The 49 percent claim: who pays no income tax, and who pays

1The claims

The first claim is quoted word for word from a post in circulation in October 2026, and the spelling and punctuation are the writer's own, because a quotation in this brief is a promise that the words are exact. The sentence states a voting intention and rests it on the figure of 49%. That figure itself has circulated since the early 2010s, usually in arguments about the income tax. The second claim is stated in its common form, because it circulates in many wordings and no single person's sentence can honestly be quoted for it. Both claims are repeated constantly, at work, at family tables, and in comment threads across the country. This brief checks the percentage against the record, and then checks what the percentage does and does not describe.

2What this argument is really about

Two different arguments share one number. The factual argument is about the current share of tax units that owe no federal income tax, and about who the households inside that share are. The values argument is about how broad the income tax base should be, and about whether payroll and other taxes count as paying a share. Underneath both sits a further question, and it is whether a share that the code produces on purpose is a fact about the people inside it or a fact about the design of the code. The record below settles the factual argument. The values argument remains open after the record is read, and it belongs to the reader.

3The record

• The current share: the Tax Policy Center estimates that about 40% of tax units will not pay federal income tax in 2025. Before the pandemic, the figure ran about 44%. (Corroborated, Tax Policy Center.)
• Where the 49% comes from: the share rose during and after the Great Recession and again under pandemic policy, reaching roughly half in the recession era, and about 61% in 2020 and 57% in 2021. The 49% figure belongs to those years. Stating it as the current, normal share is a date error. (Corroborated, Tax Policy Center.)
• Who the non-payers are (the Tax Policy Center's composition analysis of the group): about 60% of those who owe no federal income tax work, and they pay payroll taxes at 7.65% from the first dollar of wages, with no standard deduction. Most of the remaining 40% are retirees whose income is too low to owe income tax. Only about 9% of that group has its payroll tax fully offset by refundable credits. (Corroborated, Tax Policy Center.)
• The filer-side view: USAFacts, analyzing 2023 returns, found that filers earning under $50,000 filed 49.5% of all returns and paid just under 3% of total federal income taxes, and that 30.5% of filers owed no federal income tax that year. The two sources count differently, because the Tax Policy Center counts tax units and USAFacts counts returns, and the percentages from the two bases are never blended in this brief. (Corroborated, USAFacts.)
• The other taxes: payroll taxes fund Social Security and Medicare, and they are owed by design from the first dollar of wages. Sales, excise, and state taxes apply regardless of federal income tax liability. The statement that a household pays no federal income tax and the statement that it pays nothing are different statements, and the first is sometimes true while the second is almost never true. (Verified, as a matter of tax structure.)

4Where the first claim holds, and where it breaks

The first claim holds on the size of the group it points at, because about 40% of tax units owing no federal income tax in a normal year is a large share, and the breadth of the income tax base is a legitimate subject for public argument. The claim breaks on the date of its figure, because 49% is the share from the recession and pandemic years rather than the share in a normal year, and the current estimate is about 40%. The claim also rests on the income tax alone, and the record shows that most of those inside the share pay payroll taxes from the first dollar they earn, which the figure of 49% leaves out of view.

5Where the second claim holds, and where it breaks

The second claim holds on the origin of the share, because the share exists by design, produced by the standard deduction and by credits such as the Earned Income Tax Credit and the Child Tax Credit that Congress enacted on purpose, largely to keep low-wage work and retirement from being taxed into poverty. The claim also holds that the households inside the share mostly pay other taxes, payroll taxes first among them. The claim breaks if the design explanation is asked to do more than explain, because the origin of the share does not settle its size, and a share of about 40% of tax units remains large whether it was designed or not. Whether that size is the right size is a separate question, and the design fact alone cannot answer it.

6What is true in both

Set both claims side by side and they share a foundation. A large share of tax units owes no federal income tax in a given year. That share is the product of the code as Congress wrote it. The record settles those two points. What the figure means, and whether the share should be smaller or larger, is the actual debate, and it cannot be settled by a dataset, including this one.

7The number in the middle

60%. The first claim, taken at full strength, treats the share that owes no income tax as the measure of who pays. The second claim, taken at full strength, treats the share as a product of design and stops at its origin. The Tax Policy Center measured who is inside the share, and about 60% of those in it work and pay payroll taxes from the first dollar they earn.

8What would settle it, and the question it leaves

The factual part settles with the current estimate and the composition data, and both are in the record above: about 40% of tax units owe no federal income tax in 2025, and about 60% of the households in that group are workers paying payroll tax. What the record cannot settle is whether the income tax base should be broader, whether refundable credits are good policy, and whether payroll taxes should count as paying in for the purposes of the argument. Those are values questions. They belong to the reader, and they are real ones.

9What we simplified

Tax unit and filer statistics are kept labeled throughout this brief, because the two bases are not interchangeable, and percentages from the two bases differ by design rather than by error. The 60 percent and 9 percent figures come from the Tax Policy Center's composition analysis of the group, published alongside its earlier estimates; the roughly 40 percent figure is its 2025 estimate.

10Words in this brief

Tax unit

the household or individual as counted for tax purposes, and the Tax Policy Center's denominator. Commonly misused as: interchangeable with "filers" or "Americans." Each denominator produces a different percentage, and each is correct on its own terms.

Payroll tax

the 7.65% employee share, plus an equal employer share, funding Social Security and Medicare, and owed from the first dollar of wages. Commonly misused as: not a real tax on workers. It is the largest federal tax most low and middle income workers pay.

Refundable credit

a credit that can reduce tax below zero and pay out the difference, such as the Earned Income Tax Credit and part of the Child Tax Credit. Commonly misused as: something outside the tax system. It is written into the code, and it is the mechanism by which about 9% of households end up offsetting even their payroll tax.

11Sources

Tax Policy Center (share of tax units owing no federal income tax, 2025 estimate; pre-pandemic norm; pandemic-year estimates; composition of non-payers) · USAFacts (2023 filer analysis of federal income tax returns). Receipt Pack MC-0007 available on request.