Wealth Distribution Systems

This is the third brief in Category IX, which is new to this map. The categories before it ask what can be built; Category IX asks how the output is produced, claimed, distributed and counted. Within it, this slot covers mechanisms and their measured effects — the instruments that move stocks of claims on output, and what each one has actually been observed to do. Post Scarcity Economics asks whether there is a distributional problem at all; Universal Basic Abundance examines one specific transfer instrument and the experimental record of how recipients respond to it. This brief assumes scarcity binds and asks what the machinery does.

Established The framing under test is that distribution is a policy choice and the instruments to change it exist. The first clause survives; the second is where the evidence gets interesting. The instruments named in the phrase — wealth taxes, inheritance taxes, transfer payments — are attached to a minority of the observed variation. Between two-thirds and 90% of the difference in inequality between the United States and Europe is already present before any tax is levied or transfer paid, which means the machinery the debate is about is pointed at roughly a third of the gap.

Frontier And beneath the instruments sits a measurement problem that the policy debate almost never surfaces: we cannot yet agree on the size of the thing being redistributed. Two of the best teams in the field, using overlapping US tax data, differ on the top 0.1% wealth share by 4.3 percentage points — more than most of the proposals would move. Every revenue projection for a wealth tax inherits that uncertainty, and most are quoted without it.

Established Start with the measurement dispute, because every instrument below is sized against it. Saez and Zucman capitalise reported capital income into wealth stocks; Smith, Zidar and Zwick do the same but allow returns to vary across the distribution. For the top 0.1% share in 2016: Saez–Zucman's 2016 benchmark gives 18.6% and their 2020 revised series 18.4% (with a top 1% share of 36.3%); Smith–Zidar–Zwick's baseline gives 15.7% (top 1% 33.7%), rising to 20.4% under an equal-returns assumption. The gap on the top 0.1% is roughly 4.3 percentage points, and the heterogeneous-returns team characterise the rise in that share as half as large as prior estimates.

Frontier The disagreement is methodological and unusually specific, which is what makes it tractable. On fixed income, Smith, Zidar and Zwick assign the wealthy higher interest rates — the Moody's Aaa corporate bond rate, averaging 6.0% in 2000–09 and 4.2% in 2010–16 — reporting that the wealthy earn interest at rates “3.5 times higher” than average through boutique funds and private credit. Saez and Zucman reply that wherever income and wealth are both observed the wealthy earn far less — roughly 1.4–1.5% in matched estate-income data in 2016, 2.0–3.0% in the Survey of Consumer Finances — and that the rival estimates capture only 68% of the top 1%'s fixed-income claims against the Federal Reserve's distributional accounts, putting the bias at about 2.0–2.2 points of the top 0.1% share.

Frontier Two further components complete the arithmetic of the gap. On equity, weighting dividend income at 90% and capital gains at 10% misses non-dividend-paying equity; against the Forbes 400 the method captures only about 57–60% of billionaire wealth — $1.7tn against $2.4tn in 2016 — contributing about 1.7 points. On pass-through business, treating 75% of pass-through profit as labour income is argued to understate business value substantially, with the profit measure for real-estate S-corporations off “by a factor of 5”, contributing roughly 0.5–1.0 points. Saez and Zucman claim these three items account for the entire gap and that correcting them restores a rise in the top 0.1% share of 4.1 points from 2001 to 2016 against 0.9. Established Both camps agree on the direction: Smith, Zidar and Zwick report the top 1%, 0.1% and 0.01% shares rising 6.6, 4.6 and 2.9 percentage points from 1989 to 2016, and call rising concentration robust to method. The direction is settled; the level and the recent slope are not, and anyone quoting a top-share number without naming the series is overstating what is known.

Established The repeal record is the strongest single piece of evidence against the framing's second clause. On Scheuer and Slemrod's count, twelve OECD countries levied annual net wealth taxes in 1990; by 2018 four did — France, Norway, Spain and Switzerland — and France then replaced its annual wealth tax with a tax on immovable property in 2018. The OECD's stated reasons for repeal are efficiency costs, risk of capital flight, failure to meet redistributive goals, and high administrative costs. Established Two country specifics matter because they are routinely miscited. Germany's tax fell to the Federal Constitutional Court in 1995, on the ground that it treated real property and financial assets discriminatorily — a legal failure, not a behavioural one. Sweden abandoned its tax on the argument that special treatment of business equity made it regressive, and it was blamed for spurring avoidance and evasion. The instruments existed and two-thirds of the countries that had them chose to stop, which is not the same as the instruments failing on the merits, several having been badly designed with narrow bases and generous exemptions.

Established What survives raises real money, and the surviving cases have a shape. Switzerland raises 1.1% of GDP from its cantonal net wealth taxes as of 2018 — more than any other retained wealth tax, and comparable in magnitude to the revenue projected for recent US proposals. It has no business-asset exemption on the scale seen elsewhere, and valuation is done cantonally with long-standing administrative machinery. Norway retains a net wealth tax raising considerably less as a share of total tax revenue. Frontier The surviving wealth taxes are the ones embedded in decentralised, long-established, broad-based administrative systems — not the ones designed as redistributive statements. That is a pattern across four cases, not a tested proposition.

Frontier The elasticities are where this literature is strongest, and the estimates are large. In Switzerland, a one-percentage-point cut in the wealth tax rate raises reported taxable wealth by at least 43% after six years, with responses around 96% for major reforms. In Denmark, for the very wealthy, a one-point cut raises taxable wealth by 21% after eight years. And Norway supplies a unique within-country natural experiment: the municipality of Bø cut its municipal wealth tax from 0.85% to 0.35% effective 1 January 2021, and the reported elasticity is a 66.6% increase in average taxable wealth per point of cut — 71.6% among actual wealth-tax payers, and 88% on a triple-difference estimate.

Frontier The decomposition of that Norwegian result is the interesting part, and it points both ways. Migration dominates: the share of the municipality's wealth attributable to movers rose from about 1% pre-reform to 71% in 2020 and about 20% in 2021, with wealth taxpayers' likelihood of moving to it up 0.28% and, for individuals holding over NOK 10 million, up 2.56%. But national out-migration is small: out-migration of wealth taxpayers from Norway runs 1.83%–2.78% annually and wealth taxpayers are under-represented among out-movers relative to their roughly 10% share of the population, which the authors call limited out-migration. There is little or no bunching at the exemption threshold. Frontier The interpretive fork is the whole policy argument: a 43–67% reported-wealth elasticity per point of tax is either evidence the instrument destroys its own base, or evidence that most of the response is reporting and relocation within a small jurisdiction, both of which shrink as the jurisdiction gets larger. The Norwegian pattern — enormous local elasticity, small national out-migration — is consistent with the second. Established A think tank with a stated position against wealth taxation concedes the point in its own brief on the study, noting the migration remained relatively modest in absolute terms, occurred in a remote declining rural area, and may not apply to urban reforms. That concession runs against interest and gains weight accordingly.

Established Inheritance taxation is near-universal and nearly weightless, and the second half of that sentence is the finding. The OECD's 2021 study records 24 of 36 OECD countries levying inheritance, estate or gift taxes, raising on average 0.5% of total tax revenues in the levying countries in 2018. Most estates escape taxation entirely, because bases are narrow: preferential treatment for close relatives, and reliefs for residences, businesses and pensions. The avoidance channels are structural rather than marginal — splitting bare ownership from usufruct, preferential asset valuations, and more favourable treatment of in-life gifts than of transfers at death — and these reliefs disproportionately benefit the wealthiest households. Frontier Survey data also understate concentration at the top: one German study found top-1% wealth holdings at 35% rather than a previously estimated 22% once the very top was properly captured. Established The OECD's recommendation — recipient-based rather than estate-based taxes, broader bases, alignment of gift and inheritance treatment, stronger reporting — is the position of a member-funded intergovernmental body and should be read as advice, not as a finding. An instrument that exists in two-thirds of rich countries and collects half a percent of tax revenue makes “the instruments exist” true and close to vacuous.

Established Sovereign wealth funds are the mechanism that demonstrably moves money at scale, and there are exactly two live cases with different architectures. Norway's Government Pension Fund Global had a market value of NOK 19,742 billion at end-2024, with net transfers into it from petroleum revenues of NOK 409 billion in 2024. The fiscal rule caps structural non-oil deficit spending at the fund's expected real return — set at 4% at inception and cut to 3% in spring 2017 — with Parliament approving withdrawals against expected rather than realised returns. For 2025, approximately 24% of fiscal budget expenditure is financed by transfers from the fund. A resource rent converted into a permanent rule-bound flow financing a quarter of a national budget is the largest working instance of the framing's claim in existence — and it distributes through public services and the general budget, not through individual payments.

Frontier Alaska is the same idea with the opposite delivery, and the contrast is the useful part. Norway distributes a resource rent through the state; Alaska distributes it through cheques, and its dividend is estimated to reduce the state poverty rate by 2.5 to 4 percentage points annually since 1990. Only Alaska's produces a measurable individual-level poverty effect; only Norway's is large enough to reshape a national fiscal position. Speculative What neither shows is that a fund can be built without a rent. The standard proposal — a citizen's wealth fund capitalised by equity dilution or by borrowing against future returns — has never been implemented at scale anywhere, and both working cases rest on hydrocarbons.

Established And now the result that reorganises the whole slot. Blanchet, Chancel and Gethin, using distributional national accounts that combine national accounts, surveys and tax data, report pre-tax income shares for 2017: the bottom 50% receives 11.7% in the United States against 20.8% in Western Europe, 24.1% in Northern Europe and 20.1% in Eastern Europe; the top 10% takes 47.8% in the US against 34.8%, 30.8% and 35.6% respectively. Over 1980–2017 the top 1% pre-tax share rose from 11% to 21% in the US and only from 8% to 11% in Europe. Predistribution — the pre-tax distribution itself — accounts for between two-thirds and 90% of the Europe–US inequality gap. Frontier The authors add a corollary that reverses the standard comparison: after accounting for indirect taxes and in-kind transfers, the US redistributes a greater share of national income to low-income groups than any European country. That depends on including fiscal components the usual comparisons omit, it is contested, and it is flagged accordingly. The authors are affiliated with an inequality-research lab, and the headline finding undercuts the redistribution agenda usually associated with it.

Established A discipline worth imposing on the whole field: separate what has been measured from what has only been projected. Measured — Swiss wealth-tax revenue at 1.1% of GDP; reported-wealth elasticities of 21% to 67% per point across three countries and three designs; inheritance-tax revenue at 0.5% of total tax revenues; Norway's fund financing 24% of budget expenditure; Alaska's dividend removing 2.5–4 poverty points; the repeal count of twelve to four. Speculative Projected only — revenue from proposed US-style wealth taxes at 2–3% above high thresholds, which depends on which top-share series you accept and therefore inherits the 4.3-point disagreement; citizen's wealth funds capitalised without a rent; and the redistributive effect of closing inheritance-tax reliefs, which the OECD recommends and nobody has run at scale and measured.

Frontier Is the level of wealth concentration knowable with existing data? Probably not, and that is the most consequential open question here. The two leading estimates differ by 4.3 points at the top 0.1% and by more than four-fold on how much that share has risen since 2001, using the same tax records. Frontier What would settle it is administrative wealth data — an actual register — and the United States has none. Every revenue projection, every distributional table and every threshold design inherits the gap in the meantime.

Frontier Are heterogeneous returns the right modelling choice? The wealthy plausibly do earn different returns, and private credit and boutique funds are a real channel; the counter-evidence is that in every dataset where income and wealth are both observed, measured returns at the top are lower rather than higher. Frontier This is resolvable in principle with matched income-and-wealth microdata at the very top, and irresolvable in practice without it. It is worth noticing that a purely technical modelling choice — how to convert observed interest income into an implied stock — moves the headline number by more than most policy proposals would.

Frontier Are wealth taxes administratively unworkable at national scale? The repeal record says something, and Switzerland says the opposite: a net wealth tax raising 1.1% of GDP, administered cantonally, sustained for generations. Frontier The untested question is whether a large country can build valuation machinery from scratch, since the surviving cases inherited theirs. Nobody has tried, so the claim that it cannot be done is an inference from countries that stopped rather than from a country that attempted and failed.

Frontier Do wealth taxes destroy their own base through migration? Contested, and the best evidence splits by scale. Within Norway, one municipality's cut drew 71% of its 2020 wealth from movers with a 2.56% rise in in-move likelihood for the largest holders. Nationally, out-migration of wealth taxpayers runs 1.83–2.78% a year with wealth taxpayers under-represented among emigrants. Frontier The live hypothesis is that the measured elasticity is mostly reporting and relocation and shrinks with jurisdiction size — supported by the local estimate of 66% sitting far above the Danish national estimate of 21%, and conceded by an opposed think tank. What would settle it is a systematic comparison of elasticities across jurisdiction scale, which has not been assembled.

Frontier Is inheritance taxation the highest-leverage under-used instrument? The case is that it addresses the transfer point where concentration is most visible, that most estates escape entirely, and that the reliefs doing the escaping disproportionately benefit the wealthiest. Speculative The evidence for the potential is a projection: no country has broadened its base as recommended and reported the revenue. Until one does, the 0.5% figure is the only measurement in the field, and it is a measurement of the instrument as designed rather than as imagined.

Speculative Can a sovereign fund be built without a resource rent? Both working cases rest on hydrocarbons. Proposals to capitalise a citizen's fund by equity dilution, by a share of corporate profits, or by borrowing against expected returns are coherent and have never been implemented at scale. Frontier That is a genuinely open engineering-of-institutions question rather than a settled negative, and it is the single largest gap between what this field proposes and what it has evidence for.

Frontier Does predistribution dominate redistribution, and does the corollary survive? The decomposition is strong and the direction is not seriously contested. The contested part is the authors' further claim that the United States redistributes more than any European country once indirect taxes and in-kind transfers are counted — which depends entirely on the accounting treatment of health and education in kind. Frontier Agreement on that treatment would settle a reversal of the standard comparison, and there is no agreed standard.

Speculative Two fringe positions belong on the record. The first is that the measurement dispute is politically produced — that capitalisation choices are selected for their conclusions. There is no evidence for it: both teams publish code and data, and the disagreement tracks defensible modelling choices. What would test it is pre-registration of capitalisation assumptions before seeing results, which is not a norm in this field. Frontier The second, arriving from opposite ends of the political spectrum, is that distribution is not a policy choice at all — that factor shares are set by technology and global capital mobility and national policy moves them only at the margin. The size of the Europe–US predistributional gap is itself evidence against it, since it shows national institutions producing very different pre-tax outcomes under common technology and open capital markets.

Established The binding constraint on this entire field is that most countries have no wealth register. Income is observed because employers and financial institutions report it; wealth is not, because nothing obliges a comprehensive third-party statement of what a person owns. Every US estimate in this brief is therefore capitalised — inferred from reported income flows — which is exactly why two competent teams can differ by 4.3 points. Frontier The countries with functioning wealth taxes are the countries with valuation machinery, and the direction of causation there is not obvious: Switzerland administers a wealth tax because cantons have long assessed wealth, and cantons assess wealth because there is a tax.

Established Valuation of unlisted assets is the specific technical bottleneck, and it is what broke the German tax. The 1995 constitutional ruling turned on discriminatory treatment of real property against financial assets — a valuation problem, not a capital-flight problem. Sweden's regressivity complaint was also a valuation problem in another form: special treatment of business equity. Frontier Closely held businesses, private equity stakes, art, trusts and unlisted real estate are where wealth at the top actually sits and where no cheap, defensible annual valuation method exists. A wealth tax is a valuation programme with a rate attached.

Frontier The elasticity literature has a jurisdiction-size confound it cannot resolve internally. The three best estimates come from a canton-level system, a small national system and a single Norwegian municipality, and they range from 21% to 88% depending on design and scale. Extrapolating any of them to a large national proposal with unprecedented design features is what the surveying economists explicitly caution against. Frontier The estimates are excellent measurements of something, and it is not obvious that the something is the parameter policy needs.

Established And the largest bottleneck is that the instruments plausibly capable of moving the other two-thirds have no evaluation literature at all. Wage-setting institutions, union coverage, sectoral bargaining, education financing, healthcare financing, housing supply and competition policy are where the predistributional gap sits. Speculative There is no equivalent of the elasticity estimates catalogued above for any of them — no comparable natural experiments, no agreed parameters, no revenue-style arithmetic. The best-measured instruments in this field are attached to the smaller share of the problem, and that is a fact about what is easy to measure rather than about what works.

Established Nothing on this map produces a result this brief waits on, and no typed depends-on edge is claimed. What it waits on is measurement and legal machinery: a comprehensive wealth register with third-party valuation, a valuation rule for unlisted business assets that would survive constitutional review, a cross-border reporting standard that reaches trust-held and unlisted wealth, and a capitalisation source for a fund without a resource rent. All four are recorded as typed requirements below and none of them is a discovery.

Frontier From research it waits on three specific things. Matched income-and-wealth microdata at the very top, which would adjudicate the returns dispute rather than continuing it; an agreed accounting treatment for in-kind health and education transfers, which decides whether the contested reversal of the Europe–US redistribution comparison stands; and a systematic comparison of behavioural elasticities across jurisdiction scale, which decides whether the large local estimates mean anything nationally. Speculative None requires a new method. All three require data access that currently sits with tax authorities.

Established The cheapest high-value step is procedural: pre-register capitalisation assumptions before estimating. The 4.3-point disagreement is not a data problem, it is a modelling-choice problem, and the choices are made by people who know what each choice implies. Pre-specifying interest-rate assumptions, dividend and capital-gains weights and pass-through profit treatment before running the series would not resolve the dispute but would establish which parts of it are genuinely about the world.

Frontier Second: a national wealth-tax reform with cross-border migration outcomes tracked as a pre-specified endpoint. The best natural experiment available is a municipality of a few thousand people; the parameter policy needs is national. Norway's own national record is the nearest thing, and it has never been analysed as a designed evaluation with emigration, asset relocation and reported-wealth responses separated.

Frontier Third: broaden one inheritance-tax base as recommended and report the revenue. Recipient-based taxation, alignment of gift and death treatment, and scaled-back residence and business reliefs are a specific package that two-thirds of the OECD is positioned to try. The gap between the 0.5% collected and the potential asserted is entirely projection, and one country running the experiment would convert it into a measurement.

Speculative Fourth, and the one that would matter most: build an evaluation programme for predistributional instruments. If two-thirds to 90% of the gap is pre-tax, then minimum wage schedules, sectoral bargaining coverage, licensing, zoning, education financing and healthcare financing are the high-leverage instruments, and none has anything resembling the elasticity literature that wealth taxation has. Frontier This is not a single experiment; it is the recognition that the field has industrialised measurement of the smaller half of its own subject.

Established The engineering of a wealth tax is valuation, reporting and coverage, in that order, and the surviving cases show what each costs. Switzerland's cantonal system works because valuation is administered locally by institutions that have done it for generations and because the base is broad, with no large business-asset carve-out. That combination — broad base, embedded machinery, decentralised assessment — is what distinguishes the wealth taxes that lasted from the ones repealed.

Frontier Third-party reporting is the mechanism that makes income taxation work and it barely exists for wealth. Listed securities and bank deposits are reportable; closely held businesses, trusts, art and unlisted real estate are not, and that is where the disputed share of top wealth is held. Speculative International proposals for a financial register that would make cross-border holdings observable are coherent and have no implementation; without one, coverage gaps at the top are not an administrative detail but the main determinant of what the tax collects.

Established For sovereign funds the engineering is a rule rather than a valuation. Norway's is explicit and legible: transfers in from petroleum revenue, withdrawals capped at expected real return — 4% at inception, cut to 3% in 2017 — and parliamentary approval against expected rather than realised returns, producing about 24% of budget expenditure in 2025. Frontier The rule is what makes the fund a distribution mechanism rather than a savings account, and it is the transferable part of the design. The rent is the part that is not transferable, and every proposal to build such a fund without one has to solve the capitalisation problem first.

The boundaries inside Category IX. This slot covers mechanisms and their measured effects on the distribution of stocks and flows. Post Scarcity Economics asks whether there is anything left to distribute over — this brief assumes there is, and deliberately does not argue that abundance would dissolve distributional conflict, which is that brief's untested hypothesis and not a finding here. The Baumol material on relative prices of goods belongs there; the wealth-share dispute belongs here, because it is about stocks of claims. Universal Basic Abundance covers one transfer instrument and the randomised evidence on how recipients respond. Alaska appears in both, correctly and for different reasons: there as an unconditional payment with an estimated labour-supply effect, here as a sovereign-fund distribution mechanism contrasted with Norway's. The randomised-transfer literature is not imported here; the wealth-tax elasticities are not exported there.

Two further Category IX slots are being authored in parallel and this brief does not reach into either. Human Flourishing owns wellbeing measurement, which is where the question “does redistribution make people better off” belongs; this brief stops at the measured distributional effect. Human Development Metrics owns composite indices and what publishing them does to policy — distributional statistics such as top shares and poverty rates are used here as measurements, not as composite indices, and the question of whether the underlying national statistics are trustworthy is handed to that slot rather than restated.

Elsewhere on this map: Institutional Design, where the durability of a rule against the interests it burdens is stated generally; Future Federalism, whose equalisation formulas are the same consent problem in fiscal form and whose Spanish and German material bears directly on subnational capacity to tax; Long-Term Institutions, the closest thing on this map to the sovereign-fund design problem; Future Public Administration, which owns the valuation and reporting machinery any wealth tax would run on; Future Legal Systems, where the constitutional-review constraint that killed the German tax lives; and Space Resource Economies, where the question of who owns a rent before it exists is being asked in advance rather than in retrospect.

Established Almost every source in this brief is an interested party, and the direction of interest is load-bearing in each case. The two teams in the measurement dispute are each one side of it. The Norwegian elasticity study is hosted by a lab that advocates wealth taxation and reports a very large reported-wealth elasticity — against interest. The critical brief on that study comes from a think tank with a stated position against wealth taxation and concedes that the result may not generalise — also against interest. The OECD is a member-funded intergovernmental body publishing tax-policy recommendations, so its recommendation is a position. The Norwegian fund figures are a government reporting on its own fund. Frontier The predistribution result is the most striking case: its authors are affiliated with an inequality-research lab and the finding undercuts the redistribution agenda that lab is associated with, which is why it is given the weight it is here.

Established The institutional constraint that binds hardest is a register, and its absence is the reason the field argues about levels. A wealth tax requires knowing what people own; most countries have no comprehensive statement of that, and the United States — the country whose concentration figures are most disputed — has none at all. Frontier Building one is a legislative and administrative choice with real privacy and cost implications, not a research problem, and it is recorded as a typed requirement below.

Established The second is constitutional, and it is the most under-appreciated fact in the repeal record. Germany's wealth tax was struck down because its valuation treated real property and financial assets discriminatorily. That is a design defect a legislature can fix, and it is routinely cited as though it were evidence about capital flight or administrative impossibility. Frontier Whether a differently-valued tax would survive review has never been tested, because no German government has attempted one since.

Established The reliefs are where the ethical weight actually sits, and they run the opposite way to the instrument's stated purpose. Inheritance taxes exist in 24 of 36 OECD countries and collect 0.5% of tax revenue because most estates escape entirely, through preferential treatment for close relatives and reliefs for residences, businesses and pensions — reliefs that disproportionately benefit the wealthiest households and undermine the progressivity the instrument is defended for. A tax that is politically justified as a redistributive cornerstone and whose exemptions favour the top is not a neutral compromise; it is a specific distributional choice that is rarely defended in those terms.

Frontier Measurement uncertainty is itself an ethical problem, because thresholds are set on it. A wealth tax above a stated threshold, sized against a top-share series that two competent teams disagree about by 4.3 points, will collect a materially different amount and fall on a materially different population than its proponents describe. Established Quote a series or quote a range; a single confident number for US wealth concentration is not available and presenting one is a claim the data does not support.

Frontier And the predistribution finding relocates the ethical argument entirely. If two-thirds to 90% of the observed gap is present before any tax, then debating tax instruments is debating the smaller share — and the instruments that reach the larger share are wage-setting, bargaining coverage, education and healthcare financing, housing supply and market structure. Frontier Those are less legible, less quantified and much harder to evaluate, which is precisely why the debate has settled where it has. The distribution of research attention here tracks measurability rather than leverage, and that is an ethical fact about the field as well as a methodological one.

Established The terminal position is a declared tie, and both halves are measured. The instruments work where they are well designed and administratively embedded: Switzerland's net wealth tax raises 1.1% of GDP and has for generations; Norway's fund finances about 24% of budget expenditure from a rent other petro-states dissipated; Alaska's dividend removes 2.5–4 points from a state poverty rate every year. And they address only a minority of the observed inequality: two-thirds to 90% of the Europe–US gap is predistributional, inheritance taxes collect 0.5% of tax revenue where they exist, and two-thirds of the OECD countries that once levied a net wealth tax repealed it. Both are supported by the record and the site should say both.

Frontier What follows is a change of question rather than a compromise. The useful question is not which instrument redistributes most but at what point in the sequence the distribution is actually determined — and the answer the best available decomposition gives is: before the tax system sees it. That reframes the debate from rates and thresholds to wage-setting, bargaining institutions, education and health financing and market structure, none of which has the measurement apparatus that wealth taxation has. Speculative A field that reorganised itself around that finding would look very different from the one that exists.

Frontier The long-run structural question is whether the pre-tax European advantage survives further capital-market integration. The claim that distribution is not a policy choice — that technology and capital mobility set factor shares and national policy moves them only marginally — is contradicted by the size of the existing pre-tax gap between rich regions operating under common technology. Speculative Whether that remains true is the single most consequential open question in this slot, and it is answerable only by waiting.

Speculative And the durable design lesson from the two working sovereign funds is about rules rather than rents. What makes Norway's fund a distribution mechanism is a withdrawal rule tied to expected real return and approved by Parliament, and what makes Alaska's survive is that its payment is defended as a property right rather than as welfare. Handwave Every proposal to build such a fund without a resource rent asserts the capitalisation step rather than solving it, and no implementation exists anywhere at scale to argue from.

These horizons track legislative decisions, data access and the slow arrival of measurement rather than technology readiness:

• 10 yr: Frontier Expect the measurement dispute to persist unresolved in the United States, because the resolving data — a wealth register or matched income-and-wealth microdata at the top — is a legislative choice nobody is making. Established Expect Switzerland's cantonal taxes and Norway's fund to keep functioning, because both are embedded rather than newly designed. Frontier Expect inheritance-tax revenue to stay near 0.5% of tax revenue in the countries that levy it unless a base-broadening package is actually enacted somewhere, and expect proposals for new national wealth taxes to continue being costed against series that differ by 4.3 points at the top 0.1%.
• 25 yr: Speculative The plausible split is that cross-border reporting improves — because it serves income taxation, which governments care about more — and that better reporting incidentally makes wealth partially observable, which would change this field more than any rate decision. Speculative A citizen's wealth fund capitalised without a resource rent is possible at this horizon and would be genuinely new evidence; equity dilution and profit-share proposals are the live designs and none has been implemented. Handwave Claims that automation-driven capital income shares will force such a fund by a stated date carry no falsification condition and belong here as assertions.
• 50 yr: Speculative At this range the binding question is whether national institutions can still produce very different pre-tax distributions under deeper capital-market integration. If they can, predistribution remains the high-leverage lever and this slot's instruments stay secondary. If they cannot, the tax-and-transfer machinery becomes the only remaining instrument by default — which would be an argument for it arrived at by elimination rather than by evidence. Handwave Both branches are extrapolations from a single decomposition covering 1980–2017.
• 100 / 250+ yr: Handwave Beyond useful forecasting. The longest continuous instruments in the record are Switzerland's cantonal wealth taxes and Norway's fiscal rule, and neither is a century old in its current form. Handwave Proposals for global wealth registries or internationally coordinated wealth taxation are coherent as designs, have no implementing institution, and rest on a level of cross-border enforcement that no existing tax instrument has achieved.

• Depends on Nothing on this map. No brief in this corpus produces a result this one waits on: the mechanisms have been tried and measured, and what is missing is administrative and legal machinery rather than a discovery. No typed depends-on edge is claimed.
• Requires (not on this map) Four constraints, none of them a research result, and the first is the reason this field argues about levels rather than rates. Most countries do not have a wealth register. Income is observed because employers and financial institutions are obliged to report it; wealth is not, so US estimates are capitalised from reported income flows — which is precisely why Saez and Zucman and Smith, Zidar and Zwick can differ by 4.3 percentage points on the top 0.1% share in 2016, and by more than four-fold on how much that share has risen since 2001, using the same tax data. Every revenue projection for a proposed wealth tax inherits that gap. The second is valuation, and it is what actually killed the German tax: the Federal Constitutional Court struck it down in 1995 for treating real property and financial assets discriminatorily — a design defect, routinely miscited as evidence about capital flight. Closely held businesses, private equity stakes, trusts, art and unlisted real estate are where top wealth sits and where no cheap, defensible annual valuation exists; Sweden's regressivity complaint was the same problem wearing different clothes. The third is coverage across borders: without a reporting standard reaching trust-held and unlisted holdings, the base is defined by what is observable rather than by what is owned, and the observability gap is concentrated exactly where the tax is aimed. And the fourth is a capitalisation source. Both working sovereign funds — Norway's, at NOK 19,742 billion and financing about 24% of 2025 budget expenditure under a 3% expected-real-return rule, and Alaska's, whose dividend removes 2.5 to 4 points from the state poverty rate annually — rest on a hydrocarbon rent. A citizen's fund capitalised by equity dilution, by a profit share or by borrowing against future returns has never been implemented at scale anywhere, and every proposal asserts the capitalisation step rather than solving it.
• Enables A working distribution mechanism changes the political feasibility of transitions with concentrated losers — energy, housing and industrial programmes across this map all have that structure — but no such relationship has been measured, and asserting one from two resource-rent cases would be exactly the extrapolation this brief refuses elsewhere. No typed enabling edge is claimed.
• Adjacent Public finance and optimal tax theory, which supply the elasticities; national accounting, which supplies the distributional accounts; tax law and constitutional review, which supply the valuation constraint; sovereign fund management, which supplies the two working cases; and within this map Institutional Design, Future Federalism, Long-Term Institutions and Universal Basic Abundance.

Established “The top 0.1% owns X per cent of US wealth.” No single confident number is available. For 2016, the leading estimates run from 15.7% to 20.4% depending on the return assumption, with a 4.3-point gap between the two main baselines and a four-fold disagreement about how much the share has risen since 2001. Both camps agree concentration rose substantially since 1989 — top 1%, 0.1% and 0.01% shares up 6.6, 4.6 and 2.9 points. Frontier Quote a series or quote a range. A figure without a series attached is overstating what is known, and the policy proposals sized against those figures differ by less than the disagreement between them.

Established “Wealth taxes have been tried and failed.” The repeal record is real — twelve OECD countries in 1990, four by 2018, with France then converting its annual tax to a property tax — and it is mostly a record of badly designed taxes with narrow bases and generous exemptions. Switzerland's raises 1.1% of GDP and has done so for generations. Established Germany's fell to a constitutional ruling about discriminatory valuation, not to capital flight, which is a design defect a legislature could fix and which is cited as though it were an economic verdict.

Frontier “Wealth taxes cause capital flight.” The best within-country natural experiment finds a very large local response — a Norwegian municipality's wealth share from movers going from about 1% to 71%, in-move likelihood up 2.56% for holders of over NOK 10 million — and a small national one: out-migration of wealth taxpayers at 1.83–2.78% a year, with wealth taxpayers under-represented among emigrants. Frontier Relocating between municipalities is not capital flight, and the think tank that opposes wealth taxation concedes in its own brief that the case occurred in a remote declining rural area and may not apply to urban reforms. The honest statement is that the elasticity is large, that its composition varies with design, and that its national analogue is unmeasured.

Established “Inheritance tax is a major redistributive instrument.” It exists in 24 of 36 OECD countries and raises, on average, 0.5% of total tax revenues. Most estates escape entirely; the avoidance channels are structural — splitting bare ownership from usufruct, preferential valuations, and more generous treatment of in-life gifts than of transfers at death — and the reliefs disproportionately benefit the wealthiest. Frontier The potential may be large and the OECD recommends broadening the base, but nobody has done it at scale and reported the result, so the potential is a projection and the 0.5% is the measurement.

Speculative “A citizen's wealth fund would work like Norway's.” Norway's fund exists because a petroleum rent was captured and rule-bound; Alaska's dividend exists for the same reason. No sovereign fund has ever been built at scale without a resource rent. Handwave Equity-dilution and borrow-against-returns proposals are coherent designs whose capitalisation step is asserted rather than solved, and treating the Norwegian outcome as evidence for them skips the only part that has never been done.

Frontier “Europe is more equal because it redistributes more.” The decomposition says otherwise: two-thirds to 90% of the gap is predistributional, with the bottom 50% receiving 11.7% of pre-tax income in the US against 20.8–24.1% in Europe, and the top 1% pre-tax share rising 11% to 21% in the US against 8% to 11% in Europe. Frontier The authors go further and claim that once indirect taxes and in-kind transfers are counted, the US redistributes a greater share of national income to low-income groups than any European country — a reversal of the standard comparison that depends on including fiscal components those comparisons omit. That corollary is contested and is flagged as such rather than adopted, but the main decomposition is not seriously disputed and it is the finding that reorganises this slot.

Speculative “The measurement dispute is manufactured.” Alleged occasionally from both directions and unsupported: both teams publish code and data, and the disagreement tracks defensible modelling choices about interest rates, dividend weights and pass-through profits. Frontier What would test the allegation is pre-registration of capitalisation assumptions before results are seen, which is not a norm in this field — so the claim is currently unfalsifiable rather than refuted, and it is recorded here on that basis.

Established And the mirror-image misconception: that redistribution is futile. Norway finances a quarter of its budget from a fund built out of a rent that other petro-states dissipated. Alaska's dividend removes 2.5–4 points from a state poverty rate every year. Switzerland demonstrates that a net wealth tax can be administered indefinitely at 1.1% of GDP. Established Both hypotheses — that these instruments work where they are well designed and administratively embedded, and that they address only a minority of observed inequality — are supported by the record, and this brief declares the tie rather than picking.