Tokyo new condo prices crossed ¥100 million on average for 23-ward units in 2024. They kept climbing through 2025. You are now deciding whether to buy in 2026 — and the question is simple. Is this a bubble that pops, or a cycle that plateaus?
The honest answer: nobody knows. But the signals can be read. At RE : public, we work with foreign buyers who want a second opinion before signing. Here is how we frame the data right now.
What the price data actually shows
The Real Estate Economic Institute (不動産経済研究所) reported the 2024 average new condo price in Tokyo's 23 wards at roughly ¥119 million. That is more than double the 2013 level. Used condo prices tracked by the Real Estate Information Network System (東日本不動産流通機構, REINS) show a slower but steady climb — roughly 8 to 10% year-on-year through most of 2024 and into 2025.
Two things are happening at once:
- New supply is concentrated at the top. Tower projects in central wards (Minato, Chiyoda, Chuo) skew the new-build average upward. Developers are not building cheap stock.
- Used inventory is tight. Days-on-market for central Tokyo used condos remains low. Sellers are not panic-listing.
So the headline number ("¥100 million average") is real but misleading. It is a mix shift, not only a price shift. That distinction matters when you assess risk.
The bubble checklist
A bubble is not just "prices went up a lot". A bubble is a self-reinforcing loop where buyers buy because prices rose, financed by credit that ignores income fundamentals. Run the checklist against Tokyo 2026:
1. Credit conditions
The Bank of Japan (日本銀行) ended negative rates in March 2024 and has been raising the policy rate slowly. Variable mortgage rates from major banks still sit near 0.3 to 0.7% for prime borrowers in 2025. That is loose by global standards but tighter than two years ago.
Signal: Mixed. Not a clear bubble signal yet, but the direction of travel is up.
2. Loan-to-income ratios
Flat 35 (フラット35) data from the Japan Housing Finance Agency (住宅金融支援機構) shows the average loan-to-annual-income multiple for Tokyo condo buyers has pushed past 7x in recent years. Historically, 5x to 6x was the norm. Buyers are stretching.
Signal: Elevated risk. This is the data point we watch most carefully.
3. Investor share
Pre-sale tower projects in central Tokyo have seen heavy participation from domestic high-net-worth buyers and foreign capital (Taiwan, Hong Kong, mainland China, Singapore). In some Minato-ku projects, owner-occupier share is reportedly below 50%.
Signal: Concentrated investor demand is a known bubble ingredient. But foreign buying in Tokyo is still a small share of total transactions nationally.
4. Price-to-rent ratio
Central Tokyo gross rental yields have compressed to roughly 3.0 to 3.8% on new towers. Ten years ago, 5% was achievable. When yields compress this far, you are paying for expected capital gain, not income.
Signal: Classic late-cycle tendency.
5. Construction pipeline
Here is where Tokyo differs from a textbook bubble. New condo supply has been falling, not rising. Annual new condo launches in the 23 wards are roughly half the 2000s peak. Construction costs (labor, materials) are up sharply, and developers have been disciplined.
Signal: Bearish for a supply-glut crash. This is the strongest argument against the bubble thesis.
The cycle case
Property markets cycle. Tokyo has been through at least three major cycles in living memory — the late-1980s asset bubble, the early-2000s recovery, and the post-Abenomics climb from 2013. Cycles tend to run 7 to 10 years on the way up.
If you assume we are in a cycle (not a bubble), the base case looks like this:
- Prices plateau rather than crash.
- Central Tokyo and the 23 wards behave differently from suburbs and regional cities.
- Used condos in good locations hold value better than peripheral new builds.
- The downside is time, not price — meaning your exit window may stretch from 5 years to 10.
The cycle thesis is supported by:
- Demographics in central Tokyo. Population in the 23 wards is still growing, even as Japan overall shrinks. The Tokyo Metropolitan Government (東京都) projects 23-ward population to peak later than the rest of the country.
- Tight new supply. See above.
- Yen weakness. For foreign buyers earning USD or EUR, Tokyo property is roughly 30% cheaper in their home currency than it was in 2020. That structural discount has not closed.
The bubble case
The bubble thesis is not crazy either. The arguments:
- Wage growth has not kept up. Real wages in Japan have been flat to negative for most of the last three years. If locals cannot afford ¥120 million condos, the marginal buyer is the investor — and investors leave faster than residents.
- Rate normalization risk. If the BOJ raises rates faster than expected, variable-rate borrowers face payment shock. Roughly 70% of Japanese mortgages are variable rate.
- China capital outflow rules. A meaningful share of recent Tokyo demand has come from Chinese buyers. Policy changes in Beijing could shut that tap quickly.
- Tower concentration. A handful of mega-projects (Azabudai Hills, Toranomon, Shibuya) define the top of the market. If sentiment turns on towers specifically, the headline index falls fast.
Signals to watch in 2026
Rather than predict, we monitor. These are the indicators we track for RE : public clients:
- REINS used condo days-on-market. When this lengthens by more than 30% year-on-year, sellers are losing pricing power.
- New launch contract rate (初月契約率). The industry rule of thumb: above 70% in the first month is a healthy market. Below 60% sustained is a warning.
- BOJ policy rate path. Each 25 bp move matters for variable mortgages.
- Yen exchange rate. A sharp yen rally (say, back to 120 to the dollar) removes the foreign-buyer discount and changes demand mix.
- Land price index from MLIT (国土交通省). The official 公示地価 release each March is a lagging but reliable cross-check.
None of these alone is decisive. Together they shape an analysis result you can act on.
What this means for you as a foreign buyer
You are not the marginal buyer driving the headline. You are buying for reasons that include lifestyle, currency diversification, residency utility, or long-horizon investment. So the bubble-or-cycle question matters differently for you than it does for a leveraged domestic flipper.
A few practical points:
Location asymmetry is widening
The gap between central five wards (Chiyoda, Chuo, Minato, Shinjuku, Shibuya) and outer Tokyo has widened. In a downturn, central holds better. In an upturn, central leads. If your goal is risk control rather than maximum upside, the central premium has historically been worth paying.
New vs. used tendency
New condos carry a developer premium of roughly 15 to 25% over comparable used stock. That premium evaporates the day you take possession. Used condos from the 2000s and 2010s in good locations often offer better risk-adjusted entry, but require more due diligence on management association (管理組合) health and repair reserves (修繕積立金).
Floor plan and size
Sub-30㎡ studios in central Tokyo are an investor product. They are liquid in good markets and illiquid in bad ones. Family-sized units (60㎡+) have a broader buyer pool, including domestic owner-occupiers, which is a more stable demand base.
Financing as a foreigner
If you do not have permanent residency (永住権), your mortgage options are narrower. Shinsei, SMBC Prestia, and a few regional banks lend to non-PR foreigners under conditions. Rates and LTV terms vary. Build this into your timing — financing constraints can be the real bottleneck, not price.
How RE : public approaches the question
When a client sends us a property they are considering, we do not tell you whether the market is a bubble. We give you a reference estimate based on:
- Comparable transactions in the same building and nearby buildings (REINS-derived data).
- Building-level factors: age, structure, management association financials, repair reserve adequacy.
- Location factors: station distance, ward-level price trends, redevelopment plans.
- Macro context: where the cycle indicators sit on the day you ask.
The output is a range, not a single number. We tell you what the tendency of the data suggests, where the risks concentrate, and what we would want to verify before you sign. We do not take commission from sellers or developers. That is the point of a second opinion.
A framework for your decision
Strip away the bubble-vs-cycle debate and ask three questions:
- Holding period. If you can hold 10 years, short-term cycle calls matter less. If you may need to sell in 3 years, they matter a lot.
- Leverage. A 50% LTV purchase tolerates a 20% price drop without damage to your equity story. A 90% LTV purchase does not.
- Use value. If you will live in it, the residence yield (rent you do not pay) is real and currency-stable. That cushions any paper loss.
The buyers who get hurt in any downturn share a profile: high leverage, short horizon, single asset, bought on momentum. The buyers who do well share the opposite profile.
Our read for 2026
The signals do not look like 1989. Supply is disciplined. Credit is loose but not reckless. Demographics in central Tokyo still support demand.
The signals do look like late cycle. Yields are compressed. Loan-to-income multiples are stretched. Investor share is high in towers.
Our working assumption — not a forecast — is that 2026 brings price deceleration rather than collapse, with widening differentiation between central and peripheral, between family-sized and studio, and between well-managed older buildings and over-priced new towers. The risk is not zero. The risk is also not 1990.
This is not investment advice. The final decision is yours.
Get a second opinion before you sign: republic-of-real-estate.com