Who Really Pays Their “Fair Share” of Taxes?

“Pay your fair share” may be one of the most repeated phrases in American politics, yet it remains one of the least defined. It sounds measurable without ever supplying a measurement. It conveys a moral judgment before anyone has examined the books, identified the taxpayer, defined the tax base, or calculated what that taxpayer already contributes.

Fair according to whom? A fair percentage of income? A fair percentage of national tax revenue? A fair amount relative to wealth? A fair contribution after accounting for government benefits received? The phrase usually arrives without answers because its political usefulness depends on remaining elastic. Any amount already paid can be declared insufficient when the standard itself is never fixed.

A serious analysis must reverse the normal rhetorical process. Instead of beginning with the conclusion that wealthy Americans are failing to contribute enough, it should begin with the accounting. Who earns the income? Who remits the taxes? What percentage of total revenue comes from each income group? What effective rates do those groups pay? Who receives refundable credits, cash assistance, medical benefits, food assistance, housing support, and other transfers? Which households are net contributors to the federal tax-and-transfer system, and which are net recipients?

Once those questions are separated and measured, the conventional story becomes much harder to maintain. The United States does not have a federal system in which poor and middle-income households carry wealthy households. It has a substantially progressive and redistributive federal system in which upper-income households provide most federal revenue, pay considerably higher effective federal tax rates, receive very little means-tested assistance, and finance a large share of the benefits distributed to households below them.

This does not settle every philosophical argument about taxation. Data cannot tell us what is morally “fair,” because fairness is a normative judgment rather than a column in an IRS table. The data can, however, establish who is paying, who is receiving, and which political claims are incompatible with the actual structure of the system.

The Three Ledgers That Must Be Kept Separate

Most tax arguments become confused because they combine three different ledgers.

The first ledger measures tax liability. It includes individual income taxes, payroll taxes, corporate income taxes attributed to households, excise taxes, and—depending on the analysis—state and local taxes.

The second ledger measures transfers and direct benefits. These include Medicaid, the Children’s Health Insurance Program, the Supplemental Nutrition Assistance Program, Supplemental Security Income, housing assistance, refundable tax credits, health-insurance subsidies, child-nutrition programs, energy assistance, and other income-qualified programs.

The third ledger contains collective public goods and broadly available services. National defense, courts, policing, fire protection, roads, contract enforcement, monetary stability, financial regulation, public education, scientific research, and infrastructure all fall somewhere within this broader category.

These ledgers cannot simply be mixed together. A Medicaid benefit can be assigned to a household with reasonable precision. The value of an aircraft carrier, a federal court system, or a stable dollar cannot. A public-school student consumes a service that can be estimated, while a business owner benefits indirectly from an educated labor force. A highway may be used by a family, a delivery driver, a manufacturer, and a multinational logistics company on the same day.

The strongest argument is therefore not that wealthy households receive literally nothing from government. They benefit from the legal, institutional, monetary, and physical framework within which wealth can be created and protected. The demonstrable conclusion is narrower and more important: high-income households receive very little in direct, means-tested benefits while supplying most of the federal taxes that finance the system.

What the Individual Income-Tax Data Actually Show

The individual federal income tax is the most visibly progressive part of the American tax system. The IRS’s tax-year 2023 percentile data show that the burden is not merely larger in dollar terms at the top. It is disproportionately concentrated there relative to the distribution of adjusted gross income.

The top 1 percent earned approximately one-fifth of the adjusted gross income reported on individual returns, but paid nearly two-fifths of all individual income taxes. The top 10 percent earned less than half of reported AGI while paying more than 70 percent of the tax. The entire bottom half of the filing population reported about 12.3 percent of AGI and supplied only about 3.3 percent of individual income-tax revenue. IRS ¹

Those figures do not demonstrate that every feature of the tax code is ideal. They do demonstrate that the federal individual income tax already imposes a sharply disproportionate share of the burden on higher earners.

The phrase “the rich should pay more because they earn more” also fails to describe what is occurring. High earners do not simply pay more dollars because their incomes are larger. They generally surrender a higher percentage of their taxable income and provide a share of tax revenue substantially greater than their share of reported income.

For the top 1 percent, the tax share was approximately 1.86 times the group’s share of AGI. For the top 10 percent, it was roughly 1.48 times the group’s income share. The bottom half’s tax share, by comparison, was only about one-quarter of its AGI share.

A proportional system would produce tax shares much closer to income shares. The American individual income tax does not do that. It deliberately concentrates liability toward the top.

The Bottom Half Is Not Necessarily Paying “Nothing”

The bottom half’s small share of individual income taxes should not be twisted into the claim that lower-income Americans pay no taxes. They pay payroll taxes, sales taxes, gasoline and other excise taxes, property taxes directly or through rent, utility taxes, licensing charges, and numerous state and local levies.

Payroll taxes are particularly important. An employee can owe little or nothing in net federal income tax while still having Social Security and Medicare taxes withheld from every paycheck. Excise taxes can also consume a larger percentage of a low-income household’s resources because necessities and consumption represent a larger share of its income.

The distinction remains essential, however, because political rhetoric commonly speaks about “taxes” while displaying only whichever category produces the preferred conclusion. One argument points exclusively to federal income taxes because they are highly progressive. Another points to payroll or sales taxes because they are comparatively less progressive. A useful analysis must disclose exactly which taxes are being counted.

The claim that many low-income households are “net negative taxpayers” is most accurate when referring to the individual federal income tax after refundable credits. It should not automatically be extended to every tax those households pay.

What Happens When All Major Federal Taxes Are Included?

The Congressional Budget Office provides a broader measure by combining individual income taxes, payroll taxes, corporate income taxes attributed to households, and federal excise taxes. Under that wider accounting, the overall federal system remains substantially progressive.

CBO’s latest complete distributional study covers 2022. It found an average total federal tax rate of 1.4 percent for the lowest income quintile, 9.5 percent for the second quintile, 13.4 percent for the middle quintile, 17.6 percent for the fourth quintile, and 25.9 percent for the highest quintile. The top 1 percent paid an average federal rate of 31.5 percent. CBO calculates these rates by dividing each group’s allocated federal taxes by its income before means-tested transfers and federal taxes. Congressional Budget Office ²

The table shows why precise language matters. The lowest quintile had a negative 10.1 percent individual income-tax rate, largely because refundable credits exceeded pre-credit income-tax liability. Once payroll, corporate, and excise-tax incidence was included, the group’s total federal rate became a positive 1.4 percent. The same distinction applies to the second quintile, which had a slightly negative individual income-tax rate but a positive overall federal rate. Congressional Budget Office ²

The payroll-tax pattern also explains why an extraordinarily wealthy household can have a lower payroll-tax rate than a middle-income household. Social Security payroll taxes apply only up to the annual wage base, and much of the income at the top comes from business and capital sources rather than wages. This feature makes payroll taxation less progressive than the individual income tax, but it does not overturn the progression of the federal system as a whole.

In 2022, the highest income quintile paid approximately 70.2 percent of the four major federal taxes allocated by CBO. The top 1 percent alone paid approximately 27.3 percent. The lowest quintile supplied about 0.3 percent, the second quintile 3.9 percent, the middle quintile 8.6 percent, and the fourth quintile 16.8 percent. Congressional Budget Office ²

The concentration is even greater within the individual income tax. CBO attributed 87.6 percent of individual income-tax payments to the highest quintile and 40.3 percent to the top 1 percent. The lowest quintile’s share was negative because refundable credits exceeded its income-tax liability.

The claim that wealthy households somehow avoid financing the federal government is not a serious description of these numbers. The federal government is financially dependent upon upper-income households.

The Other Side of the Ledger: Who Receives Means-Tested Transfers?

Tax burden is only half of the analysis. A household paying $5,000 in taxes and receiving $15,000 in direct benefits occupies a different fiscal position from a household paying $100,000 and receiving $1,000.

CBO defines means-tested transfers as cash payments and in-kind services delivered through federal, state, and local assistance programs for which eligibility is determined primarily by low income or limited assets. Its major categories include Medicaid and CHIP, SNAP, Supplemental Security Income, housing assistance, Temporary Assistance for Needy Families, child-nutrition programs, low-income energy assistance, certain Medicare subsidies, Affordable Care Act cost-sharing subsidies, and state and local general-assistance programs. Congressional Budget Office ²

The distribution of these benefits is heavily concentrated toward the bottom:

The bottom two quintiles received approximately 73 percent of all means-tested transfers. The highest quintile received about 4.2 percent, while the top 1 percent received three-tenths of 1 percent.

Some benefits appear in higher quintiles because households are ranked using income adjusted for household size, income can fluctuate from year to year, certain medical benefits may reach households under particular circumstances, and broad statistical categories never divide the population with absolute purity. The pattern is nonetheless unmistakable.

The Fiscal Position of Each Income Group

The most revealing comparison subtracts federal taxes from means-tested transfers. This does not capture every benefit of government, nor does it include every state and local tax. It shows the direct redistributive relationship measured by CBO: income-qualified transfers received minus major federal taxes paid.

The lowest quintile received, on average, about $18,600 more in means-tested transfers than it paid in the major federal taxes allocated by CBO. The second quintile received roughly $4,000 more. The middle quintile crossed into net-contributor territory, paying approximately $7,000 more than it received. The highest quintile paid about $105,000 more, while the average top-1-percent household paid approximately $847,600 more than it received in means-tested transfers. Congressional Budget Office ³

This comparison gives the phrase “fair share” an entirely different complexion. The households most frequently accused of not contributing enough are, on average, the households financing the system after direct transfers are taken into account.

The lowest quintile’s means-tested transfers were equal to approximately 72.3 percent of its income before transfers and taxes. Medicaid and CHIP alone were valued at approximately 48 percent of pre-transfer income, SNAP at 10.1 percent, SSI at 4.8 percent, and other transfers at 9.4 percent. For the highest quintile, total means-tested transfers equaled approximately four-tenths of 1 percent of pre-transfer income. For the top 1 percent, the figure was about one-tenth of 1 percent. Congressional Budget Office ²

The tax-and-transfer system is not accidentally redistributive. Redistribution is embedded in its construction. Lower-income households receive a larger percentage of their resources through government transfers, while higher-income households face higher federal tax rates and receive almost none of those direct benefits.

Refundable Credits Are Transfers Administered Through the Tax Code

The word “credit” can conceal what is economically occurring. A nonrefundable credit can reduce a tax bill to zero, but it cannot ordinarily produce a payment beyond the taxpayer’s liability. A refundable credit can.

When refundable credits exceed a household’s pre-credit income-tax obligation, the IRS does not simply forgive the remaining tax. The excess becomes a net payment to the household. CBO explicitly treats such cases as producing a negative individual income-tax rate. Congressional Budget Office ²

The lowest quintile’s refundable credits were equal to approximately 12 percent of its income before transfers and taxes in 2022. These credits include the earned income tax credit, refundable portions of the child tax credit, premium tax credits, and certain education credits. Their introduction and expansion helped drive the lowest quintile’s average individual income-tax rate from around zero in the late 1970s to negative territory in many later years. Congressional Budget Office ²

A household receiving a $4,000 refundable credit after incurring $1,000 in pre-credit income-tax liability has not paid negative dollars in every category of taxation. It has received a $3,000 net payment through the individual income-tax system. Calling the entire amount a “tax cut” may be legally conventional, but the portion exceeding liability functions economically as a transfer.

This distinction matters because public discussions often present every taxpayer as though each were depositing money into the same common account. Many households are doing exactly that. Others are paying certain taxes while simultaneously receiving a larger amount through credits and transfers.

“The Wealthy Pay a Lower Percentage Than Their Secretaries”

This claim has circulated in several forms for decades. Occasionally it describes a real anomaly in a particular year. More often it depends on an unannounced change in the tax being measured, the income being counted, or the period being examined.

A salaried employee may face a higher payroll-tax rate than a billionaire whose income comes primarily from capital. A billionaire may also experience a year in which enormous unrealized asset appreciation is accompanied by relatively little taxable income. A founder whose stock rises by $10 billion has become wealthier on paper, but current income-tax accounting does not automatically include that $10 billion as taxable income if the shares were not sold.

A calculation that divides taxes paid by salary, interest, dividends, business income, and realized gains is an income-tax calculation. A calculation that divides taxes paid by the increase in a person’s total net worth is using a broader Haig-Simons-style economic-income concept. The second approach may be used to argue for a different tax base, but it cannot be presented as though it were the same measurement.

This substitution often produces the headline that a billionaire paid only a single-digit “true tax rate.” The numerator may contain federal income taxes actually paid, while the denominator includes billions of dollars in unsold stock appreciation. The result can be mathematically correct under the author’s chosen definition while remaining fundamentally incomparable to the effective tax rate quoted for an ordinary wage earner.

The wage earner’s house appreciation, pension growth, retirement-account appreciation, increase in business value, and other unrealized gains are rarely inserted into the denominator of the comparison. One household is measured using taxable or realized income while the other is measured using estimated wealth accumulation. The apparent contrast is created partly by comparing two different accounting systems.

Treasury proposals for minimum taxation of extremely wealthy households have made this distinction explicit by proposing to include unrealized gains in the tax base. Such a proposal is not simply an increase in the income-tax rate. It changes when appreciation is treated as taxable and moves the system toward mark-to-market or prepayment taxation for certain assets and taxpayers. U.S. Department of the Treasury ⁴

Unrealized Gains Are Economically Important—but They Are Not Cash Income

None of this requires pretending that unrealized gains are meaningless. A rising asset value can strengthen a person’s balance sheet, expand borrowing capacity, increase control over resources, and create economic options unavailable to someone living paycheck to paycheck.

The tax question is whether an unsold and fluctuating increase in value should be treated as current taxable income. That is a policy choice with serious consequences.

An asset worth $100 million in December may be worth $60 million several months later. A mark-to-market system must decide how losses are credited, whether taxpayers receive refunds for declining valuations, how private companies and illiquid assets are appraised, how valuation disputes are resolved, and how an owner pays the tax without selling part of the asset.

The liquidity issue is not imaginary. A founder may be extremely wealthy according to the estimated market value of a closely held business while receiving far less in annual cash. Farms, private companies, real estate holdings, art, mineral interests, and intellectual property do not always have a continuously observable market price.

Taxing such gains may still be defended as a policy. The policy should be described honestly as taxation of accrued wealth appreciation rather than evidence that conventional taxable income was somehow omitted from an ordinary effective-rate calculation.

Deferral and the Basis Step-Up Are Legitimate Areas of Debate

The realization principle allows a taxpayer to delay capital-gains taxation by continuing to hold an appreciated asset. Deferral has economic value because a dollar of tax paid years from now costs less in present-value terms than a dollar paid today.

The criticism becomes stronger when appreciated property is held until death. Under current federal rules, the basis of inherited property is generally reset to its fair market value at the decedent’s death or an alternate valuation date. Appreciation occurring during the decedent’s lifetime can therefore disappear from the heir’s future capital-gains calculation, although estate-tax rules may apply separately to sufficiently large estates. IRS ⁵

This is a real feature of the tax code and should not be waved away. It supports arguments for carryover basis, realization at death, or other reforms designed to prevent accrued gains from escaping income taxation permanently.

It does not support the broader claim that high-income households as a class pay little tax. CBO’s all-federal data already include realized capital gains and allocate a substantial corporate-tax burden to capital owners. The top 1 percent still faced an average federal rate of 31.5 percent and supplied more than one-quarter of the major federal taxes in 2022.

A tax code can contain preferences, deferrals, and avoidance opportunities while remaining highly progressive overall. Those propositions are not mutually exclusive.

Marginal Rates, Effective Rates, and Tax Shares Are Different Measurements

A large amount of tax commentary depends on moving among three measurements as though they were interchangeable.

The marginal rate is the rate applied to the next dollar of taxable income within a particular bracket. A taxpayer in the highest statutory bracket does not pay that rate on every dollar. Lower portions of taxable income are taxed through the lower brackets first. The IRS’s published bracket schedules make that graduated structure explicit. IRS ⁶

The effective rate is the tax paid divided by the selected measure of income. The result depends on whether the denominator is AGI, taxable income, cash income, income before transfers and taxes, consumption, or accrued economic income.

The tax share is the group’s portion of total tax collections. A group can pay a high effective rate and a large tax share, but the concepts remain distinct.

A fourth measurement, the net fiscal position, compares taxes paid with transfers received. This is often the most relevant measure when asking who finances redistribution, yet it is one of the least commonly included in political arguments.

The phrase “the wealthy pay 40 percent” might refer to their share of a particular tax, not a 40 percent effective rate. The phrase “the poor pay 10 percent” might include payroll and excise taxes as a share of income, not the share of national revenue they provide. Any analysis that does not identify the measurement is rhetorically convenient and analytically incomplete.

The Very Top Does Not Follow a Perfectly Smooth Progression

Accuracy requires acknowledging an irregularity in the IRS data. Although effective income-tax rates generally rise across broad income groups, the progression is not perfectly monotonic among the extremely wealthy.

For tax year 2023, the average individual income-tax rate for the top 1 percent was approximately 26.7 percent. The rate for the top 0.001 percent was about 23.6 percent. The top 0.01 percent averaged approximately 25.2 percent. IRS ¹

Several factors can contribute to this pattern: a larger share of income from long-term capital gains and qualified dividends, charitable deductions, business losses, timing of realizations, tax-exempt income not included in taxable income, and the volatility of capital income from one year to another.

This means the data do not support the absolutist claim that every additional step up the wealth ladder produces a higher effective income-tax rate in every year. They support the broader conclusion that high-income households pay much higher rates and vastly larger shares than lower-income households.

The difference matters. A rigorous argument does not need to pretend that the tax code is a perfectly smooth staircase. The overall gradient remains unmistakable.

Tax Expenditures Complicate the Benefits Question

Direct transfers are not the only way government confers financial advantages. Tax deductions, exclusions, preferential rates, deferrals, and credits are commonly described as tax expenditures because they reduce revenue relative to a specified baseline.

Higher-income households receive substantial dollar benefits from several tax expenditures. Preferential treatment of long-term capital gains and qualified dividends, deductions for charitable contributions, some retirement preferences, and certain business provisions can deliver large benefits to households at the top. CBO found that higher-income households received a larger dollar share of the major tax expenditures it examined for 2019, although the value relative to household income could be larger for lower-income groups in some cases. Congressional Budget Office ⁷

The earned income tax credit, by contrast, is concentrated among lower-income workers. The exclusion of employer-sponsored health-insurance premiums benefits a much broader range of households, while preferential capital-gains rates are concentrated heavily toward the upper end of the distribution.

A complete article should therefore avoid defining “benefit” only as a welfare check. The tax code distributes benefits through exclusions, deductions, credits, deferrals, and preferential rates as well as through direct spending.

The presence of those tax expenditures does not erase the underlying payment data. Tax liability reported by the IRS is already calculated after taxpayers have used the lawful deductions, credits, and preferences available to them. Even after those provisions, the top 10 percent paid more than 70 percent of individual income taxes in 2023.

Do Wealthy Households Use Fewer Public Services?

The answer depends entirely on the service.

For Medicaid, SNAP, SSI, housing assistance, income-qualified energy assistance, and refundable credits, the answer is generally clear. High-income households are either ineligible or receive very little. The direct benefit distribution is overwhelmingly concentrated in the lower quintiles.

Public education is more complicated. Higher-income families are more likely to use private education, and Census data show that private K–12 students come from households with higher median incomes than public-school students. In 2021, median household incomes for private kindergarten, elementary, and high-school students ranged from approximately $103,600 to $112,800, compared with approximately $79,900 to $89,500 for public-school students. Census.gov ⁸

That difference supports the proposition that some higher-income households pay school taxes while declining to consume public K–12 education directly. It does not mean affluent families universally avoid public schools. Many high-income households live in expensive districts precisely because of the public schools and consume an unusually costly local educational service through high property values and taxes.

Public transit also resists a simple income hierarchy. Lower-income households may be more dependent on transit because they have fewer vehicle alternatives, while affluent residents of dense urban centers may use rail and commuter systems extensively. Roads are used by families, employers, freight carriers, emergency vehicles, and customers across the income distribution.

Public defenders and income-qualified legal services are directly concentrated among people unable to afford private representation. Courts and contract enforcement, however, may provide substantial value to businesses, investors, landlords, lenders, and property owners.

Police and fire protection serve the entire jurisdiction. Higher-value property may mean a wealthy household has more financial value being protected, while high-crime neighborhoods may consume more police activity. Neither observation produces a clean household-level dollar allocation.

The strongest factual claim is therefore that high earners receive the least means-tested and income-qualified assistance. The claim that they receive the least total value from all government activity is considerably harder to prove and may be false under some methods of assigning institutional benefits.

Collective Public Goods Cannot Be Divided Like SNAP Benefits

National defense protects the country, its people, its territory, its commercial routes, and its economic system. A person cannot opt out of that protection, and an analyst cannot assign a precise household share without making contestable assumptions.

The same problem arises with courts, monetary stability, contract enforcement, disease surveillance, basic research, regulatory institutions, and interstate infrastructure. These systems create conditions within which economic activity occurs.

A wealthy investor may benefit more in dollar terms from property-rights enforcement because more property is being protected. A low-income worker may benefit more relative to income from public safety or a stable currency. A corporation may depend on roads and ports for every stage of its supply chain while never appearing as a household user of those facilities.

Any attempt to prove that one group receives the “least” from collective public goods will depend heavily on the allocation rule chosen. Assign the same dollar value per person, and the distribution appears equal. Assign value according to income or property protected, and more benefit flows upward. Assign value according to usage or dependence, and the pattern changes service by service.

This uncertainty is a reason for precision, not a reason to abandon the analysis. Direct transfers and tax liabilities remain measurable even when collective goods are not.

State and Local Taxes Must Be Discussed Separately

The federal system is not the entire American tax system. States and localities rely on different combinations of income, sales, property, excise, business, and severance taxes.

Sales and excise taxes often consume a greater share of lower-income households’ income because those households spend a larger percentage of available resources. Property taxes can also be burdensome relative to income, including for renters who bear part of the cost through rent.

State income taxes vary from sharply progressive to flat-rate systems, while several states impose no broad individual income tax. Local tax burdens also depend on home values, consumption, location, business ownership, and local service structures.

These taxes make the combined federal, state, and local system less progressive than the federal individual income tax standing alone. They do not justify speaking as though federal income-tax figures are false. They answer a different question.

A rigorous national article should make its scope explicit: the clearest and most current comprehensive distributional data concern federal taxes and means-tested transfers. State and local incidence requires a separate model and can vary dramatically by jurisdiction.

Annual Income Is Not Lifetime Income

Income quintiles are not permanent castes. A medical student, temporarily unemployed engineer, retired millionaire, small-business owner reporting a loss, and chronically low-wage worker can all appear in a low annual-income group for entirely different reasons.

A successful entrepreneur may report a massive capital gain in one year and little in the next. A retiree may own a valuable home and investment portfolio while reporting modest taxable income. A worker receiving overtime or a one-time bonus can move upward temporarily.

Annual distributional data remain essential because taxes and most transfers are administered annually. They should not be mistaken for a complete lifetime accounting.

A lifetime analysis would include taxes paid throughout a career, Social Security and Medicare benefits received in retirement, public education consumed during childhood, transfers received during unemployment or disability, inheritance taxes, and the changing value of public services over decades.

Such an accounting might show that some households move from net-recipient status early in life to net-contributor status in their peak earning years and back toward recipient status in retirement. The annual CBO data provide a cross-section of the system, not the entire biography of every household.

Social Security and Medicare Are Not Ordinary Means-Tested Welfare

CBO’s income-before-transfers-and-taxes measure includes major social-insurance benefits such as Social Security and Medicare. This is important because retirees across the income distribution can receive them.

These programs combine contributory and redistributive characteristics. Workers and employers pay dedicated payroll taxes, benefits depend partly on earnings histories, and benefit formulas replace a larger percentage of prior earnings for lower-wage workers. Medicare also pools health risks and provides benefits whose value can exceed an individual’s lifetime contributions.

It would therefore be misleading to place Social Security and Medicare in exactly the same category as SNAP or housing assistance. A high-income retiree receiving Social Security is not receiving an income-qualified transfer, although the program’s formula and financing still have distributional effects.

The direct tax-versus-means-tested-transfer comparison used here intentionally does not pretend to be a complete lifetime actuarial balance for Social Security and Medicare.

Dollars and Percentages Answer Different Moral Questions

Suppose Household A earns $40,000 and pays $4,000 in taxes. Household B earns $4 million and pays $1 million. Household A has paid 10 percent, while Household B has paid 25 percent.

The higher-income household has paid 250 times as many dollars and two and a half times the effective rate. It will almost certainly receive fewer means-tested benefits. A political argument can still declare the $1 million insufficient, but it cannot honestly claim that Household B contributed less under either the dollar or percentage measure.

Another example produces the rhetorical result critics often seek. Suppose Household B’s assets also appreciate by $10 million. Divide the $1 million tax payment by the $14 million combination of income and unrealized appreciation, and the calculated rate falls to 7.1 percent.

The tax did not change. The denominator did.

Whether the $10 million should be taxed currently is a legitimate debate. Presenting 7.1 percent as directly comparable to Household A’s 10 percent tax on cash income hides the methodological choice that created the result.

Percentages are useful, but they do not erase dollars. The difference between paying $4,000 and paying $1 million matters to the financing of government even when both figures are expressed as rates.

What Would a Neutral Definition of “Fair Share” Look Like?

A neutral definition would have to establish the principle before identifying the taxpayer.

One possibility is proportionality: each household pays the same percentage of income. The federal system exceeds that standard at the upper end because effective rates rise with income.

Another possibility is benefit taxation: each household pays according to the government benefits it receives. That standard would place a smaller burden on many affluent households for direct transfers, but potentially a larger burden if institutional protections were assigned according to property, business activity, or wealth protected.

A third possibility is ability to pay: households with greater economic resources pay higher percentages. The federal tax system is already built substantially around this principle.

A fourth possibility is equal sacrifice: taxes should impose a comparable loss of well-being rather than an equal dollar or percentage payment. This approach is highly subjective because the marginal value of income cannot be directly observed or agreed upon.

A fifth possibility is openly redistributive: tax rates should be set not only to finance government but to reduce the concentration of income or wealth. Under that philosophy, no existing contribution necessarily satisfies the demand because the objective is a different distribution of resources rather than the funding of a defined level of spending.

Most “fair share” rhetoric moves among these principles without announcing the transition. It invokes ability to pay, suggests benefit taxation, displays percentages, emphasizes wealth inequality, and concludes with a redistributive demand.

The ambiguity is functional. Once a standard is made explicit, it can be tested.

What the Data Establish

The data do not establish that every tax preference is justified. They do not establish that every billionaire faces a higher rate than every secretary in every year. They do not establish that unrealized gains must remain untaxed forever or that capital gains should receive their current treatment.

They establish several conclusions that should no longer be treated as matters of speculation.

Higher-income households pay substantially higher average federal tax rates than lower-income households. The top 1 percent paid an average total federal rate of 31.5 percent in CBO’s 2022 analysis, compared with 1.4 percent for the lowest quintile.

Upper-income households supply most federal revenue. The highest quintile paid approximately 70 percent of the major federal taxes allocated by CBO, while the top 10 percent of individual filers paid more than 70 percent of federal individual income taxes in the IRS’s 2023 data.

Lower-income households receive most means-tested transfers. The bottom two quintiles received approximately 73 percent of those benefits in 2022.

The direct tax-and-transfer system makes the bottom two quintiles net recipients on average and makes the middle, fourth, and highest quintiles net contributors. The average top-1-percent household paid hundreds of thousands of dollars more in federal taxes than it received in means-tested transfers.

Refundable tax credits cause the lowest quintile’s average individual income-tax rate to become negative, although payroll and other federal taxes bring its total federal rate above zero.

Arguments claiming that billionaires pay extraordinarily low percentages often use increases in net worth, including unrealized gains, as the denominator. That is a broader economic-income calculation rather than the same taxable-income measure generally applied to wage earners.

The federal tax code contains real preferences and deferral opportunities for capital income. Those provisions can be examined and reformed without denying that the overall federal system remains strongly progressive.

The Question the Slogan Avoids

The central fact is not difficult to understand. The American federal tax-and-transfer system collects disproportionately from the top and distributes disproportionately toward the bottom.

A person may support more redistribution. A person may favor a wealth tax, higher capital-gains rates, realization at death, lower estate-tax exemptions, or mark-to-market taxation for publicly traded assets. Those are policy positions that should be defended according to their economic effects, administrative feasibility, constitutional implications, and incentives for investment and capital formation.

The policy should not be smuggled into the conversation through the unsupported premise that affluent Americans currently contribute little.

The top 1 percent pays nearly two-fifths of federal individual income taxes. The top 10 percent pays more than seven-tenths. The highest quintile pays roughly 70 percent of the major federal taxes. The bottom two quintiles receive nearly three-quarters of means-tested transfers. The average highest-quintile household pays more than $100,000 beyond what it receives in such transfers, while the average lowest-quintile household receives roughly $18,600 beyond what it pays.

Those figures do not dictate what tax policy must be. They establish the point from which an honest tax-policy discussion must begin.

“Fair share” is not an accounting term. It is a political conclusion looking for a denominator.

Before demanding more from the people already financing most of the federal government, the burden rests on the claimant to define what amount would finally be considered enough, what measure of income is being used, which taxes are included, which benefits are being counted, and what economic consequences are acceptable in pursuit of that undefined standard.

Without those answers, the phrase is not an analysis of the tax system. It is a permission slip for an unlimited claim on someone else’s earnings, capital, and property.