Compendium

Public Economics and Taxation

30 min read

Public Economics and Taxation

Public economics studies the role of government in the economy — what it produces, what it regulates, what it taxes, what it transfers — under the standard normative criteria of efficiency, equity, simplicity, and political feasibility. Modern public economics combines welfare theorems and externality theory to characterise what governments should do, optimal tax theory to characterise how they should raise revenue, and a large empirical literature using administrative data and quasi-experimental designs to estimate the elasticities that those normative theories require. This note covers the canonical theorems, the main tax instruments and their incidence, the empirical findings on labour-supply, capital, and corporate responses, and the contemporary debates over wealth taxation, universal basic income, and industrial policy.

See also

1. The role of government

The First Welfare Theorem states that a competitive equilibrium under standard convexity assumptions is Pareto efficient. The Second Welfare Theorem states that any Pareto-efficient allocation can be supported as a competitive equilibrium given suitable lump-sum redistribution. Two standard reasons that governments nonetheless intervene:

  • Market failure. Externalities, public goods, market power, asymmetric information.
  • Distributional concerns. Even an efficient outcome may be socially unacceptable; redistribution is a primary government function.

Government activity in developed economies is large by any historical standard. As of 2024 OECD data, general-government expenditure as a share of GDP is roughly 39% in the United States, 44% in the United Kingdom, 49% in Germany, 56% in France, and 47% across OECD countries on average. Tax-to-GDP ratios run roughly 27% in the US, 33% in the UK, 39% in Germany, and 45% in France. The composition differs sharply: the US relies more on income and payroll taxes and less on consumption taxes than the European average.

2. Public goods

A pure public good is non-rival (one person’s consumption does not reduce another’s) and non-excludable (the producer cannot prevent consumption). Classical examples: national defense, lighthouses, basic research, the rule of law. Goods can be partially public — congestible (public roads), club goods (cable TV), open-access common-pool resources (fisheries).

2.1 The Samuelson condition

Samuelson 1954 Review of Economics and Statistics 36 derived the efficiency condition for public-good provision:

∑_i MRS_i = MRT.

That is, the sum of individuals’ marginal rates of substitution between the public good and a private good equals the marginal rate of transformation in production. This contrasts with private goods, where each consumer’s MRS equals the MRT. Because each consumer benefits from the public good’s quantity simultaneously, the sum across consumers measures social marginal benefit.

2.2 Free riding and revelation

The private-incentive failure: an individual considering voluntary contribution to a public good faces marginal cost equal to the contribution but only personal marginal benefit, ignoring the benefit to others. Free riding is the equilibrium prediction.

Mechanisms that elicit truthful preferences:

  • Lindahl pricing (Lindahl 1919). Personalised tax prices proportional to marginal benefit. Theoretically efficient but informationally demanding.
  • Vickrey-Clarke-Groves (VCG, Vickrey 1961 Journal of Finance; Clarke 1971; Groves 1973). Truthful revelation as a dominant strategy. Pivotal-bidder taxes. Heavily used in auction design (FCC spectrum auctions, sponsored search).
  • Median voter (Black 1948). Under single-peaked preferences, the median voter’s preference defeats all others in pairwise majority. Predicts political outcomes track the median voter’s preferred level of public spending — empirically a weak approximation in practice.

2.3 Empirical evidence on public-good provision

Bergstrom-Blume-Varian 1986 Journal of Public Economics 29 formalised the voluntary-contribution model; experimental and field studies (Ostrom Governing the Commons 1990; List-Lucking-Reiley 2002 American Economic Review 92) documented when private provision succeeds and fails. The Andreoni 1990 Economic Journal “warm glow” model rationalises private donations to public goods through impure altruism.

3. Externalities and Pigouvian taxation

Pigou 1920 The Economics of Welfare identified externalities — uncompensated costs or benefits accruing to third parties — and proposed corrective taxes. The Pigouvian tax sets t = marginal external damage at the optimum, internalising the externality.

3.1 Coase theorem

Coase 1960 Journal of Law and Economics 3 — given well-defined property rights and zero transaction costs, parties will negotiate to the efficient outcome regardless of which party initially owns the right. The Coase theorem’s policy implication is not laissez-faire (transaction costs are rarely zero) but a focus on transaction-cost-minimising institutional design.

3.2 Carbon pricing

The canonical contemporary Pigouvian application. The social cost of carbon (SCC) is the present-value damage of an additional ton of CO2 emissions; current US government estimates (Interagency Working Group, EPA 2023) place the SCC at roughly 51 figure (in 2020 dollars). The EPA also computes SC-CH4 and SC-N2O.

Major carbon-pricing schemes:

  • EU Emissions Trading System (ETS). Cap-and-trade for power, heavy industry, intra-EU aviation since 2005; extended to maritime in 2024 and to road transport and buildings under ETS2 starting 2027. Allowance price was €82/tCO2 in late 2024, having peaked near €100 in 2023 (down from €25 pre-2018). Phase IV (2021–2030) caps fall ~4.3% per year through 2030.
  • California Cap-and-Trade (2013+). Linked with Quebec since 2014. Allowance prices around $30/tCO2 in 2024.
  • Regional Greenhouse Gas Initiative (RGGI). Northeast US power-sector cap-and-trade; ~$15/tCO2 2024.
  • British Columbia carbon tax (2008+). Started at C80 (~US$60) in April 2024. Revenue-neutral by statute through income-tax reductions and credits.
  • Canada federal backstop (2019+, OBPS + fuel charge). Federal carbon tax for non-compliant provinces; price scheduled to reach C$170/tCO2 by 2030.
  • Sweden carbon tax (1991+). Highest in world at roughly SEK 1 460 (~$140) per tCO2 on heating fuels in 2024.
  • UK ETS, Switzerland ETS, New Zealand ETS, South Korea ETS, China national ETS (2021+, power sector only initially).

The EU Carbon Border Adjustment Mechanism (CBAM, transitional phase October 2023, full enforcement January 2026) imposes a carbon-equivalent levy on imported steel, aluminium, cement, fertilisers, electricity, hydrogen — addressing carbon leakage and providing political cover for higher domestic carbon prices.

3.3 Congestion pricing

Vickrey 1969 American Economic Review 59 proposed time-of-day road pricing; first implemented Singapore Area Licensing 1975, ERP 1998; London Congestion Charge 2003; Stockholm 2006 (after 2006 referendum); Milan 2012; Manhattan 2025 (initially scheduled 2024, postponed). The Singapore ERP charges Sg9 for passenger cars during peak periods, expected to raise $1B annually and reduce CBD traffic ~17%.

3.4 Pollution and air-quality regulation

Acid Rain Program under 1990 Clean Air Act Amendments was the first large-scale US emissions trading (sulfur dioxide); reduced SO2 emissions ~50% by 2007 at far lower cost than command-and-control alternatives (Schmalensee-Stavins 2013 Review of Environmental Economics and Policy 7).

EPA’s Mercury and Air Toxics Standards (MATS), Cross-State Air Pollution Rule (CSAPR), Good Neighbor Rule. Recent Sackett v EPA (2023 US 651) narrowed Clean Water Act jurisdiction over wetlands.

4. Optimal income taxation

4.1 Mirrlees model

Mirrlees 1971 Review of Economic Studies 38 founded modern optimal income taxation. The model: heterogeneous individuals differ in productivity (wage rate) w, derive utility from consumption c and disutility from labour l, earn z = wl. Government observes z but not w or l separately, sets a tax schedule T(z), maximises social welfare:

∫ W(u(c(w), l(w))) f(w) dw,

subject to government budget ∫ T(z(w)) f(w) dw = R and individual incentive compatibility (each type w chooses (c, l) optimally given the schedule).

Mirrlees’s key insight: the optimal marginal tax rate at income z balances the equity gain (marginal redistribution) against the efficiency loss (behavioral response of types above z). Saez 2001 Review of Economic Studies 68 re-derived the formula in elasticity-friendly form:

T’(z) / (1 - T’(z)) = (1 + 1/e) · A(z) · G(z),

where e is the elasticity of taxable income, A(z) is a Pareto-thickness term capturing the density of types above z, and G(z) is the social marginal-welfare weight on those above z.

Implications:

  • Optimal marginal tax rate is zero at the top of a bounded distribution (Sadka 1976; Seade 1977). With a bounded skill distribution there is no one to redistribute to at the very top.
  • Optimal marginal rates near the top are large for unbounded (Pareto-tailed) distributions. Saez computed roughly 50–80% top rates for US data, depending on elasticity assumptions.
  • U-shaped optimal marginal rates. High marginal rates at both bottom (phase-out of transfers) and middle/top; lower in between.
  • Diamond-Saez 2011 Journal of Economic Perspectives 25 synthesis. “Top tax rates: a scientific debate” — recommends top US marginal rates of 50–70% for revenue maximisation.

4.2 Elasticity of taxable income

Feldstein 1995 Journal of Political Economy 103 introduced the elasticity of taxable income (ETI) as the sufficient statistic combining all behavioral responses to taxation: labour supply, effort, evasion, deductions, income shifting between tax bases. Long-running debate over its magnitude:

  • Saez-Slemrod-Giertz 2012 Journal of Economic Literature 50 surveyed: 0.12–0.40 for the broad ETI.
  • Chetty 2009 American Economic Journal: Economic Policy 1 — distinguishes structural ETI (compensated labour supply) from total ETI; bunching at kink points yields micro-elasticity estimates often near 0.2.
  • Kleven-Schultz 2014 American Economic Journal: Economic Policy 6 using Danish administrative data found long-run ETI of 0.2 with little evasion margin.

4.3 Tagging and categorical transfers

Akerlof 1978 American Economic Review 68 argued for transfers contingent on observable characteristics correlated with low ability (disability, age, family status). The EITC is a major US application. Pure tagging is constrained by horizontal-equity concerns and administrative complexity.

4.4 Mirrlees Review (Mirrlees-Adam-Besley-Blundell-Bond-Chote-Gammie-Johnson-Myles-Poterba 2010, 2011)

Comprehensive review of UK tax design over five years, published by IFS. Recommendations included: integration of tax and benefit schedules, broader consumption tax base, neutral capital-income taxation, removal of distortions across tax bases.

5. Optimal commodity taxation

5.1 Ramsey rule

Ramsey 1927 Economic Journal 37 — minimise deadweight loss raising a given revenue R by levying commodity taxes inversely proportional to compensated demand elasticities (the inverse elasticity rule, applicable under separable preferences and no income tax). General Ramsey rule: optimal tax structure equalises proportional changes in compensated quantities across goods.

5.2 Diamond-Mirrlees production efficiency

Diamond-Mirrlees 1971 American Economic Review 61 — under standard assumptions, optimal taxation does not distort production decisions: only consumer prices are taxed; intermediate-input and investment taxes should be zero. Implies (i) no taxes on intermediate inputs (a justification for VAT design), (ii) producer prices equal marginal social costs.

5.3 Atkinson-Stiglitz theorem

Atkinson-Stiglitz 1976 Journal of Public Economics 6 — when utility is weakly separable between labour and a vector of consumption goods, and a nonlinear income tax is available, optimal commodity taxes are uniform (or equivalently, can be replaced by income tax). Justifies broad-based consumption taxes (VAT) over differentiated rates. The result has been pushed in many directions; deviations from weak separability or restricted income-tax instruments justify non-uniform commodity taxes (e.g., higher tax on goods complementary to leisure).

5.4 VAT vs sales tax vs income tax

VAT applies a tax at each stage of production with credit for input VAT — economically equivalent to a retail sales tax under uniform rates, but more enforceable because the credit mechanism creates a paper trail and self-enforcement (Keen 2007 Journal of Public Economics 92, “VAT, tariffs, and withholding”). VAT introduced in France 1954 (Maurice Lauré); now used by ~170 countries. The US is the only major economy without a national VAT.

Empirical comparisons:

  • VAT. Broad base, neutral between goods/services, exports zero-rated, imports taxed at border. Self-enforcing credit chain. Common rates: 19% Germany, 20% France/UK, 25% Sweden/Denmark, 10% Australia/Japan, 5% Canada GST (plus provincial).
  • Retail sales tax. US state-level taxes; rates 0–~10% combined state/local. Narrower base (exemptions for food, prescription drugs, services); harder to enforce because last-stage retailers bear all enforcement burden. Wayfair v South Dakota 2018 585 U.S. ___ permitted states to require remote sellers to collect sales tax — fundamental change for e-commerce.
  • Income tax. Tax base = labour + capital income, with deductions. Progressive rate schedules. Withholding for wages; quarterly estimated payments for the self-employed.

The Mirrlees Review (2011) endorses a broader, more uniform VAT combined with redistribution through the income-tax/benefit system rather than differentiated VAT rates.

6. Tax incidence

The party who legally remits a tax (statutory incidence) need not bear its economic burden (economic incidence), which depends on the relative elasticities of supply and demand.

6.1 Partial-equilibrium incidence

For an excise tax on a competitive market with linear supply and demand:

share borne by consumers = e_S / (e_S - e_D), share borne by producers = -e_D / (e_S - e_D),

where e_S > 0 is supply elasticity and e_D < 0 is demand elasticity. Inelastic side bears more.

6.2 Payroll tax incidence

Statutorily split between employer and employee in most countries (US: 7.65% employer + 7.65% employee = 15.3% FICA combined). Economic incidence: empirical evidence (Saez-Schoefer-Seim 2019 American Economic Review 109; Gruber 1997 Journal of Public Economics 65) suggests labour bears most of the long-run burden — labour supply is inelastic relative to labour demand. Therefore the statutory split is roughly fictional in the long run.

6.3 Corporate tax incidence

A major active question. The pre-2017 consensus that capital bore much of US corporate-tax burden (Harberger 1962 Journal of Political Economy 70 in a closed economy; Auerbach 2006 surveyed) has been challenged by open-economy models where capital is internationally mobile and labour bears a substantial share. Empirical estimates:

  • Suárez Serrato-Zidar 2016 American Economic Review 106 — exploiting US state corporate-tax differences, found ~30–35% of corporate-tax burden borne by workers via wage effects.
  • Fuest-Peichl-Siegloch 2018 American Economic Review 108 — German municipal trade tax variation; ~half of corporate-tax burden borne by workers.
  • Saez-Zucman 2019 Brookings Papers on Economic Activity — argue for incidence-on-capital-owners model in the US given large foreign-owned share of US corporate equity.

6.4 Tax shifting and capitalisation

Property taxes are partially capitalised into house prices (Tiebout-Yinger 1995). Estate taxes affect inter vivos transfers. Carbon taxes are partly capitalised into capital values of carbon-intensive assets (Hsiao 2024 on coal-plant valuation).

7. Deadweight loss

Excess burden of taxation: the welfare loss above and beyond revenue raised. For a small tax t on a good with linear demand and supply,

DWL ≈ (1/2) · t^2 · |e_D| · e_S / (|e_D| + e_S) · (p_0 · q_0 / |e_D - e_S|).

Hausman 1981 American Economic Review 71 developed the exact (not Marshallian-approximate) compensating-variation measure. Modern empirical estimates of total deadweight loss of the US tax system: roughly 20–50% of revenue raised, with high uncertainty driven by labour-supply elasticity assumptions.

Bunching designs (Saez 2010 American Economic Journal: Economic Policy 2; Chetty-Friedman-Olsen-Pistaferri 2011 Quarterly Journal of Economics 126) identify ETI from excess mass at kink points or notches — clean modern method.

8. Tax bases in detail

8.1 Personal income taxation

US federal individual income tax brackets (2024 single filer, taxable income):

  • 10% up to $11 600
  • 12% 47 150
  • 22% 100 525
  • 24% 191 950
  • 32% 243 725
  • 35% 609 350
  • 37% above $609 350

Standard deduction 29 200 joint (2024); QBI deduction (Section 199A) 20% for pass-through income.

The Tax Cuts and Jobs Act (TCJA, December 2017, PL 115-97) reduced top rate from 39.6% to 37%, doubled standard deduction, capped SALT deduction at $10 000, cut corporate rate from 35% to 21%, lowered estate-tax-exemption thresholds. Most individual provisions sunset at end of 2025 — major fiscal-policy decision pending.

Most countries: progressive rate schedule plus deductions, with rate variation (top rates: 37% US federal, 45% UK, 45% Germany, 55% Denmark, 50% Australia). Flat tax adopted in some Eastern European countries (Russia 13% until 2021, then 15% for high incomes; Estonia 20%).

8.2 Payroll taxation

US Social Security (OASDI): 12.4% on wages up to $168 600 (2024 wage base), split 6.2/6.2 between employer/employee. Medicare: 2.9% (1.45/1.45) on all wages, plus additional 0.9% Medicare on high earners (since 2013 ACA). Self-employed pay both shares directly (15.3% + 2.9%, deducting half from income).

Social Security finances retirement (PIA formula based on AIME, primary insurance amount), disability (SSDI), and survivors benefits via PAYGO. Trust Fund projected to be depleted ~2033 (OASI alone); SSDI in better shape. Annual Social Security trustees report.

Comparable European systems: higher payroll-tax rates (40–50% combined employer-employee in France/Germany), funding broader benefits (unemployment, health, family).

8.3 Corporate income taxation

US corporate rate: 21% federal (since TCJA 2018, down from 35%), plus state corporate taxes averaging ~4%. Combined effective rate around 25% statutorily but lower for many firms via deductions and credits.

International:

  • OECD average statutory corporate rate. Around 23% in 2024, down from ~32% in 2000.
  • Ireland. 12.5% (with 15% minimum from 2024 under Pillar Two for largest firms).
  • Hungary. 9% (lowest in EU).
  • Bermuda, Cayman Islands, Jersey. 0% (corporate income tax havens).
  • UAE. Introduced 9% corporate tax June 2023, ending zero-rate regime.

The Inclusive Framework’s Pillar Two global minimum tax (15% effective rate on large MNEs, agreed October 2021 at OECD/G20; in-force in 145 jurisdictions during 2024–2026) is the largest international tax-coordination project since the OECD Model Convention. Pillar One (reallocation of taxing rights for largest digital and consumer-facing firms) remains stalled as of late 2024.

8.4 Capital-gains and dividend taxation

US: long-term capital gains (assets held >1 year) taxed at 0/15/20% depending on income; short-term at ordinary rates. Net Investment Income Tax (NIIT) of 3.8% on top of that for high-income taxpayers since ACA. Qualified dividends taxed at long-term capital-gains rate.

Step-up in basis at death eliminates accrued capital gains from the income tax base — a major US tax preference. Biden 2021 American Families Plan proposed limiting step-up; not enacted.

European systems vary: UK 10/20% (basic/higher), Germany 25% flat capital-gains tax (Abgeltungsteuer).

8.5 Estate, gift, and inheritance taxation

US federal estate tax: 40% on estates above 5M); sunsets back to ~$7M in 2026 absent legislation. Gift tax integrated with estate (lifetime exemption shared). Generation-skipping transfer tax (GSTT). State estate/inheritance taxes vary (12 states + DC have estate tax, 6 states have inheritance tax).

European: UK 40% above £325 000; Germany progressive 7–50% depending on relationship and amount; France progressive 5–60%.

Substantial behavioural responses: Kopczuk 2007 Journal of Public Economics 91 found large mass at the estate-tax filing threshold suggesting tax avoidance through valuation discounts. Goupille-Lebret-Infante 2018 American Economic Journal: Economic Policy 10 found French regional inheritance-tax variation generated migration responses.

8.6 Property taxation

State and local in US (~$700B 2023 collected). Computed as assessed value × millage rate; rates 0.3–2.5% of assessed value depending on state. Property taxes are economically efficient (immobile base) and salient. Proposition 13 (California 1978) capped assessed-value growth at 2%/year and reassessment only on sale; created intergenerational redistribution and lock-in effects.

Land-value taxation (LVT) is the canonical efficient property tax: incidence on landowners only, no behavioural distortion on improvements. Henry George Progress and Poverty 1879 proposed LVT as the “single tax” (universal funding from land rents). Modern advocates: Stiglitz-Henry George Theorem (Stiglitz 1977; Arnott-Stiglitz 1979 Quarterly Journal of Economics 93). LVT implemented partially in Estonia, Denmark, Pennsylvania (Pittsburgh, Harrisburg, Allentown).

8.7 Wealth taxation

A levy on accumulated net worth, distinct from taxing the income generated by wealth. As of 2024 only a few OECD countries operate net-wealth taxes (Switzerland — cantonal, ~0.1–1%; Norway — 0.85% on net wealth above NOK 1.7 M with extra 0.4% above 20 M; Spain — patchwork with re-introduction since 2022; Colombia — variable). Most European wealth taxes (France, Germany, Sweden) were abolished in 1990s–2000s after high administrative cost and capital flight.

US proposals:

  • Warren wealth tax (2019 campaign). 2% on wealth above 1 B (later 6% under Ultra-Millionaire Tax Act). Saez-Zucman estimated $2.75 T over 10 years.
  • Sanders wealth tax. Higher rates and lower thresholds.
  • Biden Billionaire Minimum Income Tax (2022 budget). Effective minimum 20% on income including unrealized capital gains for households worth >$100M. Not enacted.

Empirical wealth-tax evidence:

  • Brülhart-Gruber-Krapf-Schmidheiny 2022 American Economic Journal: Economic Policy 14 — Swiss canton variation, large elasticity of declared wealth to tax rate (~30%) but the response is largely re-valuation and avoidance rather than real saving.
  • Jakobsen-Jakobsen-Kleven-Zucman 2020 Quarterly Journal of Economics 135 — Danish wealth-tax reform; smaller real-economic response but substantial behavioural avoidance.
  • Saez-Zucman 2019 Brookings and 2022 AEA Papers and Proceedings defend wealth-tax feasibility; Sarin-Summers 2019 Brookings express skepticism about revenue estimates.

8.8 Carbon taxation

Already covered in §3.2. Among the most-supported policies by economists (IGM Forum 2019 — 89% of US economists support carbon tax) and the most-resisted politically. Canadian Liberal government’s federal backstop (CCRebate, formerly Climate Action Incentive) returns most revenue as per-capita rebates — administratively simple, progressive.

8.9 Financial-transaction taxes

UK stamp duty on shares (0.5%); France 0.3%; Italy 0.1%; Spain 0.2%; recent G7 FTT debates. Tobin 1972 originally proposed a global currency-transaction tax to reduce speculation. Empirical evidence (Becchetti-Ferrari-Trenta 2014; Capelle-Blanc 2017 on French FTT) — revenue raised, modest market liquidity reduction.

8.10 Excise taxes

Tobacco, alcohol, gambling — designed for sin-tax (correctional) and revenue purposes. US federal tobacco excise: $1.0066 per pack of cigarettes since 2009. Cigarette demand elasticity ~−0.4 to −0.7; tax increases produce meaningful smoking reductions especially among youth (Chaloupka-Yurekli-Fong 2012 Tobacco Control 21).

Sugar-sweetened beverage (SSB) taxes: Berkeley CA 2014 (1¢/oz), Philadelphia 2017 (1.5¢/oz), Mexico 2014 (1 peso/L), UK Soft Drinks Industry Levy 2018. Allcott-Lockwood-Taubinsky 2019 Quarterly Journal of Economics 134 derived welfare-optimal SSB tax balancing internalities, externalities, and regressivity.

9. Transfer programs

9.1 EITC

The US Earned Income Tax Credit, enacted 1975, expanded by every administration through 2009. 2024 maximum credit ~63 B paid to ~25 M households in 2023. Stylised facts:

  • Labor-supply extensive margin response. Eissa-Liebman 1996 Quarterly Journal of Economics 111 — 1986 and 1990 expansions raised single mothers’ employment substantially. Modern replications (Hoynes-Patel 2018 Journal of Human Resources 53) confirm.
  • Intensive margin response. Smaller; some evidence of bunching at phase-in completion (Saez 2010 AEJ: Economic Policy 2).
  • Income gains and child outcomes. Hoynes-Miller-Simon 2015 AEJ: Economic Policy 7 — EITC expansions improved infant health. Bastian-Michelmore 2018 Journal of Labor Economics 36 — improved schooling and earnings of children exposed.
  • Errors. EITC has the IRS’s highest improper-payment rate (~25%) — administrative simplicity vs accuracy trade-off.

9.2 Child Tax Credit

US CTC was 1 600 in 2024 dollars). ARP 2021 (American Rescue Plan) expanded to 3 600, fully refundable, monthly payments. Curran-Collyer 2022 AEA Papers and Proceedings 112 estimated the temporary expansion cut child poverty in half (CPS data). Expansion lapsed end of 2021 amid Senate impasse. Late-2024 bipartisan tax bill proposed restoring partial expansion; not enacted by year-end.

9.3 Universal Basic Income

A periodic, unconditional cash transfer to every adult. Recent experiments:

  • Finland 2017–18. 2 000 unemployed received €560/month for 2 years vs control. Kangas-Jauhiainen-Simanainen-Ylikännö 2020 — modest employment effects, large well-being and trust improvements.
  • GiveDirectly Kenya. Ongoing 12-year study in 50 villages providing $0.75/day per adult unconditionally. Banerjee-Niehaus-Suri 2019 Annual Review of Economics 11 reviewed early findings; positive consumption and asset effects.
  • Stockton SEED (2019–21). $500/month to 125 low-income residents for 24 months. West-Castro-Baker-Bidadanure 2021 — improved employment (counterintuitive vs labor-supply theory) and well-being.
  • Sam Altman’s OpenResearch / Y Combinator Research. 50/month control. Vivalt-Rhodes-Bartik-Broockman-Krause-Miller 2024 working paper — modest reduction in labor supply, ~1.3 hours/week; substantial increase in well-being, financial security, and decision-making time.

Theoretical critique (Hoynes-Rothstein 2019 Annual Review of Economics 11): a US-scale UBI would cost ~$3 T/year (~14% of GDP); funding requires substantial tax increases or cuts to existing transfers. Targeted transfers (EITC, CTC, SNAP) are more redistributive per dollar.

9.4 SNAP

US Supplemental Nutrition Assistance Program (formerly Food Stamps). ~$120 B in 2023 to 41 M recipients. Hoynes-Schanzenbach 2009 American Economic Journal: Applied Economics 1 — county-level rollout 1961–75 RDD identified large effects on diet and adult health.

9.5 Unemployment insurance

Joint federal-state in US. Replacement rate ~50% of pre-UI earnings, capped; duration 26 weeks regular + extensions. CARES Act 2020 added $600/week federal supplement (PUC), Pandemic Unemployment Assistance for gig workers, and 13-week PEUC. Chetty 2008 Journal of Political Economy 116 — UI’s liquidity-vs-distortion trade-off; Hendren 2016 American Economic Review 106 — moral-hazard cost.

10. Distributional analysis

10.1 Measurement of inequality

  • Gini coefficient. Standard summary; 0 = equality, 1 = maximal. US household-income Gini 0.41 (2022 Census); EU average 0.30; Scandinavian 0.25; Brazil/South Africa 0.50+.
  • Top-income shares. Piketty-Saez 2003 Quarterly Journal of Economics 118 (and updates at WID.world) using tax records — top 1% US pre-tax income share ~21% in 2020. Auten-Splinter 2024 Journal of Political Economy dispute these via post-tax, transfer-inclusive concepts, finding lower top shares.
  • Wealth concentration. Saez-Zucman 2016 Quarterly Journal of Economics 131 — capitalisation of capital-income flows; top 0.1% US wealth share ~22% by their estimates. Smith-Zidar-Zwick 2023 Quarterly Journal of Economics 138 argue for lower estimates via alternative imputations.
  • WID, LIS, SCF, PSID. World Inequality Database, Luxembourg Income Study, Survey of Consumer Finances, Panel Study of Income Dynamics — the main microdata sources.

10.2 Distributional analysis of tax/transfer policy

Three major US agencies produce distributional analyses:

  • CBO. The Distribution of Household Income annual report; uses CPS plus tax data to allocate federal taxes (individual income, payroll, corporate, excise) and transfers across income deciles.
  • Treasury Office of Tax Analysis. Internal analyses; Distribution tables in budgets.
  • Joint Committee on Taxation (JCT). Congressional tax-bill scoring; Overview of the Federal Tax System.
  • Tax Policy Center (Urban-Brookings). Microsimulation model used for outside analyses; widely cited (e.g., 2017 TCJA analyses).
  • Penn Wharton Budget Model. Comparable academic-affiliated model.
  • Institute on Taxation and Economic Policy (ITEP). Particular focus on state-level incidence.

CBO’s after-tax-and-transfer income concept is the most-cited single measure of US distributional outcomes.

10.3 Fiscal incidence in low- and middle-income countries

Commitment to Equity (CEQ) Institute methodology (Lustig 2018 Commitment to Equity Handbook) standardises fiscal-incidence analyses across countries. Findings: in many Latin American countries the combined tax-and-transfer system is approximately neutral or even regressive in absolute terms; in OECD countries it is strongly redistributive.

11. Behavioural public economics

Standard public economics assumes rational optimising agents; behavioural public economics relaxes this.

11.1 Tax salience

Chetty-Looney-Kroft 2009 American Economic Review 99 — supermarket experiment posting tax-inclusive vs tax-exclusive prices; demand response substantially larger with tax-inclusive prices. Implies “ductile” tax base when tax is non-salient; revenue/distortion calculations under-state distortion when tax is salient.

11.2 Defaults and nudges

Madrian-Shea 2001 Quarterly Journal of Economics 116 — 401(k) auto-enrollment dramatically increased participation. Thaler-Sunstein 2008 Nudge. Pension Protection Act 2006 codified auto-enrollment options. Choi-Laibson-Madrian 2011 — defaults influence portfolio choice and savings.

11.3 Internalities

Internalities — costs to one’s future self from current behaviour. Allcott-Mullainathan-Taubinsky 2014 American Economic Review 104 derived optimal corrective taxes accounting for both externalities and internalities; applied to energy efficiency, soda taxes.

11.4 Tax morale and compliance

Slemrod 2019 Journal of Economic Perspectives 33 — compliance is influenced by social norms, perceived fairness, and trust in government beyond pure enforcement. Naturally varies across countries: Scandinavian high voluntary compliance vs lower in southern Europe.

Modern empirics:

  • Pomeranz 2015 American Economic Review 105. Chilean VAT randomised audit-letter experiment; deterrence from third-party reporting and audits.
  • Almunia-Lopez-Rodriguez 2018 American Economic Journal: Economic Policy 10. Spanish notch-and-bunching estimates of evasion.
  • Best-Brockmeyer-Kleven-Spinnewijn-Waseem 2015 Journal of Political Economy 123. Pakistani corporate-tax notch design.

12. Tax avoidance and evasion

12.1 Magnitude

Zucman 2013 Quarterly Journal of Economics 128 — ~700 B annually).

US tax gap: IRS estimates net tax gap of $688 B per year (TY 2021), or ~14% of expected tax revenue. Individual underreporting dominates; corporate gap smaller in absolute terms.

12.2 International tax competition and havens

Tax-haven mechanics:

  • Conduit havens. Netherlands, Luxembourg, Ireland — channel profits through low-tax jurisdictions via royalties, interest, intra-group services. Double Irish with Dutch Sandwich (closed 2020).
  • Sink havens. Cayman Islands, BVI, Bermuda — terminal destinations holding profits.

OECD BEPS (Base Erosion and Profit Shifting) project 2013–2015 produced 15 Action Items. Pillar One/Pillar Two extends the framework. Country-by-Country Reporting (CbCR) since 2017 requires MNEs to file revenue and tax data per jurisdiction.

Cross-border VAT trades exploit zero-rating of intra-EU supplies and credit on domestic purchases — fraudulent traders disappear without remitting VAT collected. Estimated tens of billions of euros per year in EU losses. Reverse-charge mechanism on susceptible sectors (carbon allowances, mobile phones, integrated circuits) and the OSS (One Stop Shop) for e-commerce VAT mitigate.

13. Federalism and intergovernmental finance

13.1 Tiebout sorting

Tiebout 1956 Journal of Political Economy 64 — under mobility, heterogeneous preferences, fixed numbers of local jurisdictions providing distinct public-good bundles, and other conditions, individuals sort into the jurisdictions matching their tastes. “Voting with feet” achieves efficient public-good provision. Empirical applications: school district sorting, US state migration.

13.2 Fiscal federalism

Oates 1972 Fiscal Federalism — decentralisation theorem: in absence of cross-jurisdiction spillovers or scale economies, decentralised provision dominates centralised (preferences vary across jurisdictions). Mismatches generate cross-subsidy and political pathologies; assignment of revenue sources and spending responsibilities to levels of government is the core question.

Federal grants:

  • Block grants. Lump-sum to lower levels; recipients have flexibility.
  • Categorical (matching) grants. Federal pays a share of qualifying spending; lower-level can choose level.
  • Closed-ended vs open-ended matching. Open-ended (Medicaid pre-2014, traditional unemployment-insurance match) creates spending incentives at lower level; closed-ended (most highway grants) does not.
  • Flypaper effect. Empirically, grants increase recipient spending more than equivalent income increase would predict — “money sticks where it hits.” Hines-Thaler 1995 Journal of Economic Perspectives 9 survey.

13.3 State-local public finance

US federal grants account for ~30% of state revenue and ~5% of local revenue. State sales taxes (38 states + DC), individual income taxes (41 states + DC), property taxes (universal local + some state), corporate income taxes (45 states + DC), excises. State balanced-budget rules constrain countercyclical fiscal policy.

14. Recent US fiscal policy

14.1 Inflation Reduction Act 2022 (PL 117-169)

79B in IRS enforcement (intended to close tax gap), partly rescinded in subsequent agreements; corporate alternative minimum tax (CAMT) of 15% on book income for firms with >230B+ announced FDI in clean energy by mid-2024 per Atlas Public Policy).

14.2 CHIPS and Science Act 2022 (PL 117-167)

$52 B in subsidies for US semiconductor manufacturing + 25% advanced-manufacturing investment tax credit (Section 48D). Major investments: Intel Arizona/Ohio, TSMC Arizona, Samsung Texas, Micron NY/Idaho. Industrial policy revival.

14.3 Bipartisan Infrastructure Law 2021 (PL 117-58, IIJA)

550B in new spending including 66B passenger rail, 55B water, $39B transit. Tax provisions modest (changed digital-asset reporting).

14.4 TCJA sunset 2025

Most individual provisions of TCJA expire at end of 2025; full extension cost ~$4.6 T over 10 years per CBO 2024. Top of the 2025 fiscal-policy agenda — either bipartisan reform or partisan reconciliation extension.

14.5 Federal deficit and debt

CBO 2024 long-term projections: federal deficits ~6% of GDP indefinitely; debt held by public reaches 120% of GDP by 2034 and 166% by 2054 under current law. Driven by population aging (Social Security, Medicare) and rising interest costs. Major fiscal-policy challenge: bending the trajectory requires some combination of revenue increases and benefit reforms that have been politically unpalatable.

15. Comparative tax systems

15.1 Nordic dual income tax

Sweden, Norway, Denmark, Finland tax labour income progressively (top marginal 50%+) while taxing capital income at a flat rate (~30%). Rationale: capital is internationally mobile and tax-elastic; labour less so. Sørensen 2007 Annual Review of Economics analysis.

15.2 Flat taxes

Russia 13% flat 2001 (replaced 12/20/30% progressive); produced large compliance gains (Ivanova-Keen-Klemm 2005 Economic Policy 20) attributable principally to reform of enforcement; modest revenue gains relative to predictions. Reverted to mild progressivity 2021 with 15% rate above income threshold.

15.3 Cash-flow / consumption-based corporate tax

Auerbach-Devereux-Keen-Vella 2017 Tax Law Review 70 — Destination-Based Cash-Flow Tax (DBCFT): tax on cash flow rather than profit, destination-based with border-adjustment. Not adopted but informed Pillar Two design. Hall-Rabushka 1995 The Flat Tax and related “X-tax” proposals.

15.4 Land-value taxation case studies

Pittsburgh’s split-rate property tax (1913–2001) — higher rate on land than improvements. Stimulated construction without measurable revenue loss (Oates-Schwab 1997 National Tax Journal 50). Largely abandoned because of administrative reassessment difficulties.

16. Open problems

  • Top of the income distribution. Optimal-tax theory disagrees with political feasibility; the gap between revenue-maximising and politically-feasible top rates remains large. Tax avoidance, capital flight, and labour-supply responses at the top are intensely debated empirically.
  • Wealth-tax administration. Asset valuation, especially for closely held businesses and illiquid assets. Lock-in and migration responses.
  • Pillar Two implementation. Domestic top-up taxes, undertaxed-profits rules, transitional safe harbours. Pillar One remains stalled.
  • Carbon-tax revenue recycling. Per-capita rebates vs reductions in distortionary taxes vs investment in transition. Theoretical optimum depends on labour-supply elasticities and political feasibility.
  • UBI vs targeted transfers. Cost, labor-supply response, and political durability of universal vs means-tested transfers.
  • Behavioral public-finance design. Default architectures, simplification, salience — how to incorporate behavioral facts into formal optimal-tax theory.
  • Tax administration and the IRS. Funding levels (IRA 2022 enforcement boost partially rescinded), modernization, equitable enforcement across the income distribution (recent evidence of audit-rate disparities — Elzayn-Smith-Black-Hertz-Yeoman 2023 found Black taxpayers audited at 3–5× rate of non-Black taxpayers).
  • Fiscal sustainability. Long-run path of US debt, Social Security, Medicare; comparable challenges across OECD with aging populations.

Further reading

  • Auerbach, A. J. and M. Feldstein (eds.) 1985+. Handbook of Public Economics (5 vols.); current Vol 5 (2013) co-edited Auerbach-Chetty-Feldstein-Saez.
  • Atkinson, A. B. and J. E. Stiglitz 1980 (and 2015 revised ed.). Lectures on Public Economics.
  • Mirrlees, J. et al. 2011. Tax by Design. IFS Mirrlees Review.
  • Salanié, B. 2011. The Economics of Taxation (2nd ed.).
  • Gruber, J. 2022. Public Finance and Public Policy (7th ed.).
  • Saez, E. and G. Zucman 2019. The Triumph of Injustice.
  • Piketty, T. 2014. Capital in the Twenty-First Century.
  • Piketty, T. 2020. Capital and Ideology.
  • Hines, J. R. 2007 (and update). “Corporate Taxation and International Competition.” Tax Policy and the Economy.
  • Chetty, R. 2009. “Sufficient Statistics for Welfare Analysis: A Bridge Between Structural and Reduced-Form Methods.” Annual Review of Economics 1.
  • Slemrod, J. and J. Bakija 2017. Taxing Ourselves (5th ed.).
  • Diamond, P. A. and E. Saez 2011. “The Case for a Progressive Tax: From Basic Research to Policy Recommendations.” Journal of Economic Perspectives 25.
  • Mankiw, N. G., M. Weinzierl, and D. Yagan 2009. “Optimal Taxation in Theory and Practice.” Journal of Economic Perspectives 23.
  • IPCC AR6 WG3 Chapter 13 (Stavins et al. 2022) — “National and Sub-National Policies and Institutions.”
  • Tørsløv, T., L. Wier, and G. Zucman 2023. “The Missing Profits of Nations.” Review of Economic Studies 90.
  • Saez, E., J. Slemrod, and S. H. Giertz 2012. “The Elasticity of Taxable Income with Respect to Marginal Tax Rates.” Journal of Economic Literature 50.
  • Hines, J. R. 2017. “Business Tax Burdens and Tax Reform.” Brookings Papers on Economic Activity.

Adjacent

Graph View