There is a recurring fixture of most debates about public policy that might surprise you at this stage of the book: it’s the argument that the progressivity of the tax system doesn’t matter. The government, according to this view, can always achieve the redistribution it desires through public spending. As long as spending helps the more vulnerable among us, how taxes are levied is irrelevant; collecting revenues is what matters. This view is prevalent in the United States and in Europe and has informed most of the tax advice provided by the International Monetary Fund and the World Bank over the last decades. Governments in Asia or Africa have been encouraged to raise their value-added taxes—levies that burden the poor more than the rich—to fund social programs. Progressive income taxes? Inheritance taxes? Wealth taxes? These are unnecessary, and perhaps even politically dangerous.
This strategy is not entirely without merit: value-added taxes can generate a large flow of revenue that can help fund education, health care, and other public goods that raise standards of living. The problem lies with the underlying view of the development process that informs these well-meaning experts. Development is not primarily a matter of mechanically collecting taxes to fund spending, no matter how useful this spending may be. Development is about building trust in institutions, including, most importantly, governments. When governments take more from the poor than from the wealthy, sustained trust becomes impossible.
This insight is essential for understanding the history of taxation, from the tax revolts of the Middle Ages to the 2018 “yellow vests” movement in France. It will likely remain relevant in the future.1 Take environmental taxes. Putting a price on carbon is critical to combat climate change, but since spending on fuel and other carbon-intensive goods absorbs a greater share of income for the poor than for the rich, carbon taxes are typically regressive. To offset this pain, fighting climate change will require additional progressive taxes. Governments that forget this basic truth will learn it the hard way.
Or take health care. In the United States, two of the most comprehensive efforts to introduce universal health insurance—the Clinton proposal of 1993, and the Vermont single-payer health care project of 2014—failed not for lack of general support, but largely because there was no palatable, fair funding solution. That’s what happens when only spending matters, and not how the money is raised. Often, no spending occurs. Since the failure of Clinton’s 1993 universal health care plan, thousands of Americans have died for lack of insurance;2 millions have lived with the fear of losing theirs.
As we saw in Chapter 7, the United States could collect up to four points of national income in additional taxes by taxing the rich more. This would be enough to provide health care to the millions of currently uninsured Americans. But it could also be the starting point of a more ambitious expansion of the US social state, centered on the public funding of health care for all and education for all from early childhood to university. This expansion of the US social state would require funding above and beyond the extra taxes collected from the rich. In this chapter, we show one potential way to generate this funding.
THE RISE OF THE SOCIAL STATE
Why do most people think that the government should fund health care and education? For the same reasons they believe that retirement ought to be funded largely by the government. An adequate standard of living is recognized as a fundamental human right. In practice, without education, old-age income support, and health care, the right to an adequate standard of living cannot be fulfilled.
Before the twentieth century, families—rather than the government—did provide support for the elderly and the sick. Parents paid for the education of their children; children took care of their aging parents. Religious organizations aided those without family support. But that all occurred in a context where educations were short for most, medical care rudimentary, and the elderly could not expect to live long. As technology improved, life expectancy rose, and medical science progressed, the cost of education, retirement, and health increased—making collective funding necessary.
In the United States, the income of the working class (half of the population) is $18,500 a year per adult in 2019. That’s at a time when America spends 20% of its national income on health care, or $15,000 per adult. All advanced economies spend at least 10% of their national income on health, even when they work hard to control costs.3 If the United States emulated them and curbed health spending to 10% of national income, this would still amount to $7,500 per adult, a sum which is out of reach on a $18,500 income.
Can’t the poor purchase discounted health care and education? No, because cheap health care, like cheap education, means in practice no education and no health care when you need it. The view that health care services are like haircuts or restaurant meals—services for which there is a product tailored to any budget—is a myth. The poor need as much health care as the rich, and as much education. Almost all American kids, no matter their background, graduate from high school after at least twelve years of education; and we all wish that more kids from disadvantaged backgrounds would attend college. Everybody needs their broken legs mended. No advanced economy succeeds in delivering decent health care and education on the cheap.
That’s why all developed countries have, over the course of the twentieth century, gradually entrusted their governments with the task of funding education (including pre-kindergarten child care), old-age support, and health care. Of course, there is a legitimate debate about how much governments should spend, how much individuals should contribute, and how to regulate providers of education, retirement, and health services. But there is no example of a successful model for retirement, education, and health care that is not heavily funded through taxes or mandatory contributions equivalent to taxes. In all advanced economies, total taxes have increased from less than 10% of national income around 1900 to between 30% and 50% of national income today, primarily to fund the three pillars of the social state: education for the young, retirement benefits for the old, and health care for all.4
PRIVATE HEALTH INSURANCE: A HUGE POLL TAX
The United States is no different. Social Security, created in 1935 and funded by payroll taxes, now spends about 6% of national income each year on retirement and disability benefits. Mass secondary and higher education has always been organized primarily through government and funded by general tax revenue.5 Even though tuition for higher education is high and student loans burden many, government still funds about two-thirds of all education spending in America. It also provides health insurance for the poor (Medicaid), the elderly (Medicare, starting at age sixty-five), and veterans.
The US social state, however, has major holes. The government spends very little on child care and early education, placing it near the bottom in international rankings on that metric. The world’s richest countries guarantee mothers more than a year of paid maternity leave; the United States guarantees them nothing. Except for some US cities at the vanguard, there’s no public school before age five; no public nursery in America. Other wealthy countries have long understood that education—including early education—is better and more efficiently provided by the community than by the market; the United States, not yet.
Because the cost of child care is prohibitively high—the annual cost of day care can easily reach $20,000 per infant—many families resort to parental care. In practice, this task falls primarily on mothers. The absence of government spending, in effect, imposes a huge tax on women’s time—the most archaic possible form of tax. This tax has profound impacts on women’s careers and deepens gender inequality. Earnings for American mothers fall 31% on average after the birth of a first child relative to fathers. That’s how, despite the fact that women are more educated and more likely to graduate from college than men, massive gender disparities in earnings remain.6 Isn’t it absurd, even from a pure efficiency point of view, to devote resources to higher education, but then fail young mothers at a critical time in their careers by not providing early education to their children?
The other specificity of the United States compared to other advanced economies is that public health insurance is far from universal in America. About half of all health care spending (10% out of 20% of national income) is publicly funded. A large fraction of the population must purchase private insurance. The system of private insurance excludes millions of Americans and imposes a mammoth burden on workers.
As we saw in Chapter 5, private insurance premiums are akin to a huge private tax. Although most workers get insurance through their employers—and thus employers nominally foot the bill—the premiums are a labor cost as much as payroll taxes are. Just like payroll taxes, premiums are ultimately borne by employees. The only difference is they are even more regressive than payroll taxes, because the premiums are unrelated to earnings. They are equal to a fixed amount per employee (and only depend on age and family coverage), just like a poll tax.* The secretary literally pays the same dollar amount as an executive.
Poll taxes, unsurprisingly, are not popular. When Margaret Thatcher imposed a poll tax in 1988 to replace real estate property taxes, she faced an unprecedented revolt and was ousted from office in 1990. No government would out of the blue impose a poll tax to fund health care; it would be a crushing burden on moderate-income families. And yet in essence that’s what the United States does today: employers are administering a huge poll tax on behalf of the government. Since the passage of the Affordable Care Act in 2010, employers with fifty or more employees are legally required to provide health insurance to their workers or pay a penalty of $2,500 per employee in 2019. Given how big the average annual health insurance premium has become ($13,000 per covered worker), this system is unsustainable.
To illustrate the magnitude of this poll tax, let’s look at the distribution of US tax payments once we include mandatory private health insurance premiums. As we saw in the first chapter, considering regular taxes only the US tax system looks like a giant flat tax that becomes regressive at the top. But with the health care poll tax added, it is in fact frankly regressive: once private health insurance is factored in, the average tax rate rises from a bit less than 30% at the bottom of the income distribution to reach close to 40% for the middle class, before collapsing to 23% for billionaires.
9.1 THE US TAX SYSTEM: FLAT . . . OR MASSIVELY REGRESSIVE?
(Tax rates including compulsory health insurance, 2018)
Notes: The figure depicts the average tax rate by income group and its composition by type of tax in 2018. All federal, state, and local taxes are included. The figure adds quasi-mandatory employer-sponsored health insurance premiums as an additional tax paid by enrolled workers. Including these quasi-taxes, the US tax system is sharply regressive, with the working class and especially the middle class paying more than the very rich. Complete details at taxjusticenow.org.
The poll tax hammers the working and the middle class. At the bottom of the distribution, it’s not as onerous as sales and payroll taxes. But that’s because many working-class Americans do not get health insurance from their employers. They either face the burden of obtaining coverage themselves, rely on a family member to cover them, enroll into Medicaid, or go uninsured. The Affordable Care Act increased the pool of Americans eligible for Medicaid and subsidized the purchase of private insurance for low-income people who weren’t otherwise covered, but the law still left about 14% of the adult population uninsured in 2019.7 And it provided no relief for workers who fund their health care through a poll tax whose cost for the middle-class class far outweighed that of the income tax.
FUNDING THE SOCIAL STATE: BEYOND PAYROLL TAXES AND THE VAT
How do other rich countries—where health insurance is universal or near-universal and the provision of public child care much more prevalent—fund these essential social needs? Generally speaking, health insurance is funded by payroll taxes or general government revenue such as the value-added tax. While better than nothing, this style of funding is not ideal.
Payroll taxes are fairer than a poll tax, since they are proportional to wages, at least up to some limit. But they have a big limitation: They are typically imposed on labor income only. Capital income is exempt. Certain countries have tried to expand their payroll taxes to include some capital income into the base, but despite that effort labor carries the bulk of the burden of funding health care.8
If people have access to health care whether they earn labor or capital income, there’s no reason why only labor should contribute. And, as we’ve seen, capital income is taxed less and less (despite rising faster than national income in most countries), while labor is taxed more and more (despite rising more slowly than national income, and sometimes stagnating). In that context, exempting capital from funding rising health expenditures seems neither sensible nor sustainable.
On top of payroll taxes, all advanced economies except the United States have large value-added taxes. The principle of the VAT emerged in the early twentieth century, invented independently by a German industrialist, Wilhelm von Siemens, and the American economist Thomas Adams. France was the first country to experiment with VAT implementation in 1948, adopting it more broadly in 1954. The concept caught fire in the 1960s and was adopted by most countries in the decades that followed.9 The VAT replaced earlier consumption taxes such as excise taxes on specific goods, sales taxes, and turnover taxes (which are similar to VAT with the difference that they tax intermediate goods).10
The VAT has clear advantages over the consumption taxes it replaced and that still exist in the United States today. It taxes services as well as goods. It does not create cascading taxes over the chain of production—as turnover taxes do—because the cost of intermediate products purchased by firms can be deducted from the value of the goods and services they sell. It is harder to evade than sales taxes because the tax is collected at each stage of production, not only at the time of the final sale. That’s why, following France’s lead, the VAT has been broadly adopted across the world.
To some observers, the solution is obvious: The United States should adopt a VAT to fund the expansion of its incomplete social state. In our view that would be a mistake. The VAT has two big flaws: it’s regressive and its tax base, although larger than that of payroll taxes, is too small.
The VAT is regressive because it taxes consumption, not income. The working class and many in the middle class cannot afford to save: they consume all their income and, during some rainy days, more than their income. The VAT hits these groups hard. As one moves up to the top of the income pyramid, by contrast, consumption becomes smaller and smaller relative to income. There’s only so much you can consume, even if you spend lavishly, and the ultra-wealthy barely pay any VAT relative to their incomes. At some point even the rich consume their savings, but that can be decades after income has been earned (if savings are used to fund retirement) or centuries after (if savings are passed to successive generations of heirs). The fundamental injustice of consumption taxes, relative to income taxes, is that the well-off can postpone them by saving, while the poor pay cash on the nail. “Justice too long delayed is justice denied”:11 this is true also when it comes to taxation.
Contrary to what is widely believed, the VAT exempts a significant fraction of the economy. Finance, education, and health care, three of the largest sectors in our modern economies, are typically exempt. Finance has contributed more than any other sector to the upsurge in income inequality in the United States; health care is high on the list too.12 Introducing a new tax that exempts these sectors would not exactly advance the fight against inequality. VAT excludes finance because there’s no easy way to compute “value-added” in the financial industry. For regular businesses, value-added is equal to sales to customers minus cost of intermediate inputs. The financial sector manages your funds (bank accounts, mutual and pension funds) by taking a cut on the returns, and it lends you money (credit cards, student loans, mortgages) at a high rate. But it does not explicitly and separately charge for its services.
Finance, health care, and education together were small sectors when value-added taxes were first introduced in the 1950s. Since then, however, they have grown fast. Moreover, because the VAT is (correctly) perceived as regressive, necessities such as food receive preferential rates. For all these reasons, France and Germany for example, which have standard VAT rates of 20% and 19% respectively, only raise about 8% of their national income via VAT.13 In other words, the VAT only draws on 40% of their national income. In the United States—where the health care and finance sectors are larger than in Europe, but where people save less of their income overall—the base of the VAT would be a similarly low fraction of total national income. To raise 6% of national income in revenue, America would need to apply a 15% VAT rate.
The limitations of VAT and payroll taxes mean they aren’t up to the job of funding the social state at a time of high inequality. These two options were popular in Europe in the postwar decades when inequality reached a historical nadir, but they are now outdated. We need to innovate.
FUNDING THE SOCIAL STATE IN THE TWENTY-FIRST CENTURY: THE NATIONAL INCOME TAX
The United States can leapfrog the VAT. It can pave the way in the creation of the fiscal institutions of the twenty-first century—as it did during the twentieth century. How? By creating a national income tax.
The basic idea is simple: the national income tax is a tax on all income, whether it derives from labor or from capital, and whether it originates from the manufacturing sector, finance, nonprofits, or any other sector of the economy. The tax does not exempt saving, which is highly concentrated among the well-off and is more effectively encouraged by government regulations (such as automatic enrollment in pension plans and financial regulation) than tax breaks. To keep administration simple, the national income tax has a single rate and offers no deductions.
Let’s be clear: the national income tax is certainly not meant to replace the income tax, or any other progressive tax for that matter. It is meant to supplement progressive taxation and to replace regressive taxes that impose an unfairly high burden on the American working class and middle class, chief among which are private insurance premiums—the most regressive levy.
The national income tax is a true flat income tax. The “flat tax” proposed by the economists Robert Hall and Alvin Rabushka in 1985 and embraced by many conservatives is in reality a consumption tax at a flat rate, like a VAT, but it’s often disguised as an income tax to make it more appealing.14 The national income tax is more comprehensive and fairer, since it does not discriminate across different uses of your income (consumption versus saving).
To see how the tax would work, it is important to bear in mind that national income is the sum of labor income, business profits, and interest income. Concretely, taxing national income means taxing each of these income flows.
For labor income, the national income tax would be administered and remitted by employers. All employers—whether for-profit businesses, not-for-profit organizations, or governments—would pay a tax proportional to the full labor cost of all their employees. This would look like an employer payroll tax but levied on a larger base, including all fringe benefits and with no cap. All employee compensation, already reported on corporate and business tax returns and amounting to 62% of national income, would be covered by the national income tax.
Next, all businesses—from mom-and-pop restaurants to giant corporations—would have to pay the national income tax on their profits. The base would be the full amount of profits with no deductions or exemptions. Businesses would depreciate their capital assets to reflect normal wear and tear, but would not be allowed to deduct any tax paid. Business profits are already measured for income tax purposes on corporate or business tax returns.
The national income tax would also be levied on interest income. The interest businesses pay on their loans and bonds is deducted from business profits; the corresponding interest received by lenders must be taxed. For businesses, interest received is already included in profits. This leaves only interest received by individuals and nonprofits to be added to the tax base, which does not present any administrative difficulty. Foreign dividends received by individuals and nonprofits, as well as other income forms received from abroad, also would be liable for the tax.
Because a tax defined this way taxes all income sources only once, there is no need to tax US dividends (corporations have already paid the tax on their profits), retirement income (for labor income, including retirement contributions, has already been taxed), or any government transfer such as social security payments or unemployment benefits. That’s a key difference with the VAT: the national income tax does not burden people who live off transfer income, who tend to be at the bottom of the income distribution. This makes the national income tax much more progressive than a VAT.
Our computations show that the national income tax has a base nearing 100% of national income. The rents that homeowners pay to themselves (which are part of national income, but not easy to tax) are excluded from the base; but the national income tax base does not deduct mortgage interest payments. In practice, tax evasion would make the base of the national income tax slightly lower than 100% of national income. The informal economy, including employees receiving wages off the books or self-employed workers paid in cash, would not be reached, and some businesses would underreport profits. According to the available estimates, these activities would reduce the base by about 7% of national income.15
Because it’s so broad, the national income tax could raise substantial revenues with low rates. It would be a stable source of tax revenue—since national income does not vary much from one year to the other—which is important to fund the core, long-run missions of the social state. Occasionally it’s argued that a carbon tax could provide some of the funding for health or childcare, but that’s a mistake. Carbon taxation is, of course, necessary to fight climate change. But its goal should be only this: fight climate change. It should not aim at collecting revenue in the medium run, but instead aim at eradicating future carbon emissions. A successful carbon tax should eventually yield zero revenue.
If the national income tax is such a great idea, why hasn’t it been proposed and implemented before? Probably because of international tax competition, since the national income tax does increase the taxation of corporate profits. However, with proper taxation of multinationals, as discussed in Chapter 6, concerns about tax competition would fall away.
UNIVERSAL HEALTH INSURANCE, NOW
The national income tax opens a world of possibility. In the United States, it might be used to fund universal health insurance, child care, and a more equal access to higher education, for instance through more funding for public universities. Higher education is particularly unequal in America, where only 30% of young adults from poor families attend college by age twenty-two (versus close to 100% for the rich),16 and where students are burdened with sky-rocketing loans that impede wealth building for the middle class. The national income tax could also be used to replace the states’ archaic sales taxes, which are very regressive, and to provide states with a tool to fund their own social state, should the federal government fail to act. Other countries could implement a national income tax to reduce their VAT or payroll taxes on labor earnings, thereby making their tax systems less regressive.
For example, in America a national income tax at a rate of 6%, combined with a greater taxation of the rich, would generate about ten points of national income in government revenue. With six points going to health care, one point to universal child care, and half a point to higher education, America would acquire a social state worthy of the twenty-first century. The remaining revenue could be used to eliminate the archaic sales taxes (and the Trump tariffs) that currently hammer the working class.
9.2 FUNDING THE SOCIAL STATE OF THE TWENTY-FIRST CENTURY
Tax revenue |
||
|
Type of tax |
Revenue (% of national income) |
Wealth tax |
2% rate above $50 million |
1.2% |
|
3.5% rate above $1 billion |
|
Income tax |
Full taxation of dividends and capital gains |
1.7% |
|
60% top marginal income tax rate |
|
Corporate tax |
30% effective US corporate tax rate |
1.2% |
|
25% country-by-country minimum tax |
|
National income tax |
Flat 6% rate |
5.6% |
Total |
|
9.8% |
Spending |
||
|
Type of spending |
Cost (% of national income) |
Health care for all |
$8,000 for currently covered workers |
6.0% |
|
$8,000 for the currently uninsured |
|
Education for all |
Public child care and early education |
1.0% |
|
Free tuition for public universities |
0.5% |
Sales tax cut |
Eliminate sales taxes and Trump tariffs |
2.3% |
Total |
|
9.8% |
Notes: Our proposed reform funds health care for all and education for all (from early child care to university), and eliminates regressive and archaic sales taxes (but keeps excise taxes, which primarily fall on gasoline, alcohol, and tobacco). This reform is funded by extra taxes on the rich (progressive wealth taxation, more progressive income taxation, and enhanced taxation of corporations) and a national income tax, fairer and broader than a VAT. Complete details at taxjusticenow.org.
Although it is difficult to quantify the economic effects of a healthier and more educated workforce, the evidence suggests that the effect on growth would be positive. Freed of the risk of losing their employer-provided health insurance, more people might start businesses. More college graduates would boost productivity. Universal child care would increase women’s labor force participation. In turn, higher incomes would increase tax collection, eventually reducing the government deficit.
If a 6% national income tax was used to fund health care, here is how it would work. A rate of 4.5% would be enough to fund standard health insurance covering all medical needs for all workers who currently pay contributions through their employers. It would also allow the extension of Affordable Care Act exchange subsidies to all participants regardless of their family income. Increasing the rate to 6% would be enough to cover the 30 million Americans who are uninsured today—and achieve true universal health insurance.
With a national income tax of 6% funding health care, most Americans would come out ahead. Of course, such a tax would reduce labor earnings by 6%. But a large swath of workers pay more than 6% of their income today in health insurance. Suppose you earn $40,000 and your employer currently pays $12,000 for your health care. In reality your labor income is $52,000, but 23% of it is eaten by the health insurance poll tax. Covered workers would come out ahead if their insurance premiums are less than 6% of their total labor income, which is the case for over 90% of workers with employer-sponsored health insurance. On the flip side, high wage earners and individuals with capital income would pay more.
The main practical objection to embracing what amounts to a “Medicare for All” program is that currently covered employees do not want to give up the private insurance they know for a new public insurance program. One way to address this issue involves giving workers the option to keep their current plan. Suppose that you earn $40,000 and that your employer contributes $12,000 to a private plan today. Imagine again that the public health insurance is worth $8,000. In this case, the government would pay your employer $8,000. The cost, for your employer, of subscribing to your preferred private plan would fall from $12,000 to $4,000. The law would mandate that employers pass on to workers the $8,000 received from the government; your take-home pay would thus increase by $8,000—a 20% jump. The alleviation of health insurance costs would be neutral for employers and directly show up on workers’ paychecks.
9.3 A PROGRESSIVE TAX SYSTEM FOR THE TWENTY-FIRST CENTURY
(Tax rates, percentage of pre-tax income)
Notes: The figure depicts average tax rates by income groups in 2018, treating employer-sponsored health insurance premiums as taxes. The reform scenario abolishes all sales taxes and employer-sponsored health insurance premiums, introduces a national income tax at a rate of 6%, creates a progressive wealth tax, and increases the corporate income tax and the progressivity of the individual income tax. Complete details at taxjusticenow.org.
Figure 9.3 illustrates what the US tax system would look like under our proposed reform: taxation of the rich up to the top of the Laffer curve, a national income tax (fairer and broader than a VAT), the abolition of sales taxes, and the disappearance of the mammoth health insurance poll tax.
All social groups from the bottom up to roughly the ninety-fifth percentile would pay less than today when including health insurance premiums. The working class (which pays a lot in sales taxes) and the middle class (for which health care costs are currently prohibitive) would see their disposable income boosted. Around the median, the tax rate would fall from 38% to 28%: about thirteen points of health insurance premiums would disappear, as would three points of sales taxes, all of which would be replaced by six points of national income tax.
Would this tax system harm growth? Would it be the end of America as we know it? History does not suggest so. As we’ve seen, similar levels of tax progressivity were reached in the 1950s, before the collapse in taxes at the top, the explosion of health care costs, and the rise of payroll taxes transformed the US tax system into an engine of injustice. It’s through collective spending on education, health, and other public goods that rich countries have become wealthy, not through the deification of a tiny minority of ultra-rich. If history is any guide, the prosperous nations of the future will continue to be those that invest in the success of all.