Senators Elizabeth Warren (D‑MA) and Bernie Moreno (R‑OH) propose removing the Social Security payroll tax cap, subjecting earnings above $184,500 to the 12.4 percent combined employer-employee payroll tax. Senators Chris Van Hollen (D‑MD) and Cory Booker (D‑NJ) each propose exempting more wages from income taxes at the bottom while raising taxes on higher earners. President Trump has pursued a similar strategy of expanding tax exemptions, while President Joe Biden and Vice President Kamala Harris both pledged not to raise taxes on anyone earning less than $400,000.

Each approach shifts more of the tax burden toward the top. One problem with this approach is that there are not enough high-income Americans to finance the current federal budget deficit, let alone fund additional spending or tax cuts.

One simple way to illustrate the mathematical impossibility of raising taxes only on rich people is to ask an intentionally extreme question: How much income is actually left to tax at the top? Not as much as popular proposals usually assume.

Using IRS data, the post below shows an upper bound for income-tax increases on high earners. In 2023, if the government had confiscated every dollar earned over half a million dollars, it still would have run a budget deficit.

What’s left to tax?

Using IRS data from the 2023 tax year (the most recent available), we can illustrate the difficulty of raising a lot more revenue from a narrow segment of the population.

In 2023, taxpayers filed 161 million individual income tax returns, reporting $15.3 trillion in adjusted gross income (AGI). AGI includes wages, capital gains, personal business income, and other forms of income, minus adjustments for things like student loan interest and retirement contributions.

The IRS reports this information by different income groups, separating taxpayers into buckets with AGIs above and below $200,000, $500,000, $1 million, and $10 million, among others. Table 1 shows the total AGI and income taxes paid, including federal taxes and an estimate of state-level taxes, by each group.

In 2023, taxpayers earning over $1 million reported $2.5 trillion in total AGI and paid $747 billion in federal and state income taxes. To estimate state income taxes, we apply average rates by income group from the Institute on Taxation and Economic Policy. The 799,094 tax returns in the $1 million+ group accounted for 0.5 percent of all returns and paid an average federal and state income tax rate of 29.5 percent.

In theory, Congress could devise a way to reach every dollar of untaxed millionaire income. But most proposals to raise taxes on high earners instead start by increasing marginal tax rates. Under a graduated income tax, a higher rate imposed above $1 million applies only to income exceeding that threshold, which exempts the taxpayer’s first $1 million from additional taxes.

The IRS data show that for the $1‑million-and-above group, there is $1.7 trillion in AGI above the threshold. Applying the group’s average tax rate implies they have already paid roughly $512 billion in taxes on their above-threshold income. That leaves $1.2 trillion after taxes.

If Congress confiscated every one of the remaining $1.2 trillion after-tax dollars earned above $1 million, the resulting revenue would have fallen nearly $600 billion short of covering the cost of the 2023 $1.8 trillion calendar-year deficit. Dropping the taxable income threshold to $500,000 would also have fallen just short of covering the same year’s deficit. And these estimates make the wildly unrealistic assumption that a 100 percent marginal tax rate would have no behavioral or other economic effects.

Figure 1 extends the improbable assumption over 10 years, assuming that high-income Americans would continue to earn income when facing 100 percent income tax rates. It adjusts the 2023 data by projected income and household growth to show untaxed income over the next 10 years. Confiscating all income earned over $1 million would cover only about 80 percent of the Congressional Budget Office’s (CBO) projected $24.4 trillion federal deficit over the same period.

The Committee for a Responsible Federal Budget produces a more realistic projection of future deficits that assumes Congress extends many expiring tax and spending programs (which the CBO is required to assume are not renewed). At the more likely deficit figure of $29.4 trillion, even lowering the income threshold to $500,000 does not cover the next decade’s budget shortfall.

Lowering the taxable income threshold further to $200,000 expands the pool of untaxed income, but it does not make confiscatory tax rates economically plausible.

Common sense and economic incentives make clear that Congress cannot raise marginal income tax rates anywhere close to 100 percent and expect taxpayers to continue earning and reporting the same income. A recent report by economists at the Joint Committee on Taxation estimates that combined state and federal income tax rates are already near their revenue-maximizing level. Raising top income tax rates further would result in revenue gains of about 0.1 percent of GDP, equivalent to at most $400 billion over the next decade.

Conclusion

Taxing incomes at 100 percent marginal rates is not a realistic policy proposal. Taxes significantly higher than what we have today would radically change how much people work, invest, and realize as income, as well as how much income they report to the government. The point of this exercise is to show that “just tax the rich” proposals fail, even under arithmetic that is the most favorable possible.

The DSA platform claims that the bill for all this will be sent to “the richest individuals and corporations.” Tally up that bill, and it ballparks between $71 trillion and $212 trillion in new spending over the next decade. Confiscating every dollar of high-end wealth and corporate profits would cover only a fraction of those costs. The DSA agenda necessitates high taxes on middle-class Americans.

$200 Trillion in New Spending

Totaling up nine of the largest proposals in the DSA platform would mean new federal spending equivalent to between 18 percent and 53 percent of GDP.

Table 1 reports various low-end and high-end estimates of proposals for programs that approximate the DSA’s vague descriptions. Each proposal’s original spending estimate is converted to a share of GDP and then applied to the 2027–2036 projected GDP, so all estimates are in current dollars.

Medicare-for-All-style proposals for universal healthcare would increase federal spending by $40 trillion to $75 trillion over 10 years. Reparations, a federal jobs guarantee, infrastructure, green energy investment, larger retirement benefits, free housing, paid family leave, and no-cost college would increase spending by tens of trillions of dollars more. In total, the DSA’s new spending would cost between $71 trillion and $212 trillion over the next decade.

This exercise is inherently imperfect, which is why the estimates vary so widely and should be understood as orders-of-magnitude estimates. They likely overstate the cost where programs overlap with each other or existing spending. They understate the cost by failing to fully capture behavioral responses, broader economic effects, and the comprehensive scope contemplated by the DSA. Each estimate comes from different authors using different methods and assumptions, and builds on a similar methodology by David Burton.

Internationally High Spending

In the US, federal, state, and local governments spent almost 40 percent of GDP in 2024. The average across the European Union is 49 percent, ranging from 58 percent in Finland to 22 percent in Ireland.

Using the lower-bound estimates, the DSA agenda would raise US spending to more than 57 percent of GDP. Among large, industrialized countries, only Finland would have a larger government. France comes in a third of a percentage point under the US’s low estimate. Add the high-end estimates, and US government spending would reach 92 percent of GDP.

No comparable country on Earth spends anywhere close to that amount. The DSA agenda’s spending could give the government a claim on national output much closer to estimates of state control under Soviet-style communism than to today’s European welfare states.

Who Pays?

The federal government is projected to collect $70 trillion in taxes over the next decade, roughly 18 percent of GDP. Paying for the DSA agenda would require roughly doubling federal revenue at the low end and quadrupling it at the high end, in addition to the revenue needed to cover the Congressional Budget Office’s $24 trillion projected ten-year deficit.

The DSA suggests that the richest Americans and corporations will pay for all these new outlays. The problem is, there simply aren’t enough resources at the top to make this plan work.

The 400 wealthiest Americans were worth a record $6.6 trillion in 2025. Confiscating all of their wealth would cover only about 9 percent of the low-end revenue requirement and 3 percent of the high-end estimate. Their wealth could be seized only once, and attempting to liquidate trillions of dollars in assets would, in turn, drive their value down.

Domestic corporate profits after federal taxes are projected to be about $35 trillion over the next decade. Seizing every additional dollar of corporate profits would fund half of the low-end estimate and 17 percent of the high end. This also assumes that firms continue operating normally while the government takes every cent of profit. Without a profit motive, businesses would cease to exist.

Higher earners are also not a source of vast untapped revenue. A recent report by economists at the Joint Committee on Taxation concluded that raising top federal income tax rates to their revenue-maximizing level would result in revenue gains of less than 0.1 percent of GDP, equivalent to roughly $400 billion over a decade at today’s projected GDP levels.

The entire wealth of the richest Americans, plus every dollar of corporate profit and maximum top income tax rates, still leaves the DSA agenda between $29 trillion and $169 trillion short.

The only remaining source of revenue large enough to cover the DSA agenda is the same one every large European welfare state relies on: the middle class. France and Finland don’t fund their large governments by only taxing billionaires. They impose high income, payroll, and consumption taxes on ordinary households.

To cover the DSA’s high-end spending estimate and current deficits, every $1 the federal government collects today would need to become about $4.36. Mechanically applying that increase to individual income-tax rates would push the 24 percent bracket above 100 percent and the top rate above 160 percent.

The DSA is promising Americans a world in which someone else will pay for potentially hundreds of trillions of dollars in new benefits. The problem is that there aren’t enough rich people or corporations to pay for Democratic Socialism. Eventually, the bill will come for the rest of us.

The announcement fits nicely with a piece I wrote for the Washington Examiner in May about a growing tension in Republican tax policy between the traditional goal of a broad-based, low-rate tax and the Trumpian impulse to use the tax code to reward particular activities and important constituencies. The piece never made its way onto Liberty Taxed. Now seems like a good time to revisit this ongoing tension.

The full piece is republished below:

Republicans’ massive 2025 tax bill was built on two opposing theories. One should be familiar to anyone following Republican politics over the past several decades: lower tax rates, a broader tax base, and better treatment of investment. The other is newer: explicitly using the tax code to reward politically important interest groups.

That contradiction lies at the center of the One Big Beautiful Bill Act. It also reveals a deeper question facing the American Right: Does it still believe tax reform is possible, or has it instead turned to using the tax code for social engineering, once the primary province of the Left?

For decades, Republican tax policy had a clear North Star. From the Hall-Rabushka flat-tax plan in the 1980s to Herman Cain’s 9–9‑9 plan and reform proposals from then-presidential hopefuls Marco Rubio and Ted Cruz, the debate was usually about how best to move toward a flatter, simpler, more neutral tax system. The logic was so infectious that even Democrat Jerry Brown’s 1992 second-place finish to Bill Clinton included a 13% flat tax. Details differed, but low rates on a broad base were the shared goal of tax reform.

The basic insight across all these proposals was straightforward. The tax system should raise the necessary revenue to fund a limited government while causing the least possible damage to economic growth, work, and investment. To do this: keep tax rates low, treat similar people similarly, and limit the carve-outs that turn the tax code into Swiss cheese.

Much of the 2025 tax law still reflects the older philosophy. It preserves lower marginal tax rates from the first Trump term’s tax reform, maintains a larger standard deduction, improves incentives for business investment through permanent expensing, and extends some limits on itemized deductions. These are significant achievements but mostly extensions of the 2017 reforms.

The most economically important components are the lower rates and investment tax cuts. But they are not the political selling point.

The big campaign banners and TV ads highlight “no tax on tips,” “no tax on overtime,” “no tax on Social Security,” “no tax on American-made car loan interest,” and government-funded child investment accounts. In total, the law adds or expands almost two dozen targeted preferences aimed at sympathetic constituencies.

That may be smart short-term politics. But it is poor fiscal policy and will become increasingly unpopular over time.

A broad tax base with low rates treats everyone the same. But every new carve-out moves in the opposite direction. A teacher and a waiter who make similar incomes now face very different tax burdens. Up to $25,000 of the waiter’s income isn’t taxed. The teacher pays full freight.

Vice President JD Vance’s now-infamous, miserable “childless cat ladies” comment and his related suggestion that they should pay higher taxes are a perfect illustration of how perverse the tax code already is. Thanks to the child tax credit and half a dozen other child-related tax subsidies, the childless cat lady already faces a tax penalty for choosing not to have children.

Complexity breeds unfairness, and unfairness fuels lobbying for new carve-outs. Temporary breaks get extended, excluded groups demand special treatment, and industries fight to keep existing favors.

This is how tax systems decay: one exception at a time. It’s not a new problem. It’s precisely the dynamic that earlier generations of tax reformers sought to solve.

It should worry fiscal conservatives for another reason. As more income is exempted, pressure grows to raise rates on those still paying or create new taxes elsewhere.

This dynamic is already visible on the Left. Sens. Cory Booker (D‑NJ) and Chris Van Hollen (D‑MD) have both proposed exempting large amounts of income from tax, paired with higher taxes on high-income earners and family-owned businesses. On the Right, a similar impulse has driven many previously anti-tax crusaders to embrace century-high tariff rates and celebrate the billions in revenue they generate.

As more and more income is carved out of the tax base, the remaining taxpayers must bear a larger share of the burden. When current taxes fall short, lawmakers will add new ones: a carbon tax, a value-added tax, or new taxes on wealth and assets.

These preferences endure because their true cost is easy to disguise. Supporters campaign for them as tax relief for ordinary workers, but someone must pay. If the goal is to help workers, simple, transparent systems are better than complex, opaque subsidies embedded in the tax code.

The fundamental question facing Republicans is philosophical. Is the tax code supposed to fund a constitutionally limited government with as little economic harm as possible? Or is it simply another political tool to direct rewards and punishments? Those are very different views of what the fiscal system is for.

When the temporary “no tax on [fill in the blank]” policies expire at the end of 2028, Republicans will need to confront these two conflicting views of reform, which are really two philosophies of the role of the state in private affairs.

The old Republican consensus isn’t dead yet — far from it. Its core principles shaped most of what ended up in the One Big Beautiful Bill Act. But it now also shares space with a more populist instinct that sees every grievance, industry, or voting bloc as a candidate for special treatment.

The good news is that the older path remains open. Republicans can still return to the broad-base, low-rate, flat-tax North Star.

Californians will vote on a 5 percent tax on billionaire wealth this November. In Congress, Sens. Elizabeth Warren (D‑MA) and Bernie Sanders (I‑VT) are pushing plans for an annual wealth tax. And abroad, Norway and Spain have expanded their wealth taxes, and some countries want to impose a global wealth tax.

Our new Cato study, “Failures of Wealth Taxation,” discusses why these efforts are misguided. The study describes how a dozen European countries have tried wealth taxes but then repealed them in failure. Wealth taxes raise little revenue, drive out investment and entrepreneurs, and become riddled with loopholes.

Here are a few highlights:

Countries repealed wealth taxes. Austria, Denmark, Finland, France, Germany, Iceland, Luxembourg, the Netherlands, and Sweden repealed their wealth taxes as the economic damage and administrative costs were too high. Colombia, Norway, and Spain still have wealth taxes but face similar problems.

Low wealth tax rates are not low. A wealth tax is based on the value of assets, not annual returns, thus a seemingly low 3 percent annual wealth tax on an asset earning 5 percent is equal to a 60 percent income tax.

California’s proposed tax could lose money. Proponents say the initiative will raise $100 billion, but after taxpayer responses, it may only raise $40 billion. Indeed, the state would likely lose money overall because the income tax base would shrink from departing billionaires.

Workers pay a price. Counter to the views of wealth tax advocates, capital and labor are complements. Taxing wealth means taxing the business assets of high-end portfolios. Those assets support worker productivity and wage growth, so taxing wealth is ultimately an anti-worker policy.

Wealth is good. Rather than being concentrated, wealth in America is dispersed across the economy in productive business assets. Among the top 0.1 percent of households, 73 percent of net wealth is equity in private or publicly traded companies, and almost three-quarters of American billionaires are self-made. All wealth benefits the economy, and policymakers should reduce the barriers to wealth creation for all families.

Tax consumption instead. America does not need new taxes but rather to fix the complex and unequal tax systems already imposed. Congress should shift the federal tax code toward a consumption base to reach every dollar the wealthy spend without penalizing the investments that benefit workers and the economy.

Read the full study here.

The app prominently features a tool to estimate the account’s value based on parental contributions, and it pushes people to “make a contribution” by starting “with as little as $1.”

The problem—as I’ve described in the Wall Street Journal and in a much more detailed Cato Policy Analysis—is that the growth can happen in any investment account; it’s the tax treatment that matters.

Nowhere on the future-value page does it say the government will tax you on the gains as ordinary income. You have to dig deep into the app to find this information. It also doesn’t say there are likely other investment accounts where the after-tax value of your deposits could be higher.

By leaving out the tax treatment, the app is misleading. The gain on money withdrawn from a Trump Account is taxed as ordinary income (the rate on your paycheck). The same gains on an after-tax contribution made to a 529 plan and spent on college are tax-free. Gains in a taxable brokerage account face the lower long-term capital gains rate. That rate could be zero for lower-income Americans.

In an apples-to-apples comparison, Trump Accounts are one of the least tax-advantaged ways to save for your kid’s future. By pushing people to put their own after-tax money into Trump Accounts, the app will very likely make some Americans worse off.

So, while watching Argentina eke out a win over Egypt in the World Cup, I vibe-coded a fix for the Trump Account app’s core defect. (If you see anything amiss, let me know.)

The widget below makes the same assumptions as the Trump Account app, but it adds the tax effect, along with a side-by-side comparison with a 529 plan and a taxable brokerage account. For identical contributions, spent on qualified expenses, the Trump Account finishes last because its earnings are treated as ordinary income. Click here or on the image below to see the interactive version.

A few results don’t fit neatly in the widget. Even with the $1,000 government deposit in the Trump Account (and nothing extra in the others), the handout doesn’t cover the tax bill in many scenarios. This is especially true at higher tax brackets and when parents contribute a lot of their own money. The widget also leaves out the 10 percent penalty on non-qualified Trump Account withdrawals. That penalty pushes the Trump Account further behind an unrestricted taxable account, which you can tap anytime, for anything, at capital gains rates. It also does not show the tax benefit from pre-tax employer contributions to Trump Accounts and many other complexities.

Congress could fix Trump Accounts by allowing families to deduct their contributions, following the tax treatment of traditional IRAs. Ideally, it would also scrap the subsidy and the restrictions, so Americans can save and spend their own funds on whatever they need without penalties from Washington.

Read more about Trump Accounts and how to fix them here.

Our chapter in the new book, A History of Repeated Injuries, argues that the standard understanding of taxes at the founding glosses over a richer story with implications for fiscal systems today. If English taxes hadn’t been discriminatory, a revolt over taxation without consent might never have happened.

Far from a revolt over taxes being too high, the colonists’ Tea Party was a protest against a targeted tax cut. By exempting the British East India Company from taxes on tea, Parliament undercut domestic sellers to create a tax-subsidized monopoly. The colonists revolted, throwing the company’s tea into the harbor. The Boston Tea Party was more a protest against a corporate bailout than a broadly burdensome fiscal system.

In 1763, the average British citizen paid about 26 shillings a year in taxes. The average New England colonist paid about one shilling, and many understood that taxes would have to rise if the independent colonies were to fund their own defense. In addition to other grievances, we rebelled over the way taxes were imposed, not the size of the bill.

War debts forced the young states to raise domestic taxes to pay bondholders, and it was widely understood that government debt enriched the wealthy urban elites who held the bonds. The taxes that funded repayment, meanwhile, fell disproportionately on poorer rural farmers. The result was resentment and repeated violent rebellion. Yet the protesters went out of their way to make clear they were objecting to the inequities of the tax design, not the principle of raising revenue in general.

This is not a uniquely American conclusion. A recent empirical assessment of the French Revolution found that unrest was concentrated on the high-tax side of France’s internal fiscal borders and was strongest where fiscal disparities were largest. The authors conclude that the discontent that toppled the monarchy “was fueled not only by tax levels but also by how inequitable the system appeared.”

Similar tensions played out in debates between anti-Federalists and the Federalist supporters of the new Constitution, which included language intended to ensure uniform treatment under the law, including tax law. The widely held concern that taxes could become oppressive, in both magnitude and design, led to constitutional safeguards. Unfortunately, these protections were insufficient to keep the anti-Federalists’ warnings of an increasingly intrusive federal fiscal system from materializing with the introduction of the Sixteenth Amendment and the modern income tax.

Two hundred and fifty years ago, the Founders taught us something simple: a tax system earns consent when it treats people equally, and it invites discontent when it doesn’t. We’ve drifted a long way from that standard. Today’s tax code is a sprawling instrument of privilege and punishment. Just look at the two most recent large tax bills passed by Congress. The Inflation Reduction Act included a trillion dollars in energy subsidies for politically connected industries, and the One Big Beautiful Bill Act added a dozen new carve-outs that reward important political constituencies at the expense of everyone else.

The challenge today is not simply to lower taxes but to design a system that is transparent, neutral, broad-based, and resistant to political manipulation. That system is a flat consumption tax.

My reform idea is straightforward: let parents deduct their contributions to Trump Accounts, so the investments are actually shielded from the tax code’s double tax.

You can read the full piece here.

I end the op-ed by arguing that Congress should also relax the lock-up rules and penalties that apply to Trump Accounts and instead allow the savings to roll into a more flexible Universal Savings Account at age 18, rather than a retirement-focused IRA.

However, part of my argument was left on the cutting-room floor. The $1,000 government subsidy is the root of the Trump Account’s complexity. Once you start handing out other people’s money, Congress attaches strings. That’s not a bad impulse on its own, but the problem is that the strings attach to personal contributions as well, not just the government deposit. It’s the resulting lock-up rules and tax penalties for non-authorized access that make the accounts less useful. It is also the root of the confused tax treatment that makes Trump Accounts the least tax-advantaged savings account available to families.

Beyond making the accounts functionally useless outside the government transfer, the subsidy also fails to meet most of the other goals its proponents claim for it.

For example, it is unlikely to mint a new generation of capitalists who come to see the benefits of markets by sharing in their growth. It’s a good story, but there’s not much evidence to back it up. As I detail in my recent report, in the US and around the world, the growth of private retirement savings is not associated with rising conviction in markets. If anything, the opposite is true.

It is also not free money. The government is borrowing at increasingly high interest rates, betting that stock market returns will compensate. This shifts market risk onto taxpayers and adds to the already large fiscal pressures facing the US government. And those risks are unlikely to stay small. Trump Accounts currently cost about $3.5 billion a year. Once the accounts exist, every future Congress faces incentives to increase deposit sizes, add new types of matching incentives, and expand eligibility.

Kill the handout, and the Trump Account structure can be easily simplified. No seed money means no lock-up to protect it, and no penalties to enforce the lock-up.

Government subsidies are neither a way to inoculate American youth against socialism nor a magic arbitrage machine whereby Washington borrows ad infinitum, invests the gains, and leaves everyone better off. There is no free lunch. Congress should focus on lowering taxes and simplifying the rules that keep people from saving their own money, not on redistributing everyone else’s.

You can read my full Cato policy analysis on Trump Accounts here.

The proposals differ in detail, but they cite the same story: Labor’s share of national income has declined, and AI is accelerating workers’ losses by shifting more economic activity to capital. The tax code makes all of this worse by undertaxing capital relative to labor.

Every part of that story is wrong.

Capital Versus Labor?

The capital versus labor story is ubiquitous. In an October 2023 interview before he became Treasury secretary, Scott Bessent claimed that “capital has gotten treated better to the detriment of labor” since the 1980s.

The AI optimists tell a version of the same story. Sam Altman, CEO of OpenAI, fears that AI could “break capitalism” through a “shift of leverage from labor to capital” and undermine the role of traditional work in our economy. Anthropic CEO Dario Amodei regularly cites similar concerns.

In theory, productivity-increasing technologies can replace or complement human labor. New technologies have always replaced some jobs, but in the process, they have created entirely new industries, expanded overall output, and enhanced the value of human inputs and, thus, their wages.

As I explain in a new piece for GIS Reports, the labor is losing to capital story does not show up in the data:

In standard economic models, output is attributed to the combination of labor, capital and technology. Each component can be thought of as earning a share of national income. If, over time, capital became more important for economic output, capital’s share of national income would increase. Empirical evidence does not support this claim.

Figure 1 (below) uses data from the United States Bureau of Economic Analysis to show that the labor share of net income (net of taxes and depreciation, which better captures income actually available to workers and capital owners) is within its historical range, fluctuating above and below the average of 69 percent. Labor’s share rose gradually from the mid-20th century through the early 2000s, declined modestly thereafter and has since returned near its historical average.

The entire framing that labor and capital are locked in a zero-sum fight over income misinterprets how production has historically worked. Empirical research consistently finds that investment in new technologies is complementary, primarily augmenting, not replacing, work. One recent review suggests that a 1 percent increase in capital per worker raises wages by about 3 percent.

Certainly, the future could be different. But even in a world where capital’s share of national income rises, it does not necessarily follow that workers are made worse off. Workers can still benefit from increasing wages in a growing economy, even if capital’s share is growing faster. If everyone is getting richer, then who is getting richer faster is, at most, a secondary concern. The primary economic problem is not one of simply dividing existing resources but of making more goods and services from fewer resources.

Even on Its Own Terms, a Compute Tax Fails

Suppose you believed that this time is actually different and workers are about to be displaced by machines, without any offsetting new jobs. Economist Brian Albrecht concludes that “a compute tax is about as bad as a tax can be.”

First, a tax on inputs to the AI process, such as graphics processing units, power generation, and data centers, is a tax on intermediate goods. Because AI is increasingly an input into other firms’ production, taxing AI’s inputs raises costs before any final good or service is produced. Those higher costs then pyramid through downstream products. The result is a hidden, arbitrary tax on final outputs that varies with the number of production stages and rewards firms for minimizing taxable inputs rather than for producing efficiently. These types of intermediate taxes are some of the most economically damaging ways to raise revenue.

Second, most of the inputs to AI that could be taxed are capital, and capital taxes are notoriously shifted from owners to workers through lower investment, slower productivity growth, and, ultimately, lower wages. As Albrecht notes, “The features of AI that people worry about (easy substitution between capital and labor, mobile capital, self-replicating infrastructure) are exactly the features that make capital taxation counterproductive.” For workers to succeed in an AI economy, they will need more investment in newer and better tools, not less. Taxing the capital behind those tools would slow the very process that can raise wages and expand opportunity.

Third, the AI capital tax base is too small and too elastic to raise significant revenue. Beyond punishing the AI technology directly, proponents often frame the taxes as raising revenue for new programs or tax cuts for impacted workers. One optimistic scenario projects $1.4 trillion in AI spending next year. Even at a 100 percent tax rate, with no behavioral effects, the revenue wouldn’t even cover the current federal budget deficit, let alone new income support programs, additional redistribution, or tax cuts for workers. And anything near 100 percent tax rates would dramatically shrink projected AI spending and, thus, expected revenue. AI development wouldn’t stop; it would just shift offshore. The result would be less domestic investment and fewer American workers building and using the next generation of productivity-enhancing tools.

Is Capital Undertaxed?

The popular narrative behind AI taxes is that the tax code already overtaxes labor and undertaxes capital. Putting numbers to this, Economists Daron Acemoglu, Andrea Manera, and Pascual Restrepo estimated that labor faces an effective tax rate around 25 percent while equipment and software face rates closer to 5 percent. Garrett Watson at the Tax Foundation details a number of reasons why we should be skeptical of this analysis, as do the commentators on the paper itself, noting numerous methodological flaws. I will highlight several places where this comparison misses the mark.

On the input side, wages are 100 percent deductible the year they are paid (called “expensing”). It is often claimed that extending the same treatment to capital is a tax subsidy. It is not a subsidy; it is equal treatment.

Expensing brings deductions forward in time, shifting taxes paid into the future. The low capital tax figure reported by Acemoglu et al. is largely driven by failing to properly account for this artifact of timing. Following the 2025 tax package, equipment, research and development, and manufacturing structures are eligible for immediate deductions and are at parity with labor. Nonmanufacturing structures are still disfavored and must be depreciated over three to four decades. So, on this margin, there is no systematic bias toward capital.

The only significant tax asymmetry that remains in a firm’s decision between capital and labor is the employer-side payroll tax. Economists generally assume that the economic cost of the tax falls overwhelmingly on workers through lower take-home pay and not on a firm’s owners. Given equal outputs, firms compare total compensation costs to total machine costs. If payroll taxes are largely capitalized into wages, the tax is not a penalty on firms employing humans; it is, however, a tax that changes worker behavior.

The same logic applies on the output side, where the relevant comparison is between the taxes that distort workers’ decision to work and savers’ decision to invest. Labor income is taxed by the income and payroll taxes. Across both taxes, the US Treasury Department estimates the average federal tax rate on labor at 17 percent. The top marginal statutory rate is 40.8 percent.

The return on invested capital in public markets is subject to taxes at the entity and investor levels. Corporate profits are taxed at 21 percent, and the remainder is taxed again at rates up to 23.8 percent when paid out as dividends or realized as capital gains. Combined, the top statutory federal tax rate is roughly 40 percent on the return to corporate equity. Economist Kyle Pomerleau finds that average effective tax rates on investment differ by business form and financing source but generally fall between the mid-20 percent and mid-30 percent range.

By most careful measures, capital and labor face roughly similar overall tax burdens. But the comparison is somewhat incoherent. A tax on the normal return to capital (the return just sufficient to make the investment worthwhile) is a second tax on saved wages, distorting when to consume rather than whether to work. A tax base that is neutral to the saving-consumption decision would exempt the normal return and tax supernormal returns at the same rate as wages.

This is also why a new tax on AI is unnecessary to capture its gains. Existing taxes on corporate profits and capital income already capture the types of supernormal returns that could come from high-growth AI scenarios. There are many ways to improve the tax system to ensure supernormal returns are fully taxed and increase parity among wages, consumption, and saving. But removing full immediate expensing or indiscriminately raising tax rates on capital income, tokens, or compute will do more harm than good.

AI will most certainly change the labor market. But it has not yet changed the complementary relationship between workers and the tools that make them more productive. Taxing the machines is simply a softer version of the Luddite English textile workers who sought to destroy the machines that were necessary for our modern world. An AI or capital tax would fall on the very workers the policy is intended to help.

The resolution sets out a number of important principles that DSTs violate: net taxation rather than gross taxation, no double taxation, legal certainty, and taxing rights tied to physical presence.

Opposition to DSTs has been a recurring feature of congressional bipartisanship, and it is shared across administrations. Trump’s trade office opened the first investigation in 2019, which the Biden administration continued. In early 2025, Trump issued an order to renew investigations into a half-dozen additional European DSTs.

Washington’s consensus centers on the discriminatory mechanics of DSTs. They tax gross revenue instead of profit, so it applies even if the company is losing money in the jurisdiction. It reaches firms with no physical presence in the country levying the tax, and the tax is typically designed to catch American platforms while leaving domestic firms untouched.

DSTs are interesting because they are most strongly opposed by American politicians. But their costs fall mostly on consumers. This point came up repeatedly at a Cato webinar I hosted last month with European tax economists. The DST is assessed and remitted by large digital platforms, but the economic incidence falls on consumers in the jurisdiction imposing it.

Digital companies have been transparent about how they handle these levies. Google adds jurisdiction-specific surcharges to advertisers’ invoices that closely follow local DST rates. Apple adjusted developer proceeds in the UK, France, and Italy to reflect DST costs and raised app prices in countries like Turkey, where the combined tax burden was large enough to warrant a price change. Meta had been the holdout, absorbing the cost of DSTs for years. That changes on July 1, 2026, when Meta introduces “location fees” to pass DST costs directly to advertisers across several European countries.

A recent study confirms the same pattern for Amazon, which now charges a standalone Digital Services Fee in countries where it faces a DST. Economists Dominika Langenmayr and Rohit Reddy Muddasani find that third-party sellers pass those higher fees on to consumers. In France, Spain, and the UK, they find that consumers ended up bearing between €/£1.50 and €/£2.50 for every euro or pound the tax actually raised. They find Italy is an exception.

American politicians are still right to push back against these taxes. A tax’s economic damage routinely exceeds the revenue it raises. So even when the cost is fully passed through to consumers, US firms still lose through forgone sales and thinner margins. This is especially true when the tax explicitly exempts their foreign market competitors.

The bigger problem, and the one at the core of the House resolution, is that DSTs erode the international income tax system that has shielded cross-border trade and multinational firms from punitive and duplicative taxes for more than half a century. DSTs are not the only threat to that order. The return of tariffs is a direct threat. The Organisation for Economic Co-operation and Development (OECD) has also spent the better part of the last decade diligently undermining the protections of physical presence and rules against double taxation.

The Rep. Estes resolution’s principles are good: net over gross taxation, no double taxation, certainty, and taxing rights tied to physical presence. They should apply as broadly as possible. The same principles that condemn European DSTs, impugn tariffs, the US’s multiple overlapping corporate minimum taxes, and emerging proposals to tax AI.

Congress should oppose bad taxes wherever they appear. The principles behind Washington’s consensus against DSTs are a good starting point.

The basic idea of Trump Accounts is a good one: help more Americans start saving early in life and allow contributions to a child’s account from parents, relatives, employers, and charities. But the design falls short. My new Cato Policy Analysis explains how Trump Accounts work, why many claims about them are overstated, and how Congress could turn them into a simpler, more useful savings vehicle.

The biggest surprise is that Trump Accounts are often not tax advantaged for normal family contributions.

Parents, relatives, and friends can contribute up to $5,000 per child each year, but these contributions are not deductible. The money must be invested in low-cost US index funds and is locked up until the child turns 18. At that point, the account converts to a traditional IRA. Withdrawals are generally taxed at ordinary income tax rates, and withdrawals before retirement that do not qualify for a short list of exceptions face an additional 10 percent penalty.

That structure creates a problem. Family contributions go in after tax, just like money put into a regular brokerage account. But unlike a brokerage account, the investment gains are later taxed as ordinary income instead of at the lower capital gains rate. For many families, that means putting their own money into a Trump Account could leave their child worse off than using a normal taxable brokerage account, let alone a 529 plan.

A simplified example from the report shows this gap. Over 30 years, a $5,000 pre-tax investment could grow to more than $40,000. In a Trump Account funded with after-tax family contributions, the final, withdrawn value is $24,496. That is $2,451 less than the same investment in a regular taxable brokerage account. Table 2 of the report shows the full comparison across account types.

Employer contributions and sophisticated tax-planning strategies can improve the result. But that is exactly the problem. A savings account meant to help more families build wealth should not require a tax adviser.

The other major problem is the $1,000 government subsidy. Direct government transfers require rules about eligibility, approved uses, penalties, and administration. They also add billions of dollars in new fiscal costs and create political incentives for Congress to expand the subsidy in future political cycles.

Congress should keep the early-life account structure and the ability to receive contributions from multiple sources. But Congress should eliminate the $1,000 subsidy, simplify the rules, and make the accounts genuinely tax neutral.

One option is to allow all contributions to be deductible, let the funds grow tax-deferred, and tax withdrawals as ordinary income. At age 18, the account should convert not into a traditional IRA but into a Universal Savings Account. That would allow unrestricted withdrawals without extra penalties or government-approved uses.

Canada and the United Kingdom already have similar flexible savings accounts. They are simple, broadly used, and especially attractive to younger and lower-income savers.

That is the lesson Congress should take from Trump Accounts. The country does not need another complicated, restricted savings account layered on top of the existing maze of qualified savings accounts. It needs fewer, simpler, more flexible accounts that let Americans save for their own priorities.

Read the full report here.

I will argue against the resolution that billionaires should pay a higher share of federal taxes. National polling on this question runs about three to one for the resolution. However, I strongly believe that wealth is good and that someone has to stand up for unpopular groups.

I’ll start by being clear about my priors because this is, after all, a normative question.

When I earn a dollar, when you earn a dollar—from wages, from a side business, or from an investment—that dollar is yours. The bar for taking it is not “could this fund something useful?”

The question we should ask is: “Is this a function the federal government has to perform, and is this the least intrusive way to pay for it?” The presumption runs against the state taking your money. This holds true for me at my salary. It is true for the person making minimum wage. It is true for millionaires and billionaires. Eroding this principle for one class of taxpayers ultimately erodes it for everyone.

Instead, the debate is usually over who else we can tax and how we can tax them more. Rarely do we start by asking: “What should the feds stop doing?”

Against that backdrop, we’re running the largest intergenerational wealth transfer in history from relatively younger, poorer workers—many supporting families—to wealthier, older retirees. We are headed toward a $1.5 trillion military budget.

Against that backdrop, the question of whether billionaires should pay a higher share of taxes strikes me as precisely backward. The right question is whether the federal government should be taxing the share it already takes from anyone.

But even if you don’t share my priors and you think federal revenue should go up—I’ll argue that billionaires still shouldn’t pay a higher share of federal taxes.

Billionaires Already Pay High Taxes

Let’s start with what billionaires already pay. Because I think most Americans, when they hear this resolution, have no idea what the current bill actually looks like.

According to Treasury data across all federal taxes, the top 0.1 percent—that’s about 200,000 families—paid more than 16 percent of all federal taxes, almost double their share of national income. They paid more taxes than the lowest-income 70 percent of Americans combined, across 130 million families.

That statistic illustrates why the United States has the most progressive tax system in the developed world.

Two of the most prominent advocates for higher taxes on the rich in America and around the world are Emmanuel Saez and Gabriel Zucman. Last year, they published a study showing that US billionaires pay higher tax rates than their counterparts in the Netherlands, Sweden, Norway, and France. Other data shows that the top 10 percent of taxpayers in the United States pay more of the tax burden than the same top ten percent in any European country people point to as a model for higher taxes.

We are the outlier in how much we already concentrate the tax burden on people at the top.

Misleading Statistics

But what about the famed statistic that “billionaires pay an eight percent rate”? Or that “Warren Buffett pays a lower tax rate than his secretary.”

You are not hearing a tax rate. You are hearing a ratio of taxes paid in a year over someone’s unrealized gains. That’s not income. Putting unrealized gains—that can fluctuate by billions of dollars a day—in the denominator and actual realized income taxes in the numerator is not measuring an effective tax rate, it’s constructing a fiction that is designed to be misleading.

When you measure income the way the IRS measures it, so you can compare it to every other taxpayer, the wealthiest 92 American billionaires pay an average effective federal tax rate of 34 percent and a total tax rate, including other levels of government, of 59 percent.

The top ten percent pay total rates closer to 35 percent. And Americans who have lower incomes pay lower rates.

What is Fair?

Many billionaires already keep only 40 cents of every dollar they earn. What is their “fair share?” Should they get to keep 20 cents? 10 cents? 5? What is the number at which high tax advocates would say, yes, that’s their fair share?

I’ve never seen a consensus number, nor have I heard of a principle that would generate one. Without an underlying model, it’s hard to pursue changes in tax policy. “More” is just a vague direction. And a direction without a destination is not something I can debate against, because there is nothing on the other side of it to engage with. So it’s still an open question: beyond “more,” how much is fair?

Economic Costs

Even if you can answer the “how much” question to your own satisfaction, the economics of getting there is a lot worse than advocates let on.

Most billionaire wealth isn’t in yachts, personal houses, or jewelry; that’s only about 2 percent of it. Most wealth is in productive business capital—equity in companies that employ people, build things, make products, and fund research. When you tax that wealth, you’re taxing the buildings we all live in, the businesses that employ us, and the research labs that create life-saving drugs.

And when you tax something, you get less of it. This is why “tax the billionaires” sounds like a free lunch, but it’s not. A deep and well-established body of economic research finds that the burden of taxes on capital and businesses does not stay with the owners. A meaningful share is passed through to workers in the form of lower wages, fewer jobs, and slower productivity growth.

The historical record bears this out. Rich people move, poor people are left holding the bag.

France repealed its 75 percent super tax on high incomes in 2015 and its wealth tax in 2017. They repealed them, in part, because they cost more revenue than they raised. People and businesses left the country. Same pattern in Sweden. It abandoned its estate tax and wealth tax. Revenue went up afterward.

In Spain, which still has a wealth tax, researchers found that regions that imposed a wealth tax lost six dollars in personal income tax revenue for every new dollar of wealth tax revenue they gained. In each of these cases, high taxes on billionaires and business owners caused so much economic damage that they led to losses in total revenue.

When I look around at the best American companies, many at the leading edge of global innovation, I want more of them, and I want them to stay in the United States. Raising taxes on the people who build those businesses will drive them overseas or out of business.

The Progressive Case for Not Taxing Billionaires More

I’ll end with the argument I think should matter most to anyone who doesn’t share my point of view. Suppose you want a bigger federal government, European-style health care, fully funded family policy, and generous social insurance. How do other countries pay for all those things?

Yes, some of them pay for it by taxing the rich a bit more. But they also pay for it by taxing the poor and the middle class at much higher rates than in the US. That mechanically reduces the billionaire share of total taxes.

Every developed country with a large welfare state raises a larger share of revenue from broad-based taxes, such as the value-added tax, that falls more evenly on consumption across the entire income distribution.

This is not by accident. A welfare state funded by a small share of the population at the very top of the income distribution is unstable. The highest incomes are most volatile: in good years, it might look fine, but in downturns, when social programs need the most funding, revenue dries up. A welfare state funded by broad-based taxes on everyone is more politically and economically stable.

Franklin D. Roosevelt is said to have made this point regarding Social Security. He insisted on a broad-based payroll tax because he understood that broad funding mechanisms create a sense of ownership and durability for the program.

So, even if you want a bigger government and more revenue, it doesn’t follow that you’d want billionaires to pay a higher share to fund it. Looking around the world, it seems you’d actually want them to pay a lower share than they currently do in the United States.

The answer to the resolution is no. Billionaires should not pay a higher share of taxes. On moral grounds, on economic grounds, and on progressives’ own terms.

Billionaire Taxes

Tonight, I will debate Natasha Sarin at the Soho Forum in New York. The resolution is: Billionaires should pay a higher share of federal taxes. I’ll be arguing the negative.

If you are local, join us in person. Or watch the live stream.

Crypto Taxes

On Thursday, May 21, I’m joining Nicholas Anthony (Cato) and Jason Somensatto (Coin Center) for an in-person Hill briefing on crypto tax policy. The event starts at 12:30 in 340 Cannon. Lunch is provided.

The tax code treats crypto as personal property, not currency, so every transaction — even buying coffee — triggers a capital gains event. Regular users can rack up dozens of pages of tax filings a year. Several bills are circulating to fix it. We’ll walk through what practical reform looks like.

Register here: Crypto Use Is Taxing: How Congress Can Fix the Broken Framework.

Affordability Fellowship

Also, for Hill staff, Cato is taking applications for the 2026 Congressional Fellowship in Free Market Solutions for Affordability. It’s a nine-week summer program with weekly receptions (free food and drinks) and discussions with many of my excellent colleagues. While I will not be speaking at this fellowship, I’ve led them in the past, and they have resulted in great discussions.

Apply here: 2026 Congressional Fellowship. Applications are open through May 26.

Fix The Cause, Not The Symptom

First, if the administration wants to bring down gas prices, it should end its unconstitutional, unauthorized, and unnecessary war in Iran. That’s what’s driving the price spike. According to AAA, the national average for regular gasoline was $4.52 per gallon on May 11, 2026. That’s up from just under $3 before the conflict, an increase of more than $1.50 per gallon.

Even if we generously assume that suspending the 18.4‑cent federal gas tax is immediately and entirely passed through to consumers, it would offset only about 12 percent of the war-driven price increase. Consumers would be much better off by simply ending the war (as taxpayers funding the war, they’d also be better off).

But Full Repeal Would Provide Relief and Reform

Rather than a temporary holiday, Congress should use this moment to eliminate the federal gas tax entirely and devolve highway funding to the states.

As Chris Edwards and I argued in 2023:

Federal gas tax revenues go into the Highway Trust Fund and then are dished out to the states to use on highway and transit projects. However, since 98 percent of the nation’s streets and highways are owned by state and local governments, it would be simpler and more efficient if those governments were responsible for the funding. Having the federal government raise the funds and then return the funds to the states with regulations attached is unnecessarily bureaucratic. States have the best information to determine their local infrastructure needs.

States have already shown they can take the lead. Average state gas tax rates rose 44 percent between 2000 and 2021. States know what their infrastructure needs are, and they have the fiscal tools—gas taxes, sales taxes, user charges, debt, and privatization—to meet them without a federal middleman.

Congress will have to pass a highway bill reauthorization before September 30, 2026. It should use that opportunity to permanently repeal the gas tax. A federal gas tax repeal would permanently lower prices and be a genuine step toward decentralizing government power. A temporary holiday is just a political gesture.

Next week, we will discuss these new developments and what they tell us about the uneasy future of taxing the digital economy.

Join me on Wednesday, May 6, at 10 a.m. ET for an online conversation: Digital Services Taxes at a Crossroads: European Perspectives on What Comes Next. I will be joined by Matthias Bauer (ECIPE), Cristina Enache (Tax Foundation Europe), and Andreas Hellmann (Tholos Foundation).

Register Here

We’ll cover the political prominence of DSTs compared to their limited fiscal capacity, who really pays the tax, what the UK conflict tells us about Europe’s next moves, and what a better path forward might look like, including the possible role of value-added tax reform.

The conversation will draw on recent work by our panelists. Matthias Bauer and Dyuti Pandya recently argued that Europe’s existing VAT systems contain far more untapped fiscal capacity than digital services taxes could ever provide. Cristina Enache has advanced a similar view, and in testimony before the European Parliament last fall, made a compelling case that the revenue rationale for these taxes is far weaker than domestic politicians believe. The Tholos Foundation has also published a report framing DSTs as a fundamental violation of established tax principles. Finally, last year, I surveyed several problems with DSTs and warned that some proposed alternatives could be even worse. Since then, the DST landscape has only become more complicated.

If you work on tax policy, trade, or the digital economy—or you just want a clearer picture of where Europe goes from here—I hope you’ll tune in!

Register here: Digital Services Taxes at a Crossroads: European Perspectives on What Comes Next

But the report’s headline claim is simply wrong. The profits and taxes in ITEP’s headline come from two different accounting systems, built for different purposes, reporting to different audiences, using different rules.

You Can’t Compare Accounting Profits to Taxes

ITEP’s method compares two numbers that are calculated under different rules for different purposes.

The $105 billion in US income reported is financial accounting profits that companies report to investors under Securities and Exchange Commission (SEC) rules, in accordance with Generally Accepted Accounting Principles (GAAP). The reported taxes (or lack thereof) are current federal income tax expenses that reflect an estimate of tax payments owed to the IRS in that company’s 2025 fiscal year. These estimates follow tax rules set out in the Internal Revenue Code but can still differ substantially from what’s actually reported on a company’s Form 1120. The two systems—GAAP and the tax code—define income, deductions, and timing differently. Using one system in the numerator to the other in the denominator tells us nothing meaningful about profits or taxes. It instead serves to intentionally confuse.

A simple example shows why. Imagine a company earns $10 million and spends all of it on a new piece of manufacturing equipment in 2025. Under financial accounting rules, the firm spreads out the cost of that purchase over 10 years, deducting $1 million per year. So in 2025, the firm’s accounting books show $9 million of profit on $10 million of pre-investment earnings.

Under current tax law, the firm can deduct the full $10 million in the year the investment is made. The firm’s taxable income in 2025 is zero, but it will have no additional deductions left over in future years.

ITEP’s method would report that this hypothetical firm earned $9 million in profit and paid zero federal income tax. In reality, it spent every dollar it earned investing in new manufacturing capacity. Its economic profits and taxable income match; they are both zero.

This isn’t tax avoidance; the tax code and accounting rules simply recognize costs at different times. Comparing the two misunderstands the lumpiness of business income, and this misunderstanding is at the heart of nearly every company on ITEP’s list.

The Tax Breaks Are the Point

ITEP helpfully catalogs the tax provisions that reduced these companies’ tax and taxable income to zero. They don’t reveal a scandal; they are a list of popular, bipartisan policies. Companies using them are doing what the law intends, often for good reason.

Accelerated depreciation and expensing. The most common tax break is accelerated cost recovery. The purest form of this policy is expensing, which allows businesses to deduct the cost of an investment when they actually spend the money, rather than requiring the cost to be spread out over the asset’s useful life as GAAP requires. The accounting treatment makes sense for reporting to investors who want to understand the firm’s long-run financial position by smoothing over lumpy capital expenditures. But immediate expensing is the economically correct treatment of business costs and has been supported by economists across the political spectrum for decades. Some version of accelerated depreciation has been in every major tax bill since the early 2000s, signed by presidents of both parties, including the Democrats’ Inflation Reduction Act.

Expensing does not change the nominal amount of tax payments; it changes only the timing, shifting taxes paid into the future when the firm is economically profitable and moving deductions forward in time. The bipartisan appeal is that this keeps taxes low on businesses that are continually reinvesting their profits into new productive capacity.

Research and development credit. Bipartisan majorities have supported the R&D tax credit since its introduction in 1981. It is not particularly well designed, but it nonetheless retains widespread support.

Foreign-derived intangible income deduction (FDII). This provision lowers the effective tax rate on income from US exports. The 2025 tax bill reformed and expanded it into a broader export incentive now called foreign-derived deduction-eligible income (FDDEI), effective for tax years beginning in 2026. The original FDII deduction is widely credited with encouraging major US firms, such as Google, to repatriate hundreds of billions of dollars in intellectual property, shifting taxable income back to the US that would have otherwise been reported and taxed abroad. The Biden administration included its own domestic intellectual property incentives in its annual budget proposals.

Stock option deductions. When a company grants an employee stock options, there is no tax event until the employee exercises them. At exercise, the employee reports the difference between the purchase price and the market price as ordinary taxable income. The tax code allows the company to deduct the same amount the employee reports as income, just like wages. The tax break ITEP identifies arises because GAAP requires companies to estimate and report the cost of stock options when they are granted, instead of what employees ultimately realize at exercise. This is not a tax problem; it is simply a function of two systems measuring different things for different purposes.

The Right Corporate Tax Rate Is Zero

The corporate income tax should be eliminated entirely. The corporate tax is the most economically costly major source of revenue, and it comes with high compliance costs. The economic cost of the tax is borne mostly by workers in the form of lower wages, but shareholders and consumers also pay a hidden price. Other taxes on individual income, consumption, or land raise more revenue, at lower rates and with less economic damage.

The goal should be for all companies (not just 88) to pay zero federal income tax. But ITEP’s list doesn’t show that we’ve moved closer to the zero tax ideal. Many of these companies have paid substantial taxes in the past and will pay even more in future years. The ITEP report takes a snapshot in time and claims that timing differences reveal something meaningful about the tax code. They don’t.

Direct File was conceived with grand ambitions. It was going to simplify tax filing, save taxpayers money, and finally prove that the agency that collects your taxes should also be in charge of preparing them for you. The Trump Administration shut the program down last year, and for good reason. As I argue in my April 15 letter to the editor of the New York Times (republished below), it did none of the things it promised.

Because letters must be short, I left out the also not-so-small matter of legality. Congress didn’t authorize a government tax-prep service. The Inflation Reduction Act included $15 million for a task force to study the feasibility of a direct e-file system and to deliver a report to Congress. Under the Biden Administration’s direction, the IRS took that study as license to build a permanent program out of whole cloth, spending more than $41 million. A classic example of bureaucratic mission creep.

Direct File expired under the weight of its own contradictions: costly, duplicative, conflicted, ineffective, and illegal. Good riddance.

As someone who does tax policy for a living, I often find myself explaining to people why I find the topic so engaging. Sometimes I slip and say: I love taxes! It’s actually the intersection of taxes with just about every policy area and economic sector that I love—there’s just always something new to learn. Paying taxes, on the other hand, I hate.

In the spirit of April 15, my annual update of “five charts for Tax Day” is here. I cover where revenue comes from, who pays and how much, and why the US’s relatively low tax burden compared to Europe is not sustainable without meaningful spending restraint. Cutting spending is the only way out of this mess.

I’m also on the newest episode of the Cato Podcast, talking taxes with Chris Edwards.

Here are a few of my favorite Tax Day reads from this season:

America’s Other Tax Day is April 16th
Jared Walczak, Tax Foundation (Subscribe to his Substack, The SALT Road)

Walczak reminds us that collectively, Americans work 106 days just to pay our total tax bill across all levels of government. That puts “Tax Freedom Day” on April 16—just one day after Tax Day this year. It’s mostly a coincidence that the days are back-to-back, but starting tomorrow, you’ll be working for yourself again rather than the government.

Correcting the Top 10 Tax Myths
Jessica Riedl, Brookings Institution

Riedl updates her 2024 piece with Freakonomics, dispelling persistent misconceptions in tax policy:

• Tax cuts pay for themselves

• Tax cuts will starve the beast

• The middle class pays higher tax rates than the rich

• Those old 91% tax rates raised large revenues

• Europe raises more revenue by taxing the rich more

• Tax cuts for the rich drive deficits

• Taxing millionaires and corporations can eliminate the deficit

• Most of the 2017 tax cuts went to corporations and the wealthy

• Repealing all post-1980 tax cuts would painlessly reduce deficits

• US corporate taxes are far below international norms

Simplifying the Tax Code Is a Pro-Growth Policy
Jack Salmon, Mercatus Center

Salmon estimates that tax compliance costs total roughly $546 billion annually (nearly 2 percent of GDP). That’s more than just paperwork. It’s time, legal complexity, and economic manipulation. These costs are a symptom of deeper structural flaws in the tax code that would be fixed if Congress pursued true structural tax reform.

Bitcoin Taxes Make No Sense
Nicholas Anthony, Cato Institute

We’ve established that the tax code is complicated. One small corner of that complexity is the taxation of cryptocurrencies. Anthony shares his tax return, noting that it runs more than 70 pages long because he’s required by the IRS to detail a capital gain and loss for every one of his Bitcoin transactions.

Think April 15 Is the Deadline? Maybe Not.
Joe Bishop-Henchman, National Taxpayers Union Foundation

A reminder that the tax code isn’t just complex, it’s occasionally absurd. Electronic payments must be scheduled by 8 p.m. ET on April 14 to count as timely (hope no one is waiting until the last minute to hit send on that tax payment!). A check mailed on April 15 (and arriving days later) still qualifies. “This is outdated and dumb.”

What the IRS Data Is Showing This Filing Season
Erica York, Tax Foundation

Early IRS data show average refunds are up almost 11 percent year-over-year, reaching $3,571, or $350 dollars more than last year. Much of this increase reflects changes from the One Big Beautiful Bill Act. But remember, bigger refunds are not a gift! It’s the government returning the hard-earned money they over-collected from you throughout the year. And they return it without interest.

US Tax Reform Timeline, 1945-Present
Fredrick Hernandez and Andrew Lautz, Bipartisan Policy Center

A comprehensive timeline of major tax changes since World War II. Hernandez and Lautz provide a useful reminder that the tax system we live with today isn’t the product of any single, coherent design, but a hodgepodge of discrete provisions shaped by decades of political dealmaking, interest-group pressure, and growing partisanship.

Occasional marketing appeal: If you enjoy this newsletter, please share it with a friend or colleague. Help me grow the community of people who hate taxes, but love tax policy!

Share Liberty Taxed: A Blog on US Tax Policy

That’s a lot of money, and government data show that most of it is paid by the highest-income Americans. It is also not nearly enough to cover the $7 trillion Congress spent last year.

Here are five charts on where revenue comes from, who pays and how much, how average US taxes compare to those around the world, and why today’s spending puts future taxes at risk.

Most Federal Revenue Comes from Income Taxes

Personal income taxes raise just over half of the federal government’s revenue. However, for all but the top 10 percent of income earners, Americans pay more in payroll taxes on average than in income taxes each year. Payroll taxes account for 33 percent of federal revenue.

Figure 1 shows that the remaining revenue comes from corporate income taxes, customs duties, and other sources, including excises, estate taxes, and other fees.

Unlike federal taxes, more than half of state and local tax revenue comes from property and sales taxes. While it varies significantly by state, income taxes often account for less than a quarter of subnational revenues.

The Rich Pay the Highest Tax Rates

The United States has the most progressive tax system in the developed world, driven primarily by high federal income tax rates and large exemptions for lower-income taxpayers.

At the federal level, the top 10 percent of income earners pay more than 60 percent of all taxes and 72 percent of income taxes, shares that have been increasing over time.

Figure 2 reports estimates from the US Treasury’s Office of Tax Analysis, showing that average federal tax rates rise with income, accounting for income, payroll, corporate, and other taxes.

The lowest-income 20 percent of earners, measured by adjusted family cash income, face average tax rates that are either negative or close to zero. A negative tax rate means the taxpayer is a net beneficiary of the tax system, likely receiving refundable tax credits, such as the earned income tax credit (EITC), child tax credit (CTC), and Obamacare tax credits.

At the top of the distribution, the top 10 percent of income earners pay an average tax rate of 27.3 percent, which is 7 percentage points higher than the next-highest group. Treasury breaks the highest-income earners into narrower segments, showing that the top 0.1 percent pay the highest average tax rate of 33.4 percent.

The federal income tax system is even more progressive. The latest IRS data on income taxes for the 2022 tax year show that higher-income Americans pay a disproportionate share of income taxes and that the system has become more progressive over time.

Figure 3 shows that as a share of adjusted gross income (AGI), the top half of income earners paid 97.1 percent of federal income taxes. The top 1 percent earned 22.4 percent of total income and paid 40.4 percent of all the income taxes. The top 10 percent earned 49.4 percent of the income and paid 72 percent of the income tax.

Since 2001, average income tax rates have fallen for all five income groups. During this same time, the share of income taxes paid by the top five percent increased from 52.2 percent to 61 percent, while the share paid by all other taxpayers declined. According to the National Taxpayers Union, the top one percent’s share of federal income taxes has more than doubled since the 1980s, reaching 40.4 percent in tax year 2022.

America’s Tax Advantage and the Threat of Bigger Government

The United States remains a relatively low-tax country compared to its global peers. Among 22 high-income nations in the European Union, the US ranks dead last in total tax burden (federal, state, and local) for the average single worker.

Figure 4 shows the tax rate paid by an average single worker in the US and across Europe. Belgium, Germany, Austria, France, and Italy confiscate more than half of their workers’ pretax compensation. An average US worker sends less than a third of their income to the government.

Middle-class Europeans face high payroll and consumption taxes that dramatically reduce take-home pay and work incentives. The tradeoff is clear: higher taxes make you poorer. In high-tax countries, people work fewer hours and are less likely to work full-time because the government takes a larger share of each additional dollar they earn.

The American tax system is far from perfect. But keeping our overall tax burden low is a key reason why US workers are more productive, more entrepreneurial, and ultimately better off than their counterparts in high-tax countries.

Looking only at current taxes obscures the fact that the US federal government has run a budget deficit every year since the early 2000s, financing higher spending levels through government debt.

Figure 5 shows that spending on interest and other mandatory programs (such as Social Security and health entitlements) will permanently surpass revenues next year, according to the Congressional Budget Office’s recent budget outlook. This leaves Congress to finance all other spending with additional debt.

The current US fiscal trajectory is unsustainable. Eventually, taxes will need to rise, or spending will need to be cut. Large, European-style welfare states cannot be sustainably financed when a narrow sliver of income earners pay the lion’s share of taxes—big government requires high taxes on the middle class.

Instead of raising taxes, Congress should reduce spending to maintain America’s outlier status as a country where the government confiscates less of your paycheck. The only way to ensure taxes remain low is to cut spending and maintain a smaller government.

This updates a previous blog.

The next reconciliation package should finish what the One Big Beautiful Bill Act (OBBBA) started. It should finish construction on the wall around the welfare state by drawing a clearer line between immigration and access to public benefits.

The 2025 law made meaningful progress. It restricted immigrant eligibility for programs including the Supplemental Nutrition Assistance Program (SNAP), tightened access to health subsidies, and required parents to have a Social Security number to claim the child tax credit (CTC). These are steps in the right direction, but more needs to be done.

Now, the Treasury Department is considering reclassifying refundable tax credits (credits that are direct payments, rather than tax cuts) as “federal public benefits.” That move could restrict eligibility for some categories of noncitizens who currently file taxes and qualify under existing law.

The Treasury proposal is directionally correct policy. But major eligibility rules should be designed by Congress rather than regulatory fiat. Reconciliation 2.0 should codify these proposed changes to make them more durable and expand them to other programs to provide needed offsets for a package that promises to be expensive (but doesn’t need to be, as most of the new proposed spending is unnecessary).

Immigration Is a Fiscal and Economic Positive

It’s important to acknowledge a key empirical point: On average, immigrants contribute more in taxes than they consume in benefits from all levels of government. Even lower-income immigrants tend to use fewer public resources than similarly situated native-born Americans. Under current law, immigration is a fiscal and economic asset.

But that does not mean the current policy mix is optimal. A system that limits access to public benefits can strengthen those positive fiscal effects and ensure that immigration policy is aligned with work, contribution, and upward mobility, not dependence.

What’s Left to Fix?

Despite recent reforms, significant access to federal benefits remains for immigrants. Eligibility rules operate on a spectrum, and recent reforms have only moved partway along it. Requiring a valid work-eligible Social Security number is a meaningful step that generally ensures a claimant is authorized to work in the United States. Congress could go further by limiting eligibility for major refundable credits and other transfer programs to lawful permanent residents or just US citizens. That would draw a clearer line between temporary labor market participation and full eligibility for taxpayer-funded benefits.

For example, major benefit categories still allow relatively broad access under current law, often with carveouts that weaken otherwise clear eligibility rules.

• Tax credits: Recent reforms strengthened the Social Security number requirement for the CTC and narrowed eligibility for Obamacare tax credits, denying access to those with temporary protected status, refugees, asylum seekers, and migrants with parole. Here, Congress could codify the Treasury’s proposed federal public benefit rule for refundable tax credits, but an even stricter restriction to only US citizens would be better. Requiring self-certification of citizenship for all refundable tax credits—earned income tax credit, CTC, Obamacare credits, adoption credit, and American opportunity tax credit—would strengthen existing protections offered by Social Security number requirements, with minimal additional administrative burden compared to complex pre-certification requirements. Based on data from Cato’s recent report on the fiscal effect of immigrants, noncitizens received $456 billion in federal refundable tax credits over the last 30 years.

• Cash and income support programs: Following the OBBBA, Supplemental Security Income, Temporary Assistance for Needy Families, and SNAP are now generally limited to citizens, lawful permanent residents (often after a waiting period), lawful children (SNAP), and a narrow set of exempt groups such as certain Cuban and Haitian entrants. Social Security retirement benefits are limited to applicants with at least 40 quarters of work history in the United States, except for nationals of countries with which the United States has a totalization agreement. Illegal immigrants are ineligible, but other lawfully present noncitizens can qualify. Additional savings could be achieved by further restricting benefits to only US citizens and excluding work history prior to naturalization to qualify for Social Security. Over the last 30 years, noncitizens have received $218 billion in federal cash welfare, another $125 billion in SNAP and federal school lunch, and $501 billion in Social Security benefits.

• Health programs: The OBBBA significantly narrowed Medicaid, Medicare, and Children’s Health Insurance Program (CHIP) eligibility, limiting federally funded coverage primarily to citizens, lawful permanent residents, and a small set of exempt groups while excluding many previously eligible populations such as asylees and refugees. The final bill omitted key enforcement provisions from the House version, including reduced federal funding for states that cover unauthorized immigrants and a requirement to withhold federal funding until eligibility is verified. These loopholes should be closed, and eligibility should be narrowed further to citizens only. Over the last 30 years, noncitizens have received $638 billion in federal Medicaid and CHIP funds, and they have benefited from another $437 billion in Medicare funds.

• Housing subsidies: Public housing, Housing Choice Vouchers, Section 8 project-based rental assistance programs, and rural rental assistance allow benefits to flow to most permanent lawful immigrants, while mixed-status households may still receive prorated benefits. Other programs, such as many grant-based housing and homelessness programs and the low-income housing tax credit, do not condition tenant eligibility on immigration status at all. Congress should standardize eligibility rules and, at a minimum, apply the same limits adopted in the OBBBA—restricting eligibility to citizens, lawful permanent residents, and a narrow set of exempt categories. Ideally, eligibility would be limited to citizens alone. Noncitizens have received $62 billion in federal housing and rent assistance over the last 30 years.

• Education and workforce: The College Assistance Migrant Program, High School Equivalency Program, and Migrant Education Program provide academic, financial, and support services to lawful immigrant and seasonal farmworker populations. Federal student aid extends grants and loans to a broad set of noncitizens, including refugees, asylees, and certain visa holders. These programs should be eliminated or limited to US citizens. Unemployment insurance is limited only to workers available to legally work, which may not include immigrants here without work authorization but may include other lawfully present immigrants. Over the last 30 years, noncitizens have received $181 billion in unemployment insurance payouts.

• Other programs: The Federal Communications Commission’s Lifeline program and Head Start have historically relied on income-based eligibility with limited immigration status verification. Recent proposed regulatory changes would tighten access by reclassifying these programs as federal public benefits. Congress should codify these changes and further narrow eligibility to ensure access is tied to citizenship.

A Pro-Immigration Reform

The United States benefits enormously from immigration. Most people come to work, build businesses, raise families, and contribute to our communities, not to access public benefits. But good policy can reinforce those aims with clearer rules.

In a better-designed fiscal system, most federal benefit programs would be smaller or would not exist at all. But to the extent that they do exist, they should be tightly targeted. Limiting access to public benefits helps ensure that immigration remains a fiscal strength and may reduce political backlash, while helping keep labor markets open and welcoming.

While not every change listed above is eligible for the reconciliation process, lawmakers can go a long way toward finishing building the wall around the welfare state in Reconciliation 2.0.

Under longstanding tax rules, gambling winnings are taxable income. If you hit it lucky in your office March Madness pool, the Internal Revenue Service (IRS) expects its cut. The tax code has historically allowed professional gamblers and taxpayer who itemize to deduct their full losses against their winnings. This recognizes the basic principle that you should be taxed on your net income, not your gross receipts.

Starting this year—thanks to the One Big Beautiful Bill Act—gamblers can only deduct 90 percent of their losses. That means a gambler who breaks even over the year will still owe tax on income they never actually earned.

The gambling change may seem small, but for professional gamblers, it could be disastrous. It also reflects a broader feature of the rest of the US tax code, which systematically limits how taxpayers can use losses across the economy.

The Tax Code Makes Losing More Painful

A neutral tax system should measure and tax real income, which is what’s left after paying the costs of earning it. Those costs aren’t just wages, investments, and supplies; they are also the risk of loss. Gambling makes this intuition clear. If you win $10,000 on a lucky run but lose $10,000 over the rest of the year, your real income is zero. But under a rule that limits losses, the IRS still taxes you as if you came out ahead.

As with gambling, many investments lose money, leaving businesses and their investors in the red at the end of the year. Allowing full and immediate deduction of losses ensures the tax code treats good years and bad years symmetrically. Otherwise, the tax code could end up taxing profits that don’t actually exist yet, whether it’s a gambler who just breaks even over the year or a business that hasn’t yet recovered its start-up costs.

When loss deductions are delayed, capped, or denied, the government systematically overstates taxable income, effectively penalizing the risk-taking that generates the income in the first place.

Like gambling losses, business losses should be fully usable. In an ideal system, a start-up business that invests heavily in its first five years without turning a profit could bank those losses to fully offset future-year profits until the firm is truly profitable over its full lifecycle.

But current tax law limits the use of net operating losses (NOLs) from prior years. Following changes in 2017, most businesses can no longer carry losses back to prior profitable years and, instead, must carry them forward (without indexing for inflation). C corporations can only offset 80 percent of their taxable income with past losses. Pass-through businesses face limits that can be even more restrictive. The excess business loss rule caps the amount of loss that can offset other income in a given year, and the passive loss rules restrict the use of losses from certain activities against other income. Like gamblers down on their luck, any business that faces these limits will owe taxes on phantom profits before it has fully recovered losses.

A similar problem exists for individual investors. If a taxpayer sells an asset at a loss, those capital losses can offset capital gains in that year. But if losses exceed gains, only $3,000 of losses can be deducted against other sources of income each year. The rest must be carried forward.

In real terms, delayed deductions carried forward to future tax years lose value due to inflation and time. By not allowing losses to offset other sources of income and by not indexing carried forward losses, the IRS taxes profits immediately and fully, only partially recognizing losses. The longer losses must be carried forward, the less they are worth, which means start-up businesses that take longer to succeed face higher effective tax rates.

These rules can shape real economic behavior. Similar to gambling, entrepreneurship and investment are inherently risky. Often, multiple projects fail before one returns a profit. When the tax code limits loss recovery, it raises the effective tax rate on the riskiest activities. The ability to deduct losses fully and promptly is critical so the tax system does not discourage entrepreneurial risk-taking.

Fixing the Problem

Congress should reverse the limitation on gambling losses. Taxpayers should be allowed to deduct 100 percent of their losses.

But the real reform agenda should go much further. All forms of losses should be allowed to offset ordinary income, and any losses carried forward should be indexed for inflation. Businesses should also not face arbitrary limits on NOLs.

If policymakers are serious about growth, innovation, and economic dynamism, they should start by fixing how the tax code treats failure. Whether it’s a busted bracket in March or a failed start-up, the tax system shouldn’t just share in the upside; it should account for the downside too.