Foundationspublic finance
Progressive vs. Regressive Taxes: Who Pays What Share?
Overview
Should the rich pay higher rates (progressive), everyone the same rate (proportional), or do the poor pay more as a share (regressive)? The question structures every tax debate. But 'progressive' describes the system's burden distribution, not just rate schedules — and measuring it requires incidence analysis. This lesson defines the terms, measures real systems, examines the flat-tax and optimal-progression debates, and connects to the equity-efficiency frontier.
Learning objectives
- Define progressive, proportional, and regressive taxation
- Distinguish marginal from average rates
- Measure system-wide progressivity with incidence
- Evaluate the US and Nordic systems' actual progressivity
- Debate optimal progression: the equity-efficiency frontier
Definitions
A tax is progressive if the average rate rises with income (the rich pay a larger share); proportional (flat) if the rate is constant; regressive if the average rate falls with income (the poor pay a larger share). Note: it's about average rates across incomes, not the rate schedule's shape alone. A flat-rate tax with a basic exemption is progressive in effect (average rates rise from zero). Sales taxes are typically regressive (the poor spend larger shares of income); income taxes with graduated brackets are progressive by design. Crucially, judge the system's total burden — income + payroll + sales + property — not one tax in isolation.
Marginal rate = tax on the next dollar (drives incentives). Average rate = total tax ÷ income (drives fairness judgments). Progressivity debates need both; incentive debates need marginal.
Measuring the real system
Statutory schedules mislead; incidence-corrected burdens reveal. The US: federal income tax is progressive (top 1% pay ~40% of it), but payroll taxes are regressive (capped), sales taxes regressive, and corporate incidence contested — the total system is mildly progressive overall, less than the income-tax schedule suggests. Nordics: famously progressive outcomes via large transfers, but their tax systems are surprisingly flat — heavy VAT and payroll taxes on everyone fund universal benefits. The lesson: progressivity lives in spending as much as taxing. A flat tax funding generous transfers can be more progressive than a graduated tax funding little.
High redistribution with relatively flat taxes: the Nordics tax everyone heavily (VAT ~25%) and redistribute via universal services. Progressivity through spending, not rate schedules.
The flat-tax debate
Proposals (Hall-Rabushka, Eastern European adoptions): one rate, broad base, large exemption. Case for: simplicity (postcard returns), efficiency (uniform marginal rates minimize distortions), transparency. Case against: less progression than graduated systems (the exemption does limited work at the top), and the real complexity is defining the base, not the rates. Evidence from flat-tax adopters (Russia, Baltic states): revenue and compliance improved initially — but from base-broadening and better administration, not the flat rate per se. Most economists: the rate structure matters less than the base; a clean graduated tax beats a loophole-ridden flat one.
Tax debates obsess over rate schedules while the base (what's taxed, what deductions exist) drives both revenue and fairness. A flat rate on a Swiss-cheese base is neither simple nor fair.
How progressive should it be? The optimal-tax lens
Mirrlees (1971): optimal marginal rates balance redistribution gains against incentive costs — depending on the income distribution's shape, labor-supply elasticities, and society's inequality aversion. Modern calibrations (Saez, Piketty, Stantcheva): with today's inequality and modest top elasticities, optimal top rates may be 60–80% — far above current US levels. But the results hinge on contested parameters: how much do top earners really respond (real effort vs. avoidance?), and how much does society value equality? The framework disciplines the debate: state your elasticity beliefs and equity weights, and the optimal schedule follows. Disagreements become empirical, not theological.
Modern optimal tax: you don't need the full model — key elasticities and the income distribution largely determine optimal rates. The debate narrows to measurable parameters.
Beyond rates: what really determines fairness
Three underappreciated determinants. One: transfers dominate — the bottom half's net position depends more on benefits received than taxes paid. Two: lifetime vs. annual incidence — students and retirees look poor annually but not lifetime; consumption taxes look less regressive lifetime. Three: the pre-tax distribution — market inequality (wages, capital ownership) matters more than tax tweaks; predistribution (education, labor markets, antitrust) may beat redistribution. The mature view: judge the whole fiscal system — taxes plus transfers plus public services — over lifetimes, against the market distribution it modifies. Rate schedules are the visible tip of a deeper structure.
Key points
- Progressive = average rates rise with income; regressive = they fall. Judge average rates and the whole system, not single schedules.
- The US system is mildly progressive overall; Nordics achieve redistribution via flat-ish taxes funding universal transfers.
- Flat taxes' real effects come from base-broadening, not the single rate — the base matters more than the schedule.
- Optimal-tax theory (Mirrlees, Saez): optimal top rates depend on elasticities and inequality aversion — an empirical debate.
- True fairness = taxes + transfers + services over lifetimes, against the market distribution.
Common misconceptions
✓ Truth: Federal income tax is progressive, but payroll caps, sales taxes, and contested corporate incidence make the total system only mildly progressive — far less than the bracket schedule implies.
✓ Truth: Equal marginal rates still produce unequal burdens relative to ability — and a flat rate with exemptions is actually progressive in average rates. 'Flat' describes the schedule, not the fairness.
✓ Truth: Beyond some point, avoidance, reduced effort, and migration erode the base — the Laffer-curve logic. Where the peak is depends on elasticities; it's an empirical question, not an ideological one.
Practice questions
Q1. A tax is progressive if:
- The rich pay more dollars
- The average tax rate rises with income
- The marginal rate is constant
- It's levied on income
Progressivity is about average rates across the income distribution. A flat-rate tax with an exemption is progressive in this sense — average rates rise from zero.
Q2. The 'Nordic paradox' in tax progressivity is that:
- Nordics have no taxes
- Nordic redistribution is high but achieved through relatively flat taxes funding universal transfers — progressivity via spending, not rate schedules
- Nordics are regressive
- Only the rich pay taxes there
Heavy VAT and payroll taxes on everyone fund generous universal benefits. The system's progressivity lives in transfers and services, not graduated income-tax brackets.
Q3. In optimal-tax theory (Mirrlees-Saez), the optimal top marginal rate depends mainly on:
- The deficit level
- The income distribution's shape, top earners' behavioral elasticities, and society's inequality aversion
- The inflation rate
- Trade balances
The sufficient-statistics logic: Pareto tail thickness, elasticity of taxable income, and welfare weights pin down optimal top rates — turning ideology into empirical debate.
Click a question to reveal the answer and explanation.
Frequently asked questions
What is the difference between marginal and average tax rates?
Marginal: tax on the next dollar earned (drives work/investment incentives). Average: total tax ÷ total income (drives fairness assessment). A 37% top marginal rate coexists with much lower average rates.
Why are sales taxes regressive?
Lower-income households spend (vs. save) a larger share of income, so a uniform consumption tax takes a larger share of their income. Exemptions for necessities soften but don't erase this.
What is 'predistribution'?
Reducing market inequality before taxes — via education, minimum wages, unions, antitrust — rather than redistributing after. Advocates argue it's more efficient and politically durable than tax-and-transfer.
Do the rich really pay most income taxes?
In the US, the top 1% pay ~40% of federal income tax — but that's one tax. Their share of total taxes (including payroll, sales, property) is lower, and their share of income is the relevant benchmark.