IT IS NO secret that inequality grew in much of the rich world in the 1980s and 1990s and has remained relatively high since. In 1980 the top 1% of Americans commanded 9% of pre-tax income, rising to 16% by 2022. Over the same period Europe’s top 1% increased their share from 8% to 12%. In fact, the increase in earnings inequality and its corollary, the stagnation of middle-class incomes, has been so noticeable and so widespread that it is often posited as a possible explanation for the rise of populism around the world.
Yet there is a much less noticed, countervailing trend: as pre-tax incomes have become more unequal, tax codes have become more progressive. The effect has more than outweighed the growth in inequality in much of the rich world, and almost kept pace with it even in America. Today’s taxman, it turns out, is less like the Sheriff of Nottingham and more like Robin Hood. He has largely stopped the rich getting richer and the poor getting poorer. But redistribution on such a scale is an inefficient process, which generates lots of economic and political friction. And the further it proceeds, the greater the likelihood that the merry taxmen will run out of stagecoaches to rob the wealthy.
Levelising levies
A simple measure of progressivity involves comparing the distribution of income both before and after tax. By this measure America redistributes about twice as much as in the 1960s (see chart 1). Germany and Japan, the next biggest rich economies, also redistribute a lot more than they used to. So do Britain and Canada. Indeed by our estimate, seven in ten countries have more progressive tax-and-benefit systems than in 1990. The ones that have become less progressive tend to be dysfunctional (Belarus, Eritrea, Haiti) or were exceptionally redistributive to begin with (Norway, Sweden).
Most of the increase in inequality within rich countries associated with the late 20th century occurred in the 1980s. Since then, governments have become much more interested in alleviating disparities. From 1990 to 2023 the average Gini coefficient of pre-tax income, a measure of inequality, in the G7 increased by on average four percentage points. Yet, thanks to more redistribution, the average Gini coefficient of post-tax income across the G7 has fallen slightly.
The vigour of the Robin Hood state varies across countries. In America more redistribution has merely offset rising pre-tax inequality. According to the Congressional Budget Office, for example, 2022—the most recent year for which detailed data are available—was the fourth-most unequal year on record, as measured by the Gini coefficient of taxable income, behind only 2021, 2020 and 2012. But inequality after taxes and transfers was lower than in 2000, 2005-07, 2012, 2014, and 2017-18 and 2021. It was also only a shade higher than in 1986. Even as the share of America’s taxable income going to the top 1% of earners has soared, their share of post-tax income has risen only slightly (see chart 2).
Elsewhere in the G7, Robin Hood is even more energetic. In Britain, France and Japan the top 1% have seen little growth in post-tax income since the 2000s, even as the bottom 50% have benefited from hefty gains. In Britain post-tax inequality, as measured by the Gini coefficient, has fallen by three percentage points since 1990. In France it is down by four points. In Canada the 1% have done terribly in recent years. In Italy both plutocrats and average Joes have done pretty badly. Only in Germany have the rich clearly pulled away.
It is hard to reconcile these trends with the popular history of tax policy. This says that the rise of “neoliberalism” destroyed the post-war settlement, in which marginal tax rates on the very rich were exceptionally high. There is a grain of truth to the story. For years Britain’s top income-tax rate was 83%, and 98% on “unearned income” such as dividends. From 1951 to 1963 America’s top income-tax rate was at least 91%. In Sweden in the 1970s the marginal rate in effect exceeded 100% for some people. And across the rich world, inheritance and estate taxes were high. That sounds more Robin Hoodish than what exists today.
Impostor impost
In those days, however, the rich found ways not to pay. For instance, many billed personal expenses, such as food and transport, to their company, lowering their taxable income. This is why the protagonists of “Mad Men”, a television show about advertising executives in America in the 1960s, spend lavishly on the company card. During the “progressive” 1950s the effective federal income-tax rate on the top 1% of taxpayers (meaning the rate they actually paid) was around 20%, and fell as low as 13.8% in the 1960s.
Today, avoiding tax is much harder. The effective tax rate on high earners has risen markedly. In Britain, for the top 1%, it is now nearly 40%, up from 34% in the 2000s. “Whisper it quietly,” wrote the Institute for Fiscal Studies (IFS), a British think-tank, at the tail-end of the Conservative government in 2024, “but this Tory government has taken a serious chunk out of the incomes of the 1%”. Trends are similar in Canada, where the top 1% now have an effective income-tax rate (including provincial ones) of 39%. In Spain over the past 20 years, the effective income-tax rate levied on the 1% has risen from 30% to 34%. Italy and Japan levy a higher average effective tax rate on high earners (defined as earning 250% of their country’s average wage) than they did a decade ago.
Under Presidents Ronald Reagan, George W. Bush and Donald Trump, America has passed tax reforms that are widely seen as having benefited the rich. Yet the average federal income-tax rate paid by the top 1% today, 27.6%, is not far off the historical high, in 2000, of 28.9%. Far more dramatic has been the fall in corporate taxes. The benefit of such cuts is notoriously difficult to apportion between (rich) shareholders and (poorer) workers. Democrats say it is company-owners who benefit when taxes on companies fall; Republicans say it is workers, because investment and wages rise. Wonks tend to side with the Democrats on the matter, meaning that corporate-tax cuts have acted as a big counter to the Robin Hood state in America.
All the same, a study in 2023 by Thomas Coleman and David Weisbach, both from the University of Chicago, found that America’s tax system “has become more progressive and more redistributive over the last several decades, with much of that change occurring in recent years”. A study by Thomas Piketty of the Paris School of Economics and his colleagues found that the “reduction of inequality implied by redistribution is significant in France and the US and increased throughout the entire 20th century”.
Even the reduction, and in some cases abolition, of inheritance taxes has made less difference to inequality than you might think. Economists argue over the extent to which bequests increase or reduce inequality—sometimes, after all, poor people inherit a lot of money. In addition, at no point in recent history have wealth transfers been sufficiently large to have much impact on income inequality. In 1990, the earliest year for which comparable data is available for the members of the OECD, a club mostly of rich countries, they raised 100 times more in income taxes than in estate, inheritance and gift taxes combined. We estimate that if America had kept the estate tax as high as it was in 1960 (a rate of 77% on inheritances over $10m), the overall tax burden on the 1% today would be 46.0%, not 45.3%.
It is possible that the extremely rich—not the 1%, or even the 0.1%, but the 0.01%—have escaped Robin Hood’s ambush. Vastly wealthy people often earn a large fraction of their income as capital gains, which many governments tax at a lower rate than income. Some rich people borrow against their assets to fund consumption. This “buy, borrow, die” wheeze reduces tax bills because loans are not taxable as income. Others classify their earnings as profits rather than wages to take advantage of lower corporation-tax rates. And those stretching the rules to minimise their tax are especially likely to get away with it at the moment in America, at least, since the current administration has made swingeing cuts to the enforcement division of the Internal Revenue Service.
A paper by Arun Advani of Warwick University and others suggests that in 2018 the very richest British people (those earning many millions of pounds a year) paid around 25% tax when income is measured broadly, compared with about 35% for the merely very wealthy. In other research Akcan Balkir of the University of California, Berkeley, and his colleagues find similarly low effective tax rates for the 400 or so richest households in America. Shigeki Kunieda of Chuo University looks at very wealthy Japanese people, finding that beyond an annual salary of about ¥100m (roughly $650,000-700,000 a year), effective tax rates decline.
Yet the literature on plutocrats’ tax tricks is small and contested. In America “buy, borrow, die” is a minority pursuit, find Edward Fox of the University of Michigan and Zachary Liscow of Yale University in a recent paper. Gerald Auten of the Treasury Department and David Splinter of the Joint Committee on Taxation, a non-partisan arm of Congress, recently published a study in the Journal of Political Economy that analysed the effective tax rates of different parts of the income distribution through a very broad lens that encompasses taxes on wages, profits, capital gains and bequests. In this work and updates to it, they find that the very richest Americans tend to pay the very highest shares of their income to the taxman (see chart 3). Someone in the top 0.01% pays about 50% tax overall, they conclude. In addition, the very richest people face higher effective tax rates today than in the 1960s or 1970s.
This result chimes with another finding in Mr Advani’s paper, which rather than focusing on effective tax rates in 2018 instead calculates an average from 2014 to 2018. Over this longer period, Britons earning many millions of pounds a year paid a similar rate to the merely rich. Their apparent windfall in 2018 stemmed from the difficulty of reflecting the complicated tax system for those selling business assets.
In other work Mr Splinter makes a number of adjustments to Mr Balkir’s methodology. He cites research that finds that the wealth of the “top 400” earners is actually distributed among more than 2,000 adults, since the extremely rich tend to support several family members. Analyses that do not capture the tax paid by these relatives and instead look at only the main beneficiary of a family fortune tend to yield unrealistically low effective tax rates. Mr Splinter finds that in America the rate for the 400-odd richest taxpayers is in fact closer to 40%.
Paying their duties
Whether the rich are paying a reasonable rate or not, today’s combination of higher pre-tax inequality and higher tax rates on top earners has left governments astonishingly dependent on the wealthiest for their revenue. Britain’s 1% account for nearly 27% of the income-tax take, up from 22% in 2000. In Australia the 1% account for 17-20% of income tax, up from 16% in the early 2010s. A recent official report in South Korea created a stir when it revealed that the top 1% of taxpayers account for 40% of income tax, up slightly in recent years. In America, too, the 1% paid 40% of income tax in 2022, up from 33% in 2001.
Governments are using the rich’s riches to do a number of things. One is to cut taxes on everyone else. “The average earner in the UK now has the lowest effective personal tax rate since 1975,” boasted Jeremy Hunt, Britain’s then-chancellor (finance minister), in 2024. The trends are similar elsewhere. The tax burden on America’s bottom, second, third and fourth quintiles—80% of the population—is far lower today than it was in the 1960s or 1970s. In particular, the expansion of the earned-income tax credit for low earners in the 1990s made the system more progressive.
At the same time, benefit payments to low- and middle-earners have boomed. Across the rich world social transfers now account for about 22% of GDP, up from 17% of GDP in the mid-1990s (see chart 4). A lot of this is pensions. State support for the elderly is more generous than it was, even as the number of old people has increased. In the EU, for instance, the median state pension is now worth 60% of the earnings of a 50-something, up from 55% in 2010 (when the data begin).
Payments for working-age folk have grown too. More than half of working-age people in Britain now receive more in state spending (defined to include such perks as health care from the National Health Service) than they pay in tax, up from a third in the 1990s, according to the official statistics office. In America subsidies for health insurance were expanded during reforms under Presidents Clinton, Obama and Biden. The subsidies offered by the Affordable Care Act were greatly expanded during the pandemic. Although that expansion, always intended to be temporary, has lapsed, Democrats last year forced a government shutdown in a failed attempt to extend it. They are likely to try again, in keeping with Milton Friedman’s dictum, “Nothing is so permanent as a temporary government programme.” These days an American in the bottom income quintile receives means-tested benefits worth 100% of their earned income, up from 50% in the late 1970s.
Is the tax system now too progressive? As the pre-tax incomes of the rich grew, they were always going to pay a higher share of overall taxes. Keeping post-tax inequality steady—assuming this is a legitimate goal—typically also requires governments to increase the effective tax rates the rich pay, too.
Plenty of politicians are campaigning to raise taxes on the rich yet higher. Zohran Mamdani, New York’s new mayor, wants a 2% additional tax on those earning more than $1m a year. Rhode Island and Washington state are considering a similar tax. Britain recently changed the rules for wealthy foreigners, meaning that many now have to pay British tax on their worldwide income. Last year a furious public debate ensued in France about a proposed wealth tax. And the citizens of California may soon vote on the introduction of a “one-off” 5% tax on the assets of billionaires resident in the state. Squillionaires such as Sergey Brin and Peter Thiel have moved business out of the Golden State in protest. On September 7th opponents held a “march for billionaires” in San Francisco to oppose the tax. “Vilifying billionaires is popular. Losing them is expensive,” the organisers said.
The big question is whether, in the pursuit of equality, governments are raising taxes on the rich so high that they are becoming counter-productive. At some point swingeing levies will start to discourage work and so harm the economy. Some see the exodus of billionaires from California as proof that it has gone too far. Others argue the same of the small number of wealthy foreigners who have left Britain in response to its tax changes.
Yet in many places there are few signs of high earners slacking, even as effective tax rates have risen. The working hours of high-earning professionals have grown in recent decades, not shrunk. Even with an effective tax rate of 50% or so, millionaires and centimillionaires in California seem to be working harder than ever. Some rich people, from Mark Cuban (an American businessman) and J.K. Rowling (a British author) even take a grim pleasure in how much tax they pay.
Some economists therefore argue that rich countries should tax the rich even more heavily. In a world where pre-tax inequality rises, perhaps as AI takes over the jobs of humbler workers, the rich will have to shoulder an even bigger share of the burden. A paper by Mr Piketty, Emmanuel Saez of the University of California, Berkeley, and Stefanie Stantcheva of Harvard University concludes that the optimal tax rate on the rich could be over 80%. Are Mr Piketty and his band of merry men correct? On current trends, the world may soon find out.
