How to tax the rich, explained
It’s no accident that the Democratic Party went from wanting a 39.6 percent top tax rate to wanting much more.
In just the past few months, at least three major Democratic Party figures, two of whom are presidential contenders, have proposed large tax increases targeted at the richest Americans:
Rep. Alexandria Ocasio-Cortez (D-NY) floated a big increase in top marginal income tax rates in an interview on 60 Minutes. Sen. Elizabeth Warren (D-MA) proposed an annual wealth tax. Sen. Bernie Sanders. (I-VT) proposed a drastic hike in the estate tax.
The Democratic urge to tax the rich isn’t a wholly new development. In the 2016 general election, Hillary Clinton embraced Sanders’s proposal for a 65 percent top estate tax rate, and both Bill Clinton and Barack Obama were able to raise the top income tax rate by a few points.
But Sanders, Warren, and Ocasio-Cortez’s proposals are significantly more ambitious; Warren, for instance, is proposing a kind of tax the federal government has never imposed before, breaking totally new ground.
Right now these are somewhat disjointed proposals that don’t add up to a coherent picture of how the tax system ought to look; Ocasio-Cortez’s, especially, was dashed off in the course of an interview (though it got debated as if it were a carefully constructed white paper).
But over the 2020 primary and in the new majority-Democratic House, we should expect these ideas to start to harden into a broader agenda for dramatically raising rates on top earners.
At the end of the process, the Democratic Party is likely to wind up with a much more aggressive party consensus on taxes than has prevailed since the 1980s.
This isn’t a change that came out of nowhere. It comes out of a growing dissatisfaction with the timidity of Democratic efforts to tax the rich in the Clinton and Obama eras, and a growing intellectual trend in economics that has given a sound theoretical basis to tax plans that soak the rich.
But it’s not a trend that’s necessarily going to be easy to translate into policy. There are some important questions to sort out: What tax rate should we apply to the richest to maximize revenue? What rate is so high that it begins to discourage work — and the government gains less revenue? How feasible is a tax on wealth (as opposed to income), and what obstacles stand in the way? These and many other details need to be reckoned with as Democrats begin to think about turning campaigning into real policy.
The Democratic proposals to tax the rich
Before we get into it, let’s recap the basic plans — or loosely sketched ideas — for taxing the rich currently on offer from prominent Democrats:
- Rep. Alexandria Ocasio-Cortez (D-NY) suggested a top rate of 70 percent on incomes over $10 million, which would actually be somewhat below prevailing rates under Dwight Eisenhower, and around where top rates were from Lyndon Johnson to Jimmy Carter.
- Warren’s plan would set up a progressive wealth tax, with a normal rate of 2 percent on wealth over $50 million and a top rate of 3 percent on wealth over $1 billion. That might sound small, but because it’s levied on wealth, not income, and every year rather than at death, it could wind up hitting billionaires harder than high income or estate taxes.
- Sanders’s plan would raise the top estate tax rate to 77 percent. That’s the same top rate that existed from 1941 to 1976; this is higher than the 65 percent top rate he proposed in the 2016 race.
- Sen. Cory Booker (D-NJ) has proposed a 65 percent estate tax rate (higher than at any point since 1982), a higher capital gains tax rate, and applying capital gains taxes to assets held at death, all to pay for his baby bonds bill.
- Rep. Jan Schakowsky (D-IL), a lefty House Democrat who’s been proposing a top income tax rate of 49 percent since at least 2011, says she’s working with Ocasio-Cortez to formulate a new bill that might feature even higher top rates.
- And if Sanders excitedly releasing his estate tax plan after Warren’s wealth tax proposal is any indication, the 2020 race will involve a leftward arms race as candidates attempting to court voters worried about income inequality try to one-up each other’s plans to tax the rich.
A recent history of taxing the rich
High tax rates targeting the rich used to be the norm in the United States. As recently as 1963, the top income tax rate was 91 percent, a rate that persisted, plus or minus a point or two, from World War II through Truman and Eisenhower and the early Kennedy years.
A mere quarter-century later in 1988, the top rate had toppled to 28 percent. How’d they get there?
Both parties played a role. First, under across-the-board tax cuts backed by John F. Kennedy and signed by Lyndon B. Johnson in 1964, the top rate fell to 70 percent. Then Ronald Reagan’s 1981 tax cuts slashed top rates for both the income and estate tax — the former down to 50 percent, the latter to 55 percent by 1984. Finally, a bipartisan coalition in Congress, along with Reagan, put together the 1986 tax reforms, which cut back on a number of major income tax deductions but also took the top rate down to 28 percent.
A desire to slash deficits led George H.W. Bush to conclude that a 28 percent top rate was infeasibly low, prompting him to boost it to 31 percent when he famously violated his “no new taxes” pledge in 1990. Shortly after taking office in 1993, Bill Clinton added 36 percent and 39.6 percent brackets.
Fights over the top income tax rate ever since have taken place in a very narrow band between 35 and 39.6 percent. George W. Bush cut it to 35 in 2001; Obama forced House Republicans to agree to restore it to 39.6 percent in late 2012; Donald Trump and his congressional allies brought it back down to 37 percent in 2017.
The Affordable Care Act did include a small 0.9 percent surtax on wages for wealthy individuals, which put their effective top rate (including the existing 2.9 percent Medicare payroll tax) at 43.4 percent by the time Obama left office; Trump’s tax cuts took that number down to 40.8 percent.
The estate tax has seen a similar trajectory, stagnating at 55 percent between the first Reagan tax cuts and George W. Bush. The 2001 tax cuts then set it on course to be totally abolished — and it was for one year, 2010, a truly great year for rich people to die in — only to see it resurrected with lower top rates (35 percent and 40 percent) after the 2010 and 2012 budget battles.
Obama proposed other measures to increase taxation on top earners that Congress never gave a hearing, but they were modest compared to the ideas that Sanders, Ocasio-Cortez, and Warren are floating now. His two principal ideas were the “Buffett Rule” — a minimum tax rate of 30 percent for people with $1 million-plus in income, inspired by Warren Buffett — and a cap on itemized deductions and exclusions. The Buffett Rule would’ve dramatically increased the tax rate on investment income, but only to the mid-30s range. Neither plan would’ve raised the top rate on wage income to anything like its historic highs.
Obama certainly wanted to increase taxes on the rich more than he did. But even his most ambitious proposals were not really scaled to substantially cut inequality or raise a huge amount of revenue. The new generation of Democratic proposals are.
The intellectual basis for taxing the rich
So what accounts for the new aggressive support for taxing the rich among leading Democrats? There’s the Great Recession, in which bankers were bailed out as high unemployment persisted for years; and the Occupy movement, which made economic inequality a core national issue in 2011 and onward.
But an important intellectual precondition was the work of a handful of French economists focusing on distributional issues. Berkeley’s Emmanuel Saez and the Paris School of Economics’ Thomas Piketty have been using tax records to analyze income inequality for decades now, but a handful of publications by them and collaborators in recent years made high marginal rates intellectually respectable.
Is 73 percent the magic number?
One paper in particular looms large here. In 2012, Saez and Peter Diamond, an MIT professor who had recently won the Nobel in economics, released “The Case for a Progressive Tax,” which estimated that the optimal top tax rate in the US — the rate that maximizes both revenue and the welfare of Americans — is about 73 percent. That’s much higher than the top rate in the US, even taking into account state and local taxes in high-tax places like California and New York.
Saez and Diamond’s calculation relied on the idea of diminishing marginal utility — rich individuals benefit less from additional money than low- or middle-income people do — and on their estimation that top earners aren’t very responsive to top tax rates.
The 73 percent figure relies on a “mid-range estimate” of the elasticity of income in response to taxation for the rich that is quite low (0.25). Elasticity measures how much one factor (in this case, the income earned by the rich) changes in response to another (in this case, the tax rate). In this instance, low elasticity means raising taxes on the rich would cause them to work less and hurt economic growth — but not by that much, and that effect is easily outweighed by the benefits of higher tax revenue spent to benefit the non-rich through welfare state programs.
Economists who dispute the Saez-Diamond result tend to argue that 0.25 is too low of an elasticity estimate, and that a top rate in the realm of 73 percent would significantly reduce the rich’s work effort. A recent paper, for instance, argued that the revenue-maximizing rate is more like 49 percent, when you take into account the prospect that high tax rates could discourage young people from developing economically useful skills.
Capital: to tax or not to tax
Saez and Diamond also argued that capital income — income from things like capital gains, corporate profits, dividends, etc. — should be taxed, which broke with previous models of optimal tax theory. (Our current capital gains top rate is 23.8 percent.) Those models had suggested the proper tax rate on capital income was zero, on the grounds that it discouraged savings: If you spend money on an investment, your profits are taxed, but if you spend money on food or a house or what have you, you don’t get hit with a capital tax — so a capital tax’s presence pushes you to spend more and save less.
But Saez and Diamond dispute whether these model findings are relevant to lived reality. If you’re a lawyer with your own practice, are your billable hours capital income (profits accruing to your law firm) or wage income (earnings accruing to you)? There’s ambiguity there, and taxing capital income less than labor income can thus lead to gaming. The rich have a strong incentive to reclassify labor income as capital income — the way that hedge fund and private equity managers do already using the carried interest loophole — which suggests we should tax capital income to prevent this kind of avoidance.
This argument is important for estate taxes as well, as they are also a kind of capital tax. A 2013 paper by Piketty and Saez made this argument explicitly, arguing that the optimal inheritance tax rate was something like 50 to 60 percent, and possibly even higher for the very rich.
Piketty and Saez’s inheritance paper also notes that what tax rate is “optimal” depends a lot on your moral philosophy. Most optimal tax papers tend to assume a kind of crude utilitarianism: Taxes should be used to maximize the public welfare, roughly estimated, so you should only take money from the rich so long as doing so boosts overall welfare.
But maybe we don’t care as much about overall welfare but instead the welfare of the worst-off. Piketty and Saez show that the optimal inheritance tax is much higher if you use a “Rawlsian” social welfare function, where the goal is for taxes to maximize the interests of the worst-off, not of everyone.
Piketty also, of course, promoted a specific kind of capital tax in his best-seller Capital in the 21st Century: a tax on accumulated wealth, with top rates of 2 percent or more. It’s no coincidence that Saez and his Berkeley colleague Gabriel Zucman, who both work closely with Piketty, advised Warren on her proposal for a wealth tax topping out at 3 percent. In the context of Piketty’s book, the wealth tax is meant to prevent the permanent intergenerational accumulation of wealth, which he argues will be inevitable due to the rate of interest on capital exceeding the rate of economic growth (his famous r > g claim).
Wage bargaining and the “third elasticity”
The last key component of this literature is a paper by Piketty, Saez, and Harvard professor Stefanie Stantcheva, which introduced the idea of a “third elasticity.”
The first elasticity in their model is the one described above, representing the effect of top tax rates on the work effort of top earners. Maybe Apple CEO Tim Cook would be less likely to pursue new projects in search of a big bonus if he thought that 70 to 80 percent of that bonus would be confiscated. They argue this elasticity is fairly small, in the 0.2 to 0.25 range, in keeping with Saez’s article with Diamond.
The second elasticity is the elasticity of avoidance: how much tax dodging increases as tax rates go up. Piketty, Saez, and Stantcheva argue that this elasticity is even smaller.
The third elasticity is where the real action is. High tax rates, the authors note, do not merely reduce work effort and increase tax avoidance; they also decrease wage bargaining. When people like Cook’s tax rate is low, they have a strong incentive (as do other top Apple executives) to fight for increased compensation for themselves where possible. But when top tax rates are high, they’re less likely to fight as hard, and more likely to use that money to pay workers in lower tax brackets, invest, etc.
Bargaining effects like these, Piketty, Saez, and Stantcheva argue, are bigger than the effects on actual work effort. By reducing bargaining like this, high income tax rates actually perform a social good. That implies the top tax rate could be very high indeed — their preferred parameters result in a top rate of 83 percent.
All of this adds up to a comprehensive intellectual case for the proposals that Ocasio-Cortez, Sanders, and Warren are floating: much higher income and estate tax rates, and perhaps an annual wealth tax as well.
The ideas of the French school of inequality economics are controversial within the profession. Plenty of tax economists on the center and right tend to believe top rates are worse for growth than Saez et al. conclude, and that taxes on investment are especially costly. In particular, many doubt that there’s a solid empirical basis for the idea of a third elasticity.
But while top rate skepticism was once the dominant view within economics, the French school has shown that it’s possible to defend much, much higher rates, including on capital income, using fairly standard methods and empirical evidence.
Why taxing the rich can be hard in practice
So let’s say you buy the French school evidence. Time to all get on board higher income, estate, and wealth taxes, right? Maybe — but there are some practical complications worth sorting out first.
The US obviously knows how to levy very high taxes on rich people’s wage income. But capital income is trickier. After the income tax was first adopted in the 1910s, capital gains were initially treated as regular income and taxed at the same rate. But that quickly changed, and from 1921 onward, the top rate on capital gains never exceeded 40 percent, and was often quite a bit lower than that.
There are two major reasons for that. The first is fear of the negative effects on savings of high capital gains rates. But arguably more important was fear of “lock-in.” Because capital gains taxes are only levied when an asset is sold, very high rates on capital gains can lead investors to simply not sell stocks at all for long periods. That, among other issues, reduces federal revenue.
NYU tax law professor David Kamin has a great paper, simply titled ”How to Tax the Rich,” explaining some of the difficulties here. Right now the top capital gains rate is 23.8 percent. But because of lock-in, the revenue-maximizing rate for capital gains is more like only 28 to 32 percent, according to the Joint Committee on Taxation (JCT) and the Treasury Department (as summarized by Kamin) — not too much higher than the current top rate. Some economists argue that JCT and Treasury are too pessimistic, but as JCT will be scoring any bill raising capital gains rates, its verdict is the one that matters.
Kamin estimates that you could get, at most, $10 billion to $15 billion more a year from raising capital gains rates to their revenue-maximizing rates, and another $10 billion from treating dividends like regular income. That’s not chump change, but in terms of the federal budget, it’s not a whole lot either.
To get around the problem of “lock-in,” you’d want to move to something called a “mark-to-market” capital tax. Instead of only taxing gains on assets when they’re sold, you’d tax gains based on the assets’ estimated value every year, even if the stocks in question are just sitting in a Vanguard account. That approach would allow for much higher rates on capital gains — but it would be a huge reform, and perhaps difficult to achieve.
Wealth taxes face an even bigger hurdle in the form of Article I, Section 9, Clause 4 of the Constitution:
No Capitation, or other direct, Tax shall be laid, unless in Proportion to the Census or Enumeration herein before directed to be taken.
Helpfully, the Constitution does not define what a “direct tax” is. But as UChicago’s Daniel Hemel explains, it’s often assumed to include land taxes, and before the 16th Amendment it was assumed to include income taxes as well. (The 16th Amendment enabled federal income taxes but did not expressly allow wealth or property taxes.)
That means wealth taxes might, under the Constitution, have to be “apportioned” among the states according to their populations. (“That means that if California amounts to 12 percent of the U.S. population, Californians must pay 12 percent of any direct tax — no more, no less,” Hemel writes.)
Many legal scholars, notably including Dawn Johnsen of Indiana University and Duke’s Walter Dellinger in a paper last year, believe that wealth taxes do not run afoul of the Constitution. When unveiling her wealth tax plan, Warren released letters signed by a number of constitutional and tax law experts, including Johnsen and Dellinger, Yale’s Anne Alstott and Bruce Ackerman, Stanford’s Pam Karlan, UChicago’s Aziz Huq, and UT Austin’s Calvin Johnson, arguing for the tax’s constitutionality.
But some credible tax scholars on the left like Hemel think a wealth tax does raise real constitutional issues. It’s also worth being much more cynical on this issue: Do you really think, in your heart of hearts, that the conservative majority of the Supreme Court will miss an opportunity to strike down a wealth tax passed by a Democratic president?
And then there are more boring logistical questions about wealth taxes: How do we locate rich people’s wealth? Is it possible to value real estate accurately every year? Is the US really going to get aggressive about tax havens? These administrative issues, a recent OECD report found, have led many European countries to abandon wealth taxes in recent years.
Finally, there’s estate taxation. The estate tax currently uses a principle called “step-up basis,” which cuts down on the revenue it can bring in.
Here’s how it works. Imagine you’re living in New York in the 1980s and buy an original Basquiat painting for $200. By the time you die and leave that painting to your daughter, it’s worth $40 million. If your daughter then sells it, she’ll pay capital gains tax — but only on the difference between the sale price and $40 million. The tax code defines heirs’ gain relative to the value of an asset when they inherit it, rather than its value when their bequeathers originally bought it. Tax Policy Center head Len Burman has called this ”arguably the biggest loophole in the tax code for high-wealth households.”
The Booker estate tax bill would replace this with realization upon bequest or gift. That would mean that your hypothetical daughter would pay tax on your nearly $40 million gain as soon as you died. Kamin estimates that this would raise at least $40 billion a year, in part because people currently hoarding assets until death, when the gains become tax-free, would now have a reason to sell them off early. That’s good for overall economic efficiency as well.
But arguably the whole estate tax could use a restructuring. A number of tax analysts, notably NYU Law’s Lily Batchelder, have proposed moving to an inheritance tax, where the living beneficiaries of an estate, rather than the deceased person’s estate itself, would include inheritance money in their taxable income and perhaps pay an additional surtax on it.
All of which is to say: Democrats have gotten to the first stage in the tax-the-rich process. They’re sketching out broad ideas that poll well and offer red meat to primary and caucus voters for 2020. And they haven’t yet committed to approaches to handle these implementation details, like tracking wealth, taxing capital income as it’s earned, and taxing inheritances rather than estates.
That’s normal. The 2016 primary featured a bevy of Republicans proposing deficit-exploding tax cut plans that primarily benefited the wealthy, and ultimately Trump and congressional Republicans arrived at a tax reform that borrowed from those plans in certain structural ways but also cut top tax rates much, much less. Democratic proposals will face a similar process of loose proposals, analyst scrutiny, and legislative compromise.
At the same time, Republicans suffered when it turned out that they couldn’t get their whole party on board with key elements of tax reform that leaders like Paul Ryan considered crucial, like “border adjustment” for corporate taxes. So Democrats in Congress might benefit from sorting out the details sooner rather than later, lest they find themselves similarly flat-footed once they’re in power.
Sign up for the Future Perfect newsletter. Twice a week, you’ll get a roundup of ideas and solutions for tackling our biggest challenges: improving public health, decreasing human and animal suffering, easing catastrophic risks, and — to put it simply — getting better at doing good.