Eliminating the payroll tax ceiling would generate a fizz of new revenues but be followed by higher benefit costs and fewer taxes available to meet them.
June 2026 was a busy month for the Social Security program. On June 9, Social Security’s Trustees released their annual report, which showed a 16 percent increase in Social Security’s long-term deficit and estimated that the retirement trust fund would be exhausted in 2032. These figures send a clear and sobering message: We badly need bipartisan Social Security reform.
But bad is not the answer to badly. On June 23, Senators Bernie Moreno (R., Ohio) and Elizabeth Warren (D., Mass.) announced a joint plan to address Social Security’s $30 trillion funding gap by raising taxes on Americans earning more than $184,500. The proposed Moreno-Warren tax increase would be massive; by eliminating the payroll tax cap, it would boost the top effective tax rate by twelve percentage points and raise, on paper, $3 trillion over ten years. The plan’s authors claim it would keep Social Security solvent for a generation.
But the Moreno–Warren Social Security plan’s advertised fiscal improvements are overstated, for four reasons.
First, with many high-income households staring down marginal tax rates above 60 percent, the plan would create massive incentives to shift compensation to non-taxable forms of income, such as fringe benefits, stock options, or by becoming an S corporation.
Second, the proposal almost surely would harm the economy. The Congressional Budget Office’s (CBO) models will likely forecast that the proposed tax increases will reduce economic growth and, with it, revenues. On the other hand, the CBO has projected that fixing Social Security by reducing benefits, even with crude across-the-board cuts, would substantially increase economic growth. The Penn Wharton Budget Model, which uses a similar methodology to the CBO, finds that Social Security reforms relying heavily on taxes lead to lower growth than benefit cut–oriented reforms.
Third, the CBO and the Social Security Administration (SSA) assume that employers will reduce employee wages dollar-for-dollar to cover their increased tax bills. Those lost wages will no longer be subject to state or federal income taxes or Medicare taxes, potentially offsetting the assumed revenue increases by one-quarter or more. Illustrating this point, a 2006 study by two progressive economists concluded that 24 percent of the revenue gains from eliminating the Social Security payroll tax ceiling would be lost via reduced revenues from federal and state income taxes and from payroll taxes. As both state income tax rates and Medicare tax rates on high earners are higher today than they were in 2006, the revenue shortfall caused by the Moreno–Warren proposal may be even greater.
Fourth, in exchange for higher taxes, the plan requires Social Security to pay out increased benefits. In the long-term, roughly half the tax increase is offset by higher benefits for seniors whose pre-retirement earnings already put them in the top 5 percent of earners. These increased benefits explain why the SSA finds that eliminating the payroll tax ceiling addresses only 28 percent of Social Security’s ultimate cash flow deficits.
This fourth factor points us back to a historical episode that should serve as a cautionary tale to policymakers.
In the 1970s, Congress indulged a litany of Social Security benefit increases. Between 1970 and 1981, Social Security’s average “replacement rate,” which represents benefits relative to pre-retirement earnings, rose by 64 percent. Officials at the time took notice: Social Security’s chief actuary warned President Nixon in 1970 of SSA officials who wished the program “to be virtually monopolistic in economic security . . . As a subsidiary matter, they would eliminate the private pension system.”
As a result of rapidly rising benefits, Social Security was hurtling toward insolvency. So, Congress acted in 1977, if feebly. Under the Social Security Amendments of 1977, benefit increases for near-retirees were grandfathered in and only partially rolled back thereafter. Furthermore, new retirees in 1986 still received benefits 40 percent higher than what the 1970 benefit formula had dictated.
Congress made up the difference with tax increases. Under the 1977 legislation, the combined payroll tax rate rose from 9.9 to 10.8 percent. But the payroll tax ceiling rose even further. In 1970, the maximum taxable salary of $6,186 was equivalent to 1.3 times the national average wage. That same multiple of the average wage today would be $94,975, on par with many other developed countries.
But through the 1970s, the payroll tax ceiling was increased to pay for rising benefits. By 1986, the maximum taxable salary had increased to 2.4 times the average wage, where it remains today. Imposing higher payroll tax rates on more wages increased Social Security’s revenues from 3.3 percent of GDP in 1970 to 4.7 percent of GDP in 1986. There wasn’t any logic based on economic wisdom or considerations of fairness; the Social Security tax was applied to higher wages simply because Congress had overpromised on benefits and needed the money.
But, over time, a higher tax ceiling raises benefits. I have complained of Social Security paying a high-wage couple nearly $100,000 in annual benefits. That amount, which is far more than the maximum benefits paid in other Anglo countries — for example, Canada, Australia, and New Zealand — derives from the payroll tax ceiling increases of the 1970s. Absent those reforms, maximum benefits would have stood at around $70,000.
Eliminating the payroll tax ceiling, as Moreno and Warren propose, would repeat this foolhardy process. It would generate a fizz of new revenues but be followed by a hangover of higher benefit costs and fewer taxes available to meet them.
What Social Security reform desperately needs is what Congress seemingly lacks: the policy chops to understand the purpose of a social insurance program — namely, to protect poor seniors, survivors, and the disabled. Such programs should not function as mere replacements for what households and private markets can better provide on their own.
I don’t expect far-left politicians to understand these distinctions. But I would hope that self-professed conservative Republicans would.