Dossier mode
The 70-Year Collapse of the Top Tax Rate
The same investigation, restaged one beat at a time. Drive it with the arrow keys, space, or autoplay. Nothing is cut from the piece — long runs are split across frames. Read the full investigation or open the The Austerity Myth hub.
The Ceiling Came Down.
The top tax rate was 91 percent through the Eisenhower boom. Today it is 37. A seventy-year project, pursued by both parties, sold every time as the key to growth — here is what it actually bought, and who it bought it for.
The statutory top rate was deliberately cut by more than half over seventy years; each cut was sold on a growth promise the record does not support; revenue barely moved as a share of GDP, so the cuts changed who pays and how after-tax income concentrated at the top — not the government's topline take.
The rate history is FACT; the effective-rate fence (nobody paid 91% on all their income) is FACT; the inequality link is a fenced correlation; the capital-gains shelter and the flat ~17%-of-GDP revenue line are the honest limits that keep both the left and right versions of this story accurate. The long history behind The Interest Trap.
The statutory top rate fell from 94% (1944) to 37% today — cut by more than half.
The documented spine. The top marginal rate peaked at 94 percent in 1944–45, stayed above 90 percent through 1963 (91 percent for most of the Eisenhower years), then stepped down: 70 percent after the 1964 Kennedy–Johnson cut, 50 percent under Reagan in 1982, a low of 28 percent in 1988, back to 39.6 percent under Clinton, and 37 percent today. Source of record: IRS Statistics of Income, compiled by the Tax Policy Center.
The 91% was never what the rich actually paid — effective rates were far lower.
The 91 percent applied only to income above a very high threshold, amid a code riddled with shelters. So the top 1 percent's effective federal income-tax rate in the 1950s was near 17 percent (Tax Foundation), while the effective rate on income in the top brackets ran about 49 percent in 1958–81 versus ~35 percent today (Tax Policy Center). Both true at once: almost no one paid near 91 percent on their whole income, and the rich still paid materially more than now. We won't quote 91 percent as what they handed over.
Federal revenue held near 17% of GDP through every one of these rate regimes.
Whether the top rate was 91 percent or 28, total federal revenue hovered around 17 percent of GDP for the whole postwar period. That cuts both ways: the right is wrong that high rates would strangle the economy's taxable capacity, and the left is wrong that simply restoring a 91 percent headline would flood the Treasury — the money was always sheltered out of that bracket. The realistic revenue lever is the base and the treatment of capital, and reversing the Bush/Trump cuts — not the headline rate.
Every major cut was sold as self-financing growth — and the dividend didn't show up.
From Kennedy's advisers to Reagan's supply-siders to the 2017 authors, the pitch was constant: lower the top rate and growth will replace the lost revenue. The record doesn't support the strong version — the economy grew fastest and most broadly in the high-rate postwar decades, the deepest cuts were followed by widening deficits, and JCT/CBO repeatedly found the cuts didn't pay for themselves. PROBABLY TRUE, not FACT, because growth has many causes — but seventy years in, the promise is unmet.
As the top rate collapsed, the top 1%'s share of income more than doubled.
Over roughly the decades the top rate fell from 70 percent to the 30s, the top 1 percent's share of national income more than doubled (Piketty/Saez; confirmed in CBO's distributional work). Graded a correlation, not a proven cause — globalization, technology, financialization, union decline, and executive-pay norms all pushed the same way. But a lower top rate is not a small force among them: when the penalty for taking enormous pay falls, enormous pay becomes more worth taking.
The real shelter is how the rich get paid: a wage taxed at 37%, a capital gain at about 20%.
The top rate on wages is 37 percent; on long-term capital gains it's 20 percent plus a 3.8 percent net investment income tax — well below the wage rate. And gains are overwhelmingly top-end: an IRS SOI analysis finds about 75.7 percent of capital gains flow to the top, and CBO notes gains are 'generally taxed at lower rates than other income.' The people with the most income increasingly receive it in the form taxed most lightly — so the headline wage rate barely touches them. The gap between the wage rate and the gains rate is the door the money walks through.
'Starve the beast': cutting revenue was, for part of the movement, the strategy — not a side effect.
For an influential strand of the conservative movement the slide was a deliberate fiscal strategy, stated out loud: cut taxes first, let deficits build, then use the deficits to force spending cuts. FACT that the strategy was openly articulated and pursued ('starve the beast' was its advocates' own phrase); whether a given cut was motivated by it is case-by-case. It reframes the history — sometimes the point was precisely that the cuts wouldn't pay for themselves, so the debt could force cuts that couldn't win on their own.
The line we hold.
- FACT: statutory top rate 94% (1944) -> 91% (Eisenhower) -> 70% (1964) -> 50% (1982) -> 28% low (1988) -> 39.6% (Clinton) -> 37% today; effective rates always far below the headline (top 1% ~17% income tax in the '50s; ~49% on top-bracket income vs ~35% now); federal revenue held near 17% of GDP across every regime.
- FACT: the top rate on wages is 37% while long-term capital gains top out near 20% (+3.8% NIIT), and ~75.7% of capital gains flow to the top — the shelter is in how the rich get paid, not the headline rate.
- PROBABLY TRUE (fenced correlation): as the top rate fell, the top 1%'s income share more than doubled; and the promised self-financing growth dividend did not arrive. Many forces drive both; the tax cut is one, not the only one.
- FACT: 'starve the beast' was an openly stated strategy — cut revenue to force spending cuts later; the unoffset cuts seeded the debt whose interest bill now argues for austerity.
Why it matters.
If federal revenue held near 17 percent of GDP whether the top rate was 91 percent or 28, then seventy years of cutting it did not, on the whole, starve the government of revenue as a share of the economy — it redistributed, upward. It let after-tax income at the very top rise far faster than everyone else's, shifted the system toward lightly taxed capital and away from wages, and, where the cuts weren't paid for, converted public revenue into private wealth financed by public debt. The 91-percent era built a middle class; the 37-percent era built a billionaire class. That isn't a slogan — it's what the income-share data show, and it's the thing the headline-rate fight is structured not to notice. The debt the unoffset cuts created is now used, through its interest bill, as the argument for cutting everything else.
Help us fill these lines.
- OpenA clean decomposition of how much of the top 1% income-share rise is attributable to tax policy versus globalization, technology, financialization, and union decline.
- OpenA full menu, scored, of raising the effective rate on the top via the base and capital treatment rather than the headline rate — and what each piece raises.
- OpenThe documentary record of 'starve the beast' as applied to specific cuts: which were understood by their authors as deliberately unaffordable.