A century of top income tax rates, from 7 percent in 1913 to the 2018 law

Imagine telling a taxpayer in 1913 that within thirty years Washington would claim ninety-four cents of the last dollar a rich man earned. He would have assumed you were describing a foreign despotism, and a slightly theatrical one. Yet that is the road the American income tax took, and it is a road with plenty of turns, reversals and detours.

The top federal income tax rate is the most argued-over number in American politics. It is also one of the least understood, because the debate usually treats it as a measure of fairness when it is really a measure of what the government wanted to do and what the country was willing to tolerate. Here is the record, in order, with a line on why each change happened and how few people the highest rates actually reached. I will argue at the end what the pattern teaches.

First, two terms

A top marginal rate is the rate on the last dollar of income in the highest bracket. It is not what a rich person pays on all of his income. If the top rate is 37 percent and begins above some threshold, only income above that threshold is taxed at 37. The rest is taxed at lower rates.

An effective rate is total tax divided by total income. It is almost always lower than the top marginal rate, and in the years of the highest marginal rates it was much lower, because the code was full of ways to move income out of the top bracket. Keep both in mind, because the timeline below uses the first and the argument depends on the second.

The timeline

1913: 7 percent

The Sixteenth Amendment was ratified in February 1913, and Congress passed the first modern income tax that October, as part of the Underwood-Simmons tariff act. It had a normal rate of 1 percent and surtaxes that climbed, so that the top combined rate reached 7 percent on income above $500,000. The tax reached well under one percent of Americans. For a working family it was a rumour, and for a clerk it was a story about the rich. The detailed story of that bargain is in The Sixteenth Amendment and the 1913 bargain that built the income tax.

1918: 77 percent

The First World War changed everything. The Revenue Act of 1918 pushed the top rate to 77 percent, because the government needed money for a war on a scale it had not imagined. The tax was framed as temporary and as a duty of wartime citizenship. It was a trial run for the argument that would return in the next war: the nation is in danger, therefore the government may take what it needs.

1925: 25 percent

After the war, rates came down. Andrew Mellon, the Treasury Secretary through most of the 1920s, argued that high rates drove capital into tax-exempt securities, and that lower rates would draw it back out and into the taxable economy. By 1925 the top rate stood at 25 percent. Economists still argue about how much of the revenue growth in the 1920s came from the cuts and how much from the boom, but the incentive logic was clear, and it echoes the discussion in where tax cuts pay for themselves and where they honestly do not.

1932: 63 percent

The Great Depression gutted revenue. The Revenue Act of 1932, signed by Herbert Hoover, raised rates sharply across the board, with the top rate going to 63 percent. Think about that for a moment: in the worst contraction in the country's history, a Republican president signed a large tax increase because the budget had to be balanced. The conventional wisdom in 1932 was that a balanced budget calmed confidence. Many economists since have judged the increase to have been a mistake in the middle of a collapse. For readers who want the longer debate, see Six myths about the 1929 crash and the New Deal recovery, tested.

1944: 94 percent

The Second World War produced the highest number in the history of the American income tax. By 1944 the top marginal rate was 94 percent on the very highest slice of income, and the base of the tax had been widened to reach millions of ordinary workers for the first time, with withholding in place from 1943. This is the most important change on the list, and it is not the 94 percent. It is that the income tax stopped being a tax on the few and became a tax on the many. The 94 percent was the headline; the withholding was the revolution. I cover that part in withholding began as a wartime emergency in 1943 and simply never left.

1964 and 1965: 77, then 70 percent

After the war, the top rate stayed above 90 percent for nearly twenty years. In the Eisenhower years it was 91 percent. President Kennedy proposed cutting rates in 1963, arguing that a tax system designed to prevent a depression was now strangling a boom, and that high rates on the top slice raised little revenue while discouraging work and investment. The Revenue Act of 1964, signed by Lyndon Johnson in February of that year after Kennedy's death, took the top rate from 91 to 77 percent, and then to 70 percent in 1965. A Democratic president was making a supply-side argument in the language of the day.

1982: 50 percent

The Economic Recovery Tax Act of 1981 cut individual rates across the board, over three years, and the top rate came down from 70 to 50 percent, effective in 1982. The 1970s had produced inflation that pushed ordinary families into higher brackets without raising their real income, a quiet tax increase known as bracket creep, and the 1981 law also indexed brackets to inflation starting in 1985. The indexing may be the single most durable reform in the entire record, and it receives far less attention than the rate cuts. A later 1982 law reversed part of the revenue loss, and I would not pretend the deficits of the decade came from nowhere.

1988: 28 percent

The Tax Reform Act of 1986 was a bipartisan deal that did something almost unimaginable by modern standards: it lowered the top rate and broadened the base at once, closing a long list of shelters and preferences in exchange for lower rates. The top rate fell to 28 percent in 1988. A few years earlier a top rate of 28 percent would have been science fiction. The lesson, which conservatives should keep in front of them, is that a low rate and a clean base go together.

1993: 39.6 percent

The pattern broke the other way. Bill Clinton's 1993 budget raised the top rate to 39.6 percent, a figure that came back in 2013. In between there was a 1990 deal, under George H. W. Bush, which had raised the top rate to 31 percent after the president had promised not to raise taxes. That broken promise cost him dearly, and every Republican since has remembered it. The 1993 increase passed without a single Republican vote in either chamber.

2003: 35 percent

George W. Bush signed a 2001 tax cut that phased in lower rates and a 2003 law that accelerated them. The top rate came down to 35 percent, and the top rates on dividends and capital gains fell to 15 percent. These laws were written to expire, under the budget rules of the time, and the question of their expiration dominated tax politics for a decade.

2013: 39.6 percent

The American Taxpayer Relief Act, passed in the first days of January 2013, made most of the Bush-era rates permanent but restored the 39.6 percent top rate for single filers with taxable income above $400,000, and a higher threshold for married couples. It is worth noting how the 2013 law drew the line: the top rate applies only to income above the threshold, not to the first dollars of anyone's income.

2018: 37 percent

The Tax Cuts and Jobs Act, signed in December 2017 and effective for 2018, kept seven brackets and lowered the top rate to 37 percent, along with the corporate rate cut to 21 percent that I cover in the corporate rate cut to 21 percent and the inversions it brought to an end. It also nearly doubled the standard deduction. The individual rate cuts were written with an expiration date, as the 2001 and 2003 cuts had been, which brings the story back to the question that has been asked since 1913: for how long, and at what rate.

The steelman: "the rich paid more when growth was strong"

The most common argument against cutting the top rate points to the 1950s and early 1960s, when the top marginal rate was above 90 percent and the economy grew strongly. The argument goes: One, the top rate was above 90 percent for most of two decades. Two, the postwar boom was the strongest period of American middle-class growth. Three, therefore a very high top rate is compatible with growth, and may even have contributed to it by funding roads, schools and research.

That is a fair statement of the best case. I want to answer it in numbered points.

One: the rate that was on the books is not the rate that was paid. A top marginal rate of 91 or 94 percent applied only above a very high threshold, and the code was full of devices that let high earners convert ordinary income into lower-taxed forms: capital gains, deductions, tax-exempt bonds, income deferred into retirement or into corporate form. Economists who study the period find that the share of income reported by the top group and the share of federal revenue they paid was far lower than the rate suggests.

Two: the revenue came from the middle. Even in the years of the highest rates, the bulk of the individual income tax came from the broad middle of the income distribution, because that is where most of the income is. The top bracket raised comparatively little. That is the point: a 94 percent rate on a small base is a symbol more than a revenue source.

Three: the postwar boom had causes that had nothing to do with the rate. The United States emerged from the war with its industry intact while Europe and Japan rebuilt. The country had little foreign competition and enormous pent-up demand. Strong growth under high rates is a coincidence of circumstance, not a proof of method. When the rates came down in the 1960s and the 1980s, growth was also strong, which cuts against any theory that the high rates were the cause.

Four: the best evidence is what happened when the rates changed. The 1986 reform lowered the top rate to 28 percent and broadened the base, and the share of income taxes paid by the highest earners rose afterwards, as shelters that had been worth using at 50 percent were not worth the bother at 28. The record points to a general conclusion: what a country collects from the rich depends far more on the structure of the code than on the headline rate.

What the timeline teaches

If you read the dates against the history, a pattern emerges. The top rate tracks emergencies. It rose in the First World War, in the Depression and in the Second World War, and when the emergency ended the rate came down, though never all the way back to where it started. That is a lesson about government: emergency powers and emergency taxes are easier to adopt than to retire.

A second pattern is that the people who paid the highest rates were always few. In 1913 the tax reached well under one percent of the country. Even at its peak the 94 percent bracket reached only a tiny sliver of earners at the very top of the income scale. A rate that applies to almost no one is easy to vote for, which is exactly why it is dangerous: voters who will never pay it can raise it without feeling it, and the real bill lands on a growing share of the middle class later, as it did in 1943. The history of the tax is, in miniature, the history of a tax on the few that became a tax on the many.

A third pattern is that the best reforms combined lower rates with a broader base. The 1986 law did this, and the 1964 law did the first half. When rates are high, taxpayers spend their ingenuity avoiding the tax, and lawyers and lobbyists are the winners. When rates are lower and the base is clean, the tax matters less to decisions. For the claim that fairness is measured by the top rate, there is no better evidence than a look at Who actually pays the federal income tax, read straight from the IRS tables, which shows who in fact carries the load.

The Founders' measure

The Constitution had no income tax at all until 1913. Article I, Section 9 said that "No Capitation, or other direct, Tax shall be laid, unless in Proportion to the Census or Enumeration herein before directed to be taken," and in 1895 the Supreme Court held in Pollock v. Farmers' Loan and Trust Company that a tax on income from property was a direct tax that had to be apportioned. The Sixteenth Amendment overrode that holding: "The Congress shall have power to lay and collect taxes on incomes, from whatever source derived, without apportionment among the several States, and without regard to any census or enumeration."

Notice what the amendment does and does not say. It grants a power. It does not set a rate, and it does not say that the rate must be high or progressive. The entire century of swings, from 7 to 94 and back to 37, happened in the space the amendment left open. That space is the people's to fill, through their elected representatives, and the Founders' tradition says the first duty of a representative is to ask what a tax is for before asking how much it can raise. You can read the amendment itself on the Constitution page.

The case in brief

1. The top federal income tax rate went from 7 percent in 1913 to 77 in 1918, 25 in 1925, 63 in 1932, 94 in 1944, 70 by 1965, 50 in 1982, 28 in 1988, 39.6 in 1993, 35 in 2003, 39.6 in 2013 and 37 in 2018.
2. The rate has risen with emergencies and fallen with peace and with reform, and it has never fully reverted to what it was before an emergency.
3. The highest rates reached very few people; the real expansion of the tax came with the wartime widening of the base and withholding.
4. The best reforms, in 1964 and 1986, cut the top rate and cleaned the base together.
5. Fairness is better measured by who pays and what a family keeps than by the number on the top bracket.

What you can do: look up the current federal tax brackets in the IRS's published tables, find the threshold where the top rate begins, and ask your representative one question about what the top rate is meant to pay for.