What the Deficit and the Debt Are, and Which One the Argument Is About
The deficit is this year's borrowing. The debt is all the years added up. Most fights over "the deficit" are about one while pointing at the other.
October 28, 2026 · updated October 30, 2026
In one sentence
The federal deficit is the gap between what the government spends and what it takes in during one fiscal year; the debt held by the public is the total it has borrowed and not repaid, which grows by roughly the deficit each year.
The almanac shows both as a share of the economy, because that is the only way to compare 1946 with 2026. In dollars, every year's deficit is the largest ever and the number tells you nothing.
New articles and charts arrive in one short email on Monday mornings. Get the weekly digest
The analogy
The bathtub. The deficit is the tap: how fast water is running in this year. The debt is the water level. Turning the tap down is not the same as draining the tub; a smaller deficit means the level rises more slowly, not that it falls. The tub drains only in a surplus year, and there have been five of those since 1960, all between 1998 and 2001.
The economy is the size of the tub. It grows most years, so a given water level is a smaller share of the tub over time; that is how the country carried a 106% debt out of World War II down to 23% by 1974 without ever running large surpluses. The debt-to-GDP line falls whenever the economy grows faster than the debt, which is the whole of the growth-versus-interest arithmetic on the government page.
How it reaches your kitchen table
Interest rates: through the 10-year yield. More borrowing means more Treasury bonds for the market to absorb, and the price it charges is the term premium on the 10-year, which sets mortgages and every other long rate. Economists argue about how much; the range of estimates is a tenth to a third of a point on the 10-year for each 10 points of debt-to-GDP.
Taxes and programs: at the next budget. Interest on the debt was about 3.9% of GDP in fiscal 2025 (FYOIGDA188S), more than defense, and every dollar of it is a dollar that either comes from taxes, is cut from something else, or is borrowed again. The deficit is the number that decides which.
Inflation: if it is monetized. A government that borrows in its own currency can always pay, in dollars; whether those dollars are worth what they were is the inflation question. The 2021 relief, financed by borrowing while the Fed bought the bonds, is the recent case, and supply vs. demand takes up how much of the inflation that followed came from it.
Recessions: as the automatic stabilizer. The deficit widens on its own in a downturn, as tax receipts fall and unemployment benefits rise, and that widening is what keeps incomes up. The deficit tiles' 2009 and 2020 spikes are mostly this, not new laws.
What moves it
Receipts and outlays, both as shares of the economy, and both are on the deficit chart as companions. Receipts, 17% of GDP in fiscal 2025, move with the business cycle and tax law; outlays, 23%, move with the age of the population, health care costs, interest rates and the occasional war or pandemic. The gap between them is the deficit. The argument about whether to close it from the top line or the bottom line is the oldest in American politics, and the chart shows both lines so it can be argued from the same picture.
The debt moves with the deficit, and also with interest rates: when rates rise, old debt is refinanced at the new rate as it matures, so the interest line on the chart lags the Fed by a few years and is still catching up to the 2022–23 hikes.
How to read this year's number
Which year. The fiscal year ends September 30, so the deficit figure for a year was mostly set by the budget passed the previous fall. The Treasury's monthly statement gives a running total; the almanac shows the annual figure.
As a share of GDP. A 6% deficit with the economy at full employment is a different thing from a 6% deficit in a recession. The 2015–19 average was about 3.5%; the 2026 figure, near 6% with unemployment at 4%, is the part of the current argument that both sides agree is unusual.
The debt figure to use. Debt held by the public, on the almanac, excludes what the government owes its own trust funds. The gross figure, on the chart as a companion, includes it and runs about 24 points higher (121% against 97% in the second quarter of 2026; GFDEGDQ188S); it is the figure on the "debt clock" and in most headlines. Economists use the public figure because the trust-fund debt is money the government owes itself.
Where the analogy breaks
A bathtub overflows. Nobody knows where the economy's rim is, and the countries that have gone looking for it (Japan at 230% gross) have not found it yet, while others (Greece at 130%) found it early, the difference being who the lenders are and what currency the debt is in. A bathtub also doesn't grow. The economy does, so the level that matters is the share, and a debt that grows more slowly than the economy is falling in the only sense that counts.
What people argue about
Whether it matters. One side: interest already exceeds defense, the deficit is at a level previously seen only in recessions and wars, and the Congressional Budget Office projects the debt at 175% of GDP by 2056 with interest at 7% of the economy; the tub is filling and the tap is wide open. The other: the projection assumes current law unchanged for thirty years, which has never once been true; the 2010s showed the government could borrow at nothing; and the United States has the reserve currency and a growing economy, which is the whole difference from Greece. Both read the same CBO report. The dispute is whether a projection is a forecast.
Receipts or outlays. The deficit is the gap between two lines, and each side wants to move one. Those who would raise receipts note they are low by rich-country standards; those who would cut outlays note that spending's share has risen with an aging population and will keep rising. The chart's companions put both lines on the same axis; it cannot say which should move.
Whether deficits raise interest rates. Textbook theory says yes; the 2010s said barely; 2023–26 may be saying yes again. The term-premium estimates on the 10-year page are where the argument is conducted, and the honest range is wide.
See also
- Why the government is not a household: the better analogy for the debt, and the growth-versus-interest rule.
- Rate vs. level: why "the deficit fell" and "the debt rose" are both true.
- The 10-year Treasury yield: the price of the borrowing.
- The almanac's deficit and debt charts with every overlay.
Sources
- FRED: Federal deficit as % of GDP (FYFSGDA188S), receipts (FYFRGDA188S), outlays (FYONGDA188S), debt held by the public as % of GDP (FYGFGDQ188S), gross debt (GFDEGDQ188S), interest as % of GDP (FYOIGDA188S).
- Congressional Budget Office, The Long-Term Budget Outlook: 2026 to 2056, the projection on the chart.
- OMB Historical Tables, Table 7.1, the debt back to 1940, including the 1946 peak.
- U.S. Treasury, Monthly Treasury Statement, the running deficit within the year.
Corrections
October 30, 2026: An earlier version said interest on the debt is about 3.2% of GDP and that gross debt runs about 20 points above debt held by the public. Interest was 3.9% of GDP in fiscal 2025 (3.2% was fiscal 2023), and the gap was about 24 points in the second quarter of 2026.