The US runs a federal budget surplus

Where things stand (16 Oct 2026)

FY2026 (the first year that counts) will clearly be a deficit. Treasury's own data, from the Fiscal Data API, mts_table_1, record_date 2026-08-31, shows a deficit of $1,965.6B for FY2026 through August. The September MTS was not yet in the API on 16 Oct. CBO's Monthly Budget Review for September 2026 puts the full-year deficit at $1.993T, $218B larger than FY2025 (summarised by AEI, 10 Oct 2026). CRFB (1 Oct 2026) estimates the deficit at 6.2% of GDP. On its figures, revenue was 16.7% of GDP, spending 22.9%, and net interest about $1.1T, a record 3.4% of GDP. The official September MTS and the Treasury/OMB joint statement are due in late October, which falls in the 2026-Q4 window. They will show a deficit of roughly $2T.

The outlook is getting worse:

• CBO's February 2026 baseline has deficits of 5.8–6.7% of GDP every year through 2036. Revenue stays around 17.5–17.8% of GDP while spending rises to 24.4%.
• On 20 Feb 2026 the Supreme Court struck down the emergency-powers (IEEPA) tariffs. Refunds of about $100B or more followed, and CBO said customs receipts for 2026 would come in $250B below its earlier projection (Fortune, 10 Aug 2026).
• The 10-year Treasury yield is now around 5.0–5.3%, against CBO's assumption of 4.1–4.4%. CBO estimates that rates 150 basis points higher would add $6.0T to deficits over ten years (AEI).
• The President's FY2027 budget asks for $1.5T in defense spending, up 44%. It contains no deficit-path table, and even its optimistic 3.1% growth assumption is not presented as producing balance.

Base rate

I used OMB Historical Table 1.2 from the FY2027 budget (whitehouse.gov/wp-content/uploads/2026/04/hist01z2_fy2027.xlsx). Since 1950 there were 14 years with a deficit of at least 4% of GDP that have a full five-year window ending by 2020. None was followed by a surplus within five years. Of the 7 years with deficits of at least 5%, also none did. The only fast swings to surplus were in the WWII demobilisation years, 1942–46. The best peacetime case is 1992–98: from a 4.5% deficit to surplus in six years. That needed two tax increases, the post–Cold War defense cuts and a capital-gains boom. Bernstein and co-authors, cited by CAP in June 2026, call it the only move of that size outside a rebound from recession. Some other rich countries made swings of 6 points or more in about five years, such as Sweden and Canada in the 1990s. But they started from a recession low and acted under crisis-driven consensus. The US is starting near full employment, with large deficits excluding interest and interest costs that are still rising.

Paths to YES (only Q4 of each year matters, when the September MTS comes out)

1. A very large AI-driven boom. To reach surplus by FY2031, nominal GDP would need to be roughly 30–40% above baseline. That means sustained real growth of 8% or more, beyond anything in the post-war US. Several things weaken the revenue effect:

   • Full expensing cuts corporate tax during a capital-spending boom; corporate receipts fell 23% in FY2026.
   • AI gains flow mostly to lightly taxed capital income.
   • Faster growth would push interest rates and interest costs up.
   • Pressure to compensate displaced workers would add spending.

   Dynan, Elmendorf and Sheiner (NBER, summarised by Brookings, Aug 2026) find that federal debt keeps rising even in their most optimistic AI scenario. I put this path at roughly 0.5–1%.

2. Crisis-forced consolidation. A spike in bond yields could force a grand bargain, but a 7-point swing in GDP by FY2031 would be unprecedented, and a crisis would probably cause a recession. About 0.2–0.3%.

3. Accounting or one-off items such as asset sales or loan re-estimates. These are too small, and gold revaluation counts as financing rather than receipts. About 0.1%.

Hazards and the strongest case against

The chance of a first surplus in each Q4 is 0.05% for FY2026 (in practice zero), then 0.05% for FY2027, 0.1% for FY2028, 0.15% for FY2029, 0.3% for FY2030 and 0.5% for FY2031. Other quarters add nothing new. Cumulated, that gives about 1.15% by 31 Dec 2031.

The strongest case for a higher number is that this question set is about AI, and growth is currently running near 4% on AI capital spending. Five years is long enough for surprises. A thinly traded Kalshi market, "Will Trump balance the budget?" (any of FY2025–28), traded at 11–12%. I discount that price as longshot bias in a tiny market, especially since two of its four years are already large deficits. The strongest case for a lower number is that no mainstream projection, including the administration's own, shows balance in this window. Political incentives also point toward more borrowing. I would accept 99:1 odds against.

Sources: Treasury Fiscal Data MTS API (api.fiscaldata.treasury.gov …/mts/mts_table_1); CBO MBR Sept 2026 (cbo.gov/publication/61985), via AEI (aei.org, 10 Oct 2026); CRFB (crfb.org/blogs/us-ran-2-trillion-deficit-last-year-we-estimate, 1 Oct 2026); CRFB summary of CBO's February 2026 outlook; Fortune (10 Aug 2026); FY2027 Budget (whitehouse.gov/wp-content/uploads/2026/04/budget_fy2027.pdf); OMB Historical Tables 1.1 and 1.2; CAP (23 Jun 2026); Brookings (7 Aug 2026); Octagon/Kalshi KXBALANCE page.

Question. By the last day of each quarter, has the US Treasury reported a federal budget surplus for a full fiscal year?

Definitions.

• Full fiscal year: the federal fiscal year running from October 1 through September 30, named for the year in which it ends. The Observation period convention applies; FY2026 is the first eligible year.
• Federal budget surplus: total federal receipts exceed total federal outlays for the completed fiscal year on the unified budget basis used in the Monthly Treasury Statement (MTS), including both on-budget and off-budget results and net interest. The reported surplus must be strictly greater than zero.

Resolves YES if Treasury publishes actual completed-year results showing a surplus in the September (final fiscal-year) MTS, a subsequent official revision, or another Treasury publication explicitly reporting the same unified-budget annual result. In the MTS, use Table 1, “Summary of Receipts, Outlays, and the Deficit/Surplus of the U.S. Government,” full-year total (a negative value in the “Deficit/Surplus (-)” column), or Table 2, “Summary of Budget and Off-Budget Results and Financing of the U.S. Government,” the September “Current Fiscal Year to Date” value for “Total Surplus (+) or Deficit (-)” (a positive value). Successor tables reporting the same measure count. A revision that first reports a surplus qualifies on its publication date; a later revision does not undo a YES. No attribution to AI is required.

Resolves NO if no qualifying result has been published by the quarter-end. Exclude forecasts and budget estimates, monthly or incomplete-year surpluses, rolling twelve-month or calendar-year totals, primary surpluses excluding interest, on-budget-only or off-budget-only surpluses, accrual-accounting net operating results, and reductions in debt or borrowing without an annual unified-budget surplus. A reported zero balance does not count; a rounded zero requires an explicit Treasury statement or more precise official figure establishing a positive surplus. Repetition or revision of a fiscal year ending before the opening date does not count.

Resolving source. US Department of the Treasury, Bureau of the Fiscal Service: Monthly Treasury Statement of Receipts and Outlays of the United States Government, its official data tables and revisions, or a Treasury-authored release explicitly reporting the same completed-year unified-budget result. Treasury sources only, a narrower source rule than Public confirmation (A2); CBO estimates and press reporting do not independently qualify. Apply the Resolution date (A3), Resolution window, and Observation period conventions. Publication, not the fiscal-year-end date, determines the resolving quarter.