2025 may be remembered as a turning point in the history of Japan’s hotel development pipeline. JR Kyushu cancelled its ¥43.5 billion Hakata Station Sky City project, the Nakano Sunplaza redevelopment ended with the termination of its operator agreement, and Phase II of Shibuya Scramble Square was pushed back by four years. Surging construction costs, labour shortages, and a changing interest-rate environment have combined to bring large-scale hotel projects that had been moving forward to a sudden halt. This article catalogues the major frozen and cancelled projects that surfaced in 2025-2026, and quantifies the supply outlook from 2027 onward — “to the extent it can be confirmed today” — by combining new-opening data on an OTA listing-confirmed basis (MetroEngines Research & Consulting), the construction-application-based planning pipeline from the Ministry of Land, Infrastructure, Transport and Tourism’s Building Dynamics Statistical Survey, and building-start statistics. Because data observation lead times make a simple comparison of property counts misleading, we read the structure across three axes: room count, large-project pipeline, and existing-stock profitability metrics. We systematically analyse what investors need to watch, from the implications for REITs holding existing stock, to the conditions required for projects to restart, to spill-over effects on conversion demand.
Metric Definitions Used in This Article and How to Read the Data
- ADR (Average Daily Rate): The average of publicly listed prices on OTAs. This differs from actual transacted rates (when cross-checked against REIT disclosure data, public ADR tends to run +25-30% higher than transacted ADR, because unsold high-priced plans remain on OTAs and pull the public average above the actual transacted average). Per-room rate (tax included) for two-person occupancy, averaged across all plan types (room-only through meal-included plans).
- New openings (OTA listing-confirmed basis): The count of newly listed facilities confirmed on OTAs, drawn from the universe of approximately 27,000 domestic facilities / 1.26 million rooms tracked by MetroEngines Research. OTA listings appear within several months before or after opening, and pre-opening listings account for only about 19%, so figures for recent months and forward years are “observation-in-progress” numbers that will increase as additional listings come in. Ryokan and minshuku not listed on OTAs are excluded.
- Construction planning pipeline (application basis): Counts of building plans for the lodging industry where construction confirmation applications have been filed, per the MLIT Building Dynamics Statistical Survey. Because confirmation applications are typically filed 1-2 years before opening, figures for future years are “lower-bound” values that will increase as additional applications come in.
- RevPAR: Occupancy × ADR. Revenue per available room.
- Data sources: Opening results from MetroEngines Research & Consulting (OTA listing-confirmed basis); planning pipeline from MLIT Building Dynamics Statistical Survey; construction starts from e-Stat Construction Starts Statistics (by use, lodging industry).
Major Project Cancellations and Delays Surfacing in 2025-2026
In 2025, several emblematic hotel development projects were cancelled or postponed in quick succession. The common immediate trigger was the surge in construction costs, but the underlying backdrop varies by project: labour-shortage issues, shifts in demand outlook, deteriorating profitability, and rising land acquisition costs. Below are five representative cases compiled from media reports and developer disclosures.
(1) Hakata Station Sky City Project (JR Kyushu) — Cancellation announced 26 September 2025
The plan was to build a B1 / 12-story complex with roughly 50,000 m² of total floor area, hosting offices, a hotel, and retail, on a 5,200 m² deck over the JR Hakata Station tracks. Since 2021, around ¥6 billion had been spent on preparatory works such as track relocation. However, construction costs swelled to nearly twice the original estimate, and combined with the difficulty of building over the tracks at night, the project was judged uneconomic. The cancellation is a textbook example of how the unique demands of over-track construction amplify the impact of cost inflation.
(2) Utsunomiya Station East Exit High-end Hotel — “Difficult for the foreseeable future,” 2025
The luxury hotel was positioned as the centrepiece of the Utsunomiya Station East Exit redevelopment, but in 2025 the developer consortium responsible for the facility advised the City of Utsunomiya that “construction will be difficult for the foreseeable future.” According to the City, the construction cost index has risen to roughly 1.4× the 2015 level, and the rise in project costs is squeezing profitability. The plan was predicated on demand around the new LRT corridor, but more than a year after the town-opening, the planned site still remains empty.
(3) Nakano Sunplaza Site Redevelopment — Operator agreement terminated, June 2025
The Nakano Station new north-exit redevelopment, originally budgeted at ¥181 billion, had ballooned to over ¥350 billion by September 2024, and the basic agreement between Nakano Ward and the operator was terminated in June 2025. The ward plans to release a revised plan in spring 2026 and to re-tender for a new private operator in fiscal 2027, pushing the start of construction back to fiscal 2030 and completion to fiscal 2034. The original plan included a hotel wing attached to a 7,000-seat hall.
(4) Shibuya Scramble Square Phase II (Central and West Buildings) — 4-year delay, May 2025
Originally scheduled for completion in fiscal 2027, Phase II was pushed back to fiscal 2031 in a joint announcement by the three parties in May 2025. “Changes in visitor needs” and “diversification of pedestrian-space functions” were given as official reasons, but rising construction costs and longer build periods sit behind them. Combined with the already-completed East Building, the development would create one of Greater Tokyo’s largest retail complexes, with floor plates of up to 6,000 m². Its delay is emblematic of how large-scale mixed-use developments incorporating hotel components are slipping.
(5) Meitetsu Nagoya Station Redevelopment — Demolition and new build deferred, November 2025
Originally targeting demolition starting in fiscal 2026 and Phase 1 completion in fiscal 2033, the redevelopment in front of Meitetsu Nagoya Station hit a wall on 26 November 2025 when the prospective contractor withdrew its tender citing “difficulty in securing the personnel needed to assemble the construction team.” Both demolition and new construction schedules are now undecided. The original plan called for three 180-meter towers; the revised concept removes one of the business hotel towers and trims the main building to roughly 100 m. This effectively freezes the largest station-front redevelopment in the Chukyo metropolitan area. For a city-level read on how business demand is recovering in Japan’s four major markets, see our In-Depth Weekday ADR Analysis of Tokyo, Nagoya, Osaka, and Fukuoka.
None of these five projects was undone by a single cost problem; each is the result of compounded structural pressures. The sharp rise in construction costs (per-tsubo unit cost up +41% from 2022 to 2024), shortages of construction labour, growing gaps between assumed and achievable revenue, and the shifting interest-rate environment — these four forces are advancing in parallel, and as a result, large-scale redevelopments that had previously been treated as “a given” are now stalling one after another.
The Supply Gap by the Numbers — Structural Shrinkage of the Room-Count Pipeline
Looking only at individual cancellations misses the larger picture. At the same time, comparing only the count of new openings invites mis-reading — the observation lead time can make it look as though “2027 will see zero openings, 2028 zero openings.” In this section, we place two datasets side by side — new-opening results on an OTA listing-confirmed basis (recent years are observation-in-progress) and the construction-application-based planning pipeline (forward years are a lower bound) — and read them through the lens of room count, revealing structure that a simple property count cannot.
The aggregations in this section draw on two sources.
(1) “New-opening results”: OTA listing-confirmed basis (MetroEngines Research). Pre-opening OTA listings account for only about 19% of the total, and more than half of facilities are listed only 91 days or more after opening, so figures for recent months onward are observation-in-progress numbers and will increase as additional listings come in.
(2) “Construction planning pipeline”: MLIT Building Dynamics Statistical Survey, on a construction-application basis. Applications are typically filed 1-2 years before opening, so counts from 2027 onward are expected to rise as further applications come in, and should be read as the current lower bound for the confirmed pipeline.
The picture becomes clear when read at the room level. New openings in 2018 totalled around 113,000 rooms; since 2022, the annual range has been 40,000-50,000 rooms. 2026 stands at 23,000 rooms as observed, a 42% drop in room count versus 2024. The planning pipeline is even more striking. On a confirmation-application basis, room counts run from 15,555 in 2024, to 5,397 in 2025, 4,841 in 2026, and 2,103 rooms for 2027 and 1,175 for 2028 at present. These figures will rise as new applications are filed, but compared with 2024’s level (15,555 rooms), they imply room-level shrinkage of -86% and -92%. In essence: while the property count can shift meaningfully as new OTA listings are added, the thinning of the large-project pipeline is already visible in the confirmation-application data.
Notable are the moves in January-March 2026. The levels — 211 properties in January, 115 in February, 115 in March — are at or above the same months of 2024 (125 in January, 92 in February, 158 in March). In other words, in recent months where the observation lead time has largely cleared, the pace of openings is not much different from prior years. The real supply shrinkage needs to be read not from property count but across three axes: room count, large-project pipeline, and construction-start statistics.
Construction-start statistics confirm the supply shrinkage from another angle. The national count of hotel construction starts peaked around 2,300 buildings in 2019, fell sharply during the pandemic, and has since 2022 oscillated between 2,000 and 2,800 buildings. What matters is that, viewed in floor-area terms, cost inflation has driven a downward trend in the average size per building. The shift from the pre-pandemic “large mixed-use developments” to a “mid-size / compact-focused” pattern is well underway, and this is consistent with the shrinkage in the room-count pipeline.
Where Frozen Projects Cluster — Prefecture-Level Skew in Hotel Construction Starts
Planning freezes are not occurring evenly across geography. Extracting prefecture-level hotel construction starts from the e-Stat building-start statistics reveals areas where cancellations and delays are concentrated.
Tokyo plunged from a 2019 peak of 213 starts to 29 in 2023, and recovered slightly to 67 in 2024 — still only one third of its prior peak. Fukuoka rebounded to 36 starts in 2024, but that simply matches 2017’s level (38), and how the Hakata Sky City cancellation feeds into future numbers will bear watching. Tochigi posted unexpectedly high levels — 149 in 2022 and 124 in 2023 — then slowed to 74 in 2024. The Utsunomiya East Exit freeze illustrates the investment risk of betting on LRT-corridor demand.
| Prefecture | 2024 Starts | vs. 2019 | Notable Frozen Projects |
|---|---|---|---|
| Tokyo | 67 | -69% | Nakano Sunplaza, Shibuya SS Phase II |
| Osaka | 18 | -82% | Concerns over post-Expo reversal |
| Fukuoka | 36 | -65% | Hakata Station Sky City |
| Tochigi | 74 | -5% | Utsunomiya East Exit High-end Hotel |
| Aichi | 37 | -16% | Meitetsu Nagoya Station Redevelopment |
Market Reaction in Frozen Areas — ADR Trends in Hakata, Utsunomiya, and Nakano
What price reaction has unfolded in the affected areas after each cancellation was announced? Using OTA public-rate data, we tracked monthly ADR in Hakata-ku, Utsunomiya, and Nakano-ku from January 2025 through April 2026.
Nakano-ku saw ADR climb from ¥12,600 in January 2025 to ¥19,800 in April 2026 — a +57.3% surge. Since the Nakano Sunplaza operator agreement was terminated in June 2025, a clear upward trend has taken hold. Market recognition that mid-scale hotel supply around Nakano Station will not rise meaningfully for some time may be strengthening the pricing power of existing stock.
Hakata-ku: around JR Kyushu’s cancellation announcement in September 2025, ADR ran ¥26,600 in September, ¥30,100 in October, and ¥33,100 in November — roughly +25% within three months. After a seasonal pullback into the new year, ADR in April was ¥29,900, holding +11.7% year-on-year.
Utsunomiya climbed to ¥21,100 in October 2025, then receded to ¥15,200 in December partly on demand factors, and entered 2026 trading in the ¥16,000 range. This suggests that in regional core cities, the price-lifting effect from supply constraints driven by planning freezes is limited. The supply mix for business hotels in Tokyo’s 23 wards still shows significant area-level divergence, and the way the supply-demand balance feeds into ADR is structured by location; we cover this in our In-Depth Analysis of Tokyo 23-Ward Business Hotels (March 2026).
All these moves remain at the level of correlation, not causation, but the pattern that emerges is clear: the larger the metro area, the more readily a freeze announcement is reflected in existing-stock ADR. For supply constraints to translate into pricing power, two pre-conditions must hold — continued growth in inbound visitor numbers, and the resilient domestic-demand backdrop staying intact.
How Much ADR Headroom Does the Supply Gap Create? — Sensitivity Analysis
How much room for ADR upside does the supply gap from project freezes generate? We attempt a sensitivity analysis based on assumed price elasticity. Using an elasticity of -1.0 (a standard reference value for the hotel industry), we model scenarios where the supply gap lifts occupancy at existing stock and ultimately feeds into ADR.
| Scenario | Assumed Supply Gap | OCC Lift | ADR Lift (est.) | RevPAR Lift (est.) |
|---|---|---|---|---|
| Mild (-5%) | -5% | +2.5pt | +3 to 5% | +5 to 8% |
| Moderate (-10%)Base | -10% | +5pt | +6 to 10% | +10 to 15% |
| Severe (-15%) | -15% | +7.5pt | +10 to 15% | +16 to 23% |
New openings went from 1,968 properties / 39,639 rooms in 2024 to 588 properties / 23,132 rooms in 2026 (as observed), and the 2027 planning pipeline currently stands at 2,103 rooms (a 86% drop in room count vs. 2024, though this is a lower bound that will rise as further confirmation applications are filed). This room-count shrinkage may correspond to the moderate-to-severe scenarios. Caveats apply: the analysis assumes constant demand, and any slowdown in inbound growth or correction in domestic lodging demand would offset the upside. Because observation lead times push opening counts and pipeline numbers upward over time, ongoing monitoring of forward-year figures is essential.
Investor Implications — Tailwinds for REITs Holding Existing Stock and Conditions for Restart
A supply gap is a constraint on new development opportunities, but it is also a clear tailwind for holders of existing stock. REITs whose portfolios are concentrated in urban areas are particularly well-positioned to benefit from a double effect: stronger pricing power and stable, elevated occupancy.
Looking at the latest monthly data across seven major REITs: Invincible Investment Corporation (8963), focused on urban business hotels, posted OCC 87.6% (+2.8pt YoY), ADR ¥14,500 (+6.0% YoY), and RevPAR ¥12,700 (+9.0% YoY), maintaining steady growth. Upper-midscale-focused Japan Hotel & Residential Investment Corporation (3472) is at high levels with OCC 88.7%, ADR ¥30,900, and RevPAR ¥27,800. Hoshino Resorts REIT Investment Corporation (3287) shows some month-to-month variation due to renovations, but at the portfolio level RevPAR is on a recovery track. For a complementary perspective on the market impact of the six luxury hotels opening in 2026 within this supply-gap context, see our Six Luxury Hotels Opening in 2026: Investor Perspective on Market Impact.
Conditions for Restart — Four Triggers
(1) Construction Cost Peak-out
If the 2024 per-tsubo unit price of ¥1.95 million stops rising further and flattens or modestly declines through 2026-2027, frozen projects can be re-evaluated for feasibility. Steel prices and labour costs are key.
(2) Interest-rate Environment
Stability in long-term rates and REIT distribution yields holding around 3-4% are prerequisites for re-examining projects. A sharp rise in rates would materially worsen IRR on new projects.
(3) Continued ADR Growth
If urban ADRs continue rising at 5-8% YoY, project economics can be re-cast on higher revenue assumptions. Sustained strength in inbound demand is a precondition.
Threshold ROI for Restart — Feasibility Analysis
We estimate the threshold conditions under which a large-scale frozen hotel project becomes economically viable to restart. Assuming a 100-room property with 10,000 m² total floor area, per-tsubo unit price of ¥2 million, and approximately ¥6 billion in construction cost, we compare the following scenarios.
| Item | A. Freeze-time Level | B. Moderate RecoveryRestart Threshold | C. Bullish Scenario |
|---|---|---|---|
| Construction unit cost (per tsubo) | ¥2.20M | ¥1.90M | ¥1.70M |
| Construction cost (100 rooms / 10,000 m²) | ¥6.66B | ¥5.75B | ¥5.15B |
| Assumed ADR | ¥30,000 | ¥33,000 | ¥36,000 |
| Assumed OCC | 75% | 80% | 82% |
| Annual revenue | ¥0.82B | ¥0.96B | ¥1.08B |
| GOP (assumed 30%) | ¥0.25B | ¥0.29B | ¥0.32B |
| Yield on construction cost | 3.7% | 5.0% | 6.3% |
If construction unit costs stay at current levels (¥2.2M per tsubo) and ADR and OCC remain at freeze-announcement levels, yield only reaches 3.7%, below the average REIT yield of 4-5%. The threshold for restart appears to be Scenario B, where the unit cost settles around ¥1.9M per tsubo and ADR rises to roughly ¥33,000 — a level that would require re-evaluation on a 2-3 year horizon.
Spill-over to Conversion Demand
As new-build hotel development is suppressed, demand for converting existing buildings into hotels is growing. Use of large facilities such as the old Sunplaza building in Nakano after its closure, conversion of surplus office floors in regional cities into hotels, and lifting the residential share within mixed-use redevelopments with a lodging component are all starting to function as “de-facto new supply” in an environment where new builds are constrained. Conversions, which sidestep the structural / shell works that account for roughly 40% of new-build cost, can be delivered at ¥1.2-1.5M per tsubo — about two-thirds of a new build — and have the advantage of allowing operations to start at a lower total project cost.
2027-2028 RevPAR Upside Scenario
Integrating these structural changes, the following RevPAR upside scenarios are likely for major urban areas in 2027-2028.
| Segment | Base Case RevPAR Growth (annual) | Supply-Gap Scenario RevPAR Growth (annual) | Differential |
|---|---|---|---|
| Urban Business Hotels | +5 to 7% | +8 to 12% | +3 to 5pt |
| City / Upper Midscale | +4 to 6% | +7 to 11% | +3 to 5pt |
| Luxury | +3 to 5% | +5 to 8% | +2 to 3pt |
| Regional Business | +2 to 4% | +3 to 5% | +1 to 2pt |
Urban segments benefit most from the supply gap, while regional impact is limited. What investors should keep in mind is the structural takeaway: REITs with location-concentrated portfolios have the most RevPAR upside potential during the 2027-2028 supply gap.
Conclusion — What the Cancellation Wave Says About the Industry’s Structural Turning Point
The 2025-2026 wave of project freezes is not merely a cyclical investment correction; it points to a structural turning point for the hotel development industry. The 40% rise in construction costs, the labour shortage, and the upward shift in profitability thresholds are advancing in parallel, and as a result, large-scale redevelopments that were previously treated as “a given” have started to stall. Meanwhile, inbound demand remains resilient and RevPAR on existing stock is in an uptrend. This supply-demand mismatch is what is set to surface as a “supply gap” in 2027-2028.
What this article has emphasised is not mis-reading the observation lead time of the data. On the OTA listing-confirmed opening data, pre-opening listings account for only about 19% of the total and under-represent forward months. The construction planning pipeline also grows over time as new confirmation applications come in. It is a mistake to treat a simple drop in property count as “zero supply.” Only by reading the three axes — room count, large-project pipeline, and construction-start statistics — together does the structure emerge: a meaningful shrinkage in room-level supply versus 2024, with large redevelopments in particular coming to a halt.
The implications for investors are straightforward. In the short term, the wave is a tailwind for REITs with location-concentrated, urban-focused portfolios. In the medium term, investors should monitor progress on the restart conditions (construction cost peak-out, rate stability, sustained ADR growth) while selectively backing conversion deals. Whether one positions to capture the 2027-2028 RevPAR upside scenario, or waits for the threshold conditions for restart to look right before committing to new development, the next investment decision will be sharpened by continuously tracking the five cancellations and delays catalogued above, the structural forces behind them, and the trend in additional applications coming into the planning pipeline.
Note: ADR figures in this article reflect publicly listed prices on OTAs at the time of survey and may move as check-in dates approach. New-opening data is on an OTA listing-confirmed basis with observation lead times; the construction planning pipeline is on a confirmation-application basis, and forward years will grow as additional applications are filed. Feasibility estimates and RevPAR scenarios are simplified analyses based on public data and general assumptions, and actual investment decisions require individual feasibility studies.
Related Reading
- In-Depth Analysis of Tokyo 23-Ward Business Hotels (March 2026) — Supply Expansion and Supply-Demand Balance
- In-Depth Weekday ADR Analysis of Tokyo, Nagoya, Osaka, and Fukuoka | Visualising Business Demand Recovery
- Six Luxury Hotels Opening in 2026: Investor Perspective on Market Impact
- Investment Opportunity Analysis of In-Park Hotel Development under the Park-PFI Framework
- Hotel Investment in the Era of 80% Nankai Trough Probability — Disaster-Risk and ADR Analysis of 10 Pacific-Coast Destinations
- Impact of the Osaka Expo on Hotel ADR | Verified with REIT and OTA Data
References and Sources
– Market Data (MetroEngines Research)
- MetroEngines Research & Consulting — Japan new hotel-opening data (OTA listing-confirmed basis, approximately 27,000 facilities tracked), monthly ADR data, area-level competitive benchmarks. Note that, given the lead-time characteristics of OTA listings, figures for recent months and years are observation-in-progress numbers.
– Government Statistics / Public Data
- MLIT Building Dynamics Statistical Survey — The construction-application-based planning pipeline for the lodging industry (counts and room totals). Future years should be read as lower-bound figures that will increase as further applications come in.
- e-Stat Construction Starts Statistics (by use, lodging industry) (Statistics ID: 0003114490) — National and prefectural hotel construction starts and floor area.
- MLIT Construction Cost Deflator — Reference for the annual trend in construction unit cost (per tsubo).
– REIT and Industry Reports
- Invincible Investment Corporation (8963) Monthly Operating Results
- Hoshino Resorts REIT Investment Corporation (3287) IR Information
- JLL: “Japan’s Hotel Market — Latest Trends, Relative Value vs. Global Peers, and Limited New Supply Amid Construction Cost Inflation” (TravelVoice, December 2025)
- JLL: 2026 Global Hotel Investment Outlook Report
– Project Cancellation / Delay Primary Sources
- JR Kyushu: “Notice on the Cancellation of the Hakata Station Sky City Project” (26 September 2025)
- Nikkei: “JR Kyushu cancels Hakata Station ‘Sky City’ mixed-use building on cost inflation”
- Shimotsuke Shimbun: “JR Utsunomiya East Exit unbuilt high-end hotel — consortium tells city ‘difficult for the time being’”
- Shimotsuke Shimbun: “Unbuilt hotel told to be ‘difficult for the time being’: Utsunomiya East Exit, cost inflation cited by developer”
- Nikkei xTech: “Two years after closure, Nakano Sunplaza redevelopment reset; ward to issue revised base plan in spring 2026”
- Nikkei: “Nakano Sunplaza site redevelopment plan revised; approval application withdrawn”
- Tokyu: “Once-in-a-century large-scale redevelopment — Shibuya Station district plan enters its final chapter” (9 May 2025)
- FASHIONSNAP: “Shibuya Scramble Square Phase II delayed 4 years, now targeting fiscal 2031 completion”
- Fashion Press: “Nagoya Station-area redevelopment — retail, hotel construction and Meitetsu Nagoya Station expansion plans revised”
- Kansai Sanpo: “Meitetsu postpones and revises Nagoya Station redevelopment plan”
- HotelX: “Hotel new supply plunging, severe supply shortage looms amid construction cost inflation”