Key Points:

• PWBM projects that the Social Security Old-Age and Survivors Insurance (OASI) Trust Fund will deplete in February 2033, or in February 2035 if OASI and Disability Insurance (DI) are treated as a combined fund. The 2026 Trustees Report projects depletion a few months earlier: the fourth quarter of 2032 for OASI and the third quarter of 2034 for the combined fund. In past years, PWBM projected earlier depletion dates than the Trustees; that gap has closed and slightly reversed.

• PWBM projects a 75-year actuarial deficit of 4.65 percent of taxable payroll, compared with 4.42 percent in the 2026 Trustees Report. Closing the PWBM shortfall would require raising the combined employer-employee payroll tax rate from 12.4 percent to 17.1 percent, an equivalent reduction in benefits, or some combination of both. At depletion of the combined fund, 86 percent of scheduled benefits would be payable, falling to 60 percent by 2100.

• The similar bottom lines mask offsetting demographic paths. PWBM projects lower long-run fertility (about 1.6 births per woman versus the Trustees’ assumed 1.75) but smaller gains in life expectancy at age 65 (1.6 years for men by 2100 versus 4.4 years). Smaller longevity gains keep PWBM’s projected costs below the Trustees’ through the 2050s; lower fertility pushes them above thereafter, reaching 21.8 percent of payroll in 2100 versus 20.0 percent.

Background

Each year, the Social Security Trustees publish projections of the program’s finances over a 75-year horizon, prepared by the Social Security Administration’s Office of the Chief Actuary (OACT). The Trustees’ approach is top down: it starts from assumptions about aggregates, such as the ultimate fertility rate and the rate of average wage growth, and applies them to program totals. PWBM takes nearly the opposite approach. Its microsimulation model builds projections from the bottom up, starting from individual-level data on earnings histories and family structures with rich demographic heterogeneity at the household level. The model requires demographics, labor markets, capital markets, and economic growth to reconcile with one another, so aggregate paths such as fertility, life expectancy, and wage growth are outputs of the model rather than assumptions.

PWBM is the only non-governmental group that produces independent long-range projections of Social Security’s finances. Because PWBM’s projections do not rely on the government’s inputs, the comparison provides an external check on official estimates, including whether the economic outlook underlying them became too optimistic after the passage of the One Big Beautiful Bill Act in July 2025. Agreement on the bottom line indicates that Social Security’s financing shortfall is robust to two very different modeling approaches. At the same time, differences in the underlying demographic and economic paths reveal where the projections could diverge in the future. This brief compares PWBM’s current-law baseline with the 2026 Trustees Report across the headline solvency measures and the paths that drive them.¹

Depletion Dates and the 75-Year Balance

Table 1 summarizes the headline solvency measures. PWBM projects that the OASI Trust Fund depletes in February 2033 and that the combined OASDI fund depletes in February 2035. The 2026 Trustees Report projects slightly earlier dates: the fourth quarter of 2032 for OASI and the third quarter of 2034 for OASDI. The two projections differ by less than a year on both measures.

This near-term alignment marks a change from past years, when PWBM projected earlier trust fund depletion dates than the Trustees. That gap has closed and, in this year’s comparison, slightly reversed: PWBM’s depletion dates now fall a few months after the Trustees’.

Over the full 75-year horizon, PWBM projects an actuarial deficit of 4.65 percent of taxable payroll, somewhat larger than the Trustees’ 4.42 percent. In words, making Social Security solvent over the next 75 years on a conventional basis would require raising the combined employer-employee payroll tax rate from 12.4 percent to 17.1 percent, cutting spending by an equivalent amount, or some combination of both. PWBM thus remains more pessimistic than the Trustees over the full horizon, but the difference of 0.23 percentage points of payroll is small. An independent, bottom-up projection that lands this close to the official one, and slightly later on depletion timing, offers no support for concerns that the official projections overstate Social Security’s financial health.

After depletion, benefits would be reduced to the level payable from contemporaneous revenue. PWBM projects that 86 percent of scheduled benefits would be payable when the combined fund depletes, falling to 60 percent by 2100. The Trustees project 83 percent payable at depletion and 65 percent in 2100. The lower PWBM payable share at the end of the horizon reflects faster projected cost growth in the second half of the century, examined next.

Annual Cost and Income Rates

The annual cost rate measures Social Security spending as a percentage of taxable payroll; the annual income rate measures non-interest revenue, mainly payroll taxes, on the same basis. Figure 1 plots both rates for PWBM and the Trustees through 2100.

The income rates are nearly identical: both projections start near 12.9 percent of payroll in 2026 and drift up to about 13.5 percent by 2100. The differences are on the cost side, and they reverse over time. Through mid-century, PWBM projects lower costs than the Trustees, with a gap of 1.3 percentage points in 2050 (15.6 versus 16.9 percent of payroll). The PWBM cost rate crosses above the Trustees’ around 2060 and remains higher thereafter, reaching 21.8 percent of payroll in 2100 versus 20.0 percent.

This timing pattern explains why PWBM projects slightly later depletion but a larger 75-year deficit. Lower costs in the next two decades extend the trust fund a few months beyond the Trustees’ dates, while higher costs late in the century widen the cumulative shortfall: by 2100, PWBM projects an annual deficit of 8.2 percent of taxable payroll, compared with 6.6 percent for the Trustees.

Population Growth and Aging

The diverging cost paths trace back to demographics. Figure 2 shows total population in each projection, normalized to 1 in 2025. PWBM projects faster population growth than the Trustees over the next five decades, with the population peaking in 2076 at 11.9 percent above its 2025 level and then declining. The Trustees project slower but sustained growth throughout the horizon. The PWBM population falls below the Trustees’ in the early 2080s and ends 2100 at 10.7 percent above the 2025 level, versus 13.5 percent for the Trustees.

Population aging matters more for Social Security than population size. Figure 3 shows the old-age dependency ratio, the population aged 65 and over relative to those ages 20 to 64. Both projections show the ratio climbing steeply as the baby boom generation ages, from about one-third today to about one-half by 2100. Between roughly 2040 and 2055, the PWBM ratio runs slightly below the Trustees’ (0.39 versus 0.40 in 2050), consistent with PWBM’s lower mid-century cost rate. From around 2060 onward, the PWBM ratio rises above the Trustees’ and reaches 0.51 in 2100 versus 0.49, consistent with PWBM’s higher late-century costs.

Lower Fertility Shrinks the Future Workforce

Fertility is the largest gap between the two projections. Figure 4 shows the total fertility rate. The projections start close together, at 1.65 births per woman (PWBM) versus 1.59 (Trustees) in 2025. The Trustees assume fertility recovers to an ultimate rate of 1.75 by mid-century. PWBM’s bottom-up demographic projection shows no such recovery: fertility remains near 1.6 births per woman through 2100.

Lower fertility affects Social Security’s finances with a generation-long lag. Children born over the next two decades enter the workforce starting in the 2040s, so PWBM’s lower fertility path translates into a smaller taxpaying workforce, a higher old-age dependency ratio, and a higher cost rate in the second half of the century. This single difference accounts for much of PWBM’s faster late-century cost growth shown in Figure 1.

Smaller Longevity Gains Hold Down Costs

Mortality works in the opposite direction. Figure 5 shows life expectancy at age 65 (LE-65), the expected number of remaining years for someone who reaches age 65. The projections start close together in 2025: 18.4 years for men and 21.1 years for women under PWBM, versus 18.2 and 20.7 under the Trustees. They then diverge substantially. By 2100, the Trustees project LE-65 of 22.6 years for men and 24.6 years for women, gains of 4.4 and 3.9 years. PWBM projects much smaller gains of 1.6 years for men and 1.9 years for women, reaching 20.0 and 23.1 years.

Longer lives mean more years of benefit collection per retiree. Because PWBM projects smaller longevity gains, each retiree collects benefits for fewer years than under the Trustees’ assumptions, holding down PWBM’s projected costs, especially through mid-century before the fertility gap takes over. The offset between lower fertility (raising costs) and smaller longevity gains (lowering costs) is why the two projections arrive at similar 75-year bottom lines despite different underlying demographic paths.

The longevity projection could shift. Even greater adoption of GLP-1 drugs, especially among lower- and median-income households with shorter life expectancies, could lead PWBM to increase its projected longevity gains in the future. PWBM is monitoring this adoption closely.

Wage Growth Paths

Wage growth contributes to the differences between the two projections in the short and medium run but less in the long run. Figure 6 shows annual growth in the economy-wide average wage, uncapped by the taxable maximum. The two series are identical through the early 2020s, where both reflect historical data. Going forward, the Trustees assume wage growth settles between 3.5 and 3.6 percent per year, while PWBM’s projection fluctuates around a similar level, averaging 3.8 percent per year over 2026 to 2100 versus 3.7 percent for the Trustees. These seemingly small differences compound over time for several decades but play less of a role over time. The year-to-year variation in the PWBM series reflects the bottom-up structure of the microsimulation rather than a different view of long-run wage growth.

Importantly, for comparison with the Trustees, these estimates are “conventional” in that future wage growth follows standard growth accounting rules embedded in the PWBM microsimulation model. These estimates do not include full dynamics consistent with the large fiscal imbalances in the rest of government. Traditionally, the Trustees have taken the view, either explicitly or implicitly, that Social Security finances are “off budget” and its finances should be stated conditional on the rest of the government being in fiscal balance. As we show elsewhere, U.S. fiscal policy is far away from balance. Still, since Social Security benefits are wage-indexed, wage growth projections play a larger role in the short run than in the long run. We present sensitivity analysis to wage growth assumptions elsewhere.

This analysis was produced by Seul Ki (Sophie) Shin under the direction of Kent Smetters.

Footnotes

1. In the context of Social Security, “present law” means that the calculations assume that “scheduled benefits” determined by current formula rules will be paid. In contrast, under “current law,” benefits would be reduced after the depletion date to their “payable” levels, equal to revenue collected during the same period. The Social Security system, therefore, faces no shortfall under current law. ↩