01The fix if nobody fixes it

Start with the default, because it is what happens if Congress does nothing at all. Using the combined-fund figure, here is what an 83% check looks like.

The average monthly benefit for a retired worker in September 2026 was $2,088.79 (Social Security Administration). The 17% gap is applied to it in today's dollars to show its size; the real gap would apply to whatever benefits are in 2034.

Social Security does not borrow. Once a trust fund's reserves are gone, the program can only pay out what comes in from payroll taxes and the income tax on benefits, and the law does not let it pay more than that. According to the 2026 Trustees Report, released June 9, 2026, that point arrives in 2034 for the retirement and disability funds taken together, and the money coming in would cover 83% of scheduled benefits. That is the 17% gap. The law does not spell out exactly how Social Security would handle it: the Congressional Research Service notes it could mean smaller checks or full checks paid late.

It could come sooner. The retirement fund on its own runs out in late 2032, and at that point it can pay 78% of benefits, a 22% gap, unless Congress lets it draw on the disability fund. The two funds are legally separate. The Congressional Budget Office, which uses its own assumptions, has the retirement fund running out in fiscal 2032 as well.

The trustees also put the size of the hole into a single number: over the next 75 years, Social Security is short by 4.42% of all the wages it taxes. Here is what closing that in one move would take, according to the same report:

Nobody is proposing to do it all in one move. Real plans mix taxes and cuts, phase them in over years and spare people already retired. But those numbers are the size of the job, and every year of waiting makes them bigger.

02Every option, scored two ways

Social Security's Office of the Chief Actuary keeps a running list of more than 100 possible changes to the program, each one scored against the latest Trustees Report. For each, it reports how much of the 75-year shortfall the change would close and the year the money would run out with that change made. In September 2026 the list was updated to the 2026 report's numbers.

I took the options that come up most often and matched each one with the closest national poll I could find. The bar is the actuary's score. The heavy line at 100% is the whole shortfall closed.

Scores: Social Security Office of the Chief Actuary, provisions affecting solvency, based on the 2026 Trustees Report (intermediate assumptions). Each option is scored alone; scores do not add up exactly when options are combined. "Money runs out" is the year the combined trust funds would be depleted with that change; with no change it is 2034. "Not within 75 years" means the money lasts through the actuaries' full 75-year projection, not forever: the 17% rate closes 101% of the 75-year shortfall but only 71% of the gap in the 75th year. A year shown in red means the change does not push the date past 2034. Each poll is labelled: Same proposal means the poll asked about essentially the option scored; Related means a similar but different policy, described in the label. Where no poll asked about anything close, the related result sits behind a link and is not shown as support. The University of Maryland figures come from opt-in online panels that told respondents how much of the shortfall each change would close. Full poll details are in the sources at the end.

Two things jump out. Short of a big jump in everyone's payroll tax, only one family of options gets past half of the job on its own: taxing the wages that sit above the cap. And on the other side, the options that touch benefits each close somewhere between a tenth and a third of the shortfall.

03Why most fixes do not move the date

Look at the "Money runs out" column again. Raising the full retirement age to 69 closes about a third of the 75-year shortfall, and the money still runs out in 2034. So does switching to chained CPI, cutting benefits for future retirees with other income, and counting 40 years of earnings instead of 35.

The reason is timing. A higher retirement age only affects people as they reach it, so the savings start small and build for decades. Most of the money arrives after 2034. The same is true of anything that only applies to new retirees. Those changes fix the back half of the 75 years and leave the 2034 cliff exactly where it is.

Tax changes work faster because they start collecting the year they take effect. That is why taxing wages above $400,000 moves the date to 2047 and scrapping the cap entirely moves it to 2058. Any plan that wants to avoid the 2034 cut either raises money early or pairs slow benefit changes with something that pays out right away.

04More money in

The wage cap

Social Security tax only applies to wages up to a cap, $184,500 in 2026 and $192,300 in 2027. It rises with average wages in most years. Earn a dollar more than the cap and you stop paying the 12.4% tax on it, half from you and half from your employer.

When Congress last rescued the program in 1983, the cap covered 90% of all wages earned by people paying into Social Security. Pay at the top then grew much faster than pay for everyone else, and by 2024 the cap covered 82.6%. The difference is money that used to be taxed and is not anymore.

That is why the cap options score so well. Scrapping it entirely and taxing every dollar of wages closes 58% of the shortfall if the extra tax earns no extra benefits, and 45% if it does. The more common proposal leaves a gap: tax wages up to the cap as now, nothing in the middle, and then tax everything above $400,000 again. That closes 53% and keeps the money flowing until 2047.

It is also the most popular way to raise money on the list. In a February 2026 University of Maryland survey, 79% favored the $400,000 version, including 77% of Republicans. A probability-sample survey by the National Academy of Social Insurance in late 2024 found 68% support for scrapping the cap entirely and 65% for the $400,000 version.

The tax rate

The combined Social Security tax rate started at 2% in 1937 and has been 12.4% since 1990, apart from a temporary payroll tax holiday for workers in 2011 and 2012. Raising it is the bluntest tool there is. Taking it to 17% at once would close the whole shortfall. Smaller steps close smaller shares: a tenth of a point a year until it reaches 13.0%, the version the Maryland survey asked about, closes 12%, and 62% of respondents supported it. A Cato Institute survey by YouGov in October 2025 asked about a rise to 16.05%, a little short of the 16.65% the trustees say the full fix would take, and found 55% in favor with no "not sure" option offered.

Other revenue

Wages are not the only thing that could be taxed. Applying a 12.4% tax to investment income above $400,000 closes 28% of the shortfall. Sending all of the income tax collected on Social Security benefits to Social Security, instead of splitting it with Medicare, closes 19%. That money is not free, though: it now goes to Medicare's hospital insurance trust fund, so moving it helps Social Security by opening a hole there. Bringing newly hired state and local government workers into the system, about 27% of whom were not covered as of 2023, closes 3%.

One idea that keeps coming back is investing part of the trust fund in stocks. The actuaries do score it, but they do not report a share of the shortfall closed, saying the result cannot be read that way. Putting 40% of the reserves into stocks changes nothing about the 2034 date, and if stocks are assumed to earn no more than bonds once their risk is accounted for, the effect is zero.

05Less money out

The retirement age

Full retirement age was 65 when Social Security began. The 1983 rescue raised it to 67, phased in so slowly that the last group affected, people born in 1960, reaches 67 in 2027. Raising it again to 68 closes 13% of the shortfall. Taking it to 69 and then tying it to life expectancy closes 32%. Neither moves the 2034 date.

This is where the polling falls off. The Maryland survey found 56% would accept 68 when told what it saves. The National Academy survey found 37% support and 48% opposition for raising it to 68 or 69. And when AP-NORC asked in 2023 about raising it to 70, 10% were in favor and 75% were against.

The yearly raise

Benefits rise every year with inflation, measured by the CPI-W, a price index for working households. The 2027 raise is 3.6%. Switching to the chained CPI, which assumes people swap to cheaper goods when prices rise and so grows a little more slowly, closes 15% of the shortfall. President Obama put it in his budget in 2013 and dropped it the next year.

The opposite idea, basing raises on a price index weighted toward what older people buy, adds 7% to the shortfall. It is also one of the most popular ideas on the list: 70% in the Maryland survey. When the National Academy asked about a slower-growing measure, 38% were in favor.

Who gets cut

Cuts aimed at higher earners poll better than cuts for everyone. Cutting benefits for future retirees with more than $60,000 a year of other income, $120,000 for couples, closes 16% of the shortfall. The National Academy found 41% in favor and 43% against. The Maryland survey's version, cutting benefits for the top fifth of lifetime earners, drew 65% support.

A plain cut for everyone polls about as badly as anything can. In AP-NORC's 2023 survey, 6% supported "reducing the size of the benefits people receive." In a 2024 Pew survey, 79% said benefits should not be reduced in any way.

06What people will actually sign up for

Asked one at a time, most people want the problem fixed by someone else. Asked to build a whole plan, they get more practical.

The University of Maryland has run that test several times, giving people a briefing, the arguments on each side and the score for each option, then asking them to pick a package. In its 2024 run, majorities of Republicans and Democrats alike chose the same four things: tax wages above $400,000, raise the payroll tax rate from 6.2% to 6.5%, raise the full retirement age to 68 and cut benefits for the top fifth of earners. By Maryland's scoring those four close slightly more than the whole shortfall.

Then the same people also voted for three increases: a higher minimum benefit, a boost for those over 85 and a yearly raise based on what seniors buy. Those gave back enough that the final package closed 78% of the shortfall, not all of it.

The National Academy's 2024 survey found a similar mix. Its most preferred package taxed wages above $400,000, raised the rate to 7.2% each, based raises on what seniors buy and cut benefits for higher-income retirees, and 82% chose it over the status quo. According to the Academy, it would close the gap with a small surplus, but only because of the benefit cut for higher earners. Across that survey, 85% said they would rather pay more tax than see benefits cut.

07Build your own fix

Here is the same menu as a checklist. Tick the options you would vote for and the meter adds up the actuary's scores for each one on its own. Each row shows when the change would start and links to the actuary's table for it. Options that compete with each other, like two different retirement ages, can't both be ticked. For a version that counts every dollar of the federal budget, not just Social Security, try my Balance the Budget simulator.

Each score comes from the Office of the Chief Actuary on the 2026 Trustees Report basis. The actuaries score each option on its own, and combined options interact (a higher tax cap and a tax on investment income overlap, for example), so the total here is an illustrative sum, not an official score of the package.

08The plans in Congress

Members of Congress can ask the actuaries to score a whole bill. Here is the most recent score for each full plan scored since 2023, with how much of the shortfall each closed when it was scored.

Shares are calculated from each scoring letter: the change in the 75-year balance divided by the shortfall at the time. Every letter predates the 2026 Trustees Report, when the shortfall was smaller (between 3.42% and 3.98% of taxable payroll, against 4.42% in the 2026 report), so each plan would close less on the 2026 basis. Source: Office of the Chief Actuary, legislation and proposals scored.

Most of the bills that get close to solvency raise taxes on high earners and put some of the money back into higher benefits. The plan from Rep. Steny Hoyer and Wendell Primus of the Brookings Institution is the only one on the list that mixes new revenue with real benefit cuts, including a full retirement age of 70 for the top fifth of earners, and of the plans scored as solvent for the full 75 years, it uses the most recent basis (the 2024 report).

The best-known bill, the Social Security 2100 Act, was reintroduced on June 29, 2026 with a bigger tax change than before: no wage cap at all after 2026, plus a 12.4% tax on investment income above $400,000. As of late October 2026 the actuaries had not published a score for that version. Its lead sponsor, Rep. John Larson of Connecticut, lost his August 2026 primary and leaves Congress in January 2027.

The other idea that drew attention in 2025 and 2026 came from Sens. Bill Cassidy and Tim Kaine: borrow about $1.5 trillion, invest it in stocks for 75 years and use the gains to fill the hole. It was never introduced as a scored bill. The Center for Retirement Research at Boston College concluded it would do "virtually nothing" for solvency on its own. Cassidy lost his May 2026 primary and also leaves in January 2027.

Two bills would set up a bipartisan fiscal commission with fast-track votes on its recommendations, H.R. 3289 in the House and S. 4012 in the Senate. As of October 2026 neither had passed its chamber.

Recent changes have gone the other way. The Social Security Fairness Act, signed January 5, 2025, restored full benefits to about 2.8 million people, mostly retired public-sector workers and their spouses and survivors, and widened the 75-year shortfall by 0.14% of taxable payroll. The July 2025 tax law cut income taxes, including on many retirees' benefits, which means less tax revenue flows back into Social Security. The actuaries estimated it pulled the combined depletion date forward from the third quarter of 2034 to the first quarter.

09Keep reading

10Sources

1. Social Security Board of Trustees, 2026 Annual Report, highlights (depletion dates, payable shares, 4.42% shortfall, one-move fixes)
2. Office of the Chief Actuary, provisions affecting Social Security solvency, 2026 Trustees Report basis
3. Office of the Chief Actuary, legislation and proposals scored (plan letters)
4. Social Security Administration, Monthly Statistical Snapshot, September 2026
5. Social Security Administration, contribution and benefit base (taxable maximum)
6. Social Security Administration, latest cost-of-living adjustment
7. Social Security Administration, Social Security Fairness Act implementation
8. Annual Statistical Supplement, Table 4.B1, share of covered earnings below the taxable maximum
9. Social Security tax rate history
10. Full retirement age by year of birth
11. Congressional Budget Office, 2026 long-term projections for Social Security
12. University of Maryland Program for Public Consultation, Common Ground survey questionnaire, February 2026 (opt-in panels)
13. University of Maryland Program for Public Consultation, Social Security policymaking simulation, 2024
14. National Academy of Social Insurance, Social Security at 90 (NORC AmeriSpeak, October to November 2024)
15. AP-NORC Center, March 2023 topline
16. Pew Research Center, views of Social Security, June 2024
17. Cato Institute 2025 Social Security Survey, fielded by YouGov (opt-in, many items with no not-sure option)
18. Center for Retirement Research at Boston College, on the Cassidy-Kaine proposal
19. Congressional Research Service, Social Security coverage of state and local government workers (R46961)
20. Congressional Research Service, what would happen if the trust funds ran out (RL33514)
21. Social Security 2100 Act, H.R. 9519 (2026)
22. Fiscal Commission Act, S. 4012 (2026)