125 years of U.S. federal revenue, spending, debt and GDP — charted against every major tax law since 1913, grouped by presidential term. Simple, colorful, and honest about what the data can and can't tell us.
Loading analysis…
Check the figures below: revenue vs. spending · tax-law timeline · growth vs. revenue · wealth concentration · defense spending vs. borrowing · taxing the rich · who got the 2017 tax cut · spending cuts · the tax-cut counterfactual · how the US stacks up abroad · full analysis & caveats.
Federal receipts (what came in) and outlays (what went out), in billions of nominal dollars, every fiscal year since 1901. Where the orange line rides above the blue line, the government ran a deficit that year. Gray shaded bands are NBER-dated recession years — note how often the deficit widens because revenue drops in a downturn, not because of a tax-law change.
Nominal dollars make everything look like it grew enormously just from 125 years of inflation. This is the same revenue/spending/debt data re-expressed in constant 2017 dollars (using the GDP price deflator, i.e. nominal GDP ÷ real GDP) — a fairer basis for comparing, say, 1950 to 2020.
The gap between revenue and spending each year. Red bars below zero are deficit years (borrowed money); blue bars above zero are surplus years (paid down debt). Surpluses have been rare since the 1960s.
Total federal debt only goes back to 1939 in FRED's series — earlier gross-debt figures exist but aren't in this feed, so the line starts there, not at 1901. Left: debt in current dollars. Right: debt as a share of the whole economy (GDP) — the more meaningful number, since a bigger economy can carry more debt.
Every dollar of deficit-financed tax cut (or spending increase) adds to the debt, and that debt has to be serviced — interest payments as a share of GDP, 1940–2025. This is the feedback loop that makes deficit-financed cuts more expensive over time than their headline price tag: today's interest bill is now competing with defense and most discretionary programs for budget space.
Every dot is a major federal tax law since 1913, colored by whether it mostly cut taxes, raised them, or did some of both. Hover a dot for the act's name and which president signed it. Background shading marks each presidential term.
Year-over-year percent change in the size of the economy (GDP, aqua) versus year-over-year percent change in what the government collected (revenue, blue). If tax cuts reliably "paid for themselves," you'd expect revenue growth to consistently outrun or track GDP growth right after a cut. Mostly, it doesn't — revenue is far more volatile than GDP and dips hard in recessions regardless of tax policy. Gray bands are NBER-dated recession years; thin dashed vertical lines are the same tax-law dates as the timeline chart above, colored the same way (blue = cut, red = increase, gray = mixed) — a lighter-weight combination of the two charts rather than merging them outright.
Share of federal tax collections by source. The individual income tax and payroll/social-insurance taxes (Social Security & Medicare) together have carried the vast majority of the load for decades; the corporate share has shrunk considerably since the 1950s, and customs duties (tariffs) — newly relevant with 2025's tariff increases — are visible as a thin sliver at the top.
Everything else on this page is federal-only, which understates the full tax picture — state and local governments collect income, sales, and property taxes too. This compares the federal tax take (blue) to an estimate of the combined federal + state/local take (violet), both as % of GDP. The state/local figure is derived from consolidated government NIPA receipts minus federal receipts, an approximation (see footer), not a direct state/local tax-collections series.
Share of federal spending going to entitlement-style "mandatory" programs (Social Security, Medicare, Medicaid, unemployment insurance — blue), day-to-day agency/defense "discretionary" operations (aqua), interest on the debt (red), and everything else, mainly grants-in-aid to states and subsidies (gray). Entitlements grow on autopilot as the population ages, largely independent of whoever is setting tax policy — which is why spending-side reform, not just tax policy, drives the deficit math. This is a NIPA-based approximation of the mandatory/discretionary split, not OMB's official budget-authority classification (see footer).
Every complete four-year term since 1901, with the average annual deficit (as % of GDP, where available), average yearly revenue growth, average yearly GDP growth, and which tax law(s) landed during that term. Recessions and wars during a term distort these numbers — see the analysis below.
| Term | Avg. deficit (%GDP) | Avg. revenue growth | Avg. GDP growth | Tax legislation |
|---|
Everything above is about the government's books. This section looks at something related but different: how wealth — homes, stocks, businesses, retirement accounts, minus debts — is split across American households, and whether that has anything to do with the debt. The Federal Reserve has tracked this since 1989 — that's a genuine limit of the data, not a rounding choice, so every chart below starts there rather than 1901.
Share of all household net worth held by four brackets: the bottom half, the next 40% (50th–90th percentile), the next 9% (90th–99th), and the top 1% alone — the Fed's own published percentile bands. There's no continuous public series for a narrower cut like "bottom 10%", so this uses real reported brackets rather than inventing one; splitting out the top 1% separately shows concentration more clearly than a single "top 10%" line does.
This chart is placed side-by-side with the wealth-share chart on the left on purpose: it's the same federal debt (as % of GDP) chart shown further up the page, re-plotted over the same 1989–present window so the two trends can be eyeballed together. The "is that relevant here?" answer: it's a visual setup for the correlation figure just below — both wealth concentration and debt-to-GDP have climbed over this period, and this pairing is what that number is measuring. It does not by itself prove one caused the other.
Loading…
Same "is that relevant here?" treatment as the wealth-vs-debt chart above, for a different pairing entirely: national defense spending (BEA's national-defense consumption-and-investment measure, the standard proxy for defense outlays, as % of GDP) against this page's own federal deficit (as % of GDP), 1947–present — as far back as the defense series goes.
Loading…
A simple, transparent, illustrative back-of-envelope calculation — not a policy costing — comparing what a wealth tax could plausibly raise to the actual federal deficit.
Loading…
A distributional snapshot, closing part of the gap noted below: everything else on this page is about aggregate dollars, not who received them. This uses the Tax Policy Center's own published breakdown of the single largest, most-discussed tax law on this page — the 2017 Tax Cuts and Jobs Act — by income group. It's a one-bill, one-year hand-curated snapshot, not a continuous series, for the same reason nothing else like it appears elsewhere on this page: CBO/JCT/TPC publish these periodically per major bill, not as an ongoing feed like FRED.
Loading…
Same treatment, from the spending side: an extreme, not-actually-proposed ceiling case, versus specific options the Congressional Budget Office has actually scored in its "Options for Reducing the Deficit" report.
Loading…
This is a rough, clearly-labeled illustration, not a rerun of history — nobody can rerun history, and every estimate like this has wide uncertainty. It's a simple straight-line path built from two real, published, cited figures: the CBO's own estimate of the 2001/2003 ("Bush") tax cuts' cumulative cost, and a 2023 analysis (built on CBO data) of the combined cost of those cuts, their extensions, and the 2017 Tax Cuts and Jobs Act. The shaded band is a rough ±35% margin, shown specifically to keep the uncertainty visible rather than implying false precision.
Loading…
A curated one-time snapshot from OECD and IMF publications — not a live-fetched time series like the rest of this page, because there's no FRED feed for other countries' tax, debt, or tax-rate levels. Same 8-country comparison set (Norway, Canada, UK, Sweden, France, Japan, South Korea, Hungary) as the sibling norway-vs-us-welfare site, on three OECD/IMF metrics: total tax revenue as a share of the economy, government debt as a share of the economy, and the top statutory personal income tax rate. The United States is highlighted in orange in each chart; hover any bar for the exact figure and year (mostly 2024–2025; see the footer for the one exception).
Loading…
Short answer is in the "Quick take" box at the top of the page. Below is the full supporting detail: what the data shows, the confounding factors, and what's still missing.
Every large tax cut in this dataset — 1981, 2001/2003, 2017 — was followed by a revenue dip relative to where revenue was trending, not a revenue surge. Deficits widened after each of them, even during periods of solid GDP growth. The exceptions that get cited (like parts of the 1920s or a strong stretch after 1997) had other things going on at the same time — a booming stock market, a tech bubble, spending restraint, or a different starting tax rate — that make clean attribution to the tax law alone unreliable. The new interest-cost chart above adds a second reason cuts don't pay for themselves even loosely: debt taken on to finance a cut has to be serviced, and that interest bill is now over 15% of federal spending — itself a growing claim on the budget that has nothing to do with the tax code and everything to do with the debt already on the books. Meanwhile the spending-mix chart shows why the deficit problem isn't purely a tax-policy problem either: entitlement ("mandatory") spending is now roughly half the federal budget and grows on autopilot as the population ages, regardless of who's cutting or raising taxes.
Private wealth has grown a lot more concentrated over the same decades this page's debt has been growing: since 1989 (as far back as the Fed's wealth data goes), the top 10% of households' share of total household net worth has climbed roughly 7 percentage points, while the bottom half's share has shrunk to a couple of percent of the total. That trend tracks federal debt (as % of GDP) fairly closely — see the correlation figure above — but the two almost certainly share root causes (low interest rates inflating asset prices, an aging population, financialization) rather than one simply causing the other. And the arithmetic on "just tax the rich" is sobering: even an unrealistically aggressive wealth tax on the entire top 10% would only just cover a recent year's deficit on paper, while a realistic version of an actual proposed wealth tax (Sen. Warren's, per Saez & Zucman's published estimate) covers only a modest fraction of it. Combined with the illustrative counterfactual further up the page — where the 2001/2003 and 2017 tax cuts plausibly account for a meaningful minority of today's debt, not the majority of it — the honest takeaway is that revenue (from any source, including the wealthy) and spending both matter, and neither side of the ledger closes this gap alone.