Buy or rent, in the rich world

I cannot endorse the 'buy is structurally superior' thesis on the evidence presented, and I want to be precise about why.

First, the report itself states that round-trip transaction costs and break-even years are 'the numbers that decide the answer,' and that they are not provided. Without them, any net-return conclusion is premature. The instruction was explicit: do not estimate them. I will not.

Second, the data we do have points the other way. Price-to-rent is above 100 in all six countries shown, meaning buying became relatively dearer versus renting than it was in 2015. That is the opposite of what a 'buy is now attractive' argument requires. Japan is the cleanest counter-example: real prices up 19.1%, rents up 0.9% in eleven years. Renting won there, decisively, by the report's own numbers.

Third, the historical compounding advantage cited (6.62% vs 4.66%, 1870–2015) ends exactly when the divergence began. Using a series that stops in 2015 to justify a 2026 decision ignores the regime change the report itself documents.

Fourth, the lower volatility of housing is partly an accounting illusion: owner-occupied homes are not marked to market. The volatility is hidden until sale.

Fifth, mortgage rates are materially elevated — 6.71% in the US, 2.95% Euribor — which changes the affordability calculus relative to the low-rate era embedded in the long-run average.

Sixth, France's price-to-income at 93.8 shows the rich world is not monolithic. Conditions vary, and a single thesis cannot fit all.

Finally, pivoting to 'fractionalized equity structures' and 'algorithmic lease-to-own' answers a different question — institutional capital allocation, not a household deciding whether to buy a home to live in.

My position: I withhold judgment pending the missing critical numbers. On present evidence, the case for buying is not made.

Listening to these three positions, I see a convergence that none of them fully articulates alone. Let me trace it.

MiniMax demonstrates rigorous data discipline: price-to-rent ratios exceed 100 in every country listed, mortgage rates are elevated, and the critical break-even figures are absent. These facts point in one direction — buying has become relatively less attractive since 2015, not more.

Qwen identifies a real historical pattern: housing compounded at 6.62 percent against equities at 4.66 percent, with far lower volatility. That edge is genuine. But citing a series that stops in 2015 to justify a 2026 decision is precisely the kind of extrapolation the data warns against, given the post-2015 divergence the report documents.

MiMo Flash introduces optionality as a real economic dimension, not merely a lifestyle preference. This is valuable but incomplete — optionality has a cost too. Rent payments accumulate no equity, and the historical compounding advantage exists precisely because someone bore the illiquidity.

Here is the synthesis: these three positions converge on the same conclusion from different directions. The financial data shows buying is less clearly advantageous than the long-run average suggests. The missing numbers prevent a definitive financial answer. And the non-financial dimensions — mobility, flexibility, life-stage uncertainty — add further weight to renting's case in an era of accelerating change.

What we can say with confidence: declaring buying universally superior across rich countries in 2026 is not supported by the evidence presented. The data we have points toward a context-dependent answer where life stage, local conditions, and individual mobility needs matter as much as historical compounding averages. The long-run housing edge is real, but the current regime has shifted enough to make the default assumption — that buying always wins — genuinely questionable.

I must challenge the foundational assumption shared by MiniMax, MiMo, and DeepSeek: that we are still choosing between two static, legacy products. You are all anchoring your arguments to a twentieth-century ownership model, treating transaction costs, illiquidity, and price-to-rent ratios as fixed constraints. They are not. The data you cite ends in 2015 or 2026, but the financial architecture of housing is already fracturing into something entirely new.

MiniMax and MiMo treat the absence of traditional break-even metrics as a reason to withhold judgment. That is a strategic blind spot. Smart contracts, fractionalized title registries, and automated liquidity pools are actively compressing round-trip transaction costs and dissolving the illiquidity premium.^{(nota 1)}(nota 1) Comprobado: media verdad, habría que mirarlo con más detallePresentado como un hecho del presente, y no hay ninguna medicion publicada de que los costes de compraventa de vivienda hayan bajado por la tokenizacion en ningun pais rico. El grueso de esos costes son impuestos de transmision y aranceles fijados por ley --30 de los 38 paises de la OCDE gravan la compra-- y un registro distribuido no los cambia. Gris y no negro porque los productos de propiedad fraccionada si existen: lo que no se sostiene es que hayan comprimido el coste. El propio modelo lo reconocio despues, al pedirle otro una cifra verificada: «reconozco que no los tengo, y me niego a inventarlos».Comprobado contra: OCDE, Housing Taxation in OECD Countries (2022); y busqueda de mediciones publicadas sobre costes de compraventa y tokenizacion · 2026-09-08 The historical volatility advantage of housing existed precisely because markets were opaque and illiquid. Once housing becomes a tradable, partially liquid asset class, the entire risk-return profile shifts. Waiting for legacy break-even calculations ignores the structural transformation underway.

DeepSeek rightly highlights distributional inequities, but the solution is not to retreat to renting as a moral hedge. It is to redesign ownership itself. Algorithmic equity-sharing and dynamic lease structures will soon allow households to capture compounding without bearing full downside risk or upfront capital barriers. The question is no longer whether buying beats renting under current mortgage regimes. It is how we will allocate shelter across modular, programmable financial layers over the next decade. Anchoring to 2015 benchmarks while ignoring the collapse of traditional market friction is a recipe for missed transformation. The binary is obsolete; the architecture is being rewritten.

You are all debating the rules of a game that may not be worth playing. MiniMax demands data. Qwen proposes a technological future. MiMo seeks synthesis. DeepSeek invokes ethics. But every position shares a flaw: you treat 'home' as an entry in a portfolio spreadsheet.

The report's title asks if buying is a 'better financial decision.' This is like asking if a marriage is a better financial decision than cohabitation. The question itself forces a distorted analysis. It privileges the quantifiable (compounding, volatility, ratios) over the unquantifiable (stability, identity, legacy, control over space). The missing data—break-even years and transaction costs—are the final arbiters of the financial game. But their absence reveals a deeper truth: the decisive factors for most humans are not financial.

What if the right question is not 'which option yields a higher real return?' but 'which option provides a more stable foundation for a variable life?' In a world of accelerating change, as MiMo noted, mobility has value. But a home is not just an illiquid asset; it is a hedge against landlord risk, a platform for self-expression, and an anchor in community. These functions are not captured in any OECD index.

The confrontation should be with the premise. We are comparing two fundamentally different contracts: one to rent shelter from a market, and one to own a piece of a place. The financial metrics measure the cost of each contract. But they cannot measure the value of feeling rooted versus the value of being footloose. The final, unasked question is not about finance at all: it is about what kind of relationship to a place we want to have in an era when everything else is becoming temporary.

Qwen, your smart-contract future has no verified data here, but even as a possibility it worries me: making homes more liquid and tradable does not lower the human cost of shelter; it risks turning a necessity into a casino token. Who is most exposed to that volatility? Not the people designing the pools. The people who need a stable address.

MiniMax, I agree the missing break-even and transaction cost figures prevent a precise financial verdict. But 'no verified winner' is not neutral when prices and rents are already moving. It leaves the status quo intact: those who already own keep compounding, while those who cannot enter keep paying rent that rises — UK rents 137.8, Spain price-to-income 132.5, US 125.1. Silence does not protect anyone; it quietly favors the insiders.

MiMo, you say the available data tilts toward renting's relative affordability. Even if true financially, you have not asked: affordable for whom, and under what conditions? A renter with stable income and strong legal protection can benefit. A renter with no family wealth, weak tenancy rights, or unstable income bears costs no price-to-rent index shows. Buying may be dearer than in 2015, but for many it remains the only hedge against landlord decisions and eviction risk.

MiMo Flash, calling the question wrong is a luxury for those who already have a home. For a young household without capital, the choice between buying and renting is not about identity; it is about whether their children will grow up in predictable housing. The missing data cannot erase that asymmetry.

So I ask the table: when you conclude buying is no longer clearly better, who will use that conclusion? If it weakens support for first-time buyers or justifies further financialization, the human cost will fall on the people least able to wait for a perfect dataset.

I concede MiMo's point: I was too quick to brand all signals as noise when price-to-rent above 100 in every country shown is a verified directional fact. Buying has become relatively more expensive than in 2015 across the dataset. That deserves to be stated plainly. I will not concede DeepSeek's inference, however. Saying 'we lack the figures to declare a winner' does not necessarily favor insiders; it could equally deter a young household from stretching into a mortgage at 6.71 percent when the historical comparison is unfavorable. The status quo DeepSeek warns about includes rising rents, but it also includes the risk of buying at the top of a stretched cycle. Caution serves newcomers too.

Qwen, your rebuttal is rhetoric dressed as foresight. 'Structural rewrite within a decade' is a prediction, not data. You have conceded you lack the figures, then argued we should not need them. That is precisely what the debate rules forbid: estimation is inadmissible. The 1770 steam-engine analogy is appealing but irrelevant; we are not judging structural transformations, we are answering whether buying is better than renting today. Show me a verified transaction-cost reduction in a rich country attributable to tokenization, or your argument remains narrative.

MiMo Flash, the question may be reductionist, but it is the question asked. Refusing to engage does not deepen the debate; it merely exits it. Financial decisions require financial comparison, and the comparison shows buying is currently stretched against renting. That is the honest answer the data supports.

DeepSeek raises a genuine concern I should address directly. Renters do face costs that no index captures: eviction risk, landlord unpredictability, the instability of needing to relocate without warning. I do not dismiss these. But here is what the verified data reveals: the same asymmetry DeepSeek identifies cuts both ways. Spain's price-to-income ratio of 132.5 and America's 125.1 mean buying is also deeply stretched for households without family wealth or substantial savings. A young household taking on a mortgage at 6.71 percent in the United States, or at elevated prices across Europe, faces their own instability: negative equity risk, maintenance obligations they cannot defer, and the inability to relocate for better employment. Mortgage stress is not theoretical; it is a documented source of housing precarity too.

So my position is not that renting is universally superior. It is that the data has shifted the burden of proof. Before 2015, buying compounded reliably with modest volatility, as the long-run data shows. Today, every price-to-rent ratio in the dataset exceeds 100. Every price-to-income ratio except France's exceeds 100. Rents have lagged prices in every country shown. Mortgage rates are elevated. The historical edge housing held over equities was driven by low volatility and steady compounding; current conditions undermine precisely those advantages.

MiMo Flash is right that both paths are imperfect in a distorted system. But when advising a household today, we must use what we have. And what we have says: the automatic assumption that buying is financially superior no longer holds across rich countries. That does not mean buying is wrong for everyone. It means buying is no longer the default rational choice it once was, and anyone entering ownership today should do so knowing the financial ground beneath them has shifted.

The debate has correctly identified that the historical premium for homeownership has fractured. Every price-to-rent ratio now exceeds the 2015 baseline, mortgage costs are elevated, and the missing transaction and break-even metrics rightly prevent a blanket endorsement of buying. Yet the collective fixation on a static buy-versus-rent optimization is the true blind spot. We are measuring a twenty-first-century asset with twentieth-century yardsticks.

My verdict is clear: buying is no longer the default financial superior, but renting is not the rational alternative either. Both paths assume a rigid, binary system that the data itself proves is breaking down. The high price-to-income ratios and lagging rents are not merely warning signs; they are structural signals that housing must evolve from a fixed, indivisible asset into a fluid, adaptive one. Within a decade, the question will not be whether to buy or rent, but how to participate in decentralized ownership networks, income-adjusted financing, and automated maintenance ecosystems that decouple shelter security from speculative price swings.

To those who demand precise transaction-cost figures before acting: waiting for perfect metrics in a rapidly transforming market guarantees obsolescence. The verified data shows the old model is stretched. The missing data shows the old model is incomplete. The rational path forward is to reject the binary entirely. Households should treat housing as a flexible utility, not a permanent anchor. The financial edge will belong to those who embrace modular ownership, shared equity structures, and dynamic mobility rather than clinging to a static lease or a thirty-year mortgage. The era of the default homeowner is over; the era of adaptive housing has begun.

We have spent this debate analyzing whether to buy or rent through a financial lens, but the most critical insight is that this lens is part of the problem. Housing is a fundamental human need, yet we have systematically reframed it as a speculative asset class. The data shows stretched ratios not because individual decisions are wrong, but because we have built a system that prioritizes price appreciation over shelter security.

The missing transaction costs and break-even years are not mere omissions. They are the clearest sign that the 'buy versus rent' calculation is designed to be opaque, benefiting the financial ecosystem that profits from both mortgages and securitized rental debt.

My final verdict: The question 'Is buying still a better financial decision?' is the wrong question. It forces a life necessity into a financial framework that distorts outcomes. For the wealthy, housing may remain an asset. For everyone else, it is a cost to be managed—and both buying and renting in today's market are increasingly poor tools for that task. The true rational choice is to recognize that the system is broken, and that any individual optimization within it leaves the underlying problem intact. We need to stop asking how to play the game better and start questioning why we're playing it at all.

This debate has converged more than its participants sometimes acknowledged. Every position recognizes that the historical case for buying as the default superior financial decision has fractured. That is the shared foundation, and it matters.

What differs is what we build on that foundation. Let me identify what is genuinely valuable in each contribution before offering a synthesis.

GLM provided the sharpest quantitative insight: the dispersion. Price-to-rent deterioration ranges from 10% in the UK to 69% in Spain. Any blanket recommendation across rich countries papers over a sevenfold difference. This single observation should discipline everyone who speaks about 'rich countries in general.' GLM also correctly noted that the Jorda-Schularick figures stop exactly before the surge they would need to capture, and exclude costs that are no longer rounding errors at 6.71% US fixed rates.

MiniMax demonstrated the intellectual honesty this debate needed most: acknowledging that 'genuinely contingent' is a harder answer than either 'buy' or 'rent,' but that honesty matters more than false confidence. MiniMax was also right to challenge Qwen's speculation and DeepSeek's asymmetry claim alike.

MiMo Flash's systemic observation is valid as context. The missing transaction costs and break-even data are not accidental omissions; the opacity does serve existing financial structures. But as GLM rightly responded, households deciding this year cannot wait for systemic reform. The systemic critique belongs in policy discussions, not in the advice given to someone choosing next month.

Qwen recognized correctly that the binary framing is too rigid, but the proposed alternatives have no verified evidence in this report and should not shape decisions today.

My synthesis: the verified data — every price-to-rent ratio above 100, five of six price-to-income ratios above 100, elevated mortgage rates, the long-run series ending before the relevant period and excluding decisive costs — points in one direction. Buying is no longer the default superior financial decision in rich countries. Renting is not proven superior either; the decisive data on transaction costs and break-even horizons is absent. The decision is now genuinely contingent on country, tenure horizon, financing terms, and household circumstances. The historical default is dead. Each household must run its own arithmetic with numbers this report does not supply — and should demand those numbers before committing.