative seasoned conversions; earnings stay put until 59½
• Re-size this year’s new rung to inflation-adjusted future spending

---

State notes (→ ST-01, ST-03): Conversions are ordinary income in most taxed states (0–13%+). Washington taxes none of it — WA residents run ladders at federal-only rates, one of the strongest “retirement-friendly” features no ranking captures. Sequencing with a move matters enormously: convert after leaving a high-tax state (mind sticky-domicile audits — CA especially) and before moving into one. WA’s flip side from FN-02 still applies: the estate-tax exposure a big Roth eventually creates has no income-tax offset ramp.

Sources & further reading (verified Aug 2026)

• IRC §408A; IRS Pub. 590-B (ordering rules, 5-year clocks); Form 8606 instructions
• Rev. Proc. 2025-32 via FN-02 (2026 brackets/deductions)
• OBBBA (P.L. 119-21) — rate permanence; senior-deduction phase-out mechanics
• HC-01 §5 for the ACA-side marginal-cost math (Rev. Proc. 2025-25 percentages)
• Kiplinger, Are Roth Conversions for Retirees Dead in 2026? (post-OBBBA framing); Northern Trust, Roth IRA Conversions after the OBBBA