Claiming Social Security at 62, 67, or 70

This is the largest single financial decision most retirees make, it is irreversible in practice, and it is usually made on the wrong basis — either “take it before it runs out” or a break-even calculation that quietly asks you to predict when you will die.

The mechanics are simple and worth knowing exactly. What to do with them is not simple, and depends on things a table cannot see.

What the ages actually pay

For anyone born in 1960 or later, full retirement age is 67. That is the age at which you receive 100% of what your earnings record entitles you to. Claim earlier and the amount is permanently reduced; claim later and it is permanently increased, by 8% for each year you wait, up to 70.

The gap between the two ends is larger than it first looks: $2,480 against $1,400 is 77% more, every month, for the rest of your life, and cost-of-living adjustments are applied to the larger figure too. Waiting eight years does not buy you 8% more. It buys you roughly three-quarters again.

The break-even calculation, and why it is the wrong question

The standard comparison: claim at 62 and you collect smaller cheques for eight extra years; claim at 70 and you collect nothing during them, then more forever. Total benefits received cross over at about age 80 years and 4 months. Comparing 67 with 70, the crossover is about 82 years and 6 months. (Ignoring cost-of-living adjustments, any return you might earn on early benefits, and taxes.)

Which invites the obvious reasoning: if you expect to live past 80, wait. It is not wrong, exactly. It is answering a question you cannot answer — how long will I live — and ignoring which kind of error costs you more.

Consider both mistakes honestly. Delay, then die at 75, and you leave money uncollected — but you are dead, and the risk you were insuring against did not materialise. Claim early, then live to 95, and you spend your last two decades on a permanently reduced income, at exactly the point when your portfolio has been drawn down and your care costs are rising.

Those are not symmetrical. One error is an accounting regret; the other is running short of money at 90. Seen that way, delaying is not a bet that you will live long — it is insurance against the case where living long is expensive. It is the only inflation-adjusted lifetime income most people can buy, and the price is the eight years of waiting.

The spousal argument, which is often the deciding one

For a couple, the higher earner’s decision covers two lifetimes. When one spouse dies, the survivor keeps the larger of the two benefits — including any delayed credits the higher earner accrued. Delaying therefore raises the floor for as long as either of them lives.

This flips the usual reasoning for the higher earner in a couple. Their own health is no longer the only relevant question; the survivor’s longevity matters just as much, and the strategy of the higher earner delaying while the lower earner claims earlier is common for exactly this reason.

When claiming early is the right answer

Delaying is not universally correct, and the cases against it are real:

• Health that makes long life unlikely. The insurance argument depends on the risk existing.
• No other assets to live on. Delaying means funding those years from somewhere. If the portfolio cannot carry it, the question is settled.
• An involuntary early retirement. Plenty of people stop working before they planned to, and the choice becomes income now versus income never.
• Still working before full retirement age. An earnings test withholds benefits above an annual limit if you claim early and keep earning. It is not permanently lost — your benefit is recomputed at full retirement age — but claiming early while working full-time is usually pointless.

The bridge problem, which our pages show

Delaying has a cost that a break-even table hides: the portfolio has to cover everything in the meantime, at full rate, with no other income arriving. Those are the most dangerous years in any retirement, for the reasons in sequence-of-returns risk.

This is not an argument against delaying — it is the trade being made visible. Every one of our can-I-retire pages below age 67 has a section on exactly this gap, and the reason those pages score so much worse at younger retirement ages is the bridge, not the extra years of old age.

One disclosure about our own model: it assumes Social Security begins at 67 in every simulation, at $2,000 a month, regardless of the retirement age being modelled. That is a deliberate simplification — 67 is full retirement age, a fact about the system rather than a default we chose — but it does mean our pages do not show you the delay-to-70 strategy. The methodology page lists that alongside the other assumptions.

What to do

Get your actual numbers rather than reasoning from a $2,000 example: your Social Security statement gives your own figures at 62, at full retirement age, and at 70. If you are married, get both.

Then ask the question in the insurance form rather than the investment form. Not “which option wins if I live to X?” but “which mistake could I not recover from?” For most people with the assets to bridge the gap — and especially for the higher earner in a couple — that framing points one way. For people without those assets, it points the other, and that is a legitimate answer rather than a failure.