[SS-01] Claiming Age Math: 62 vs. FRA vs. 70

Break-even is the wrong question — Social Security is longevity insurance, and for couples it is a joint-life asset

Pillar: Social Security & Medicare Enrollment · Applies to: Everyone approaching 62, and every early retiree modelling a 30-year drawdown Last verified: August 2026 · Refresh cadence: Annual (figures via FN-02) + event-driven (trust-fund legislation) Related: SS-02 Spousal & Survivor Benefits · SS-03 Taxation of Benefits · SS-04 Earnings Test · SS-05 Coordinating SS + Medicare · HC-01 ACA Bridge · HC-06 IRMAA · TX-03 Withdrawal-Order Sequencing · FN-02 Key Numbers

Not advice. Educational reference. Your benefit depends on your own 35-year earnings record, your marital history, and your health — none of which a general article knows. Pull your actual figures from your my Social Security account and model your own numbers before filing. A claiming decision is close to irreversible after twelve months.

---

TL;DR

• Claiming age adjusts your benefit on a fixed statutory schedule. At a full retirement age (FRA) of 67, age 62 pays 70% of your PIA and age 70 pays 124% — the latest check is 1.77× the earliest, for life, indexed for inflation. ✅
• Break-even is ~80½ for 62-vs-70, ~82½ for FRA-vs-70. The median 62-year-old man reaches 82.3, the median woman 85.1 — so delay wins the average case, and wins by more for women. ✅
• But break-even is the wrong frame. Social Security is the only inflation-indexed, government-backed, unlimited-duration income most households will ever own. You don’t buy insurance because you expect to win on it; you buy it because outliving your money at 92 is the scenario you can’t recover from.
• For married couples the higher earner’s claiming age isn’t a personal decision. At the first death the household keeps the larger benefit and loses the smaller. Claiming at 62 sets a 70%-of-PIA floor under a widow(er) who may live another 25 years; claiming at 70 sets it at 124%.
• Delay pays a dividend the tables don’t show: every year Social Security is off your return is a year of cheap bracket space for Roth conversions (TX-03 §4) and — before 65 — a year of ACA subsidy kept, since all of the benefit counts in ACA MAGI (HC-01).

---

1. What the Claiming Age Actually Changes

Three numbers, in order:

1. AIME — your highest 35 years of earnings, each indexed to national wage growth, averaged monthly. Fewer than 35 earnings years means zeros in the average, which is why FIRE careers dent the benefit (SS-04).
2. PIA (Primary Insurance Amount) — AIME run through a progressive formula. For workers turning 62 in 2026: 90% of the first $1,286 of AIME + 32% from $1,286 to $7,749 + 15% above. ✅ Those bend points fix to your age-62 cohort for life; only the COLA moves your benefit afterward.
3. The claiming-age adjustment — this article’s subject. It scales PIA up or down on a statutory schedule.

The adjustment is not discretionary, not re-priced each year, and not affected by markets. It is arithmetic:

• Before FRA: reduced by 5/9 of 1% per month for the first 36 months early, then 5/12 of 1% per month beyond that. ✅
• After FRA: increased by 2/3 of 1% per month — 8% per year — for anyone born 1943 or later, stopping dead at age 70. ✅

Your FRA depends only on your birth year: ✅

Birth year	FRA	Benefit at 62
1943–1954	66	75.0%
1955	66 + 2 mo	74.2%
1956	66 + 4 mo	73.3%
1957	66 + 6 mo	72.5%
1958	66 + 8 mo	71.7%
1959	66 + 10 mo	70.8%
1960 and later	67	70.0%

Everyone reading this as a pre-retiree in 2026 is in the FRA-67 cohort or within months of it, so the rest of this article uses FRA 67.

2. The Adjustment Table (FRA 67)

Applied to a $2,000 PIA — near the 2026 average retired-worker benefit of $2,071/mo. ✅

Claim at	% of PIA	Monthly	Annual	vs. claiming at 62
62	70.0%	$1,400	$16,800	—
63	75.0%	$1,500	$18,000	+$1,200/yr
64	80.0%	$1,600	$19,200	+$2,400/yr
65	86.7%	$1,733	$20,800	+$4,000/yr
66	93.3%	$1,867	$22,400	+$5,600/yr
67 (FRA)	100.0%	$2,000	$24,000	+$7,200/yr
68	108.0%	$2,160	$25,920	+$9,120/yr
69	116.0%	$2,320	$27,840	+$11,040/yr
70	124.0%	$2,480	$29,760	+$12,960/yr

Two things to read off it. The steps are not uniform — the 62→63 step is worth 5 points of PIA, the 66→67 step 6.7, and each post-FRA year a flat 8. And the endpoints are far apart: $1,400 vs. $2,480 is a 77% difference in permanent, COLA’d income.

The maximums, for calibration (2026): a worker with 35 years at the taxable maximum receives $2,969/mo at 62, $4,152 at FRA, and $5,181 at 70. ✅ → FN-02

3. Break-Even, Done Honestly

The break-even is where cumulative benefits from the later claim overtake cumulative benefits from the earlier one. On the $2,000 PIA above, ignoring taxes and discounting:

Comparison	Head start the early claim banks	Monthly advantage after the later claim starts	Break-even
62 vs. 67	60 × $1,400 = $84,000	$600	140 months → age 78 yr 8 mo
67 vs. 70	36 × $2,000 = $72,000	$480	150 months → age 82 yr 6 mo
62 vs. 70	96 × $1,400 = $134,400	$1,080	124 months → age 80 yr 5 mo

Now put that against actual mortality. From SSA’s period life table (2023 table, as used in the 2026 Trustees Report): ✅

At exact age 62	Remaining years	Implied age
Male	20.29	82.3
Female	23.08	85.1

So the median man clears the 62-vs-70 break-even by about two years, the median woman by about five. Delay is the higher-expected-value choice for the average person before any of the arguments in §4–§6 — and these are medians for the whole 62-year-old population, including those in poor health. Conditional on being healthy enough to be reading retirement-planning material at 62, your odds beat the table’s.

What the calculation quietly omits. Inflation is neutral rather than helpful to the early claim: COLAs scale both paths, and your PIA is COLA-adjusted from age 62 onward whether or not you have claimed, so delaying forfeits no inflation protection. ✅ A real discount rate pushes break-even later; taxes push it earlier for high-income households and later for low-income ones (SS-03). Neither changes the shape of the answer. And it prices only your life — for a married couple it is the wrong calculation entirely (§4).

The framing that actually works. Delaying from 67 to 70 costs $72,000 of forgone benefits and buys $5,760/yr of extra lifetime income — a payout rate of 8.0% on the money given up, inflation-indexed, federally backed, with no sequence risk and no counterparty. No retail inflation-adjusted annuity comes close to that price for a 70-year-old ◻️. Delay isn’t an investment bet you might lose; it’s the cheapest longevity insurance a U.S. household can buy, paid for with principal instead of premiums.

4. The Married Case: It Is a Joint-Life Asset

This is the section that changes decisions.

Survivor mechanics. When one spouse dies the survivor keeps the larger of the two benefits; the smaller stops. The survivor benefit is based on the deceased’s actual benefit including any delayed retirement credits, so the higher earner’s claiming age sets a floor under the survivor’s income for life. ✅ The consequence: the higher earner’s benefit is paid until the second death, and the right horizon for that decision is joint-and-last-survivor life expectancy. A 65-year-old couple has roughly a 50% chance at least one of them reaches 90. ◻️

Worked example: the same couple, two claiming plans

Both spouses turn 62 in 2026. Higher earner PIA $2,800; lower earner PIA $1,600 (above half the higher PIA, so no spousal top-up applies — SS-02 covers the case where it does).

	Plan A — both claim at 62	Plan B — lower claims at 62, higher waits to 70
Higher earner’s benefit	$1,960 (70%)	$3,472 (124%)
Lower earner’s benefit	$1,120 (70%)	$1,120 (70%)
Household, ages 62–69	$3,080/mo	$1,120/mo
Household, age 70+	$3,080/mo	$4,592/mo

Plan B gives up 96 × $1,960 = $188,160 of the higher earner’s benefit during the wait, then runs $1,512/mo ahead. Household break-even: age 80 years 5 months — the same answer as the single case, which is why break-even analysis makes delay look like a coin flip for couples too.

Now the line the break-even never shows. Say the higher earner dies at 84:

	Plan A	Plan B
Survivor’s benefit from age 84	$1,960/mo	$3,472/mo
Difference	—	+$1,512/mo = +$18,144/yr
If the survivor lives to 92	—	≈ $145,000 more, before COLA compounding

Same household, same earnings record. The difference is a claiming decision made 22 years earlier, and it lands on the spouse with less income at the age when a portfolio is least able to absorb a shortfall. It is the same compression event behind the widow(er)’s tax trap in TX-03 §2 and the widow(er) IRMAA cliff in HC-06 §4: at the first death the household loses a benefit, loses a standard deduction, and moves to single brackets at once.

The rule that falls out of it: the higher earner delays as long as the plan can fund; the lower earner claims early if cash is needed. The lower earner’s benefit is the one likely to disappear at the first death, so its claiming age matters far less.

5. The Arguments for Claiming Early, Graded

1. “I might not live that long.” — Valid, conditionally. If you are single, have no dependents, and have a genuinely impaired life expectancy (a diagnosis, not a hunch about family history), claiming at 62 is correct and the math supports it. If you are married and the higher earner, this argument runs backwards: your early death is precisely the event that makes your delay valuable, because it starts the survivor benefit sooner (§4).

2. “The trust fund is running out.” — Weak. A proportional cut hits every claiming age at once, so filing early hedges nothing (§7).

3. “I’ll take it at 62 and invest it.” — Weak on a risk-adjusted basis. The comparison isn’t “8% vs. stock returns.” It is 8% guaranteed, real, lifetime, survivor-linked and sequence-risk-free against a nominal, risky, finite-duration return you must not lose in the first decade.

4. “I need the money.” — Unconditionally valid. If the alternative is unsustainable portfolio withdrawals, high-interest debt, or forced selling into a bear market, claim. Bridging to 70 by drawing an extra ~$150k–$190k from a portfolio is a good trade only if the portfolio can carry it. The bridge is the constraint, not the arithmetic.

5. “I don’t want to spend down my portfolio to wait.” — Backwards. Spending the portfolio is how you buy the delay: you are converting a volatile, finite, taxable balance into guaranteed indexed income you cannot outlive. The portfolio is the payment mechanism, not a competing asset.

6. What Delay Buys Besides Money

For this wiki’s audience the tax-side dividend is often worth as much as the benefit increase:

• The golden window. TX-03 §4 Phase 2 identifies ages 65–72 as the cheapest ordinary income of your life: ACA constraints gone, RMDs not yet started. Social Security starting is what closes that window. Delaying to 70 keeps five to eight years of low-bracket airspace open for Roth conversions, which permanently shrinks the RMD problem at 73/75.
• Before 65, Social Security is brutal ACA MAGI. All of your benefit counts — including the portion that isn’t taxable income (HC-01 §2). A 62-year-old claiming $16,800/yr on a subsidized exchange plan adds $16,800 of MAGI with no offsetting deduction, costing cost-sharing tiers or, near the 400% FPL line, the entire credit.
• After 65, it feeds IRMAA. The taxable portion lands in AGI and therefore in IRMAA MAGI (HC-06 §1), two years forward. And provisional-income thresholds are not indexed — unchanged since 1984/1994 — so a rising share of benefits becomes taxable over time; conversions done before benefits start don’t stack against them (SS-03).
• The earnings test applies only before FRA. Working while claiming early withholds $1 of benefit per $2 over $24,480 (2026), or $1 per $3 over $65,160 in the year you reach FRA. ✅ Withheld benefits are not lost — the benefit is recomputed upward at FRA — but the cash-flow hit surprises people (SS-04).

7. The Trust-Fund Question (as of August 2026)

State it precisely, because the vague version drives bad decisions. Per the 2026 Trustees Report: ✅

Fund	Projected depletion	Benefits payable from ongoing revenue at that point
OASI (retirement & survivors) alone	Q4 2032	78%
OASI + DI combined (requires an act of Congress to combine)	2034	83%

Three things follow. Depletion is not insolvency — payroll tax revenue continues and funds ~78–83% of scheduled benefits indefinitely. A proportional cut is claiming-age-neutral — scale every benefit by 0.78 and the ratio between a 70% benefit and a 124% benefit is unchanged, so claiming early buys no protection. And the historical pattern is late legislative action — the 1983 amendments landed months before depletion. Congress-dependent, so: stated as of August 2026, track in FN-03.

The planning posture: baseline on scheduled benefits, run a sensitivity case at ~78% for ages past the early 2030s, and don’t let a projection with a moving date drive an irreversible decision.

8. Mechanics, Deadlines, and the Two Do-Overs

• Apply up to four months ahead. Benefits are paid the month after the month they cover. DRCs earned in a calendar year are generally credited the following January — so file to start in the month you turn 70, not later. ✅
• Do-over #1 — withdrawal (Form SSA-521). Within 12 months of first entitlement you may withdraw the application entirely, once per lifetime, by repaying every dollar received on your record — including family benefits and Medicare premiums withheld. The only full reset. ✅
• Do-over #2 — voluntary suspension at FRA. From FRA to 70 you may suspend and earn DRCs on the suspended months. Since April 30, 2016, suspending also suspends benefits others draw on your record (except a divorced spouse). ✅
• Deemed filing. Anyone born on or after January 2, 1954 filing for a retirement or spousal benefit is deemed to have filed for both — “restricted application” is unavailable to this cohort. ✅ Survivor benefits are exempt: a widow(er) may take a survivor benefit first and switch to their own at 70, or the reverse. The last genuinely free option in the system (SS-02).
• Medicare does not follow automatically. Not drawing Social Security at 65? You must enroll yourself (HC-02 §6, SS-05).
• WEP and GPO are gone. The Social Security Fairness Act (January 5, 2025) repealed both, retroactive to benefits payable from January 2024. ✅ Any public-sector retiree whose decision was modelled under WEP/GPO should redo it — the delay case is stronger than the old spreadsheet said.

9. Three Scenarios

A. The couple with a big earnings gap (both 62, PIAs $2,800 and $1,600, $1.4M portfolio). The lower earner claims at 62 ($1,120/mo) for cash flow; the higher earner delays to 70 ($3,472/mo). The ~$188k bridge comes from traditional IRA balances — income they wanted to realize anyway, filling the 12% and 22% brackets during the golden window (TX-03) and shrinking future RMDs. Two wins from one decision: the survivor floor rises from $1,960 to $3,472, and the drawdown that funds the wait doubles as the conversion program. Break-even at 80½ is almost beside the point.

B. The single man who should claim at 62 (no dependents, CHF diagnosis, $400k portfolio). No survivor to protect, and a documented condition — not a hunch about family history — puts his expectancy well below the 82.3 median. Claiming at 62 gives him $1,400/mo now and preserves the portfolio for the years he will actually live. The general rule bends to a specific fact. He checks one thing first: whether an ex-spouse from a 12-year marriage yields a claim worth more than his own (SS-02).

C. The FIRE couple on the exchange (both 62, retired at 51, spending $76k, expansion state; PIAs $2,000 and $1,600). They currently realize $52,000 of MAGI from conversions and dividends — 246% of FPL for a household of two, holding a 73% AV Silver plan. Claiming at 62 would add $16,800 + $13,440 = $30,240 of ACA MAGI, taking them to $82,240 — 389% of FPL: through the 250% cost-sharing line, up the applicable-percentage ramp to the full 9.96%, and to within $2,360 of the 400% cliff where the entire credit vanishes and the clawback is uncapped (HC-01). The subsidy and CSR losses alone rival the benefits claimed — before counting the permanent 30% haircut, the compressed conversion window, and the survivor floor from §4. They wait, and it is the ACA math rather than the longevity math that decides it. At 65 they re-run the same decision against IRMAA instead (HC-06).

---

💡 Pro-Tips

1. Pull your actual numbers first. Your my Social Security statement shows your PIA and your benefit at every claiming age from 62 to 70. Every generic table, including the one in §2, is a scaled approximation of that.
2. Decide the higher earner’s age and the lower earner’s age separately. They are two different questions: the higher earner’s is a joint-life insurance decision, the lower earner’s is a cash-flow decision.
3. Check for 35 full years of earnings. A FIRE career often carries zeros; one extra year of even modest self-employment income can replace a zero and lift the PIA permanently (SS-04).
4. Fund the bridge with traditional-IRA withdrawals, not taxable sales. The money you spend waiting for age 70 is the cheapest ordinary income you will ever realize — it does conversion work while it buys the delay (TX-03).
5. Before 65, price the ACA cost of claiming before the longevity benefit. All of the benefit is MAGI; the subsidy loss frequently dwarfs the check (HC-01).
6. Calendar the 12-month SSA-521 window the day you file. It is the only complete undo, and it expires quietly.
7. Divorced after a 10-plus-year marriage? Check that record before deciding anything — an ex-spousal benefit can be worth more than your own and does not require the ex’s cooperation (SS-02).
8. If you were a teacher, firefighter, or other public-sector worker, redo the math. WEP/GPO repeal in 2025 changed the answer for millions of people who last modelled this under the old rules.

⚠️ Common Pitfalls

1. Treating break-even as the decision. It answers “which is more likely to pay more,” when the question is “which failure can I not recover from.”
2. Assuming both spouses’ benefits continue after the first death. The household keeps the larger and loses the smaller, every time — so a higher earner who claims at 62 for personal-longevity reasons sets a permanent 70% floor under a survivor who may live 25 more years.
3. Believing delay forfeits COLAs. Your PIA is COLA-adjusted from 62 onward whether or not you have claimed. Delaying costs you nothing in inflation protection.
4. Claiming early to “beat the trust fund.” A proportional cut hits every claiming age equally; you would be locking in a reduced benefit and then having it reduced.
5. Claiming at 62–64 while on a subsidized ACA plan without pricing the MAGI impact — routinely the most expensive month in an early retiree’s year.
6. Claiming before FRA while still working and being blindsided by the earnings test ($24,480 in 2026), forgetting the withheld amounts return as a recomputation at FRA.
7. Filing after 70. Delayed credits stop at 70; every later month is a gift to the Treasury.
8. Letting Social Security close the conversion window early — starting benefits at 65 instead of 70 costs five years of the cheapest bracket space in a retirement (TX-03 §4).
9. Forgetting Medicare doesn’t auto-enroll when you’re not drawing Social Security at 65 — a lifetime Part B penalty attaches to that oversight (HC-02).
10. Modelling a public-sector benefit under WEP/GPO after their 2025 repeal.

✅ Actionable Checklist

Ages 55–61 (design)

• Record PIA and the benefit at 62/FRA/70 for both spouses from my Social Security
• Count covered-earnings years — identify zeros in the highest 35 (SS-04)
• Check ex-spousal (marriage ≥10 years) and survivor eligibility (SS-02)
• Public pension? Confirm the post-2025, no-WEP/GPO figure, not an old estimate
• Model the bridge: can the portfolio fund the higher earner to 70, and what does that drawdown do for conversions (TX-03)?

Age 62 (the first decision point)

• Set the higher earner’s age on joint-life logic, the lower earner’s on cash flow
• On an ACA plan? Price claiming against the FPL lines and CSR tiers first (HC-01)
• Still working? Run the earnings test against expected wages (SS-04)

Ages 63–70 (execution)

• Re-check annually against health, markets, and portfolio sustainability — revisable until you file
• From 63: coordinate bridge-funding conversions with the IRMAA lookback (HC-06 §6)
• At 65: enroll in Medicare regardless of Social Security status (HC-02, SS-05)
• Apply up to 4 months ahead; file to start in the month you turn 70 if delaying fully
• After filing: calendar the 12-month SSA-521 deadline

---

State notes (→ ST-01, ST-04): Social Security benefits are exempt from state income tax in the large majority of states, and the list of states that tax any part of them has shrunk almost every legislative session — several have repealed or phased out taxation since 2022, so any list older than a year is unreliable ◻️ (current list in ST-04). Where a state does tax benefits it is generally partial and income-gated. Two second-order effects matter more than the headline. First, in the no-income-tax states (WA, FL, TX, NV, TN, SD, WY, AK, NH), the traditional-IRA withdrawals used to fund a delay to 70 are state-tax-free, which makes the bridge materially cheaper than the same plan run in a high-tax state — the delay decision and the TX-03 fill decision share a state rate. Second, if a relocation is planned, sequence it against the bridge years: the drawdown that funds the wait is ordinary income and belongs on the low-tax side of the move (ST-03 on making domicile stick).

Sources & further reading (verified Aug 2026)

• SSA — Retirement Benefits by Year of Birth (FRA table, age-62 reduction percentages); Early or Late Retirement (5/9 and 5/12 of 1% rules); Delayed Retirement Credits (8%/yr, ceasing at 70)
• SSA — Benefit Formula Bend Points (2026: $1,286 / $7,749); Maximum retirement benefit payable ($2,969 at 62 · $4,152 at FRA · $5,181 at 70); 2026 COLA Fact Sheet (2.8%; earnings-test $24,480 / $65,160; average retired-worker benefit $2,071)
• SSA — Actuarial Period Life Table (2023 table as used in the 2026 Trustees Report)
• 2026 OASDI Trustees Report — OASI depletion Q4 2032 at 78% payable; combined OASDI 2034 at 83%
• Social Security Fairness Act (signed January 5, 2025) — WEP/GPO repeal, retroactive to January 2024; Bipartisan Budget Act of 2015 (deemed filing for those born on/after January 2, 1954; suspension rules from April 30, 2016)
• SSA Form SSA-521 (Request for Withdrawal of Application); FN-02 for annual figures; FN-03 for trust-fund legislation