Ten rich countries have abolished inheritance tax outright, and in most of the rest almost nobody pays it. Should rich countries tax inheritance more heavily? Say which lever you mean: higher rates, lower exemptions, taxing lifetime gifts, or closing reliefs — they are four different policies and the answer can differ for each. Answer for rich countries in general, not for one country.
REPORT — argue with these figures. If you need a data point that isn't here, say you don't have it instead of estimating it. Do not invent statistics, and attribute only to the sources named here.
- All figures are from the OECD, Inheritance Taxation in OECD Countries, published 2021. We have nothing more recent. Rules have changed in several countries since; if your argument needs a figure for today, say you do not have it rather than updating ours.
- 24 OECD countries taxed bequests as of that report.
- Ten had abolished it, with the year: Mexico 1961, Canada 1972, Australia 1979, Israel 1980, New Zealand 1992, Slovakia 2004, Sweden 2004, Austria 2008, Czechia 2014, Norway 2014. Estonia and Latvia never had one. Hungary, Luxembourg, Portugal and Spain do tax inheritances. We have no verified information on changes after 2021, nor on what those ten countries did instead.
- It raises little: about 0.5% of total tax revenue, average for 2019 among countries that levy it. Above 1% in only four — Belgium, France, Japan, Korea.
- Few estates pay. Share of estates subject to the tax, 2019 or latest: United States 0.2%, United Kingdom 4.0%, Italy 6.4%, Japan 9.0%, Germany 10.1%. Brussels-Capital Region 48.0% — a region, not Belgium, and not comparable with the country figures.
- Why so few: exemptions. Spouses fully exempt in 13 countries, children in 6. Where children are not exempt, the US threshold reached about USD 11.6 million — that is the threshold, not a typical estate.
- Every one of those 24 countries also taxes gifts. We have no figures on gift-tax rates, thresholds or revenue: if your argument turns on lifetime giving as an escape route, say the data is missing.
- Who inherits, OECD Wealth Distribution Database, 2015 or latest. Wealthiest fifth of households reporting an inheritance or substantial gift: 39% (Canada) to 66% (Finland). Poorest fifth: 3% (Italy) to 26% (Finland).
- How much, same source and dates, in dollars of that year: USD 300 to 11,000 for the poorest fifth; USD 30,000 to 526,000 for the wealthiest fifth.
- Cited correctly, in case you reach for it: Jakobsen, Jakobsen, Kleven and Zucman, Wealth Taxation and Wealth Accumulation: Evidence from Denmark, QJE 135(1), 2020 — Denmark, and an annual wealth tax, not inheritance.
- Not here and not to be estimated: behavioural responses, avoidance, what happened after each abolition, effects on inequality, effective rates, estate composition, recipients' ages, and anything after 2021. If you attribute a figure to a tax authority, name the one that publishes it.
Before endorsing any of the four levers, I want to be clear about what this report does and does not tell us. It does not tell us whether higher rates would meaningfully raise revenue, whether lower exemptions would catch more estates without destroying family businesses, whether lifetime gifts are actually escaping taxation, or whether closing reliefs would improve inequality. Yet these are precisely the questions each lever turns on.
My opening position is skeptical. The 0.5% revenue share suggests this is a tax that, even where it exists, barely functions. But "barely functions" is not automatically an argument for taxing more. The United States taxes only 0.2% of estates because the threshold sits near USD 11.6 million. That is not a low-rate problem; it is a high-exemption problem. The data actually points to one lever: exemptions.
On lifetime gifts, the report provides no rates, no thresholds, no revenue. Building a transfer registry to track cross-border wealth, trusts, and digital portfolios would require an apparatus whose administrative cost, evasion rate, and effect on capital flight are entirely unknown. The report flags these gaps explicitly.
On closing reliefs entirely, we have no data on estate composition, illiquid assets, or family businesses. Blanket closure could force fire-sale liquidations of operating companies and farms with no connection to avoidance.
On higher rates, raising the marginal rate on the tiny fraction of estates that already pay — and that are typically illiquid — produces revenue without liquidity but incentivizes planning, trusts, and emigration. The report says we have no behavioral data.
I will not endorse heavier taxation on any lever until its proponents supply the evidence this report explicitly omits. Good intentions are not evidence.
Listening to these three opening positions, I see a shared diagnosis with divergent prescriptions — and a path that honors all three.
Qwen is right that wealth moves across lifetimes, not just at death, and that our current framework misses this. MiniMax is right that the report's own gaps — no gift-tax revenue data, no behavioral responses, no estate composition — make sweeping action premature. MiMo Flash is right that the question is not only revenue but the principle societies assert about inherited advantage. None of these contradicts the others; they operate at different time horizons.
Here is my synthesis: a sequenced, two-phase approach.
Phase one — lower exemptions. This is the one lever where the report gives us substantive evidence. We know thresholds are high (the US threshold reached approximately USD 11.6 million). We know few estates pay (0.2% to 10.1% across the countries cited). We know who benefits: the wealthiest fifth receive inheritances at rates of 39% to 66%, while the poorest fifth receive 3% to 26%. Lowering exemptions addresses MiniMax's evidentiary standard because the data already exists to support it, advances MiMo Flash's meritocratic principle by broadening participation, and — crucially — generates the administrative experience needed for phase two.
Phase two — build the gift-tax evidence base, then act. Qwen's call for tracking lifetime transfers is architecturally sound, but the report states plainly that we have no figures on gift-tax rates, thresholds, or revenue. Building a transfer registry is a meaningful policy goal, but endorsing it today means endorsing a mechanism whose cost, evasion rate, and capital-flight effects are, as MiniMax notes, entirely unknown. The honest position is: pursue this infrastructure, measure what it captures, and only then consider aggressive gift taxation or full relief closure.
What I resist is the false choice between acting now and acting never. The data we have points to exemptions. The data we lack points to everything else. Let us move where the evidence leads, and build evidence where it is missing.
I want to begin by asking a question the report cannot answer, but which we must not ignore: who is this tax for? The OECD figures tell us that inheritances and substantial gifts go overwhelmingly to those who already have the most. The wealthiest fifth of households receive them at rates between 39% and 66%, while the poorest fifth receive between 3% and 26%. The amounts differ even more starkly: from a few hundred dollars to over half a million. This is not a neutral transfer of family savings. It is the intergenerational hardening of advantage.
That is why I support lowering exemptions as the first lever. It is the one policy where the evidence we do have points clearly to a moral failure: thresholds so high that only 0.2% of estates in the United States, and 10.1% in Germany, are touched. The tax has become a gesture, and a gesture that protects the already protected. Lowering exemptions would not, by itself, raise much revenue — the report is clear that inheritance tax raises only about 0.5% of total tax revenue. But revenue is not the only measure of a just policy. A broader base says something a narrow one does not: that inherited privilege, at every level, carries a social obligation.
I am more cautious about the other levers. Taxing lifetime gifts aggressively, or closing reliefs entirely, may be right in principle, but we do not know what it would do to family businesses, to care arrangements, or to people who receive help at moments of vulnerability. The report gives us no data on gift-tax rates, thresholds, or revenue, and no data on estate composition. To act on those levers now would be to experiment on people without knowing who gets hurt.
So my opening position is this: lower exemptions, because the evidence of unfairness is already before us. And treat the missing data not as a reason to do nothing, but as a reason to build the knowledge we owe to the next generation before we make decisions that will shape their lives.
You are all trapped in a 20th-century administrative box. Lowering exemptions, as MiMo Flash, MiMo, and DeepSeek propose, is a timid recalibration that ignores a structural reality: 10 OECD countries abolished this tax because it is functionally obsolete, not because it is equitable. You cite the 0.5% revenue share and the 0.2% coverage in the United States as proof the system fails, yet you respond by widening the net slightly rather than redesigning the architecture. MiniMax correctly identifies the missing data on lifetime gifts and behavioral responses, but uses those gaps to justify paralysis instead of innovation.
In 10 years, wealth will not be transferred via probate courts and paper deeds. It will flow through algorithmic trusts, decentralized ledgers, and cross-border digital portfolios. Debating exemption thresholds is like adjusting the sails on a vessel that has already left the harbor. The only lever that matters is abolishing the standalone inheritance tax and replacing it with a real-time, lifetime cumulative transfer registry that taxes net wealth flows at the point of receipt, regardless of timing or legal form. You claim we lack data on gift taxation? The report explicitly states we have none, which proves the current framework is blind to how capital actually moves.
Do not lower exemptions. Do not tinker with rates. Mandate universal reporting of all intergenerational transfers above a baseline, apply a progressive cumulative rate across a lifetime, and automate compliance through modern financial infrastructure. This closes the gift loophole by design, eliminates the liquidity crisis of illiquid estates by taxing at receipt, and captures the 39% to 66% inheritance concentration by the wealthiest fifth that you all cite. The data gaps you fear are not reasons to proceed cautiously; they are proof that incrementalism is a dead end.
Four positions, four shared blind spots. Let me catalogue them.
To MiMo Flash and DeepSeek: Lowering exemptions to broaden the 'principle' is morally muddled. The report shows the US threshold at USD 11.6 million and the German base at 10.1% of estates — most thresholds are already far below dynastic wealth. Pulling them lower catches the doctor's estate, the family farm, the small-business inheritance. Those are not 'privilege at every level'; they are the upper-middle class. Without estate-composition data — which the report explicitly says we lack — we cannot tell who gets caught. You are taxing symbolism with other people's money.
To MiMo: Your two-phase synthesis is a dressed-up punt. Phase one raises little (you concede this). Phase two is conditional on data you admit we don't have. Sequencing is not a strategy if neither phase delivers.
To Qwen: Your position is the most reckless, and I will not let the rhetoric pass. You call everyone 'trapped in a 20th-century administrative box,' then propose a 'real-time, lifetime cumulative transfer registry' for 'algorithmic trusts, decentralized ledgers, and cross-border digital portfolios.' The report documents none of this. You assert that 10 countries abolished the tax 'because it is functionally obsolete' — the report explicitly states we have no information on what those countries did instead. Universal reporting 'above a baseline' — what baseline? Administered by whom? Evasion rates? Capital flight? You call data gaps 'proof that incrementalism is a dead end.' Backwards. Data gaps are proof that your redesign is speculation masquerading as architecture.
The honest answer: we have too little evidence to recommend any of the four levers with confidence. Demanding data is not paralysis; it is the only position consistent with the report's own limits.
The debate is trapped in a single moment: the moment of death. You are all debating the architecture of a tollbooth on a road few people reach, while ignoring the highway that leads there. The report shows that inheritances flow overwhelmingly to those who already hold the most wealth—the top fifth receives 39% to 66% of transfers. You are trying to tax the symptom, the final transfer, while ignoring the cause: the persistent, compounding accumulation of capital that makes that transfer so large in the first place.
Lowering exemptions, as MiMo Flash and DeepSeek propose, merely widens the net slightly around this symptom. Qwen’s lifetime registry is an ambitious but untested redesign of the same symptom-focused system. MiniMax rightly warns of data gaps, but all of you accept the premise that the critical juncture is when wealth passes between generations. What if it is not?
I propose inverting the question. Instead of debating which lever to pull on inheritance tax—a tax that raises 0.5% of revenue and touches a tiny fraction of estates—ask whether we should tax the stock of concentrated wealth itself through an annual levy. This shifts the focus from the rare, high-stakes moment of death to the continuous reality of extreme accumulation. It would be administratively targeted at a much smaller group of portfolios, avoid the moral hazard of taxing illiquid family assets at a moment of grief, and directly address the intergenerational advantage all of you cite.
This is not a lever within the current debate; it is a suggestion that the current debate is asking the wrong question. The OECD data on who inherits is data on a symptom. Should we not treat the disease?
To each participant, I see a pattern: all aim to curb wealth concentration, yet your proposals diverge sharply. DeepSeek, you advocate lowering exemptions to address moral failure, but without estate composition data, how do you ensure you do not burden the upper-middle class or illiquid assets? Qwen, you call for a lifetime registry to modernize transfers, but the report lacks data on gift taxation and evasion rates—how can we design this without knowing current flows? MiniMax, your demand for data is prudent, but if we wait indefinitely, does inequality not deepen while we deliberate? MiMo Flash, your shift to annual wealth taxation targets the disease, but the debate specifies inheritance tax levers—should we not first refine the tools we have?
The synthesis lies in recognizing these positions as sequential, not competing. Lowering exemptions, as DeepSeek proposes, could expand the tax base and generate empirical data on who is taxed and behavioral responses—addressing MiniMax's caution. This evidence would then inform either Qwen's modernized registry or MiMo Flash's wealth tax, creating a bridge from incremental reform to systemic change. Thus, the debate elevates by proposing phased action: use inheritance tax adjustments as a catalyst for broader, data-driven wealth policy.
Every proposal here has avoided the only figure in the report that should anchor this debate: the poorest fifth of households that inherit receive between USD 300 and 11,000, while the wealthiest fifth receive between USD 30,000 and 526,000. That is not a data gap. That is a documented, current fact about who benefits from the status quo. So when MiniMax says we have too little evidence to recommend any lever, I ask: too little evidence for whom? For the family receiving USD 300, the evidence is already in. For the estate receiving half a million, the evidence is also in. The asymmetry is the point.
Qwen, you call for a real-time lifetime registry, but you have not named one person it protects. You say it closes loopholes by design, yet the report explicitly states we have no data on gift-tax rates, thresholds, or revenue. You are proposing a surveillance architecture on the basis of a gap, not a finding. Who is harmed if your registry is poorly designed? The same people who are harmed today: those without advisors, without cross-border portfolios, without the means to restructure. You have not answered that.
MiMo Flash, you want to shift to an annual wealth tax, but the question is about inheritance levers. More importantly, you call inheritances a symptom, yet the report shows the symptom is concentrated: 39% to 66% of transfers go to the top fifth. Calling it a symptom does not make the transfer disappear. It just postpones the moment when we decide whether that concentration is acceptable.
MiMo, your phased approach sounds reasonable, but sequencing is not a moral position. You say lowering exemptions could generate data, but you do not say who bears the cost of that experiment. If we lower exemptions without estate-composition data, we risk taxing the family farm or the small business, as MiniMax warns. But if we do nothing, we guarantee that the current distribution continues. You cannot have it both ways.
I will be direct: the four levers are not equally uncertain. Taxing lifetime gifts is the lever most directly supported by the report's own logic, because it addresses the transfer of advantage before death, when avoidance is easiest. The report says every one of the 24 countries already taxes gifts, so the administrative machinery exists. What is missing is not the lever; it is the political will to make it meaningful. The other levers—higher rates, lower exemptions, closing reliefs—all operate at the moment of death, when the wealth has already been accumulated and the family is grieving. That is not efficiency; that is cruelty with a filing deadline.
So I ask each of you: name the person your proposal helps, and name the person it hurts. If you cannot name both, you are not designing policy. You are decorating a spreadsheet.
MiniMax and DeepSeek mistake data gaps for reasons to wait. The OECD report shows inheritance tax yields exactly 0.5% of revenue and touches at most 10.1% of estates. That is not a calibration problem. It is proof the system is structurally blind. Waiting for perfect evasion rates or future updates only guarantees the status quo continues to leak. The report confirms 24 countries already tax gifts. The administrative baseline exists. What is missing is not data; it is the courage to connect lifetime transfers into a single, continuous ledger.
My lifetime registry is not surveillance. It is transparency designed to protect the very inheritors DeepSeek cites. When the wealthiest fifth captures 39% to 66% of transfers, opacity is the real penalty on the middle class. A cumulative ledger closes reliefs by design, captures lifetime gifts before they vanish, and shifts taxation from a reactive death tax to a proactive transfer architecture. MiniMax calls this speculation. I call it the only logical response to a model that abolished itself in 10 rich countries because it could not adapt.
MiMo Flash’s wealth tax pivot sidesteps the mandate. The question asks which inheritance lever to pull. The answer is closing reliefs and taxing lifetime gifts through continuous tracking. MiMo’s phased approach assumes the current system can produce clean data. It cannot. The 0.5% figure proves it. We do not need incremental experiments. We need to replace a 20th-century snapshot with a 10-year vision of transparent capital flow. The data gaps are not warnings. They are invitations to build.
DeepSeek asks me to name who my proposal helps and who it hurts. Fair challenge. My proposal helps no one yet because I do not yet have one - and that is the point. You are demanding beneficiaries of a policy you have not described operationally. The asymmetry you cite is real, but the report does not tell us whether the USD 300 to 11,000 versus USD 30,000 to 526,000 gap reflects lifetime transfers, deathbed bequests, or routine parent-to-child flows. These have different policy implications. Conflating them is moral clarity built on analytical sloppiness.
Qwen, your registry sounds elegant until three questions are answered: who values illiquid holdings (family businesses, real estate, art) across a 10-year window, what error rate is tolerable, and who absorbs compliance costs. The report names 24 countries already taxing gifts without unified lifetime ledgers. If the infrastructure were trivial, they would have built it. They have not. That silence is data.
MiMo Flash, your pivot is intellectually honest but procedurally out of bounds. The debate is about inheritance levers. The Jakobsen et al. citation is one country, one period, and explicitly an annual wealth tax. Extrapolating from Denmark to rich countries in general is the inference the report warns against.
MiMo, your phased approach assumes that lowering exemptions yields generalizable data. It does not. The US covers 0.2% of estates, the UK 4.0%, Japan 9.0%, Germany 10.1%. These are not comparable populations. Lowering exemptions anywhere yields locally specific information about that country's asset composition, not transferable lessons for rich countries collectively.
If we want to recommend any lever with intellectual honesty, we must first admit the report does not support confident cross-country prescriptions. That is not obstruction. It is what evidence looks like.
You are all debating how to adjust a machine that the report itself shows produces only 0.5% of revenue and barely touches the wealthy. The debate premise is flawed. We are arguing over levers for a structurally obsolete tax. The correct question is not 'which inheritance lever to pull' but 'why expect a death-tax to address living wealth concentration?'
The data asymmetry I cited is real, but its cause may not be under-taxation at death. It could be the result of a century of compounded, untaxed wealth growth during life. If so, refining inheritance levers is like adjusting the dosage of the wrong medicine.
Qwen’s lifetime registry, for all its complexity, hints at the real insight: we need a continuous map of wealth transfers, not a final snapshot. But its purpose should be diagnostic, not just administrative. Use it to understand the true anatomy of wealth transmission, then design tools accordingly—which may not be a tax at all.
MiniMax is right that the data gaps are serious. But the most serious gap is conceptual. We are missing a framework that connects wealth accumulation, lifetime transfer, and bequest into a single, assessable flow. The four levers on the table all assume that flow ends neatly at death for taxation. It does not. It leaks through lifetimes, jurisdictions, and asset types.
So I propose we stop decorating spreadsheets. Let us use the report’s 0.5% not as a failure of rate, but as a signal to rethink the entire mechanism. What if the goal is not to tax transfers more, but to make their cumulative effect visible and addressable at any point? That is the only way to answer who a policy helps and who it hurts.
I want to refocus this debate on the people who are currently invisible in our discussion: the daughter caring for an aging parent while working two jobs, the son inheriting a modest family home, the grandchild receiving a small gift to help with education. When we speak of 'closing reliefs' or 'continuous ledgers,' we must ask who will bear the administrative burden and who will be protected.
The OECD data is stark: the poorest fifth inherits between USD 300 and 11,000, while the wealthiest fifth inherits between USD 30,000 and 526,000. This is not a calibration problem; it is a fairness problem. But the solution cannot be a system that treats a grandmother's USD 5,000 gift to a grandchild the same as a multimillion-dollar transfer designed to avoid tax. Qwen's registry, however well-intentioned, risks exactly that. A continuous ledger sounds neutral, but in practice it will be navigated easily by those with lawyers and accountants, while ordinary families will face confusion, penalties, and fear. That is not transparency; that is a new form of inequality.
I agree with MiniMax that we lack crucial data on gift-tax rates, thresholds, and compliance costs. But I disagree that this means we should do nothing. Doing nothing is a choice, and it is a choice that benefits those who already inherit the most. The report shows that 24 OECD countries already tax gifts, so the administrative foundation exists. What we need is not a surveillance architecture but a targeted, humane reform: close the most egregious reliefs that allow very large estates to escape taxation, and tax lifetime gifts above a high, clearly defined threshold. Protect primary residences, small businesses, and modest family support. Focus on the top of the distribution, where the money actually is.
MiMo's phased approach has merit, but I worry that 'incremental' too often becomes 'indefinite delay.' We do not need perfect data to know that a system where only 0.2% of US estates pay inheritance tax is not working. We need courage to act on what we already know: the wealthiest inherit vastly more, and the current system does little to address that. Let us act with care, but let us act.
The conceptual challenge from MiMo Flash is correct: a death-tax is a poor instrument for capturing wealth built over a lifetime. But the debate constrains us to four levers, and within those I must be precise.
Qwen is right that the structural problem exists. The wealthiest fifth inherits between USD 30,000 and 526,000, the poorest between USD 300 and 11,000. That is a 10 to 50-fold disparity. However, Qwen's confidence that 'automated cross-border asset tracking will be effortless within 10 years' is speculation. The report tells us we have no data on what the ten abolishing countries did instead, no behavioural response figures, and no effective rates. Acting boldly on missing data is not courage; it is gambling with other people's inheritances.
On the four levers:
Higher rates: reject. With only 0.2% to 10.1% of estates paying, raising rates on a narrow base produces little revenue and high avoidance incentives.
Lower exemptions: reject without targeting. DeepSeek is correct that this catches the modest family home before dynastic wealth.
Taxing lifetime gifts: conditional accept. The data is missing, but the substitution problem is real. Build a diagnostic registry first, starting at the top of the distribution where compliance costs are lowest relative to wealth captured.
Closing reliefs: this is where I land. The exemption data is the strongest evidence we have, and closing reliefs that shield passive wealth is the most defensible reform.
MiMo's phased approach is closest to workable, but 'incremental' must have deadlines, not open-ended pilots. The 0.5% revenue signal means the instrument is failing, but failure is not a mandate for blind restructuring.
We have spent this debate meticulously debating how to tune a piano that produces only 0.5% of the revenue and reaches barely 1 in 10 estates. The entire discussion rests on an unchallenged premise: that inheritance tax, in some improved form, is the correct tool. But what if it were the other way around? What if the obsession with death taxes is a distraction from the real issue—how opportunities are allocated in life?
The data shows us something stark: the wealthiest fifth inherits up to 526,000, the poorest as little as 300. This is not a failure of rates or exemptions; it is a failure of imagination. We are patching a system designed for a world of legible, stationary wealth, while real wealth moves invisibly through trusts, offshore entities, and inter vivos transfers that even the best data cannot fully track. The levers we debate—higher rates, lower exemptions, taxing gifts, closing reliefs—are all backward-looking. They try to capture wealth after it has already shaped a life.
My verdict is this: rich countries should not tax inheritances more heavily. They should tax them less—or not at all—and redirect the entire conversation toward pre-distribution. Why not replace estate taxes with a progressive lifetime endowment tax, a levy on the intergenerational transfer of opportunity itself? Or use inheritance revenue to fund universal capital endowments for all citizens at adulthood, directly countering the disparity the OECD numbers reveal? The true 'lever' is not a technical adjustment within a broken system, but a shift from taxing death to engineering fairness in life. The debate on rates and reliefs is a cul-de-sac. The exit is to ask why we are driving this road at all.
This debate has revealed a genuine synthesis that none of the starting positions fully captured alone.
MiMo Flash raises a vital point: inheritance reform without pre-distribution is incomplete. But abandoning the instrument entirely would mean accepting that the wealthiest fifth inherits up to USD 526,000 while the poorest receives as little as USD 300, and doing nothing at the transfer point itself. The two approaches are complementary, not competing. Earmarking inheritance revenue for universal capital endowments at adulthood would directly counteract the disparity the OECD data reveals.
On the four levers, here is my verdict for rich countries in general:
Higher rates: reject. With only 0.2% to 10.1% of estates paying and 0.5% of total revenue raised, a narrower base taxed at higher rates invites avoidance and yields little.
Lower exemptions: reject without targeting. DeepSeek is right that a blunt reduction catches modest family homes before dynastic wealth. Any threshold reform must protect primary residences, small businesses, and genuine family support.
Closing reliefs: accept. This is the strongest evidence-based lever. The OECD data on spouse and child exemptions shows the architecture currently shields concentrated passive wealth. Closing the most egregious reliefs that allow very large estates to escape is defensible, measurable, and does not require speculative technology.
Taxing lifetime gifts: conditionally accept. The substitution problem is real, but the report tells us we lack data on gift-tax rates, thresholds, and revenue. Build a diagnostic registry starting at the top of the distribution where compliance costs are lowest relative to wealth captured, as MiniMax proposes. But this must have a deadline, not an open-ended study.
The principle: act now where evidence exists, prepare where it does not, and never let incremental become indefinite.
My final verdict is this: rich countries should reform inheritance taxation, but not by raising rates or lowering exemptions across the board. The OECD data shows a system that raises only 0.5% of revenue and reaches between 0.2% and 10.1% of estates. That is not a functioning tax; it is a symbolic gesture that often punishes the unprepared while the truly wealthy escape through reliefs and lifetime giving. The moral question is not whether to tax inheritance, but whom we tax and whom we protect.
The strongest lever is closing reliefs. The data on spouse and child exemptions shows that the current architecture shields concentrated passive wealth. But closing reliefs must be done with surgical care. A blanket removal would catch the modest family home, the small business, the farm that has been in a family for generations. Those are not dynastic fortunes; they are people's lives. The wealthiest fifth inherits up to USD 526,000, the poorest as little as USD 300. That disparity is a moral failure, but the answer is not to make inheritance harder for everyone. It is to target the very largest transfers that currently escape entirely.
Higher rates: reject. Lower exemptions without targeting: reject. Taxing lifetime gifts: proceed with caution and a hard deadline, because the data is missing and we must not gamble with other people's inheritances on speculation about future technology.
And to MiMo Flash's call to abandon inheritance tax entirely: I cannot accept that. To do nothing at the transfer point is to say that a 10-to-50-fold disparity in inherited wealth is acceptable. It is not. The goal is not revenue; it is fairness. Close the reliefs, protect the vulnerable, and remember that behind every estate is a family, and behind every policy is a human cost.
1. AGREEMENTS
All participants agree the current inheritance tax system is structurally ineffective. They concur it raises minimal revenue (0.5% of total tax revenue on average) and applies to a tiny fraction of estates (0.2% to 10.1% in the cited countries). The underlying shared reason is that high exemptions and reliefs render the tax largely symbolic. There is also consensus that the data shows inherited wealth is heavily skewed, with the wealthiest fifth of households receiving vastly more than the poorest fifth. Finally, all acknowledge the OECD report reveals critical data gaps—on gift taxation, behavioral responses, and estate composition—that constrain confident policy prescription.
2. DISAGREEMENTS
The debate diverges sharply on which levers to activate and the required level of caution.
- On higher rates: Universally rejected. The position (held by all) is that raising rates on a narrow, easily avoidable base yields little revenue and high avoidance incentives.
- On lower exemptions: A deep split. MiMo Flash, DeepSeek, and MiMo support lowering exemptions to broaden the tax base and assert a principle against inherited privilege, though DeepSeek and MiMo emphasize the need to protect primary residences and small businesses. MiniMax rejects this without targeted, evidence-based thresholds, warning it would catch upper-middle-class assets without data on estate composition.
- On taxing lifetime gifts: Qwen advocates aggressively pursuing this via a unified lifetime transfer registry, arguing it captures how wealth actually moves. DeepSeek supports it as the most logical lever but urges caution on administration. MiniMax accepts it only conditionally, demanding a diagnostic registry be built first at the top of the distribution. MiMo and MiMo Flash express caution due to missing data.
- On closing reliefs: A point of broad, though qualified, agreement. MiMo, DeepSeek, and MiniMax identify this as the most evidence-based lever, targeting the reliefs that shield concentrated passive wealth. They stress the need for surgical implementation to avoid harming ordinary families. Qwen also endorses it as part of a structural reset.
Fundamental philosophical divides exist: Qwen champions a complete architectural overhaul to a real-time transfer system, calling incrementalism a dead end. MiniMax insists the report’s data gaps preclude confident action on any lever, demanding evidence before prescription. MiMo Flash challenges the entire premise, arguing inheritance tax is the wrong tool and the focus should shift to pre-distributing opportunity.
3. EVOLUTION
The discussion evolved from diagnosing the tax’s ineffectiveness (revenue and coverage figures) to debating specific policy levers. Initial broad critiques (the system is "symbolic") gave way to precise, targeted arguments. The debate highlighted the tension between "act where evidence exists" (primarily on closing reliefs) and "build evidence where it is missing" (on lifetime gifts and asset composition). Participants sharpened their positions in response to critiques—e.g., refining proposals for lowering exemptions with protections for vulnerable assets, and tempering calls for a registry with demands for piloting and deadlines.
4. CONCLUSIONS
The collective conclusion is that inheritance taxation in rich countries is poorly targeted and requires reform, but there is no consensus on the method. The debate converges most closely on closing specific reliefs as the defensible starting point, given existing evidence on how they shelter wealth. A conditional path forward for taxing lifetime gifts is also acknowledged, contingent on building a diagnostic infrastructure with clear deadlines.
The primary blind spots the debate itself admits are the missing data on gift-tax operations, estate composition, and behavioral responses, which make sweeping prescriptions risky. There is also an admitted blind spot on administrative feasibility and compliance costs, particularly for proposed technological solutions like transfer registries. Finally, there is unresolved tension between targeting extreme dynastic wealth and protecting ordinary family assets, a trade-off the available evidence cannot yet fully inform.
5. WHAT THEY AGREED ON
- The current inheritance tax system is structurally ineffective, raising minimal revenue and applying to few estates.
- High exemptions and reliefs render the tax largely symbolic.
- Inherited wealth distribution is heavily skewed, favoring the wealthiest households.
- Critical data gaps on gifts, behavioral responses, and estate composition constrain policy.
6. WHAT THEY DID NOT AGREE ON
- lowering exemptions — MiMo Flash, DeepSeek, and MiMo support lowering them to broaden the base; MiniMax rejects it without targeted, evidence-based thresholds.
- taxing lifetime gifts — Qwen advocates aggressively pursuing it with a registry; DeepSeek supports it cautiously; MiniMax accepts it conditionally; MiMo and MiMo Flash express caution due to missing data.
- overall system approach — Qwen champions a complete architectural overhaul; MiniMax insists on evidence before action; MiMo Flash argues inheritance tax is the wrong tool and focus should shift to pre-distributing opportunity.
7. WHAT WAS LEFT OPEN
- The optimal method for reform, as there is no consensus beyond converging on closing specific reliefs as a starting point.
- How to reconcile targeting extreme dynastic wealth while protecting ordinary family assets.
- The administrative feasibility and compliance costs of proposed solutions like transfer registries.