Most analyses of a lodging market stop at “how many rooms does the area have?” In practice, the more important question is how many buildings those rooms are split across. Two markets can both have 2,000 rooms, yet one where a single property holds 900 of them and one where rooms are spread over 70 properties react completely differently when one new hotel opens. In this article we aggregated Japan’s lodging facilities property by property (one building = one unit) at the municipality level and calculated two concentration metrics: “largest-property share” and “top-3 share.” We then set these against estimated settled ADR and estimated occupancy, overlaid building-plan data, and identified the markets where a single new hotel rewrites the market structure.

Median Largest-Property Share Is 12.3% — Distribution Across 288 Municipalities

The analysis covers 19,424 properties tracked by MetroEngines Research that have at least 10 rooms and were recently confirmed to list sales inventory on OTAs. Extracting municipalities with at least 1,000 total rooms and at least 5 properties as “markets” yields 288 municipalities, comprising 12,719 properties and 1,133,723 rooms. The median market has 31 properties and 2,082 rooms.

Calculating the largest-property share for these 288 markets gives a median of 12.3%. In other words, “in a typical town, the single largest hotel holds roughly one-eighth of the market.” The median top-3 share is 30.2%, meaning the three largest properties hold about 30% of the market. The distribution is skewed toward low values: 105 municipalities (36.5% of the total) are below 10%, while 16 municipalities have a largest-property share above 30%. Widening the threshold to 20% or more gives 46 municipalities.

Source: MetroEngines Research & Consulting (N=288 municipalities, 12,719 properties)

Concentration Is Really About Market Size — Rank Correlation of −0.66

Why does concentration differ between markets? Most of the answer lies in market size. The rank correlation between total room count (log) and largest-property share is −0.661: the larger the market, the lower the largest-property share. Indeed, the 46 markets with a largest-property share of 20% or more average just 16.9 properties, while the 105 markets below 10% average 76.1. Concentration is not a case of “someone monopolizing the market”; it is the mathematical consequence of having few properties.

Source: MetroEngines Research & Consulting (N=288 municipalities)

Distribution of largest-property share — 288 municipalities nationwide (1,000+ total rooms, 5+ properties, as of August 2026)

Market size band (total rooms)	Municipalities	Median largest-property share	Median top-3 share	Median estimated settled ADR
1,000–2,000 rooms	136	16.0%	39.6%	¥8,800
2,000–5,000 rooms	95	9.6%	24.4%	¥9,100
5,000–10,000 rooms	32	6.2%	16.3%	¥8,000
10,000+ rooms	25	6.3%	13.7%	¥11,100

Source: MetroEngines Research & Consulting (N=288 municipalities)

Top 15 Municipalities by Largest-Property Share — One Property Holds Over 30%

The table below lists the 15 most concentrated markets. At the top is Shimukappu Village, Hokkaido (北海道占冠村), where one large resort holds 47.6% of all rooms in the village and the top three properties hold 95.8%. It is an extreme case of a standalone resort location.

Among urban markets, Mihama Ward, Chiba City (Kaihin-Makuhari) stands out. Against a total market of 4,440 rooms, a single property with one of the largest room counts in Japan — APA Hotel & Resort Tokyo Bay Makuhari (アパホテル&リゾート〈東京ベイ幕張〉) — accounts for 2,007 rooms, a 45.2% share. The structure allows one building to absorb the demand from large events at Makuhari Messe, a configuration backed by real drawing power. Similarly, Naeba Prince Hotel (苗場プリンスホテル) holds 1,216 rooms, or 37.2%, of Yuzawa Town, Niigata, and APA Hotel Shin-Osaka Ekimae (アパホテル〈新大阪駅前〉) holds 660 rooms, or 36.0%, of Higashiyodogawa Ward, Osaka. Each of these properties “takes on a substantial portion of its town’s lodging demand single-handedly.”

Looking at the largest property in each of the 46 concentrated markets (largest-property share of 20% or more) by segment, business hotels are the most common at 22, followed by resort hotels (9), city hotels (8), deluxe hotels (4), and ryokan (3). Two patterns sit side by side under the same label of “concentration”: large business hotels anchoring supply near stations and expressway interchanges in regional cities, and a single large resort shaping the town in resort areas.

Estimated settled ADR is the median of monthly values for June–August 2026. Estimated OCC is for August 2026. Markets with too few OTA-listed properties to ensure estimation accuracy are shown as “—”. Source: MetroEngines Research & Consulting

Bottom 15 Municipalities by Largest-Property Share — Where a Single Property’s Impact Disappears

In contrast, some markets have extremely low largest-property shares. Chuo City, Tokyo has 186 properties and 26,858 rooms, yet its largest property has 497 rooms — a share of just 1.9%, with the top three at only 5.2%. Taito City has 280 properties and a 2.0% share, and Chuo Ward, Osaka has 247 properties, 37,826 rooms and a 2.4% share. In these markets, adding one 500-room hotel moves total market room count by only about 2%.

This is not limited to big cities. Hakone Town, Kanagawa has 176 properties and 6,269 rooms with a largest-property share of 3.2%, and Yamanouchi Town, Nagano (Yudanaka and Shibu Onsen) has 84 properties and 2,813 rooms at 4.0%. Hot-spring towns lined with many small and mid-sized ryokan are, structurally, also “fragmented markets.”

Estimated settled ADR is the median of monthly values for June–August 2026. Estimated OCC is for August 2026. Source: MetroEngines Research & Consulting

Concentration Explains Neither Rates Nor Occupancy — Rank Correlations Weaker Than −0.1

This is the central finding of the article. The intuition that “towns where one property dominates the market must command higher prices” was not supported by the data. The rank correlation between largest-property share and estimated settled ADR is just −0.097, and −0.156 for top-3 share. The correlation with estimated occupancy is −0.092. In every case, there is effectively no correlation.

Splitting the 288 markets into quartiles by largest-property share, the median estimated ADR declines gently from ¥9,700 on the fragmented side to ¥8,600 on the concentrated side, but this gap most likely reflects differences in market size (larger urban markets have higher rates) rather than concentration itself. Estimated occupancy is essentially flat, ranging from 89.9% to 91.6% (as of August 2026, estimated from OTA-listed inventory, N=163 municipalities).

Estimated occupancy is calculated only for the 163 municipalities with enough OTA-listed properties to ensure estimation accuracy (note that the fourth quartile has a small base of 16 municipalities). Source: MetroEngines Research & Consulting

In short, concentration is not a metric for reading pricing strategy or the strength of supply and demand. Where concentration matters is when a new property enters the market. If a 200-room hotel opens in the median market (2,082 rooms), that one property alone accounts for 8.8% of the market. In the fourth quartile (median 1,304 rooms), it is 13.3%. Supply of a size that would be mere noise in a fragmented market becomes a structural change in a concentrated one.

12 Municipalities Where One Property Changes the Structure — Supply Impact Seen Through Building Plans

Using MLIT’s “Building Construction Statistics,” we aggregated by municipality the lodging building plans filed for completion or operation from September 2026 onward (344 projects and 54,219 rooms nationwide after removing duplicates). Of these, 120 municipalities and 43,728 rooms fall within the 288 markets in this study. The “supply impact ratio” — planned rooms divided by existing rooms — exceeds 20% in 12 municipalities. For the distribution of the same ratio on a new-opening basis with confirmed opening dates, see our separate Hotel Supply Pressure Ranking 2026-2027.

Source: Compiled by MetroEngines Research & Consulting from MLIT “Building Construction Statistics” / Existing room counts: MetroEngines Research & Consulting

*Based on building-confirmation applications as of the survey date. Project and room counts are expected to rise as further applications are filed; treat these figures as the lower bound of the currently confirmed pipeline. Source: MLIT “Building Construction Statistics” / MetroEngines Research & Consulting

A Range Based on How Much of the Pipeline Materializes — Three-Case Comparison

The supply impact ratios in the table above assume every filed plan is realized. Because building-confirmation applications are filed one to two years before opening, cancellations or downsizing may occur, while additional applications could push the numbers higher. Here we show a range for the 12 municipalities by splitting plan realization into three cases. Case A: the single largest plan is cancelled and only the rest are realized. Case B: only the single largest plan is realized. Case C: all plans are realized (the figures used in the main body of this article).

Case A = (planned rooms − rooms in largest plan) ÷ existing rooms; Case B = rooms in largest plan ÷ existing rooms; Case C = planned rooms ÷ existing rooms. Source: MLIT “Building Construction Statistics” / MetroEngines Research & Consulting

The nature of the range differs from market to market. In Konohana Ward (Osaka), Miyakojima City, Mihama Ward (Chiba), Osaki City and Ozu Town, a single largest plan accounts for most of the pipeline, so whether that one project goes ahead determines the conclusion outright. By contrast, in Minami Ward (Hiroshima), Kutchan Town and Yokkaichi City, supply builds up from multiple plans, so even if one drops out, a supply increase of around 20% remains. The former requires tracking the progress of individual projects; the latter requires assessing the absorptive capacity of the market as a whole.

The Share a New Property Can Capture — Sensitivity to Market Size × Room Count

From an investment and development perspective, the position a property of a given size can capture changes completely depending on the size of the market it enters. The table below varies existing room count and the new property’s room count, and calculates the share that property holds of the post-opening market (existing + new). The median existing room count across the 288 markets is 2,082; adding 200 rooms to that median market yields an 8.8% share. The same 200 rooms would reach only 3.8% in a 5,000-room market but 16.7% in a 1,000-room market.

Darker cells indicate higher share. Existing room counts as of August 2026. Source: MetroEngines Research & Consulting

As this article’s analysis shows, concentration itself explains neither estimated settled ADR nor estimated occupancy (rank correlations weaker than −0.1 in both cases). This table therefore does not suggest that “capturing share raises rates”; rather, it shows how much influence a single property has over the market’s inventory curve. In combinations where share exceeds 20%, that one property’s decisions to stop selling or release inventory directly affect supply and demand for the entire market.

Map of Concentrated Markets and Supply Pipelines

Plotting concentrated markets (largest-property share of 20% or more) and markets with a supply impact ratio of 15% or more on a map shows that the two barely overlap. Concentrated markets are scattered across regional mid-sized cities and resort areas, whereas large building plans cluster at specific locations tied to urban redevelopment, airports and large-scale tourism projects. Circle size represents largest-property share; blue marks markets with supply plans.

Coordinates are the locations of each municipality’s largest property. Sources: MetroEngines Research & Consulting; MLIT “Building Construction Statistics”

Four Cases — What the Next Property Changes

Mihama Ward, Chiba City (Kaihin-Makuhari): concentration eases
Against 4,440 existing rooms across 13 properties, a mixed-use development of about 1,000 rooms in front of Kaihin-Makuhari Station is scheduled to open in the second half of 2027. If realized, the market will grow to 5,440 rooms and the current 45.2% largest-property share will fall to 36.9%. The new property alone would account for 18.4% of the market. Unusually for a concentrated market, this is a case where new supply shifts the structure toward fragmentation. We examine this development’s absorption capacity in detail in Kaihin-Makuhari 1,000-Room ¥30bn Mixed-Use Development.

Konohana Ward, Osaka City: from fragmented to concentrated in one move
4,785 existing rooms across 19 properties. The largest is Liber Hotel Osaka (リーベルホテル大阪) with 760 rooms, a 15.9% share — currently a moderately fragmented market. Waiting in the pipeline are plans totalling 3,317 rooms, including about 2,500 rooms tied to the integrated resort on Yumeshima (scheduled to open in autumn 2030). The supply impact ratio of 69.3% is the highest in Japan. Once all plans are in place, the market will reach 8,102 rooms and the largest-property share will jump to 30.9%. It is one of the few cases in Japan where a single development converts a fragmented market into a concentrated one.

Miyakojima City, Okinawa: 1,726 rooms into a 98-property fragmented market
A typical fragmented market, with 5,602 existing rooms shared among 98 properties and a largest-property share of just 5.9%. But if a large resort planned in the Gusukube district (1,726 rooms) is added, that one property alone would account for 23.0% of the market. Estimated settled ADR is ¥16,500, among the highest in Japan, and estimated occupancy is also high at 89.9% (as of August 2026, same basis as above). The market’s ability to absorb demand makes the plan plausible, but it is worth noting that the structure would change to one in which a single property’s operating policy could influence price levels across the entire island.

Minami Ward, Hiroshima City and Yokkaichi City, Mie: restructuring mid-sized cities
Minami Ward in Hiroshima City has 6 plans totalling 1,687 rooms against 4,844 existing rooms across 35 properties (supply impact ratio 34.8%), and Yokkaichi City has 3 plans totalling 928 rooms against 2,784 existing rooms across 32 properties (33.3%). Both involve several overlapping properties of 300–600 rooms each; rather than dramatically shifting the largest-property share, they lift the total room count of the market by about one-third. For Yokkaichi, we analyze its distinctive demand driven by scheduled plant maintenance shutdowns (turnarounds) in Petrochemical Turnaround Year 2026.

The Largest Properties in Concentrated Markets Sit Outside Listed REITs

For the 46 markets with a largest-property share of 20% or more, we checked against our ownership master data whether the largest property is owned by a listed hotel REIT. Only one matched: in Suminoe Ward, Osaka. Nearly all of the core properties in concentrated markets lie outside listed REIT portfolios. We covered the structure in which the seven listed REITs own only a fraction of Japan’s hotel rooms in 970,000 Hotel Rooms Untouched by REITs; viewed through the lens of concentration, a further implication emerges: the more a single property sways a town’s supply and demand, the more likely it is to sit outside the public markets. Area analysis built solely on listed REITs’ monthly disclosures cannot fully capture the reality of concentrated markets.

Practical Implications

For properties operating in concentrated markets. In markets where the largest-property share exceeds 20%, a large property’s decisions to release inventory or stop selling directly move the inventory curve of the whole market. When reading your own booking curve, also observing when the largest property in your market releases inventory and when it fills up makes it easier to plan the timing of rate changes. In addition, in markets where a new property’s opening is already known, clarifying your own strengths (location, facilities, dining, room size) one to two years before that opening helps secure room to be chosen even as supply increases. We have measured how long it takes new properties’ rates to catch up with market levels in New Hotel ADR Recovery: Median 2 Months.

For properties operating in fragmented markets. In markets where the largest-property share is below 5%, there is little need to be overly wary of a single competitor opening. What matters more is cumulative supply, and tracking the annual rate of supply growth is more practical than following individual opening announcements.

From an investment and development perspective. Concentrated markets offer room to take a clear position in an area’s supply structure through the acquisition or development of a single property. Adding 200 rooms to the median market (2,082 rooms) yields an 8.8% share, and 13.3% in a market of around 1,300 rooms. On the other hand, because rate and occupancy levels cannot be predicted from concentration, the ADR and OCC assumptions underpinning a business plan need to be built individually from the market’s size band and segment mix.

Methodology and Sources

Property and room counts are based on a population of 19,424 properties (1,501,842 rooms) tracked by MetroEngines Research that have at least 10 rooms and were confirmed in the most recent survey to list sales inventory on OTAs. Of these, the 288 municipalities with at least 1,000 total rooms and at least 5 properties are analyzed. Ryokan, minshuku and simple lodging facilities not listed on OTAs are not included. Where the same property is listed under multiple names, we consolidate by name, but some duplicates may remain due to spelling variations. Properties split into separate wings or buildings are counted according to their listing units.

Estimated settled ADR is the median of monthly values for June–August 2026 (median number of properties covered: N=27.5 per municipality). The most recent month includes estimates based on sales prices listed on OTAs as of the survey date and may change. Estimated occupancy is the monthly average for August 2026; municipalities with fewer OTA-listed properties than a set threshold are excluded (calculated for 163 of the 288 markets). Building plans are based on MLIT’s “Building Construction Statistics” and aggregate lodging-use projects filed for completion or operation from September 2026 onward, excluding duplicates with the same municipality and the same room count.

Related Articles

• Kaihin-Makuhari 1,000-Room ¥30bn Mixed-Use Development — Absorption Capacity of Chiba’s Largest Pipeline
• 970,000 Hotel Rooms, 4,934 Properties Untouched by REITs — The Listed 7 Hold Just 4.5% of Rooms
• 34% of Japan’s Hotel Pipeline Is Operator Self-Build — 69 Projects and 15,189 Rooms by Developer Type, 2026
• New Hotel ADR Recovery Takes a Median 2 Months — 182 Openings Measured and How to Build It Into a Business Plan
• Central Kumamoto Room Supply +26.0% vs 2019 — Estimated Settled ADR in a Supply Growth Phase
• Hotel Supply Pressure Ranking 2026-2027 — Ozu Town, Kumamoto Tops the Ratio to Existing Stock at 30.8%
• Does New Hotel Supply Lower ADR? Tested Across 47 Prefectures — It Only Bites Below 13,000 Rooms
• Tokyo 23 Wards Hotel Supply Pressure 2026 — Arakawa at 16.0% of Existing Stock, Minato Just 4.5% Despite Leading in Rooms