You're Asking the Wrong Question

When should I take Social Security?

The Social Security conversation almost always starts the same way.

"What's the breakeven? How long do I have to live for waiting to pay off?"

It's a reasonable question and it's the wrong frame, because it treats the decision as a bet on how long you'll live. Guess right, you win. Guess wrong, you lose. And since nobody knows, people default to taking it early — at least that way you're guaranteed to get something.

Here's the better way to think about it.

Social Security isn't a bet on living a long time. It's insurance against living a long time.

Living to 95 sounds wonderful right up until you look at what it costs. Thirty years of expenses. Thirty years of inflation. Thirty years of a portfolio that has to keep producing. That's the scenario that breaks retirement plans — not dying early.

And Social Security is the only asset you own that gets better precisely in that scenario. It's inflation-adjusted, it never runs out, and it doesn't care what the market did.

You don't buy fire insurance hoping the house burns down. You buy it because the version where it does burn is the one you can't absorb.

What the numbers actually are

For anyone born in 1960 or later, full retirement age is 67.

Claim at 62 and you receive 70% of your full benefit — permanently. Wait until 70 and you receive 124% — also permanently.

That spread is larger than people register. The benefit at 70 is about 77% higher than the same person's benefit at 62. Not 77% of one year — 77% more, every month, for the rest of your life, adjusted for inflation each year.

Between full retirement age and 70, you're earning roughly 8% a year, guaranteed, for waiting. There is no other place in your financial life offering a guaranteed 8% real return.

The part almost nobody explains

If you're married, here's the thing that matters most, and it has nothing to do with your own life expectancy.

When one spouse dies, the household keeps the larger of the two benefits. The smaller one stops.

Read that again, because the consequences are enormous.

It means the higher earner's claiming decision isn't really about the higher earner. It sets the income for whichever spouse lives longer — potentially for ten, fifteen, twenty years of widowhood.

So the relevant question isn't "how long will I live." It's "how long until both of us are gone." And that's a much longer horizon. For a couple in their mid-sixties, the odds that at least one of them reaches 90 are far higher than the odds for either individually.

Delaying the higher benefit is buying insurance for the survivor. That's what it's for.

And the survivor is usually the one who needs it. Their expenses don't halve when their spouse dies — housing, insurance, property taxes, and utilities barely move. Meanwhile they lose a Social Security check and they file as single, with narrower brackets and half the standard deduction. June is entirely about that cliff.

For couples, it's two decisions

This is where most households go wrong. They treat claiming as one decision made together. It's two.

The lower earner's benefit is the one that disappears at the first death. Delaying it buys a larger check that only lasts while both of you are alive. There's often a reasonable case for claiming it earlier — it provides income now, it reduces what you have to pull from the portfolio, and its long-term value is limited anyway.

The higher earner's benefit is the one that survives. It's the household's floor for as long as either of you is alive. That's the one to protect.

So the pattern that works for a lot of couples is: lower earner claims earlier, higher earner delays as long as possible. You get income flowing sooner while still protecting the survivor.

That's not a universal rule. But if you and your spouse are planning to claim at the same age simply because it feels tidy, that's worth a second look.

Two things people get wrong

Working while collecting doesn't permanently cut your benefit. If you claim before full retirement age and keep working, benefits above an annual earnings limit get withheld — a dollar for every two dollars over. In 2026 that limit is $24,480, or $65,160 in the year you actually reach full retirement age, where the withholding is gentler.

But here's what almost nobody knows: that money isn't gone. At full retirement age, Social Security recalculates your benefit to credit back the months that were withheld. Your monthly check goes up permanently. You get it back over time, not in a lump sum.

The earnings test is a cash-flow issue, not a penalty. Plenty of people turn down work for no reason at all.

Your benefits are probably taxable, and increasingly so. Depending on your other income, up to 85% of your Social Security can be subject to federal income tax. The thresholds that determine this were set in 1984 and have never been adjusted for inflation — not once. Which means every year, a slightly larger share of retirees crosses them.

This matters for planning because withdrawing from a tax-deferred account can push more of your Social Security into the taxable column along with it. Your true marginal rate in that zone is higher than the bracket table suggests. That's the interaction July and August work on.

Two benefits people miss entirely

If you were married at least ten years and are now divorced and unmarried, you can claim on your ex-spouse's record. It doesn't reduce their benefit, they're not notified, and it doesn't matter whether they've remarried. This is badly underused, particularly by women whose earnings record reflects years spent raising children.

If you're widowed, you have two benefits available — your own and your late spouse's — and you don't have to take them in that order. You can claim one now and switch to the other later if it will be larger. That flexibility survived the rule changes that eliminated most other claiming strategies, and it's still one of the most valuable options in the system.

"Will it even be there?"

You've seen the headlines, so let's deal with them honestly rather than pretending they don't exist.

The 2026 Trustees Report projects that the retirement trust fund's reserves run out in late 2032. At that point, incoming payroll taxes would still cover roughly 78% of scheduled benefits. If Congress allowed the retirement and disability funds to be combined, the date moves to 2034 and the payable share is about 83%.

So the honest picture is a program operating at reduced capacity if nothing changes — not one that disappears. There's no scenario in these projections where benefits stop.

Now the planning question. Does that risk argue for claiming early?

Usually not, and the reason is that claiming early is itself a permanent 30% reduction. You'd be locking in a certain cut today to avoid a possible, smaller, later one that may well be resolved before it arrives — and if reform does come, historically it has phased in over time and protected people already receiving benefits.

The rational response isn't to panic-claim. It's to know your number, understand that a future adjustment is possible, and stress-test whether your plan survives one. That's a far more useful exercise than reacting to a headline.

What April is for

Get your actual numbers from your Social Security statement. Not an estimate — the real figure.

Then look at it as two decisions if you're married, with the higher earner's benefit treated as what it actually is: the survivor's floor.

And stop asking about the breakeven. Ask instead what happens to whichever of you lives longest, and whether the answer is one you can live with.