After bragging that dodging taxes made him smart, then-candidate Donald Trump eventually offered more specifics about his accounting prowess. “I have a write-off. A lot of it is depreciation, which is a wonderful charge,” Trump said in the second presidential debate. “I love depreciation.” To bolster his case that the tax system was a rigged game, he went on to invoke Hillary Clinton’s wealthy backers who, he claimed, did not pay much tax either. “Many of her friends took bigger deductions. Warren Buffett took a massive deduction.”
What this “massive deduction” Trump had in mind is not clear, but since Buffett had famously pledged to give most of his wealth away during his lifetime, the deduction for charitable giving is a likely contender. Stung by the accusation, the chairman and CEO of Berkshire Hathaway issued a statement the next day detailing his tax affairs. “My 2015 return,” Buffett said, “shows adjusted gross income of $11,563,931.” Contrary to what Trump had claimed on television, there was no massive deduction. And Buffett did pay taxes: “My federal income tax for the year was $1,845,557. Returns for previous years are of a similar nature. I have paid federal income tax every year since 1944, when I was 13.” The statement was offered as proof that the Oracle of Omaha was a responsible citizen who, unlike the reality-show celebrity, did not eschew his duties to society.
In fact, it showed just the opposite. According to Forbes, Buffett owned $65.3 billion in wealth in 2015. We don’t know his exact rate of return on his wealth, but let’s conservatively assume that it is 5%. If true, it means that Buffett’s pre-tax income amounted to at least 5% of $65.3 billion in 2015, or $3.2 billion. Out of this sum, Buffett proudly disclosed he paid federal income taxes of about $1.8 million. You do the math: while Trump bragged he didn’t pay tax, Buffett countered that he was of a different moral character, really, with an effective income tax rate of around . . . 0.055%.
Conscientious taxpayers are all alike; every tax dodger avoids taxes in its own way. Trump skirted estate tax duties on the huge fortune he inherited from his father, and then cut his income tax bill by exploiting all sorts of dodges customized to his needs by the tax-planning industry.1 Buffett followed another route. His wealth primarily consists of shares in his company Berkshire Hathaway. The company does not pay dividends. When it invests in other corporations, it forces them to stop paying dividends too. The consequences of this maneuver? For decades, Buffett’s wealth has been accumulating, free of individual income taxes, within his firm. The perpetually reinvested profits boost Berkshire Hathaway’s share price, year after year. It now costs some $300,000 to buy one share in Berkshire Hathaway, thirty times more than in 1992. To finance any consumption needs, Buffett simply needs to sell a few shares. By selling forty shares at a price of $300,000, for example, he can move $12 million to his personal bank account. He then pays tax—a modest one—on the small amount of capital gains he just realized. And that’s all.
Buffett has famously lamented that he pays too little in taxes, and lawmakers have made proposals to correct this injustice. The most famous of these efforts, advocated by Barack Obama in 2011 and Hillary Clinton in 2016, involves applying a minimum tax rate of 30% on individuals making more than one million dollars in income a year. This “Buffett rule” has become a mainstay of Democratic tax platforms. It is supposed to address the problem that because the top tax rate on capital gains (20% in 2019) is lower than on wage income (37%), Buffett (who mostly earns capital gains) is subject to rates lower than his secretary’s (who mostly earns wages). But a problem remains: the 20% tax rate Buffett faces when selling a handful of shares applies to a nanoscale fraction of his true income. Thirty percent of this nanoscale fraction would still be infinitesimal. If enacted, the “Buffett rule” would not make a meaningful difference to Buffett’s own tax bill.
By their own admission, both Trump and Buffett pay trivial amounts of taxes. Even billionaires who celebrate paying their taxes do not contribute much to the public coffers. As we’ve seen, when taking into account all taxes, the ultra-rich as a whole have lower effective tax rates than the middle class. Most proposals on the table today would do very little to remedy the problem. How do we get out of this mess?
WHY TAX THE RICH? TO HELP THE POOR
The first question to ask concerns the objective—what is the ideal tax rate to apply to the wealthy? There are several ways to think about this problem, but a good starting point is the theory of social justice formalized by philosopher John Rawls, which has broad support among social scientists. It is acceptable, according to Rawls, to have social and economic inequalities if these inequalities increase the living standards of the most vulnerable members of society.2 When applied to tax policy, this perspective suggests we should not concern ourselves with the monetary interests of the rich. We should care only about how taxing them affects the rest of the population. The goal should not be to “to make the rich pay their fair share” (a somewhat nebulous concept), but to ensure that the great wealth of some benefits the least well off.
Concretely, this means that if raising the top tax rate reduces taxes collected (for instance, because it makes the wealthy work less), the tax rate should be cut. In this case a reduction in taxes on the rich would increase the amount of revenues that governments can spend on health, child care, and other social services that improve the living conditions of the poor. Conversely, as long as increasing the tax rate generates additional revenue, the rate should continue to be raised, for higher revenues would be in the interest of the most disadvantaged members of society. The optimal tax rate on the rich is simply the rate that raises the maximum possible revenue. It’s not a controversial objective among economists. And it makes intuitive sense: everybody can agree that an extra dollar is much more valuable in the hands of a poor person than in Bill Gates’s. Taxing the wealthy a bit more is not going to prevent them from affording good child care, but if raising rates allows those who serve them coffee or clean their houses to do so, too, it’s worth it.3
With this objective in mind, taxation becomes an applied engineering problem. In the 1920s, the prodigy mathematician and economist Frank Ramsey formally proved that if all taxpayers faced the same tax rate, the rate that maximizes government revenue is inversely proportional to the elasticity of taxable income.4 What does this mean? We ran into the notion of elasticity in Chapter 5. If taxable income is inelastic, it means that when tax rates rise, reported income does not change much. In that case, the US Treasury mechanically collects more revenue by hiking tax rates. By contrast if taxable income is very elastic, then high tax rates reduce the tax base so significantly that they don’t raise much revenue and are undesirable. That’s the cardinal rule of optimal taxation, called the Ramsey rule: governments should not tax too much what’s elastic.
Ramsey’s approach was limited. It considered only a single tax rate, what is known as a flat tax, but the flat tax is a crude instrument. In principle, the income tax can be made progressive, with higher incomes subject to higher marginal tax rates. In practice, as we’ve seen, that’s how the income tax works in almost all democracies. Researchers in the late 1990s extended the Ramsey result and investigated the optimal tax rate for the rich when the income tax is progressive. As in the standard Ramsey rule, the top marginal income tax rate that maximizes government revenue is inversely proportional to the elasticity of taxable income. But with a twist: the elasticity that matters is only that of the rich. Moreover, the optimal rate now also depends on the level of inequality: the higher the concentration of income, the larger the optimal rate to be imposed on the affluent.5
THE OPTIMAL AVERAGE TAX RATE ON THE RICH: 60%
With this theory in mind, we can see that when it comes to selecting a top tax rate, the way the rich change their behavior in response is a key parameter. In the public debate, the view that the reported income of the wealthy must be very elastic—hence that they can’t be taxed too much—is often taken as self-evident. In reality things are more complicated, because elasticities are not immutable parameters. They are heavily influenced by public policies.
There are two ways, after all, that the wealthy can respond to higher taxes. The first is by changing their real economic behavior: working fewer hours, for example, or choosing less-lucrative careers. There’s not much that can be done to prevent them from doing so, it’s their right. The second—and far more common—response is tax avoidance. And tax avoidance, in contrast to more fundamental responses to taxes, can be drastically reduced by policymakers.
When companies book profits in tropical islands, when lawyers incorporate, when doctors invest in tax shelters—they are not driven by laws of nature. Such actions arise when the tax code favors certain forms of income over others, and when governments let people exploit these differences. But what’s at times tolerated, or even encouraged, can also be regulated and forbidden. When all income—whether it derives from capital or labor, whether consumed or saved, whether booked in Bermuda or in the United States, whether paid to a bank account in Zurich or in Paris—is taxed at the same rate, and when the supply of tax dodges is strictly constrained, tax avoidance can almost disappear. In that case, the wealthy cannot dodge taxes other than by reducing their true economic resources—that is, by choosing to become poorer.
People rarely volunteer to become much poorer, even for such a noble cause as eschewing taxes. The changes in real behavior provoked by the tax code are generally quite limited. It’s unlikely that Steve Jobs would have invented another iMarvel if only his tax rate had been zero. Or that Mark Zuckerberg would have opted for a career in fine arts if the Internal Revenue Code had been penned differently. Yes, Apple does shift profits to Jersey, Facebook does create shell companies in the Cayman Islands, and a sprawling industry helps the wealthy slash their tax bills. But that’s tax avoidance, which flourishes in a light regulatory environment.
For example, the 1986 Tax Reform Act—which reduced the top marginal income tax rate to 28%—led to a rise in the amount of income reported by the rich. But this increase was mostly due to changes in tax-avoidance strategies (as it became profitable to avoid the 35% corporate tax rate by organizing businesses as partnerships, subject to the individual income tax) and not an increase in labor supply.6 When tax avoidance is kept in check, the lesson from modern research is that the elasticity of taxable income is generally quite low—and therefore the optimal tax rate quite high.
How high exactly? Not as high as 100%: at that point, most people would prefer spending time with their family or tending their vegetable garden rather than work solely for the benefit of society at large. But a body of work suggests that the top marginal tax rate that collects the most revenue possible from the rich hovers around 75%. By the rich we mean the members of the top 1%, people with more than $500,000 in income in 2019.7 This estimate is the best that exists today on the basis of many empirical studies conducted over the last two decades. If there are limited tax-avoidance opportunities, the rich respond only modestly to tax changes: whenever their keep rate rises by 1% (instead of keeping 70 cents after taxes out of any extra dollar earned, they keep 70.7 cents), they work harder and increase their pre-tax earnings by about 0.25% in response.8 This means that the tax base does not shrink much when the rich are taxed more heavily, implying optimal top marginal tax rates in the vicinity of 75%.
There are a few things to consider about this result. First, we’re talking about a marginal tax rate, a rate applied only to income earned above a high threshold, $500,000 today. The associated average tax rate is lower than that, because any dollar earned below this high threshold is taxed less. It’s only for the ultra-wealthy that marginal and average tax rates are the same. Concretely, if tomorrow the marginal tax rate on income above $500,000 were increased to 75%, the average tax rate of the top 1% richest Americans would reach 60%.9 In other words, the optimal average tax rate on top bracket taxpayers is 60%—less than 60% for people at the bottom of the top 1%, up to 75% for the ultra-rich, and 60% on average among top one percenters. In many ways it is more transparent to reason in terms of average tax rates, which give a more concrete sense of the true contribution made by the various groups of the population to the community’s funding needs. Given that the average macroeconomic tax rate is around 30%, an average rate of 60% means that the top 1% richest Americans would pay twice as much in tax, as a fraction of their income, as the average person.
Second, these optimal tax rates take into account all taxes, at all levels of government. The optimal average tax rate of 60% for the rich should be seen as including not only the federal income tax, but also state income taxes, the fraction of the corporate tax paid by the affluent, payroll taxes, sales taxes, and so on. Since payroll taxes are capped and sales taxes are insignificant at the top, the optimal top marginal rate of 75% should be thought of as combining the federal income tax, any state income taxes, and the corporate income tax.
Last, and it must be said clearly: Hiking top tax rates without any other change to the tax code or to enforcement would be a bad idea. The supply of tax dodges in circulation is too large. Before we can effectively tax the wealthy more, avoidance must be curtailed. We need to create the institutions that make a robust tax system sustainable in the long run, even in the era of extreme inequality.
HOW TO STOP THE RICH FROM DODGING TAXES: A PUBLIC PROTECTION BUREAU
The first step would be to create what we’ll call a Public Protection Bureau, charged with regulating the tax-dodging industry. Just as the United States has federal agencies to regulate the financial sector (the Consumer Financial Protection Bureau), the aviation sector (the Federal Aviation Agency), and the pharmaceutical industry (the Food and Drug Administration), it should also monitor businesses that offer tax-related services and ensure that their practices are not hurting the public interest.
For as we’ve seen throughout this book, tax avoidance and evasion are spurred by the suppliers of tax dodges, not by the taxpayers themselves. Behind every epidemic of tax avoidance there’s an outburst of creativity in the market for dodges. There is, of course, a long list of loopholes in current law that ought to be closed (more on this later). But plugging these loopholes does not strike at the heart of the problem. When income tax avoidance surged in the 1980s, it was not in response to freshly introduced tax breaks, but as a direct consequence of innovation in the tax-dodging industry. When corporate tax avoidance exploded in the 1990s and 2000s, we saw the same thing: the system of transfer pricing that facilitated the new abuses had been in place since the 1920s. To curb tax injustice, we must weed out the supply of tax scams.
Unfortunately, when it comes to regulating the tax-dodging industry, the IRS brings a knife to a gunfight. That’s due to several reasons. The first is the dramatic decline in the service’s enforcement budget: Over the last decade, the IRS budget has decreased by over 20% adjusted for inflation.10 Lower budgets mean fewer auditors: in 2017, the IRS had only 9,510 auditors—down from over 14,000 in 2010. The last time it had fewer than 10,000 auditors was in the mid-1950s, when the US population was half what it is today. The second problem is compensation: given the rewards one reaps by creating a successful tax dodge, it pays much more to work for the Big Four accounting firms than to work in public service to fight tax evasion.
The last—and critical—problem is that the IRS is vulnerable to political vagaries. The risk is not primarily direct interference by the executive branch in the day-do-day operations of the service. It is more subtle—and more fundamental—than that. Congress and the prevailing administration affect enforcement: they determine the resources available for audits, they influence how aggressively the IRS challenges the tax-lowering strategies of the wealthy, they have an impact on the application of the economic substance doctrine.11 Even if these choice are not dictated by the president directly, they are shaped by the ideology that dominates in Washington. When the party in power vilifies the estate tax as an attack against sacrosanct property rights, for example, the IRS is unlikely to devote large resources to enforcing it (and indeed the frequency of estate tax audits has collapsed since 1980, as we saw in Chapter 3). This stealth erosion of tax enforcement is undemocratic. It is a threat to any progressive tax system. To avoid a new tax-sheltering frenzy à la Reagan in the twenty-first century, we need an agency that everybody—the public, the tax-accountancy industry, the IRS—can trust to apply the spirit of the law regardless of the party in power. The IRS itself will always be seen as one-sided actor, which is why an independent agency can play a useful role.
The Public Protection Bureau should have two broad missions. First, and most important, it should enforce the economic substance doctrine—the principle that makes illegal all transactions undertaken with the sole purpose of dodging taxes. That enforcement starts with collecting the necessary information. The bureau should automatically be made aware, by law, of any new product commercialized by the tax-planning industry: intragroup sales of intellectual property, investments in sham partnerships, generation-skipping trusts, and so on. In that way, it could spot the new products created to help the wealthy and corporations dodge taxes. Businesses that do not disclose their practices should face stiff penalties. And all the products that violate the economic substance doctrine should be immediately outlawed.
Second, the Public Protection Bureau would monitor foreign tax practices, and instruct Treasury to apply economic sanctions against tax havens that siphon off the US tax base. When the British Virgin Islands enables money launderers to create anonymous companies for a penny or when Luxembourg offers sweet, secret deals to multinationals, they steal the revenue of foreign nations. Nothing in the logic of free exchange justifies this theft. The commerce of sovereignty needs to be more tightly regulated, for example through taxes on financial transactions with free-riding tax havens.12
PLUGGING LOOPHOLES: SAME INCOME, SAME RATE
Another key step to curbing tax avoidance is a simple application of common sense: people with the same amount of income should pay the same amount of tax. This seems obvious, until you realize that most of the reforms of the first two decades of the twenty-first century have done the opposite. From a preferential rate for dividends in 2003 to a lower business income tax in 2018, the main preoccupation of American lawmakers has been to tax capital less than labor. The same trend can be observed in France, where the government of Emmanuel Macron adopted a flat tax for interest and dividends in 2018, and indeed in the rest of Europe.
Taxing people with the same income at the same rate is the concrete application of calls to “plug loopholes.” It has several implications.
First, it means that every income source should be subject to the progressive individual income tax: not only wages, dividends, interest, rents, and business profits, but also capital gains, which in many countries (including France and the United States) are currently taxed at lower, flat rates. There is no compelling reason to tax capital gains less than other income sources. The practice merely encourages the wealthy to reclassify their labor income and business profits into capital gains. The reason why many countries have historically resorted to this second-best policy is because tax authorities did not track the purchase price of assets (stocks, bonds, houses, etc.), making it hard to enforce a tax on capital gains. In the United States, the IRS only started to collect this information systematically in 2012. But with today’s ample and inexpensive computing capacities, progressive capital gains taxes can be enforced, including when assets have appreciated over more than a generation.13 The frequent objection that capital gains taxes impose unfairly hefty tax bills when businesses are sold (because capital gains are a one-time windfall) can be addressed by spreading payments out, as is routinely done in the context of estate taxation.
Further still, given that today’s governments now know the purchase date of assets, they could improve the tax code by removing from capital gains the mechanical effect of price inflation. In our current tax system, an asset bought for $100 in 2012 and sold for $150 in 2020 generates a taxable capital gain of $50. This makes little sense: out of the $50 increase in value, $20 corresponds to general price inflation, which is not income; only $30 corresponds to a true capital gain. The tax imposed on the $20 is equivalent to a wealth tax—an opaque and random type of wealth tax, since it’s determined by the inflation rate. This hidden wealth tax should be ditched, and only the $30 in pure capital gains should be subject to progressive income taxation. Here’s a tax cut we can all agree on!
ENDING CORPORATE TAX SHELTERS: INTEGRATION
A second application of the “equal income means equal tax” principle: the corporate and individual income taxes should be integrated—like European countries used to do and countries such as Australia and Canada, among others, still do. Integrating the corporate and individual income taxes means that once corporate profits are distributed to shareholders, any corporate tax paid by the firm is credited against the amount of personal income tax owed. Take the case of John, a wealthy shareholder subject to a marginal individual income tax rate of 50%. Assume that John owns a firm that makes $100 in profits, pays $20 in corporate tax, and distributes the remaining $80 in dividends. Here’s how an integration system works: John would include the full $100 profit (not only the $80 dividend) in his taxable individual income. He would pay $50 tax on that income, corresponding to his marginal income tax rate of 50% times $100. But in recognition of the fact that the company he owns has already paid $20 in corporate tax, his tax bill would be reduced by $20, bringing it down to $50 − $20 = $30.
Such a system recognizes the basic truth that the corporate income tax is only a prepayment for the individual income tax. It has many advantages. To start with, it dramatically reduces the incentives for firms to dodge corporate taxes. Imagine that Apple, carefully advised by the Big Four, had avoided taxation entirely: in an integrated system, its wealthy shareholders would get no tax credits and would have to pay (in the above example) the full rate of 50% on their portion of Apple’s profits. Any tax paid by Apple would reduce the bill at the shareholder level dollar for dollar. You can bet that in such a world, Apple would be instructed to slash its tax-dodging budget.
Another advantage of an integrated system is that it removes distortions. For instance, it becomes neutral for a business to be incorporated (subject to the corporate tax) or not (with all its profits passed to its owners and subject to individual income taxes, like partnerships in the United States). It makes it neutral for corporations to issue debt or equity, because both interest and dividend payments have the same tax implications. More broadly, an integrated system ensures that capital is taxed like labor—no less, but no more. In John’s case, above, we see that the total tax on $100 of profit is $50: $20 paid by the firm, plus $30 paid by John. That’s a tax rate of 50%—the same tax rate John would face if, instead of earning profits, he earned wages. It’s never a good idea to depart from the ideal of treating labor and capital the same for tax purposes, because doing so always creates opportunities for tax avoidance. And tax avoidance reduces the amount of revenue collected. If the goal is tax progressivity, it should be achieved not by taxing capital more than labor, but by taxing all income more progressively via higher top marginal income tax rates.
Although the United States has never had an integrated income tax, this system was the norm in Europe during most of the twentieth century: the United Kingdom, Germany, Italy, and France, among other countries, used to rely on it. Integration, however, gradually disappeared. Why? In a nutshell, because of a poor response to globalization. Up until the 1990s, people barely invested in foreign companies. When cross-border investments surged in the 1990s and 2000s, governments found it unacceptable to give tax credits to domestic shareholders to offset the corporate taxes levied by foreign countries. France, for example, did not give credits to the French shareholders of General Motors, which thus paid more taxes than the French owners of Renault. In 2004, the European Court of Justice ruled that the uneven treatment of foreign companies was discriminatory, leading France, among other nations, to abandon its integration system in 2005.14
The solution to this problem is simple: Foreign corporate taxes should be credited like domestic taxes. France should give tax credits to the French shareholders of US companies, and the United States should do the same with American owners of French firms. This reciprocity already exists in the case of wages earned by people working in one country but who are residents in another for tax purposes. It could easily be negotiated in bilateral tax treaties, or even better in the context of the kind of corporate tax coordination discussed in the previous chapter. Nothing in globalization prevents an integration system from working well.
There is a third implication of the “equal income means equal tax” principle. A major advantage of an integrated corporate tax is that a dollar of wage would always be taxed the same as a dollar of distributed profit. While it would be a step in the right direction, integrated taxation would still leave a thorny issue unaddressed: retained earnings—profits made by companies but not distributed as dividends—would still be taxed less than other income sources. John would be better off if his company did not pay dividends and instead reinvested its profits, for the $100 in profit would only be subject to the 20% corporate tax—and not to the individual income tax. Always and everywhere, wealthy shareholders have incentives to retain earnings within their firms; it allows them to avoid the dividend tax and save tax-free. In practice, the risk is particularly severe for closely held businesses, that is, corporations with few owners that can directly control dividend policy for their own benefit. The way to deal with this issue is by forcing closely held businesses to pass through all their profits to their owners for tax purposes. A business that is not publicly listed should always be treated as a partnership: free from corporate tax, but with all its profits subject to the progressive individual income tax of its owners. Since the Tax Reform Act of 1986, most closely held firms, including many large and complex ones, have been organized as pass-through businesses. The US experience shows that it is technically feasible to tax closely held companies at the shareholder level.15
This rule would make it impossible for the wealthy to reinvest their income tax-free, one of the most potent sources of tax injustice today. It would also destroy the shell company business that has ballooned since the 1980s, since shell companies would stop conferring any tax advantage to their creators. Shell corporations, it goes without saying, are not corporations. Recognizing them as corporations for tax purposes—with the associated tax benefits—is absurd and must stop.
HOW MUCH TAX COULD THE TOP 1% PAY?
With tax avoidance reduced to a minimum, there’s a wide consensus that increasing the amount of revenue collected from the wealthy is possible. But by how much exactly? According to our computations, by about four percentage points of national income, or $750 billion a year in 2019.
To see how we arrived at that number, recall that the members of the top 1% earn, as a group, 20% of national income. And as we saw in the first chapter, taking into account all taxes, their average tax rate is 30% today. Their tax payments add up to 20% of America’s national income times 30%—or 6% of national income. As the editorial board of the Wall Street Journal never fails to remind us, the rich do contribute to the public coffers. But contrary to what they’d like us to believe, that’s not because the effective tax rate on the wealthy is high (it’s almost the same as the average macroeconomic tax rate in the economy) but because in America the rich have a lot of income.
As we’ve seen, after cracking down on tax dodges, the average tax rate that would maximize tax collection at the top is much higher than the current rate of 30%—it is close to 60%. Admittedly, should their tax rate double, the wealthy would report less income, even with tax avoidance kept in check: thought leaders would perhaps give fewer paid speeches, company executives may retire somewhat earlier, and so on. As a result, the inequality of pre-tax income would fall; according to the best available estimates, the top 1% share of pre-tax national income would decline from 20% to about 16%.16 Let’s run the math again: should their average tax rate double, America’s affluent families would pay 60% of 16% of national income in tax, which is about 9.5% of national income. According to standard economic theory, this is, then, the maximum amount of tax that can be collected from the top 1%: 9.5% of national income. Raising the tax rate of people just below the top 1% slightly (for instance via an increase in the corporate tax rate) would generate an extra half percentage point in revenue, bringing the total amount of tax paid by the affluent to 10% of national income. That’s four percentage points more than today.
Is doubling the average tax rate at the top, from 30% to 60%, realistic? As we have seen, it’s not without precedent. The average tax rate of the top 0.1% income earners neared 60% in the 1950s and reached close to 70% for the top 0.01% in 1950, when higher corporate taxation generated massive revenue and equity ownership was still highly concentrated. The effective tax rates that prevailed among the rich at midcentury were much higher than the tax rates paid by the rest of the population. In 1950 the bottom 90% paid 18% of its income in taxes, forty points less than the top 0.1%. The notion that it would be a drastic departure to implement high, progressive tax rates at the top of the income pyramid doesn’t hold up to scrutiny.
It is true, arguably, that even in the heyday of progressive taxation, the average tax rate among the merely rich (in the top 1% but not in the top 0.1%) was more like 40%. As a result, the average tax rate for the entire top 1% was close to 50%—less than the 60% that would be optimal today. The rich, however, captured a much smaller fraction of national income at midcentury than today. When income is more concentrated, economic theory suggests the rich should be taxed more.
Implementing a more progressive tax system today is not a matter of simply going back in time. The progressive post–World War II tax system, for all its virtue, was far from perfect. In violation of the “equal income means equal tax” principle, capital gains were taxed less than ordinary income. The individual income tax had loopholes. Each of these flaws meant that the rich had scope to dodge taxation. By leveraging modern technology and drawing the lessons from the past and from other countries, it is possible to do better today.
7.1 WHEN AMERICA TAXED THE RICH HEAVILY
(Average tax rate of the top 0.1% versus bottom 90%)
Notes: The figure depicts the average tax rates paid by the top 0.1% pre-tax income earners and the bottom 90% earners. All federal, state, and local taxes are included. Taxes are expressed as a fraction of pre-tax income. The revenue maximizing tax rate for the top 0.1% is around 65% today, similar to the effective tax rates reached at midcentury. Complete details at taxjusticenow.org.
A WEALTH TAX: THE PROPER WAY TO TAX BILLIONAIRES
The proper way to tax the wealthy in the twenty-first century—and in particular to arrive at the optimal rate of 60%—involves three essential and complementary ingredients: a progressive income tax, a corporate tax, and a progressive wealth tax. The corporate tax ensures that all profits are taxed, whether distributed or not: it acts as a de facto minimum tax on the affluent. The progressive income tax ensures high earners pay more. And a progressive wealth tax gets the ultra-wealthy to contribute an amount that reflects their true capacity to pay.
Why isn’t the income tax enough? Quite simply because among the most advantaged members of society, many people possess substantial wealth while having low taxable income. Maybe they own a valuable business that does not make much profit, but which, everybody anticipates, will be immensely profitable in the future (see: Bezos, Jeff). Or, as is more frequently the case, they may structure their already profitable business so that it generates little taxable income (see: Buffett, Warren). In both cases, these billionaires can today live almost tax-free. As we saw in Chapter 5, even from the strict vantage point of economic efficiency, there is no cogent reason why the uber-wealthy should be permitted to grow their billions without contributing to their community’s needs.
Without a wealth tax, it will be hard to reach average rates of 60% at the highest reaches of the wealth scale. Raising the top marginal income tax rate wouldn’t affect the tax bill of Jeff Bezos or Warren Buffett notably, since neither of them has much taxable income in the first place. Raising other taxes, such as the estate tax, would not do either. We might take comfort in the idea that the richest man in the world, Jeff Bezos, will one day pay estate taxes on his immense wealth. But since the founder of Amazon turned fifty-five in 2019, that won’t (hopefully) happen before 2050. And let’s not mention Mark Zuckerberg, born in 1984—is it wise to wait until 2075 for him to contribute to the public coffers? The way to address this issue is by taxing wealth itself, today and not at some distant future date.17
A wealth tax will never replace the income tax; its goal is more limited: to ensure that the ultra-wealthy do not pay less than the rest of the population. Top executives, athletes, or movie stars—who earn a lot of income—can be appropriately taxed by a comprehensive income tax. It’s the super-rich—most of whom own a lot of wealth but have little taxable income—for whom a wealth tax is essential.
There are many ways to combine a wealth tax with the progressive income tax and the corporate tax to arrive at an effective tax rate of 60% among the richest Americans.* In Figure 7.2, the effective corporate tax rate is multiplied by 2 (essentially returning tax revenues to their pre-2018 reform level, which is not impossible); the income tax is made more comprehensive (by treating capital like labor) and progressive (with a top marginal income tax rate of 60%); estate tax revenues are doubled (with better enforcement); and an annual wealth tax at a rate of 2% above $50 million in wealth and 3.5% above $1 billion is introduced.
7.2 ONE POSSIBLE OBJECTIVE: RETURNING TO THE TAX PROGRESSIVITY OF THE TRUMAN-EISENHOWER ERA
(Average tax rates by pre-tax income groups)
Notes: The figure depicts average tax rates by income groups in 1950 and 2018 and under a reform scenario that increases the corporate income tax, increases the progressivity of the individual income tax, and adds a progressive wealth tax. This reform scenario would restore the tax progressivity of the 1950 tax system. Complete details at taxjusticenow.org.
7.3 THE WEALTH TAX: A KEY INGREDIENT FOR A PROGRESSIVE TAX SYSTEM
(Average tax rates by pre-tax income groups)
Notes: The figure depicts average tax rates by income groups under a tax reform scenario that increases the corporate income tax, increases the progressivity of the individual income tax, and introduces a progressive wealth tax. The progressive wealth tax is a crucial component to restoring tax progressivity at the very top. Complete details at taxjusticenow.org.
The result? A tax system that, at the top, would closely resemble that of the 1950s.
The primary departure from the 1950s tax system is the progressive wealth tax. The 1950s did not have one, and instead the system achieved its sharp progressivity mainly through a high corporate tax rate of 52%—which applied to profitable companies that, at the time, had relatively few owners, essentially individuals rather than institutional investors.18 Since more than 20% of listed US equities are owned by foreigners and 30% by pension funds today,19 even the measures described in the previous chapter (a systematic collection of the taxes not paid by multinationals), coupled with a massive hike in the corporate tax rate, would not create the same degree of tax progressivity as the 1950s. The corporate tax, which is not progressive, is too blunt an instrument to restore tax justice. And as we’ve seen, even with a much-improved income tax, it would not be possible to tax the ultra-rich properly. That’s why the wealth tax is a critically important component to any reform.
HOW TO TAX WEALTH: LEVERAGE THE POWER OF MARKETS
A progressive wealth tax is possible because in contrast to taxable income, which can be artificially reduced, wealth is well defined at the very top. Wealth is the market value of one’s assets net of all debts. Warren Buffett reports a tiny amount of taxable income to the IRS compared to his true economic income. But he cannot hide the fact that he’s worth more than $50 billion. With a wealth tax at a rate of 2% on the fortunes above $50 million and 3% above $1 billion (such as the one proposed by Senator Elizabeth Warren in 2019), Buffett would pay around $1.8 billion a year, a thousand times his 2015 income tax bill of $1.8 million.
Not all forms of wealth are easy to value. As a publicly traded company, Berkshire Hathaway has a well-defined market value; because Buffett’s wealth is fully invested in Berkshire Hathaway shares, it is easy to tax him. But the affluent can also own shares in unlisted (also called private or closely held) businesses. Other forms of wealth—like art or jewelry—are sometimes hard to value. But overall, these concerns about valuation are overblown.
Modern capitalist economies like the United States have well-defined property rights and put a value on most assets. According to our computations, 80% of the wealth owned by the top 0.1% richest Americans consists of listed equities, bonds, shares in collective investment funds, real estate, and other assets with easily accessible market values. As for the remaining 20%—mostly shares in private businesses—valuation raises fewer problems than you might think. Although not publicly listed, shares in large private businesses are regularly bought and sold. Even before Lyft and Uber went public in 2019, for instance, it was possible for rich people to invest in the ride-sharing service companies. Private companies regularly issue new stock to banks, venture capitalists, wealthy individuals, and other “accredited investors” with deep enough pockets. These transactions de facto put a value on private firms.
Admittedly, in some cases no transactions can take place for years. It’s often true for mature private businesses that are controlled by a small number of owners. Let’s look at the agribusiness giant Cargill, the largest private company in the United States, which is 90% owned by a hundred or so members of the Cargill and MacMillan families. The last time shares in Cargill were transacted was in 1992, when 17% of the company’s shares were sold for $700 million—thus valuing the entire firm at a bit more than $4 billion.20 Here is a striking case where taxing wealth might seem hopeless: What’s the value of Cargill today, almost thirty years after this last share transaction? Isn’t any estimate fraught with countless risks of abuse?
Taxing the Cargills and the MacMillans equitably is not impossible. To start with, the tax administration can take the 1992 Cargill valuation and update it based on any changes in the company’s profits since then. If the company makes three times more profits today, it’s not unreasonable to believe it’s worth about three times as much as in 1992. Of course, much more data ought to inform a reliable valuation. For example, data about Cargill’s direct competitors that are listed as public companies, such as Archer Daniels Midland and Bunge. To better assess Cargill, the IRS can consider how much a dollar of earnings made by these firms is valued by the stock market. It can study how the price-to-earnings ratios of Archer Daniels Midland and Bunge have changed since 1992. Valuing private firms in this and more sophisticated ways is what hundreds of financial analysts do every day. The United States has no shortage of expertise in this area. Drawing on standard private sector practices, it would not be terribly complicated for the IRS to come up with a reasonable estimate of Cargill’s market value at the end of each year.
But here’s the most interesting part. Let’s assume that the Cargills and MacMillans feel cheated by the IRS—that is, they feel that the tax authority overestimates the value of their firm. Perhaps Cargill has fundamentally changed since 1992, in ways that state-of-the-art valuation techniques may fail to capture. Perhaps it has weaknesses that its competitors do not share. What is to be done?
At the heart of the problem there is a missing market: while there is an active, liquid market for Archer Daniels Midland and Bunge shares, no such market exists for Cargill’s stock. The solution to this problem, in our view, is for the government to step in and create the market that’s missing. The IRS would give the option to Cargill’s shareholders to pay the wealth tax in kind—with Cargill shares—rather than in cash. If they used this option (which they would, by definition, only do if they believed the IRS valuation was exaggerated), the tax authority would then sell the shares to the highest bidders on a market open to any and all bidders—venture capitalists, private equity funds, foundations, or other families interested in acquiring a stake in the agribusiness giant.
This solution, which to our knowledge had not been proposed before, addresses a serious impediment to deploying a wealth tax. Just as a well-functioning income tax should treat all income equally, a good wealth tax should assess all forms of wealth in the same way, at their current market value. If some values are missing, the solution is to create them. And there’s nothing better to create a market value than, well, to create a market. For a wealth tax imposed at an average rate of 2%, Cargill’s shareholders would hand in 2% of their shares each year (or the cash equivalent, if they prefer to retain full control of the company). Like Buffett’s Berkshire Hathaway, there would be no getting away.21 Transforming Cargill’s shares into cash would be the government’s job.
This solution also addresses another frequent objection to wealth taxation—the problem of liquidity. Very rich people may have a lot of wealth and yet not enough income to pay their tax bills. Isn’t it unfair to force them to pay a tax when they don’t have cash in hand? To be frank, liquidity concerns are often put forward in bad faith. Most of the time, the notion that people worth a hundred million may not have enough cash to pay a million dollars in tax does not withstand scrutiny. When ultra-rich individuals pretend they have little cash, it’s usually because they choose to realize little income to avoid the income tax. They organize their own illiquidity.
But there are instances where liquidity problems are real. The most relevant case is that of a highly valued start-up that makes no profit. Generating cash each year may turn out to be complicated or costly for anyone whose primary source of wealth is shares in the company, since young firms typically do not pay dividends. In that case, allowing taxpayers to pay in kind—with shares of the companies—addresses the problem. Because the wealth of the rich mostly consists of equity, and equity (in contrast to real estate) can always be divided, it can be used to pay for the tax. If the wealth tax can be paid with assets rather than cash, a progressive wealth tax isn’t harder to implement than a progressive income tax.
The middle class already pays tax on its wealth, in the form of property taxes. The wealthy don’t, since most of their wealth consists of financial assets, which are exempt from the property tax. In the nineteenth century, the property taxes of most states fell on all assets, both real and financial—in contrast to today. In the beginning of the twentieth century, the United States pioneered the progressive taxation of property with its federal estate tax—now moribund. Before turning its back to this distinguished tradition, America was at the forefront of the democratic regulation of property via the tax system. With a progressive tax on extreme wealth, the United States would be in a position to lead again.