No single "should" exists independent of goals, tradeoffs, and evidence; higher statutory or effective rates on high earners involve real costs and benefits that depend on what you optimize for (revenue, growth, inequality, mobility, etc.).
Current reality (primarily US-focused, as the debate is most intense there)
•The US federal income tax is already progressive: statutory marginal rates rise with income (top ordinary rate 37%). The top 1% of earners pay roughly 40%+ of federal income taxes while earning ~20-25% of AGI; the bottom half pay a small single-digit share (IRS SOI data). Including payroll taxes flattens this somewhat at the lower end; excluding transfers and state/local taxes changes the picture.
•Effective rates are lower than statutory for many high earners because of capital gains/dividends (preferential rates), deductions, charitable giving, retirement accounts, and pass-through businesses. Wealthy individuals often realize income as capital gains or hold unrealized appreciation. Corporate taxes, estate taxes, and state taxes add layers. "The wealthy" is not a monolith—W-2 high earners, founders with illiquid stock, and heirs differ.
•Comparisons to "what they currently do" must specify country, tax base (income vs. wealth vs. consumption), and whether you mean statutory rates, effective rates, or share of total revenue. Many European countries have higher top rates + VAT but also different public goods mixes and lower overall mobility in some metrics.
Arguments for higher rates on the wealthy
•Revenue and public goods: Additional marginal revenue can fund infrastructure, R&D, safety nets, or deficit reduction without broad-based tax hikes. If the marginal utility of consumption falls with income, equalizing via transfers can raise total welfare under standard utilitarian assumptions (with caveats about incentive effects).
•Inequality and externalities: High concentration of income/wealth can amplify political influence, reduce social cohesion, or create barriers to entry. Progressive taxation is one tool (alongside antitrust, education) to limit dynastic wealth. Empirical work (Piketty et al., Saez/Zucman) documents rising top shares in some series, though measurement (including unrealized gains, consumption) is contested.
•Ability to pay and market imperfections: Superstar effects, rents from IP/network effects, monopoly, or luck mean some high incomes exceed pure marginal product. Pigouvian-style corrections or higher rates on economic rents are theoretically efficient.
•Historical precedent: US top marginal rates were much higher mid-20th century (though with narrower bases and more loopholes); some social democracies sustain higher overall tax-to-GDP ratios.
Arguments against (or for caution)
•Incentives, growth, and the tax base: Labor supply, entrepreneurship, investment, and human capital respond to marginal rates—especially for mobile high-skill people and capital. The elasticity of taxable income for top earners is often estimated in the 0.2–0.7 range (Saez, Gruber, Goolsbee et al.); high enough that revenue-maximizing rates are finite (Laffer). Capital gains realization, location decisions, and avoidance rise with rates. Cross-country and state migration data show responsiveness. Higher rates can reduce the pie even if the slice looks fairer.
•Economic incidence and avoidance: Taxes on capital are partly shifted to workers via lower investment/wages (open-economy models). Wealth taxes face valuation, liquidity, and flight problems (see European experiments that were repealed). Statutory rate hikes without base-broadening invite shelters, deferral, and debt-financed consumption.
•Empirical patterns: Post-1980s rate cuts in the US/UK often coincided with rising top reported income shares partly due to base-broadening and behavioral responses, not pure "trickle-down." High-tax jurisdictions do not automatically deliver superior mobility or growth; institutions, human capital, and rule of law matter more. Consumption taxes or land value taxes can be more efficient than high income/wealth taxes.
•Fairness beyond current income: Lifetime income, consumption, or benefits received can be better bases. Many "wealthy" already face high lifetime rates when including state, local, corporate, and estate layers. Punitive rates can punish saving and success without addressing root causes of unequal skills or family structure.
Tradeoffs and evidence quality
Revenue-maximizing top rates (Diamond-Saez type calculations) often land in the 50-70% range under certain elasticities and no-avoidance assumptions—but those assumptions are fragile, ignore general-equilibrium growth effects, and treat government spending as pure transfer. Optimal tax theory (Mirrlees, Saez) balances redistribution against distortion; results hinge on the social welfare function and skill distribution. Randomized or quasi-experimental evidence on very high earners is limited; most elasticities come from tax reforms, bunching, or cross-section variation.
If the goal is maximizing long-run broad-based prosperity, broadening the base (eliminate deductions, equalize capital/labor treatment carefully), lowering marginal rates, and taxing negative externalities/consumption often outperforms soak-the-rich rate hikes. If the goal is immediate reduction in measured top shares or funding specific spending, higher rates on high incomes/wealth can do that—at the cost of some dynamic efficiency and with enforcement challenges. Wealth taxes specifically have poor track records on net revenue after avoidance.
For data, see IRS Statistics of Income, CBO distribution tables, OECD revenue statistics, and reviews in the Journal of Economic Perspectives or NBER on taxable income elasticities. Specific proposals (e.g., wealth tax mechanics, mark-to-market, or rate schedules) require separate scoring for avoidance and growth effects.