You have heard this sentence a hundred times, usually right before someone proposes a wealth tax. “We used to tax the rich at 90% - and the economy boomed. The 1950s were the golden age. Bring it back.”
It is the historical proof-text of every soak-the-rich demand. A member of parliament says it. A columnist writes it. A video essay uses it as its climax. And because it has the shape of a fact - a number, a decade, a verdict - it travels without ever being checked.
Let us check it. Not to be contrarian. Because the next UK Budget is coming - the confirmed October 28 Budget, now 72 days out - and Capital Economics has already suggested it could raise £20-25bn by “shifting focus to hikes on capital, wealth and income.” When the wealth tax argument comes, the 90% claim will be its opening exhibit. It deserves a look before it gets a standing ovation.
The claim has one thing right. The number is real. The story built on it is not.
What the claim actually says
The claim is a single sentence with three parts: the rich were taxed at 90%, the economy boomed anyway, and therefore we can do it again.
The first part is true in the narrowest possible sense. In the 1950s United States, the top federal marginal income tax rate was 91% for most of the decade. In the United Kingdom in the mid-1970s, the top rate on earned income was 83% - and with the 15% investment income surcharge, the top marginal rate on investment income reached 98%, the highest permanent rate since the war, applying to incomes over £20,000. The number is not invented.
But notice what the claim smuggles in with the word “marginal.” A marginal rate is the rate on the last dollar above a threshold. It is not the rate on the whole income. It is not even the rate on most of the rich person’s income. It is the rate on the top slice - and only for those who reached the threshold at all.
The 91% bracket did not apply to the whole country, or even to most of the top 1%. It applied above a high income threshold - the equivalent of roughly two million dollars in today’s money - and only after every deduction, exemption, and preference had done its work. Most high earners never touched it. The ones who did spent their energy avoiding it, which is exactly what the tax code invited them to do.
The number that matters: effective, not marginal
Here is where the myth dies. The Tax Foundation worked through the actual tax records of the 1950s, and the result is not the 90% of legend. The top 1% of taxpayers paid an average effective rate of about 42% of income across all federal, state, and local taxes. Federal income tax alone took 16.9% of their income.
Forty-two percent. Not ninety. And here is the comparison that matters for the “bring it back” crowd: the effective rate on the top 1% today is around 36%. The golden age taxed the rich modestly harder than we do now - six percentage points, not double. The same Piketty, Saez, and Zucman data that activists cite for inequality shows effective rates on the top 1% of about 42% in the 1950s versus 36.4% today.
So when someone says “we used to tax them at 90%,” the honest translation is: “we used to put a 91% sticker on the top bracket, and the people in that bracket paid 42% - six points more than they pay now.”
That is a real difference. It is not the difference the claim promises.
What the sticker actually collected
The deeper question is what the high rates collected - and the answer is less than the sticker implies. This is the Laffer lesson, and we do not need theory to see it. The 1950s code was a maze of deductions, exemptions, and preferential treatment for capital gains. Income that could be converted into capital gains was taxed at a far lower rate, and the very wealthy converted a great deal of it. The effective rate the records show - 42% - is what survived the maze. The rest escaped through doors the code itself built.
The UK experience is even starker. The 98% marginal rate on investment income in the mid-1970s was real, and it was accompanied by the tax policy equivalent of an evacuation. The wealthy shifted income into capital gains, into perks, into anything the code taxed more gently. Tax policy scholars who have studied the era note that despite headline rates of 83% and 98%, effective taxation on the wealthy was far lower - reduced by avoidance, perks, deductions, and loopholes. The parliamentary record confirms the rate structure: 83% top rate of income tax plus a 15% investment income surcharge, reaching 98% on the highest investment incomes.
A rate that high does not mainly collect revenue. It mainly rewards whoever can avoid it - and the people with the best accountants are the ones the tax was aimed at. The revenue the Treasury actually received from the top bracket was far below what the 90% story implies. You cannot tax what people successfully move out of reach.
And the modern echo is live right now. In Scotland, the top income tax rate is 48p in the pound, and on dividends the effective marginal rate stacks to 69.5%. That is not 90%, but it is high, and it is already producing the avoidance response - people restructuring, relocating, and shifting how they take income. The mechanism the 1950s demonstrated at 91% is operating at 48p today. The Laffer lesson was not refuted; it was re-exhibited.
Why the myth refuses to die
If the numbers are this clear, why does the claim survive? Three reasons.
First, marginal is easier to quote than effective. “We taxed them at 91%” is one number, one factoid, one punch. “The effective rate was 42%, which is six points above today’s” is a sentence with qualifiers. A sentence with qualifiers does not fit on a protest sign. The Grievance Machine - runs on sentences that fit on signs. The sticker rate is its favorite exhibit because it is the loudest version of the truth.
Second, “we did it before and it worked” is the historical glue of the inequality argument. If the 90s were real and harmless, then the wealth tax is not a leap into the unknown; it is a return to a proven past. That is why the claim is wheeled out every time, from wealth tax videos to California ballot propositions. Strip the myth away and the argument loses its historical anchor - it becomes an argument about the future, where the evidence is far less comfortable for it.
Third, and most honestly, the 1950s really were good for the economy. The postwar boom was real. The mistake is the causal claim - that the boom happened because of the 91% rate. The 1950s also had pent-up demand after a world war, a Europe and Japan that needed rebuilding, an America whose factories were untouched, and the highest rate of unionization in the country’s history. The 91% rate rode along with all of that. It did not cause the boom, and the 1970s - with rates just as high - proved it, with stagnation. High rates and a booming economy coexisted once. So did high rates and a failing economy. The coexistence proves nothing.
An Aside
You might not know, but the current richest person the world, Elon Musk, takes no income at all from Tesla. His wealth is in stocks he owns in his companies - Telsa, SpaceX, equity and stock options in xAI, Neuralink, and The Boring Company. He doesn’t care what you want the top rate of income tax to be.
The honest version
To be fair to the claim’s believers, there is a real point hiding inside the myth. The postwar era was a time of genuinely progressive taxation, and the top 1% did pay more - as a share of their income - than they do now. If your argument is “the very wealthy should carry a somewhat heavier burden than they do today,” the 1950s offer a mild precedent. That is a defensible position, and the six-point difference is its honest evidence.
What is not defensible is the 90% claim. It is a sticker rate presented as a paid rate, a marginal rate presented as an average, and a historical coincidence presented as a causal law. Every wealth tax proposal now circulating - the ones heading toward the October 28 Budget, the £63bn spending plans they would fund - inherits this distortion. The argument deserves better evidence than it has been using.
The deeper principle was settled long ago, and we covered it in Taxes - Who Actually Pays: the person who bears a tax is not always the person the law names. A 91% rate aimed at the rich did not stay on the rich. It moved into avoidance, into restructuring, into the behavior of everyone trying to step around it. The sticker named the rich. The burden, like water, found its own level.
So the next time someone tells you we used to tax them at 90% - and it was fine, and we should do it again - ask two questions. Marginal or effective? And how much did it actually collect?
The first answer is “marginal.” The second is “less than you think.”
The truth is a 42% effective rate and a boom that happened for other reasons. It is a better story than the myth, because it is true - and because it tells you exactly what will happen if the 90% claim is ever tested again. The rich will not pay 90%. They will pay whatever the code, and their accountants, and the exit doors allow. They did then. They will now. The only question is whether we will pretend otherwise.