Will Social Security Run Out? What the Trustees’ Math Actually Says
The 2026 Trustees Report does not project Social Security ending. It projects a reserve running out in 2032 — and 78% still payable after.
6 min read
“Social Security is running out” is one of the most repeated sentences in American personal finance, and it misstates the official projection in a way that changes what a worker should do about it. The people charged with keeping the program’s books — its Board of Trustees, which publishes a full actuarial report every year — do not project that Social Security ends, empties, or stops paying. They project that one of its trust funds will exhaust its reserves in the fourth quarter of 2032, and that from that point the program’s continuing income will cover 78 percent of the benefits written into law. Zero and 78 percent are different planning problems. Getting from the headline to the actual numbers requires understanding what the trust funds are, where their money comes from, and what the word “depletion” legally does when it arrives.
Social Security is financed mainly by a payroll tax that most workers see only as a line on a pay stub. The combined rate is 12.4 percent of wages — 6.2 percent paid by the employee and 6.2 percent by the employer, with the self-employed paying both halves — and it applies to earnings up to an annual maximum, $184,500 in 2026. Of that rate, 10.6 points go to the Old-Age and Survivors Insurance fund, which pays retirement and survivor benefits, and 1.8 points to the Disability Insurance fund. Two smaller income streams complete the picture: a portion of the income tax that higher-income beneficiaries pay on their benefits flows back into the funds, and the funds earn interest on their reserves. Those reserves are the part of the system most often misunderstood. They are not a vault of cash and not a portfolio of stocks. By law they are invested in special U.S. Treasury securities — the accumulated IOUs from decades in which the program took in more than it paid out, lent to the rest of the federal government and backed by its full faith and credit. The balance in the funds is simply the running total of every past surplus and deficit.
For roughly the last several years the annual flow has run in reverse: the program pays out more in benefits than it collects in taxes, and the difference is drawn from the reserves. The 2026 Trustees Report, working from its intermediate — best-estimate — assumptions set in February 2026, explains why the gap widened this year. The Trustees lowered their long-run fertility assumption from 1.90 children per woman to 1.75 and reduced their immigration estimates, two changes that shrink the future workforce whose payroll taxes finance the system. They also record that the tax legislation enacted in 2025 reduced the revenue the funds receive from taxing benefits. Fewer workers per retiree, each paying the same 12.4 percent, is the whole engine of the shortfall; the 75-year actuarial deficit for the combined program now stands at 4.42 percent of taxable payroll, meaning the gap between what the law promises and what the law’s taxes collect equals about four and a half cents on every taxed dollar, measured across the next seventy-five years.
Against that arithmetic, the report’s dates are precise. The retirement and survivors fund, OASI, is projected to pay 100 percent of scheduled benefits until the fourth quarter of 2032 — one quarter earlier than last year’s report projected. At that point its reserves are depleted, and continuing income is projected to pay 78 percent of scheduled benefits, with that share gradually declining to 62 percent by 2100 if nothing changes. The disability fund is in a different position entirely: it is projected to pay full benefits through at least 2100, the end of the projection period. Because the two funds are often discussed together, the Trustees also publish a combined figure, while noting plainly that merging them would require a change in the law. On that combined basis, reserves last until the third quarter of 2034, after which continuing income would pay 83 percent of scheduled benefits, declining to 65 percent by 2100. When a single date is quoted for “Social Security,” it is usually the 2032 retirement-fund date or the 2034 combined date — and the difference between them is not pedantry, because the 2034 figure assumes Congress first passes a law letting the two funds share money.
What depletion would legally do, if it arrived, is narrower than the word suggests. A trust fund cannot pay out money it does not have; that is the entire mechanism. With reserves gone, benefits are limited to what current income supports — hence the 78 percent figure, which is not a policy choice but the ratio of projected income to scheduled benefits. What depletion does not do is stop the income. The payroll tax is collected from every paycheck in the economy, in good years and bad, and it would keep arriving the morning after depletion exactly as it did the morning before. Nobody’s benefit goes to zero under the projection; under current law, benefits would be reduced to the payable share, across the board, unless lawmakers acted first. It is also worth being precise about what kind of statement these numbers are. They are projections under one set of demographic and economic assumptions, not appointments. The Trustees run more optimistic and more pessimistic scenarios that move the depletion date years in either direction, and every annual report for decades has revised the date as the economy, fertility and immigration actually unfolded. The honest summary is not “Social Security ends in 2032.” It is “under today’s law and today’s best estimates, the retirement fund’s cushion runs out in late 2032, and taxes then cover about three-quarters of promised benefits.”
Whether that gap closes, and how, is a legislative question with a known menu. The Trustees themselves confine their advice to timing: lawmakers have many options, and acting sooner rather than later widens the range of choices and leaves more time to phase changes in. The levers themselves are the ones Congress has used before, most comprehensively in 1983, the last time a depletion date loomed closely: raise the payroll tax rate; raise or remove the $184,500 cap, so that earnings above it are taxed as well; lift the full retirement age, which reduces lifetime benefits for each cohort; slow the growth of scheduled benefits through the benefit formula; or bring in revenue from outside the payroll tax. Each lever closes some of the 4.42-percent-of-payroll gap, and every serious proposal is a combination of them in different proportions. What the report rules out is only the passive option: under current law, with no action, the reduction to the payable percentage happens automatically.
For a worker reading the projection as an input to their own planning, the Trustees’ numbers argue against both of the common reactions. Planning as if Social Security will pay nothing at all discards a projected 78 percent that rests on the most reliable revenue stream in the federal government; planning as if the scheduled 100 percent is guaranteed ignores a shortfall that has been in the official reports for decades. The distance between those two errors is where the rest of retirement saving lives — in accounts whose tax treatment, as our explainer on Roth versus traditional IRAs sets out, decides how much of your own money you keep. The Tax Desk follows the annual Trustees Reports as they appear, because each one re-prices the same question. Social Security, on the evidence of its own ledger, is not running out. It is running short — by a measured amount, on a published schedule, with the size of the shortfall stated to the decimal. Those are better facts to plan on than the headline.
Educational coverage — not tax, financial, or investment advice.
See Market Briefer first in Google
Add to Preferred Sources