A Negative Income Tax for the 21st Century

Executive Summary

This document proposes a basic income for the U.S. in the form of a negative income tax (NIT).

It proposes an income floor for all adults (18+) in the economy of $13,000 per year, indexed to the federal poverty line. This proposal leaves aside the question of how best to include minors, but no basic income is complete without either a reduced rate income floor for minors, or a child allowance passed alongside.

Under this proposal, an adult earning $0 annually receives the full $13,000. As their earnings increase, their NIT benefits are phased out with a 33% tax rate, zeroing out benefits for incomes beyond $39,400. Benefits would be provided in monthly installments.

Such an NIT is both economically and politically feasible. Within a year of passing, it could eliminate official poverty, increase the circulation of capital throughout the economy, reduce inequality, and insulate a basic degree of livelihood from potential shocks such as loss of employment, automation, pandemics, or changing life circumstances.

In the longer term, a basic income improves the conditions for social innovations to occur alongside technological ones. Raising the economic floor lowers the risks associated with experimental behavior. It lifts everyone towards a threshold of financial security beyond which we have more freedom to decide how to spend our time. This kind of free time, contrasted with time that must be spent working for money, is fertile soil for innovation.

Moving towards collective liberation from the strictures of economic anxiety could provoke an upsurge in the kinds of broader innovation we need to meet the spectrum of challenges facing our civilization. Economic freedom is an intersectional catalyst towards ecological, racial, feminist, and mental freedoms.

I calculate this NIT would cost $855 billion per year. My funding proposal details two primary categories of revenue: reforming and eliminating existing welfare and tax programs ($346 billion), and implementing new progressive taxes ($744 billion). Together, these reach a total funding capacity of $1.09 trillion, $235 billion above my funding estimate. Crucially, the plan does so without raising the tax burden on most Americans.

What Is a Negative Income Tax?

NIT is a targeted basic income that phases out as one’s earnings increase. It’s composed of two variables: an income floor, and a phaseout tax rate. The income floor sets the amount an individual with $0 of annual income receives from the program: the maximum benefit amount one can receive from the NIT.

The phaseout rate determines how much of the NIT is phased out for each dollar of earned income. The lower the rate, the larger the category of people who benefit from the NIT program, and the more work incentives are preserved (because earning additional income loses you less of the NIT benefit). But the higher the programs cost. Conversely, higher rates mean fewer people benefit and incentives for low-wage work are decreased. But the cost is lessened.

Together, the phaseout tax rate and the income floor create a third element: the breakeven point. This is the earnings level at which NIT benefits are fully phased out to $0.

Why Negative Income Tax?

How is it that in a society as wealthy as ours, economic anxiety remains a prevailing reality for so many? This question has vexed economists at least since Henry George in the late 1800’s. In our time, after centuries of vigorous capital accumulation, we are past due to leverage our wealth to provide a basic income that reduces the hold economic insecurities have upon our lives, behaviors, and minds.

NIT is both an economic and politically feasible way to implement this basic income. It’s a policy with immediate and obvious effects, followed by more subtle, and perhaps profound ones.

Within a year of passing, NIT could:

  1. Eliminate poverty by setting the income threshold at or above the federal poverty line.
  2. Reduce inequality and increase the circulation of capital throughout the economy by redistributing wealth from its accumulation at the top back towards lower income groups who are more likely to spend it.
  3. Increase the bargaining power of workers by increasing their ability to meet their basic needs outside of employment, empowering them to more readily reject undesirable or exploitative labor contracts.
  4. Guarantee and protect a basic degree of livelihood from potential shocks such as loss of employment, automation, pandemics, or changing life circumstances.

But the long-run effects reach much deeper into the core of our social relations, ways of living, and modes of innovation. With a basic income, the risk attached to experimental behavior is greatly reduced. The spectrum of viable behaviors — ways of living, even — is increased. Unconditional income enables those who’d prefer spending more of their time engaged in unpaid activities, that they nevertheless consider valuable, to do so. As a society, we gain optionality. The ways of participating in society people consider legitimate is expanded, as recent interviews with UBI recipients from a Finnish experiment confirm.

The diversity (and redistribution) of ‘free time’ is precisely the point. Free time is unbound from the imperative to earn, from the necessity to please bosses, deadlines, or metrics. It’s time in which we can do things for themselves. The quality of thought that occurs for itself, as opposed to for a job, is notably different. More autonomous. Charles Darwin, after all, produced his theory of evolution in his free time.

The innovations that spring from free time arise from a different kind of investment. Unbound from the imperative to earn, they can address a broader scope of human concerns. These social innovations function alongside technological ones that arise from market incentives, contributing a broader range to the scope of innovation.

But perhaps the most urgent motivation to establish a basic income is because doing so would make an immediate difference in the everyday lives of ordinary people. People who’ve been left out of the gains from the extensive capital accumulation their labor contributes to.

Payment Details

This proposed NIT sets the income floor at $13,000 (rounded up from the 2020 poverty line of $12,760) and uses a 33% phaseout tax rate. The breakeven point beyond which benefits reach zero is $39,393.

The chart below shows how much NIT benefit one would receive at different levels of income:

This proposal does not cover a plan for minors. But it is absolutely vital to the efficacy of the program that such a plan is concurrently passed.

Most basic income proposals include a reduced rate for minors, often hovering around 50% of the adult level. This approach is not without oddities. Especially in the context of a NIT rather than UBI. For example, since almost none of the 74 million minors in the US earn income, and even fewer earn anywhere near a hypothetical reduced income floor of $6,500, almost all children — including those of eminently wealthy families — would receive the full benefit.

So even though the program is considered a NIT, it amounts to only an NIT for adults, but a UBI for minors.

It may be preferable to complement a NIT for adults with a child allowance that phases out benefits as guardians’ earnings increase. In this way, just as NIT benefits phase out before going to wealthier adults, benefits would phase out for minors higher up the wealth distribution as well.

However, some argue that children constitute a “public good”, and child allowances should be universal. I leave this debate untouched, trusting that either method — a reduced rate NIT income floor, or a child allowance (universal or otherwise) — complement and fulfill the motivations behind this basic income proposal.

In terms of cost, some child programs are budget neutral, while the higher end of cost estimates reach up towards $200 billion.

How Much Would This NIT Cost?

I calculate this NIT, provided in 2018 to all adults over the age of 18 would’ve cost $855 billion.

In addition to my own calculations, I use four existing cost estimates for similar programs to establish a spectrum of potential costs. This spectrum runs from $539 billion to $1.09 trillion:

Each plan slightly differs, but taken together, they create a spectrum to anchor our cost expectations. But details make all the difference. For example, the disparity between Widerquist and Santens’ estimates derives largely from differences in assumed tax rates. Wiederspan et. al’s NIT proposal is for household, rather than individual income. And Philip Harvey’s estimate was set at the poverty line with a 25.6% phaseout rate, with numbers given in 2002 terms.

For the present proposal, I calculated the cost using 2018 data from the current population survey. It gives the distribution of personal incomes in increments of $2,500, providing the quantity of individuals in each earning bracket. For each bracket, I took the middle of the bracket as the income for that group. For example, for the 11,090,000 individuals in the income group between $0 — $2,500, I treat their income as $1,250.

Since cost estimates must use income data from the economy without a basic income, we must take all estimates loosely. They cannot account for how introducing a basic income would lead to a change in behaviors and earned incomes that would alter the actual cost of the program. Whether we expect the cost to increase or decrease in response to these changes is unclear. There exist clear knock-on effects in both directions.

How to Pay for NIT

For my funding calculations, I chose to use Philip Harvey’s $1.09 trillion as the funding target rather than my own. I do so to demonstrate that whatever the cost, an achievable funding proposal is well within reach.

Funding Summary

A $1.09 trillion NIT program could be funded through two categories of funding. First, eliminating or reforming existing welfare and tax programs. Second, implementing new progressive taxes.

We could eliminate redundant welfare programs such as the earned income tax credit, supplemental security income, supplemental nutritional assistance program, and temporary assistance for needy families (total revenue of $198 billion). We could eliminate wealthy-favoring tax programs like the mortgage interest deduction and the real estate tax deduction ($89 billion). And we could reduce the defense department’s budget by 10% ($59 billion).

In total these provide $346 billion towards the cost. I then offer two strategies for approaching the remaining $744 billion via progressive taxation.

We may either enact one single, large tax, or we may prefer implementing a series of smaller progressive taxes. Options for the single tax include a value added tax (raising $600 billion — $1.3 trillion in new revenue depending on exclusions), or an adaptation of Saez & Zucman’s proposed national income tax that draws upon both labor and capital. They estimate a 6% flat rate could raise $1.2 trillion. I estimate phasing the tax in on incomes above the NIT breakeven point to a top rate of 5% would be sufficient to cover the remaining $744 billion.

For the series of smaller taxes, we could fully fund the NIT with: progressive top income tax rates ($18.9 — $70 billion), carbon tax ($100 — $210 billion), wealth tax ($118–375 billion), financial transaction tax ($70 billion), and raising the effective corporate tax rate ($235 billion).

Funding Details

Repurposing Existing Programs:

A sizable portion of the expense can be offset by folding existing welfare programs that a NIT would make redundant. These programs might include: the earned income tax credit ($59 billion), supplemental security income ($58 billion), supplemental nutritional assistance program ($64 billion), temporary assistance for needy families ($16.7 billion).

It’s crucial to emphasize, however, that NIT must not replace welfare, but place it on firmer ground. An NIT that replaces all other welfare programs (of the variety Milton Friedman or Charles Murray advocate) would have harmfully regressive outcomes. The programs listed above are means-tested, meaning most recipients would naturally be elevated above their eligibility requirements anyway.

Together, these programs would provide $198 billion towards the total cost.

In addition to these reforms, we spent $540 billion in 2013 on “tax programs”: tax credits, deductions, exclusions, exemptions, deferrals, and reduced rates, most of which overwhelmingly benefit the wealthy. Indeed, these tax programs are known as “asset building” programs, lowering the tax rates on capital so that wealth can accumulate.

But in 2013, we spent over $100 billion more on these asset building tax programs than we did the entire welfare program budget (excluding medicare/medicaid):

These tax programs, which function as federal expenditures, can be reformed to better serve most Americans, and to help offset the cost of a NIT. For example, eliminating the home mortgage interest deduction — an idea with bipartisan support — is estimated to save $59.7 billion per year. Similarly, eliminating the real estate tax deduction would have freed up $29.3 billion in 2013.

Together, eliminating these two programs alone could contribute $89 billion annually, leaving a mass of wealthy-favoring tax programs still in place.

Finally, we may consider reducing the defense department’s budget by 10%, yielding an additional $59 billion, annually.

Combining all of these tax reforms, we so far have $346 billion available in annual funding for the NIT:

The remaining $744 billion can be raised through a variety of progressive taxes.

There are at least two strategies for progressive taxation reform to achieve $744 billion in additional revenue. We may either look to enact a single, large tax, such as a value added tax, or a national income tax. Or, we may prefer enacting a series of smaller progressive taxes — such as a carbon tax, financial transaction tax, wealth tax, higher top income tax rates, and higher corporate tax rates.

One Large Tax

I’ll offer two single tax options. The value added tax (VAT) is familiar to most of the developed world (166 of 193 countries with UN membership have one in place). Some economists worry, however, that it’s a regressive tax suited more to the 20th century economy than the 21st.

As a progressive alternative to the VAT tax, economists Gabriel Zucman and Emmanuel Saez proposed a national income tax in their 2019 book, drawing upon both labor and capital income. They describe it as a way to “leapfrog the VAT”, paving the way in creating “fiscal institutions of the twenty-first century.”

Value Added Tax

A 10% VAT tax is projected to raise approximately $600 billion per year, whereas the same 10% VAT applied to a broader base of goods is estimated to raise up to $1.3 trillion per year.

Empirical studies are increasingly finding that a VAT is regressive (hits those with lower incomes harder than those with higher incomes), though debate remains as to precisely how regressive. Some UBI proposals account for this regressivity by setting the payout level — or in our case, the NIT income threshold — $200 above the poverty line.

Alternatively, a basic income can justify reducing the social security tax rate. Since social security is essentially a basic income for seniors, there’s less need to maintain its payout levels going forward, since they’ll be receiving basic income from the NIT program. Lowering the tax rate will lead to lower future benefit levels (without robbing those who’ve already paid into the program of their due benefits), but the basic income will more than make up the difference. This decreased tax rate on all (for example, from 6.2% to 4%) can help offset additional costs the VAT passes on to consumers.

National Income Tax

The national income tax was proposed as a 6% flat tax on all income, labor and capital, with no exemptions or deductions. In their proposal, Saez and Zucman (2019) estimate $1.2 trillion in annual revenue.

Adapting this tax to the NIT, I calculate a 5% national income tax phasing in on earnings above the breakeven point of $39,393 would be sufficient to fund the remaining $744 billion.

Originally, the negative income tax proposed by Milton Friedman was funded in quite a similar manner. Friedman wanted to keep it simple: If you earn below the breakeven point, you receive money. If you earn above it, an income tax phases in to fund the program.

But as we learn more about the nature of income and wealth inequality, it’s clear that the wealthy make most of their income via capital income, while working classes rely on wages. Therefore, an income tax applied only to labor incomes hits working classes much harder than the wealthy. Applying the tax to all forms of income means the wealthy pay the same proportion of their real incomes towards the program as everyone else.

The primary challenges to this approach — intuitive as it may seem that the wealthy shouldn’t get to pay a lower percentage of their overall income than others — are fiscal manipulation and capital flight. The tax would effectively raise rates on corporate profits, incentivizing the movement of money abroad. International tax competition, which has become one big race to the bottom, makes this sort of taxation tricky.

Accordingly, as Saez and Zucman acknowledge, the successful implementation of this kind of tax requires reforming our approach to the taxation of multinational corporations. Incidentally, Zucman wrote both a book and a policy brief on doing just that.

A Smaller Series of Taxes

If the VAT is too regressive, and the national income tax proves infeasible for the time being, we may prefer to use this as an opportunity to implement a series of smaller taxes that raise revenues while adjusting the economy’s architecture to better serve the majority of the American people, the planet, and our collective pursuit of progress.

Progressive Marginal Income Tax Rates

Recent research suggests the revenue maximizing top tax rate on incomes falls between 68–73%. Presently, our top tax bracket is 39.6% on earnings above $400,000. Much debate has ensued over the question of raising top marginal income taxes, largely spurred by Saez & Zucman’s recent finding that in 2018, the wealthiest households paid a lower tax rate than the poorest:

Revenue estimates for raising the highest income tax rates vary from $18.9 billion, up to $70 billion annually, largely dependent upon the broader environment of tax policy.

It’s worth noting that the US pioneered high marginal income tax rates throughout the mid-20th century. Top rates reached above 90%, with effective rates sitting closer to 50%. There’s nothing unprecedented about proposing a return to higher marginal income tax rates.

Wealth Tax

Projected revenues: $118 billion — $375 billion.

A 2% tax on wealth above $50 million, cranking up to 3% on wealth above $1 billion, is projected to raise $275 billion, annually.

Carbon Tax/Cap-and-Trade

Projected revenues: $100 billion — $210 billion.

A tax of $25 per metric ton on most greenhouse gas emissions in the US could raise slightly over $100 billion annually, while a tax of $49 per metric ton of carbon dioxide could raise closer to $210 billion annually.

There are two important elements to the debate over carbon tax. First, a successful carbon tax would yield diminishing revenues, moving towards zero. Second, there’s a design question: a carbon tax sets a price on carbon and lets emissions calibrate organically, while a cap-and-trade approach sets an emissions level and lets prices calibrate organically.

Each strategy has particular strengths and weaknesses that may complement each other well in a mixed approach.

Reforming and Raising the Corporate Tax Rate

Projected revenues: $100 billion (reforming loopholes) + $135 billion (raising rates) = $235 billion

379 of the Fortune 500 companies paid an effective federal tax rate of 11.3% on their 2018 income, 9.7% less than the actual corporate tax rate of 21%. 91 of those corporations — including Amazon, Chevron, IBM — paid $0 in taxes on their 2018 income.

Reforming the corporate tax rate — largely by reconsidering deductions, allowances, and loopholes — can help raise the actual rates they pay closer to the stated tax rates they’re meant to. In 2017, a perfect application of the 21% corporate tax rate would’ve raised $421 billion. Actual revenues from the corporate income tax in 2017 were $297 billion.

But changing the stated rate matters too. The 2017 tax cuts reduced the corporate tax rate from 35% to 21%, leading to a $135 billion decline in revenue.

Financial Transaction Tax

Projected revenues: $70 billion

A FTT, beyond raising revenue, has a significant corrective effect on the unbridled incentives of the finance industry. By discouraging high-frequency, short-term trading, a FTT discourages the short-termism that has plagued the economy since deregulation loosened capital controls in the 1970’s.


The series of taxes listed above are projected to raise between $542 billion — $960 billion. Our required revenue of $744 billion falls nicely in the middle. Further options exist, such as raising capital gains taxes, eliminating more ‘asset building’ tax programs, rental fees on natural resources (such as the Alaskan model of taxing oil revenues to provide a partial basic income to all citizens), or financing a portion of the cost through deficit spending.

Administration Infrastructure

The actual administration of the NIT — scaling benefits appropriately with fluctuating incomes on a monthly basis, and distributing those payments to all recipients — presents both obstacles, and opportunities. Presently, we struggle enough with reporting our incomes annually. It is difficult to imagine citizens reporting their incomes monthly, in order to receive appropriately sized NIT checks.

But we’re well overdue to redesign our income and tax reporting systems in light of the latest digital technologies and platforms. We certainly don’t lack the public interest, nor political will, to simplify tax reporting systems.

Consider, for example, a digital income reporting platform that each citizen is responsible for updating monthly (better yet, remove the friction by automating the process, so that any income earned automatically registers on this ledger). Monthly NIT payouts can adjust in response to these income declarations. Then, when we do our annual taxes as usual, our official annual incomes are available to resolve any discrepancies.

Individuals who under-report their incomes on the monthly platform, therefore receiving more NIT than they should, can pay it back in the form of a tax penalty (and perhaps with an additional fine, disincentivizing systematic underreporting to effectively treat extra NIT payouts as a line of credit).

Barring a new digital income reporting infrastructure, there still exist feasible options. First, as economists of all stripes agree, from Milton Friedman to Thomas Piketty, the ideal place to provide NIT benefits and tax adjustments is directly on one’s paycheck. In the same way that each paycheck lists federal tax deductions, it can list one’s NIT benefits (and tax contributions) applied to that very check. But this method may grow more difficult as work itself evolves, and the ways in which we earn income transition from single-employer careers to kaleidoscopic, diversified models.

If the evolution of work prevents paychecks from providing the bulk of NIT calculations, we can still adopt an income declaration system using existing platforms. In the same way that unemployment recipients must actively “claim” weekly benefits, receiving monthly NIT could depend on actively reporting one’s income to the NIT administering agency, such as the IRS. In the same way as above, we can adjust for any discrepancies when we do our annual taxes, using fines and penalties to discourage exploitation.

Requiring recipients to actively report their incomes also builds in a frictionless ‘opting-out’ method for all those who don’t expect to receive any NIT benefit. Someone earning $200,000 annually, well over the breakeven point, would be required to take no action. By default, no one receives NIT. So only those who expect to receive payments would ‘opt-in’ by self-reporting their income. This naturally reduces the bureaucratic load.

In a proposal for a social wealth fund, The People’s Policy Project provides a visual aid that models a potentially digitized NIT platform nicely.

Atop the left screen where “value of your share” is listed, this could be the space where one’s reported income is shown. The screen on the right, rather than showing one’s “dividend redemption”, could show the NIT benefit to be deposited into the recipients account that month.

Just as Saez & Zucman see the implementation of a national income tax as an opportunity to bring our tax system into the 21st century, digitally reporting incomes provides an opportunity to update how we do our personal taxes.

Next Steps

Two major obstacles to implementing an NIT are the lack of rigorous funding and administration proposals. Here, I offered suggestions for both, upon which I hope more capable researchers will build.

There is also a lack of research into optimal phaseout tax rates. I use 33% because it’s the lowest rate commonly used in existing NIT research. But the lower the rate, the more people benefit from the NIT, and the more potent it becomes as a vehicle for robust change. Research is needed to explore the lowest sustainable phaseout rates.

This leaves four next steps:

  1. Debate over the most appropriate funding model
  2. Determine best approach for minors: cash allowance, or reduce income threshold?
  3. Develop income reporting & payment distribution methodology
  4. Research optimal NIT phaseout rates

Whatever disputes may arise over implementation details, it’s crucial to remember that all advocates of basic income — whether UBI or NIT — stand on common ground. Broader still, this common ground extends beneath the entire sweep of policies seeking to leverage our collective wealth to directly alleviate economic insecurity for all.

Rather than waiting for aggregate-GDP growth to fulfill the neoliberal promise — where economic growth is said to do more for human well-being than any direct interventions can — we can immediately begin providing what a recent proposal for universal basic services calls “a larger life for the ordinary person”.

Cutting everybody directly into the fruits of capital accumulation increases optionality for those with lower incomes in the same way that wealth does for those with higher incomes. This allows for greater diversity in what we do with our time, how we organize our lives, and ultimately, the kinds of wealth and value society produces.

A negative income tax of the variety proposed here offers a crucial piece of the puzzle for an immediate, direct, and feasible strategy towards actualizing these possibilities today, in our time.


This proposal was originally published on Oshan’s website: www.MusingMind.org

Interested in many things, like consciousness, meditation & economics. Sure of nothing, like how to exist well, or play the sax (yet). More: www.MusingMind.org.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store