Narrative-first, evidence-driven

Back to case studies

What they did

Japan’s City Planning Law (1968) created a national zoning framework applied uniformly across the country. None of the 12-13 land-use zones functions as a single-family-only district. In every residential zone, multifamily housing is permitted by right. Local governments cannot restrict below the national floor. The result: apartment buildings could go up almost anywhere, without the block-by-block political fights that define US zoning.

Why it worked (or did not)

The short version: when you remove the local veto, supply moves faster.

In Japan, a small landowner who wanted to build a four-unit apartment did not need a variance hearing, a conditional use permit, or a neighborhood association’s blessing. The national framework said yes by default. In most US cities, the default is no, unless you survive the meeting circuit. Multiply that difference by thousands of small landowners across Tokyo over decades, and you get density that grew from the bottom up, one building at a time.

That is the plausible mechanism. It is not proven, but it is credible.

The complication is that three different explanations all fit the historical data:

1. National preemption created high throughput, which absorbed demand and kept prices lower than peer cities. (plausible)
2. Japan’s post-1990 asset bubble collapse cratered land values for a decade and a half, making new construction cheap independent of zoning. (plausible)
3. Wartime destruction, postwar land reform, and cultural norms around incremental small-plot development created a distinct urban form that zoning rules alone cannot replicate. (plausible)

No peer-reviewed study has isolated the relative contribution of any one factor. (unknown)

That is not a reason to dismiss Tokyo as a model. It is a reason to carry the causal story carefully.

Mechanism evidence

What actually moved:

• Tokyo housing stock: 2.51 million homes (1963) to 7.36 million (2013), nearly tripled in 50 years. (confirmed, Foreign Affairs Review)
• Japan historically produced housing starts comparable in absolute number to the US, despite having ~40% of US population. Per-capita: roughly 2.5x more starts. (confirmed, NBER)
• Tokyo population grew 33% from 1963 to the 2010s while residential space per person doubled. (confirmed, Foreign Affairs Review)
• Minato ward population +66% over two decades. Comparable SF growth coincided with +231% home prices; London, +441%. Tokyo prices remained comparatively stable over the same period. (confirmed, Foreign Affairs Review)
• Circa 2018: Tokyo and Sydney residents allocate ~30% of income to rent. London, Paris, NYC: ~40%. (confirmed, REthink Tokyo)

Time horizon: Long (1968-2020, roughly). The supply-stability correlation held for about four decades.

Counterfactual: No control group exists. The comparison to London, NYC, and Sydney is suggestive but not controlled, those cities differ in ways beyond zoning. The stable Tokyo thesis is a structural example, not a causal experiment.

Load-bearing variable: The mechanism requires that local governments actually cannot block construction. National preemption is the linchpin. If localities find workarounds, or if exogenous cost shocks make building uneconomic regardless of permissions, the mechanism stalls.

Certainty: The supply-and-stability correlation is confirmed. The causal story attributing that to zoning specifically is plausible. The relative contribution of zoning versus bubble deflation versus emergent urbanism is unknown.

Guardrails

These are two distinct systems. One governs what gets built. The other governs what happens to people after it is built. Neither causes the other, but both matter for the full picture.

Production guardrail: national preemption

The national framework cannot be unilaterally overridden by individual municipalities. That is the anti-NIMBY mechanism. Japan has far fewer points of discretionary local review than typical US cities, no structural equivalent of the variance hearings and design review boards that can quietly make multifamily construction impossible while technically permitting it on paper. (plausible, research file labels this inferential; not directly tested in peer-reviewed literature)

Tenant stability guardrail: lease law

Japan’s Act on Land and Building Leases (1991) makes it extremely difficult for landlords to refuse lease renewal or terminate tenants. Brookings describes the limited rent adjustment at renewal as a form of rent control. (confirmed, Brookings, Jiro Yoshida)

This is not what produced high throughput, it is what protected existing renters while throughput happened. The distinction matters for US policymakers who cite Tokyo as a supply model: the supply side and the stability side are separate design choices.

This tenant protection created a market inefficiency, professional investors largely avoided the rental sector; individual tax-motivated owners filled the gap, but it also gave existing tenants meaningful stability.

Enforcement check: The zoning framework is operationally enforced at the national level. The tenant protection law is legally binding. Both appear to function as written, not merely as written.

Where it broke (or where it is under strain)

The 2023-2025 price surge.

This is the central stress-test finding. Tokyo new condo prices rose 39.4% in 2023 alone, the average exceeded 100 million yen (~$778K USD) for the first time. (confirmed, Real Estate Economic Institute) From 2015 to 2024, new condo prices rose 66% cumulative. (confirmed, Tokyo Portfolio) Tokyo residential rents rose 6.4% year-over-year in Q4 2024. (confirmed, market analyses)

The mechanism driving the surge: construction cost inflation, an aging labor force, yen at multi-decade lows, and a flood of foreign capital, $10 billion in 2024, up 45% in the first half of the year. (confirmed) These are exogenous shocks. The zoning framework remained permissive. But permissive zoning cannot build apartments when construction costs have spiked, laborers are scarce, and global capital is treating Tokyo condos as a currency play.

Post-COVID supply chain disruptions hit construction industries globally, Japan was not unique here, but the timing compounded its structural labor problem. (plausible, the research file attributes the cost spike to construction cost inflation and an aging labor force; specific materials breakdowns are not sourced in this file.) Combined with a shrinking domestic construction workforce, the industry faced simultaneous pressure on materials costs and labor availability. This is a key stress-test finding for the abundance model: the constraint Japan removed was regulatory. The constraints that overwhelmed the model were physical and macroeconomic.

Labor supply has known levers. The hybrid model already addresses several: childcare and eldercare relief increase workforce participation; immigration has historically amplified construction labor capacity in high-demand markets. Materials supply is a less-understood policy problem, the right buffers and anti-inflationary mechanisms are not well-established and warrant dedicated analysis.

Simultaneously, housing starts declined: Japan FY 2023 saw 801,760 starts, down 7% year-over-year. In 2024, starts fell below 800,000 for the first time in 15 years. Tokyo metro new condo supply hit its lowest level since 1973. (confirmed)

The akiya problem.

Japan had 9.4 million vacant homes (akiya) as of October 2023, 14% of all residential properties nationally. Tokyo: 967,800 vacant homes, 10.3% of stock. (confirmed) Nomura Research Institute projects ~30% of all Japanese homes vacant by 2033. (confirmed)

Why does this vacant stock not relieve price pressure? Perverse tax incentive: lots without structures are taxed at much higher rates than lots with buildings, even derelict ones. Owners hold vacant homes rather than demolish for redevelopment. (confirmed) The result: Japan simultaneously builds new urban housing while accumulating suburban and rural vacant stock. Aggregate supply numbers mask a geographic mismatch.

Scale check: Tokyo’s throughput model worked at city scale under favorable conditions, a growing population, low construction costs, and domestic capital. The 2023-2025 surge shows the model is not shock-proof. External capital and cost inflation can overwhelm supply-side permissiveness.

Market verdict

Mixed, and still forming.

From roughly 2000 to 2020, the market verdict looked favorable: prices stable relative to peers, rent burden lower than London and NYC, population growing without crisis-level displacement. (confirmed)

Since 2020, and especially 2023-2024, that verdict has shifted sharply. Condo prices up 64% over four years, outpacing wages and rent growth. (confirmed) Foreign buyers comprise roughly 40% of customers at some Tokyo agencies. (confirmed) Whether this is a temporary dislocation or a structural shift is genuinely unknown. (unknown) Bank of Japan interest rate normalization in 2024 may affect trajectory, but no reliable forecast is available. (unknown)

The verdict on the zoning model specifically, not Tokyo prices generally, is that it enabled supply throughput for decades. What it cannot do is override currency dynamics, construction labor markets, or global capital flows. That is not a flaw in the model. It is the honest scope of what zoning reform can and cannot accomplish.

Policy environment

The system that produced historical Tokyo affordability had at least four components running simultaneously:

1. National by-right zoning preemption. (confirmed, the most-cited mechanism)
2. Post-bubble land value deflation that kept construction cost basis low for over a decade. (plausible, difficult to disentangle)
3. Incremental small-plot development by individual landowners, not large homebuilder-driven construction. (plausible, structural to Japan’s ownership and financing patterns)
4. Strong tenant protection that gave existing renters legal stability even as ownership markets moved. (confirmed, often omitted from US discussions of the Tokyo model)

For the model to replicate elsewhere, you would need most of these to be present or substituted. Zoning reform alone, item one, is a plausible necessary condition, but not sufficient on its own.

Exploratory, not supported by research file: construction input buffers as a policy question.

If the binding constraint can shift from regulatory to physical, as Japan’s experience suggests, then supply-side housing policy may be incomplete without addressing input availability. The Strategic Petroleum Reserve exists because oil price shocks are considered systemically harmful enough to warrant a buffer. Construction materials have similar dynamics: supply shocks translate directly into housing cost spikes, which squeeze everyone. Analogous mechanisms, strategic material stockpiles, countercyclical public construction contracts, or sustained investment in construction trades pipelines, are worth examining. The Tokyo stress-test raises this as a question worth investigating; it does not answer it. Whether this is best addressed at the state, federal, or international trade level is unknown. The right levers for materials supply stabilization are not well-understood and require dedicated analysis beyond the scope of this case study.

US Analog

Closest examples: California, Montana, and Washington passed significant zoning reform packages between 2023 and 2025. California legalized ADUs by right, exempted infill from CEQA review (AB 130), and capped parking minimums. Montana legalized duplexes statewide. Washington took statewide zoning reform further than any prior state action. (confirmed, CA Governor’s Office 2024; Sightline Institute; Bush Center 2025)

Data maturity: Early. Most reforms are 0-2 years old.

What we can measure now:

• Permit counts, are applications rising after reform?
• Approval timelines, are ministerial reviews actually faster?
• Housing starts, is new construction following permitting changes?

These are the right metrics at this stage. Some early data show permit increases in jurisdictions that implemented by-right ADU rules, but the picture is incomplete.

What we must not overclaim yet:

• Rent effects. Too early. No peer-reviewed study has isolated price or rent impacts from US state-level zoning reform.
• Generalizability of the Tokyo mechanism. US housing is built primarily by large homebuilders assembling consolidated parcels. Tokyo’s throughput emerged partly from thousands of small-plot individual landowners. That ownership and development pattern is not easily replicated.
• National-level durability. Japan’s preemption is constitutionally national. US state reforms can be reversed by subsequent legislatures. The political durability of these reforms is unproven.
• Tenant stability. Japan’s near-irrevocable lease law gave existing renters legal protection that has no equivalent in US reform proposals. Citing Tokyo as a supply model without noting this gap is incomplete.

The structural principle that transfers: preempting local veto power over housing types increases throughput. That much is mechanically similar. Whether the magnitude of effect translates is unknown.

What to do with this

Personal lever

Every time a housing proposal comes up in your city or state, ask two questions: How many of these homes are approved by right, no hearing required? And what is the median time from application to permit? If neither number is tracked or published, that is the problem.

Community and policy lever

Push for three things together, they are the structural package:

• By-right legalization of missing-middle housing (duplexes, triplexes, small apartments).
• Hard time limits on discretionary review, with ministerial approval as the default.
• Published approval timelines so accountability is possible.

These are the mechanisms that drove Tokyo’s throughput. The rent effects will lag, measure permits and starts first.

Scoreboard: what to track

If you do zoning reform, you still need:

• Construction labor pipeline - training programs, immigration pathways, and childcare/eldercare relief that frees workers to enter or stay in the workforce.
• Anti-speculation tools - transparency requirements and disclosure rules, especially in submarkets with high foreign investment pressure.
• Input-cost awareness - construction materials and financing rates can overwhelm a permissive regulatory environment; the right policy levers here are not well-established and need dedicated research.
• Tenant stability basics - eviction protections, renewal rights, and legal aid so existing renters are not displaced while new supply comes online.

Zoning reform removes the political barrier. These four address the constraints that remain.

North Star verdict

Tokyo is consistent with the core supply-side argument, that removing the local veto on housing production enables more building, while complicating any claim that zoning reform alone is sufficient to keep housing affordable.

The mechanism connects to the loop: housing costs are squeeze. When housing eats 40% of income, people are stressed, options narrow, and economic insecurity deepens. Policies that increase supply attack one source of the squeeze. Tokyo showed that a national by-right framework can sustain higher throughput than comparable cities over decades. That is real and worth learning from.

But the 2023-2025 price surge shows the limit clearly. When construction labor is scarce, costs are spiking, and global capital floods a market because the yen is cheap, permissive zoning cannot compensate. Supply-side reform is not a fire extinguisher for every kind of housing crisis.

The honest framing: zoning reform removes a political barrier to building. It does not remove economic barriers, and it does not protect against exogenous capital shocks. Both matter.

System lesson in one sentence: Removing the local veto on multifamily housing is a plausible necessary condition for supply-side affordability, but it is not sufficient without addressing construction economics, and it is not insulation against currency-driven foreign capital floods.

Research gaps

[RESEARCH GAP: No peer-reviewed study has isolated the relative contribution of national zoning preemption vs. post-bubble land deflation vs. incremental urbanism to Tokyo’s historical price stability. Causal claims will remain plausible, not confirmed, until that literature exists.]

[RESEARCH GAP: No peer-reviewed estimate of rent or price effects from US state-level zoning reforms (CA, MT, WA) has been identified. This gap will close as reforms age; appropriate check-in point is 2026-2027.]

[RESEARCH GAP: Whether the 2023-2025 Tokyo price surge represents a structural break or a temporary dislocation remains unknown. The Bank of Japan rate normalization trajectory is a key variable with no reliable forecast in the research file.]

Back to case studies