WASHINGTON — The Social Security payroll tax stops at $176,100. A nurse or factory worker pays 6.2 percent on every dollar of wages through December; a portfolio manager whose salary clears that ceiling by late March pays nothing after that. The gap between those two experiences is what a bipartisan Senate proposal would close, and what Treasury Secretary Scott Bessent says Congress should not touch.
Bessent told congressional appropriators this month that Social Security faces “a growth problem and a spending problem,” not a tax problem. The framing is a direct rejection of what is arguably the most actionable fix on the table: lifting or eliminating the payroll tax cap so that higher earners pay the same rate on their full compensation. Bessent’s stated alternative is that the administration’s economic agenda, including deregulation, lower rates, and the productivity gains artificial intelligence is meant to deliver, will generate enough additional revenue to stabilize the program without touching its structure.
That bet has a fixed deadline. The trustees of Social Security, a body Bessent chairs, said in their annual report this month that the retirement trust fund will be depleted in late 2032, one year sooner than the prior projection. Once that happens, the program can pay only 78 percent of scheduled benefits from incoming payroll taxes, an automatic cut of roughly 22 percent that no president or Congress will have chosen. The Committee for a Responsible Federal Budget puts the aggregate loss to retirees at more than $345 billion.
The bipartisan pressure on Bessent sharpened last week. Senators Elizabeth Warren, Democrat of Massachusetts, and Bernie Moreno, Republican of Ohio, a pairing that almost never shares a byline on economic legislation, published a joint op-ed calling for eliminating the payroll tax cap entirely. Their proposal would require earnings above $176,100 to pay the same 6.2 percent as a postal worker’s first dollar. The Social Security Administration’s own actuaries have estimated that removing the cap, starting in 2026, would eliminate most of the projected shortfall without adjusting benefits or the retirement age.
The Treasury declined to offer a formal response or a scoring estimate for the Warren-Moreno proposal. What Bessent said to the Senate Finance Committee is that raising payroll taxes would be “the wrong medicine.” The Congressional Budget Office has not been asked by the administration to score whether its current growth projections close the 75-year actuarial deficit, which independent analysts at the Peter G. Peterson Foundation place at roughly $25.8 trillion in present value.
Within the administration, the position is not unanimous. Health and Human Services Secretary Robert F. Kennedy Jr., who also sits on the board of Social Security trustees, signed the annual report but filed an addendum recommending that Congress immediately begin raising taxes to close the unfunded obligation gap. Kennedy’s position sits to the structural left of Bessent’s growth-only approach, a divergence inside the same cabinet that the White House has not publicly addressed. The administration’s public position on Social Security is, in effect, two different positions held by two different trustees.
Bessent’s growth thesis has a historical echo. The administration’s broader economic frame, that technology investment will be disinflationary and revenue-generating and allow the government to grow its way past structural deficits, echoes the approach Alan Greenspan took at the Federal Reserve in the 1990s. Federal Reserve Chairman Kevin Warsh has said explicitly that his monetary strategy is modelled on Greenspan’s technology bet. The 1990s version produced a surplus that temporarily closed budget gaps but did not eliminate the entitlement shortfall that was already building then, and it ended when the growth assumptions unwound in 2000.
The payroll tax cap has not kept pace with wage growth at the top. As more compensation flows through equity grants, carried interest, and bonuses that arrive after the calendar-year ceiling is hit, a growing share of national income escapes the Social Security levy entirely. The cap’s coverage rate in real terms has eroded compared with the 1980s, when Congress last made major structural changes to the program under the Greenspan Commission. The Social Security trustees have documented the widening gap every year since. None of those reports has prompted legislative action.
The political terrain for the Warren-Moreno proposal is as narrow as the fiscal math is clear. Senate Majority Leader John Thune has not scheduled a floor vote. House Republicans have avoided any public accounting of where the caucus stands on lifting the cap. The One Big Beautiful Bill, the administration’s signature fiscal package that passed the House in May, contains no Social Security stabilization provision and reduces the payroll tax base at the margins by extending income tax cuts that lower reported wage income for some self-employed filers. The White House argument that anti-fraud measures at Social Security could help stabilize the fund has been undercut by the agency’s own inspector general, whose July 2024 report found improper payments were about 0.84 percent of total benefits paid, not a figure that closes a $25 trillion gap.
The inaction has a self-reinforcing logic. Every year the payroll cap sits unchanged, the proportion of high-end wages that escape the levy grows slightly as compensation at the top of the income distribution shifts toward forms the levy does not reach. Every year Congress defers, the present-value gap in the trust fund widens. The workers for whom lifting the cap would be most costly are also among the most consistent donors to the campaigns that will eventually vote on any fix.
What the administration has not published is its own 75-year solvency projection under its growth assumptions. What Congress has not done is schedule a vote on any fix. What the trustees’ report establishes is the date the automatic cut arrives if both continue as they have. Whether Bessent’s bet closes the actuarial gap before 2032 is genuinely unknowable at this distance. What is not unknowable is the cost of being wrong.
