Social Security Trust Fund Depletion: When It Runs Out and What Happens

The retirement fund is projected to run short in late 2032 — about six years from now. At that point the Trustees project incoming payroll tax would cover 78% of scheduled benefits.

That is not the number in most headlines. Most say 2034 and 83%, and those figures describe a fund that does not legally exist.

Here is what the 2026 Trustees Report actually says, fund by fund.

When does Social Security run out of money?

It depends which fund you mean, and that is the whole problem with the question. There are four, they are legally separate, and they are in very different shape.

If you are asking about your retirement cheque, the answer is OASI, and the answer is the fourth quarter of 2032 — roughly six years away, and one quarter earlier than the 2025 report projected.

If you are asking about disability benefits, there is no depletion date in the projection at all. DI can pay full scheduled benefits through at least 2100, which is the end of the report’s horizon. Last year’s report said 2099, so it improved.

Does “depletion” mean Social Security stops paying?

No, and this is the most consequential misunderstanding in the whole subject. “Depleted” means the reserve is exhausted — not that income stops.

Social Security is funded mainly by payroll tax on current workers. That tax keeps arriving after the reserve is gone. What the reserve does is cover the gap between what comes in and what is scheduled to go out; once it is empty, the programme can only pay out what comes in.

So the question at depletion is not whether cheques arrive. It is how much smaller they are:

• OASI at depletion: 78% of scheduled benefits. A benefit scheduled at $2,000 would be about $1,560 on those projections, absent legislation.
• HI at depletion: 89% of scheduled Part A benefits.

There is also no automatic mechanism for how a shortfall would be applied. The law does not specify whether it would fall as an across-the-board percentage cut, a delay, or something else — that would be a decision, and it has not been made.

Why does everyone say 2034 when the report says 2032?

Because 2034 is the number for a fund that would require an act of Congress to exist.

The Trustees do publish a combined OASDI projection, and it is genuinely useful as a summary of the programme’s overall shape. But the report’s own sentence introducing it contains the caveat, and the caveat is usually what gets cut:

The two funds could not actually be combined unless there were a change in the law.

No such law has been passed. OASI and DI are separate by statute, each pays only what it is permitted to pay, and money cannot move between them. Congress has authorised temporary reallocations before — most recently in 2015 — but each one required legislation, and none is in force now.

The practical consequence is a two-year difference in the date and a five-point difference in the cut:

Neither figure is wrong. The combined one just answers a question about a fund nobody has created.

How much would benefits be cut, and for how long?

78% is where it starts, not where it settles. The Trustees project the OASI payable share continuing to fall after depletion, reaching 62% by 2100.

Medicare Part A goes the other way in the long run — down to 85% by 2050, then back up to 93% by 2100.

The reason the two diverge is demographic rather than financial. The report attributes the rise in cost rates for both programmes mainly to the decline in fertility rates and the resulting ageing of the population, an effect its assumptions have stabilising after 2080. Health costs and Social Security costs respond to that differently over the very long run.

The underlying gap for Social Security is a persistent one, not a spike:

Costs rise by about five points of payroll; income rises by about half a point. That divergence is the entire story, and it is why the shortfall does not close on its own.

Is Medicare in the same position?

Part A is, on a similar timetable. Parts B and D are not, and the reason is structural rather than a matter of better health.

HI (Part A) is funded much like Social Security, from a dedicated payroll tax, and it faces the same kind of squeeze: depletion in Q2 2033, one quarter earlier than last year, and 89% payable at that point.

SMI (Parts B and D) is described by the Trustees as adequately financed into the indefinite future — but not because it is cheap. It is because its financing is reset every year by law: beneficiary premiums and a general-revenue contribution are set annually at whatever level covers projected costs. A fund that is topped up by design cannot deplete. The cost shows up as higher premiums and a larger general-revenue draw instead.

You can see that in the 2026 numbers: the standard Part B premium is $202.90 a month, with an income-related surcharge starting above $109,000 of modified adjusted gross income for single filers and $218,000 for joint filers, adding between $81.20 and $487.00 a month. The Part D base beneficiary premium is $38.99.

What has changed since last year’s report?

Modestly worse for the two payroll-funded funds, modestly better for disability:

The “vs last year” column is the report’s own wording. The 2025 dates themselves are not quoted here, because “one quarter earlier” is what the 2026 report states and back-calculating a prior-year date from it would be our arithmetic presented as SSA’s.

There is also a procedural signal worth knowing about. When a fund’s assets are projected to fall below 20% of annual cost within ten years, the Trustees are required to notify Congress. Those letters have been sent, to the President of the Senate and the Speaker of the House, and they are published alongside the report.

And the short-range measure has been failing for years: OASI has not been adequately financed on the Trustees’ short-range test since 2019. Its trust fund ratio was 153% at the start of 2026 and is projected to fall below 100% by the start of 2029 — meaning reserves would no longer cover even a single year of benefits.

How big is the gap, in one number?

4.55% of taxable payroll, for OASI over 75 years. That is the actuarial deficit: the size of the permanent annual gap between what the fund is projected to take in and what it is scheduled to pay out, expressed as a share of all wages subject to Social Security tax.

Two things stand out. DI is the only one of them in surplus — the disability programme is not the part under strain, despite being the part most often described that way. And the combined deficit is smaller than OASI’s alone, precisely because DI’s surplus offsets part of it — which is another way of seeing why the combined figure flatters the retirement picture.

What that number does not do is prescribe anything. A gap of 4.55% of payroll can be closed from the revenue side, the benefit side, or both, in an unlimited number of combinations. Choosing among them is a political question, and this page does not have an opinion on it.

What this page cannot tell you

What your own benefit will be. These are projections about a fund, not about a person. Nothing here models an individual benefit, and it should not be used to decide when to claim.

What Congress will do. Every projection here assumes current law continues unchanged, which is the one assumption almost certain to be wrong over six years — the programme has been amended repeatedly, usually close to a deadline. The report projects a scenario; it does not forecast politics, and neither do we.

Whether the intermediate assumptions are right. SSA also publishes low-cost and high-cost scenarios with materially different dates. This page uses the intermediate set, which the Trustees label their best estimate, and does not average across them or argue for one.

The dates will move again. The Trustees Report is annual. OASI’s date moved a quarter earlier this year, and HI’s did too. A year from now there will be a 2027 report, and the honest expectation is that these numbers will be slightly different again.

Sources