“What is Okinawa’s hotel occupancy right now?” “How full are hotels in Osaka and Yokohama?” These questions are surprisingly hard to answer with a single number. The room occupancy rate published by the Japan Tourism Agency (JTA), the occupancy that listed hotel REITs disclose every month, and the occupancy MetroEngines Research estimates from inventory listed on online travel agencies (OTAs) are all called “occupancy,” yet they differ in both base population and timeliness. This article starts from our OTA-listed-inventory estimates for September 2026 (Okinawa 86.7%, Osaka 80.4%, Kanagawa 84.2%, Fukuoka 88.3%), sets them alongside the JTA’s prefecture-level room occupancy rates (July 2026 second preliminary release; September 2025 final figures), and compiles a lookup table for all 47 prefectures plus Naha City, Hakata Ward, five Osaka City wards and two Yokohama City wards. It also explains the three occupancy definitions and when to use each, gives RevPAR benchmarks using the estimated settled ADR, and lays out a step-by-step method for looking up the occupancy of any city yourself.
Metric Definitions Used in This Article
- Occupancy (OTA-listed inventory basis, estimated): Share of rooms sold out of total rooms in the area (an estimate based on OTA sales inventory). Covers properties that list at least 30% of their total rooms on major booking sites; calculated as a monthly average from the remaining-inventory trajectory from 45 days before the stay date to the stay date itself. Because it reflects how inventory sold on OTAs is absorbed, it differs from a property’s actual overall occupancy. Rooms not listed on OTAs are counted as sold, so it runs higher than the JTA room occupancy rate. All figures are monthly values, not full-year occupancy.
- Room occupancy rate (JTA): Room occupancy rate from the Japan Tourism Agency’s Overnight Travel Statistics Survey (宿泊旅行統計調査). Rooms used divided by the property’s total rooms, covering all property types including ryokan, resort hotels, business hotels, city hotels, simple lodgings and company/group lodgings. The latest prefecture-level month is July 2026 (second preliminary release); 2025 values are final.
- Occupancy (REIT-disclosed): Room occupancy of hotels owned by listed hotel REITs, as disclosed in their monthly IR releases. Figures are quoted exactly as published by each REIT.
- ADR (average daily rate): Estimated settled rate (tax-excluded equivalent), calculated by applying a segment-specific adjustment factor to the lowest plan rate each property publishes on OTAs (2 guests per room, per-room price, tax included). Validated against property-level results disclosed by listed hotel REITs, the median error is about 7%. These are estimates and differ from each property’s actual settled prices and accounting figures. Area-level ADR is the median of the covered properties.
- RevPAR benchmark: Simple product of occupancy × ADR. An estimate calculated for this article, not actual results.
- Data sources: MetroEngines Research; Japan Tourism Agency, Overnight Travel Statistics Survey; monthly operating results of each REIT
- — Estimated occupancy on an OTA-inventory basis for September 2026: Okinawa 86.7%, Osaka 80.4%, Kanagawa 84.2%, Fukuoka 88.3% (national room-weighted average 85.4%, 15,460 properties).
- — JTA room occupancy rate: the latest prefecture-level month is July 2026 (second preliminary), at Okinawa 61.8%, Osaka 72.6%, Kanagawa 66.4%, Fukuoka 72.4%.
- — Comparing the same month, July 2026, the definitional gap is 9–28 points. Largest in Okinawa (+27.5pt), smallest in Osaka (+9.1pt); what drives it is whether the JTA population includes ryokan and simple lodgings.
- — By city/ward (September 2026, OTA-inventory basis): Hakata Ward 91.8%, Naha City 87.8%, Yokohama Nishi Ward 85.9%, Osaka Kita Ward 84.2%. JTA city-level values for July 2026: Naha 78.0%, Osaka City 75.7%, Yokohama City 69.7%, Fukuoka City 82.8%.
- — Which to use: JTA figures for levels and business-plan assumptions, REIT-disclosed figures for operator/brand results, and OTA-inventory estimates for relative comparisons and changes in the latest month.
Occupancy in Okinawa, Osaka, Kanagawa and Fukuoka — Quick Reference for the Latest Months
First, the direct answer. On MetroEngines Research’s OTA-listed-inventory estimate, September 2026 occupancy was 86.7% in Okinawa, 80.4% in Osaka, 84.2% in Kanagawa and 88.3% in Fukuoka. August was 85.5%, 80.8%, 84.3% and 89.6% respectively: Okinawa rose 1.2 points in September and Fukuoka fell 1.3 points, while Osaka and Kanagawa were roughly flat. On the JTA room occupancy rate, the latest prefecture-level month, July 2026 (second preliminary), shows Okinawa 61.8%, Osaka 72.6%, Kanagawa 66.4% and Fukuoka 72.4% — 9 to 28 points below our estimates for the same July. That is a difference of definition, not of demand, and the next section breaks it down.
| Area | OTA-inventory estimate Sep 2026 | OTA-inventory estimate Aug 2026 | JTA room occupancy Jul 2026 (2nd prelim.) | JTA room occupancy Sep 2025 (final) | Our coverage (Sep) |
|---|---|---|---|---|---|
| Okinawa | 86.7% | 85.5% | 61.8% | 62.6% | 922 properties / 37,105 rooms |
| Osaka | 80.4% | 80.8% | 72.6% | 83.4% | 436 properties / 60,604 rooms |
| Kanagawa | 84.2% | 84.3% | 66.4% | 61.7% | 480 properties / 31,978 rooms |
| Fukuoka | 88.3% | 89.6% | 72.4% | 69.9% | 428 properties / 29,820 rooms |
| Japan (reference) | 85.4% | — | 61.0% | 62.8% | 15,460 properties / 807,290 rooms |
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated; based on listings on major booking sites; lead time within 45 days; national figure is the room-weighted average of 47 prefectures) / Japan Tourism Agency, Overnight Travel Statistics Survey, time-series tables (July 2026 second preliminary; September 2025 final). For August 2026 the JTA has published only the first preliminary national figure (64.4%); prefecture-level values are not yet released.
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated) / Japan Tourism Agency, Overnight Travel Statistics Survey, July 2026 second preliminary
Comparing with the JTA’s September 2025 final figures also shows how each area moved over the year. Osaka recorded 83.4% in September 2025, among the highest in Japan, during Expo 2025 Osaka, Kansai (April 13 – October 13, 2025). July 2026 came in at 72.6%; although these are not the same month, it reads as a level where the Expo-period boost has run its course. Kanagawa moved from 61.7% in September 2025 to 66.4% in July 2026, and Fukuoka from 69.9% to 72.4%, so on the JTA basis both are running above the prior-year level. Okinawa was broadly flat on the JTA basis, at 62.6% in September 2025 and 61.8% in July 2026.
“Occupancy” Has Three Definitions — When to Use JTA, REIT-Disclosed and OTA-Inventory Figures
Even for the same Okinawa, the JTA room occupancy rate is 61.8% (July 2026) while our OTA-listed-inventory estimate is 89.3% (same month) — a 27.5-point gap. This is not about which number is more accurate; the two simply put different things in the numerator and denominator. The table below first sets out the three definitions that readers most often confuse when citing “occupancy.”
| Item | JTA room occupancy rate | REIT-disclosed occupancy | OTA-listed inventory estimate (ours) |
|---|---|---|---|
| Population | All property types including ryokan, resort, business and city hotels, simple lodgings and company/group lodgings (estimated from a sample survey) | Only hotels owned by each REIT (tens to around a hundred properties) | Properties listing at least 30% of total rooms on major booking sites (15,460 properties / 807,290 rooms nationwide in September 2026) |
| Numerator / denominator | Rooms used ÷ property’s total rooms | Rooms sold ÷ rooms available for sale (each REIT’s own definition) | (Total rooms − remaining OTA rooms) ÷ total rooms. Rooms not listed are counted as sold |
| Level | Close to actual occupancy. Roughly 55–65% nationwide | 80s to 90s % for urban portfolios | Runs higher than actual occupancy. Mid-80s % nationwide |
| Granularity / timeliness | Monthly. First preliminary (national only) at the end of the following month, second preliminary (by prefecture and major municipalities) at the end of the month after that, final figures the next year | Monthly. Disclosed in the latter part of the following month. Some REITs break down by region or property | Daily. Observable before the stay date and can be tallied as soon as the month ends. By prefecture, city/ward and segment |
| Best suited for | Annual budgets and business plans, occupancy assumptions for investment decisions, verifying year-on-year changes | Checking results by operator or brand, comparing alongside ADR and RevPAR | Supply-demand in the latest month, comparing days of the week and event dates, relative comparisons across areas and segments |
Source: Japan Tourism Agency, Overnight Travel Statistics Survey guidelines / monthly operating results of each REIT / compiled by HotelBank Editorial Team from MetroEngines Research data
To give one sense of REIT-disclosed levels: Japan Hotel REIT Investment Corporation (ジャパン・ホテル・リート投資法人, 8985) reported August 2026 occupancy of 84.1% for its 29 hotels with variable rent etc., with 93.5% in Osaka and 82.5% in Okinawa by region. Invincible Investment Corporation (インヴィンシブル投資法人, 8963) reported 87.3% for its 101 domestic hotels in the same month, and Ichigo Hotel REIT Investment Corporation (いちごホテルリート投資法人, 3463) 84.9% for its full portfolio. All run roughly 20–23 points above the JTA national figure (August first preliminary 64.4%), because their populations are skewed toward urban, limited-service hotels. Given that the JTA figure is “the average of all properties, including ryokan in the 40% range and simple lodgings in the 30% range,” the gap is a natural one.
The rule of thumb is simple. Use JTA final figures for occupancy assumptions in annual plans and investment decisions, REIT-disclosed figures to look at operator brand results, and OTA-listed inventory to read supply and demand for the latest or next month. When mixing figures with different definitions in one table, it is essential to state the definition in each column header — every table in this article follows that principle. The procedure for using the gap between REIT-disclosed occupancy and the OTA-inventory estimate to anticipate month-end results is set out in REIT vs OTA Occupancy: How to Use the Gap.
Comparing the Same Month, July 2026, the Gap Is 9–28 Points — Definitional Differences in Four Prefectures
Let us compare the four prefectures within the same month, July 2026, the latest month for which JTA prefecture-level values are available. Okinawa was 61.8% on the JTA basis versus 89.3% on ours, a 27.5-point gap; Osaka 72.6% versus 81.7% (9.1 points); Kanagawa 66.4% versus 82.3% (15.9 points); and Fukuoka 72.4% versus 88.4% (16.0 points). The gap is widest in Okinawa because the JTA population contains many resort hotels (69.5% that month), ryokan (42.0%) and simple lodgings (32.8%), while our coverage skews toward properties with high OTA listing ratios. Conversely, Osaka’s gap is a small 9 points likely because, even on the JTA basis, the spread across property types is narrow — business hotels 76.0%, city hotels 73.0%, simple lodgings 57.2% — and limited-service hotels with high OTA listing ratios make up most of the market.
| Area (Jul 2026) | JTA all types | JTA business hotels | JTA city hotels | JTA resort hotels | OTA-inventory estimate | Gap (estimate − JTA all types) |
|---|---|---|---|---|---|---|
| Okinawa | 61.8% | 79.5% | 72.3% | 69.5% | 89.3% | +27.5pt |
| Osaka | 72.6% | 76.0% | 73.0% | 67.3% | 81.7% | +9.1pt |
| Kanagawa | 66.4% | 81.2% | 68.6% | 66.1% | 82.3% | +15.9pt |
| Fukuoka | 72.4% | 79.0% | 73.4% | 62.8% | 88.4% | +16.0pt |
| Japan | 61.0% | 74.6% | 71.9% | 59.0% | — | — |
Source: Japan Tourism Agency, Overnight Travel Statistics Survey, July 2026 second preliminary (room occupancy by prefecture and property type) / compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated; July 2026, 31-day average)
Given this structure, the right way to read OTA-listed-inventory values is for “relative comparison” and “change,” not “level.” For example, Okinawa’s 86.7% in September cannot itself substitute for the JTA figure, but the facts that it rose from 85.5% in August, sits close to Fukuoka’s 88.3% and above Osaka’s 80.4% are meaningful as comparisons within one definition. Preliminary figures also get revised — the JTA national figure for July 2026 was revised from 60.8% in the first preliminary to 61.0% in the second. Revision patterns are covered in detail in our guide to reading the Overnight Travel Statistics Survey (for a verification of the revision magnitude itself, see How Much Are Preliminary Lodging Statistics Revised?).
Occupancy for All 47 Prefectures — JTA July 2026 × OTA-Inventory September 2026
So that readers can look up any prefecture by name, we list values for all 47. JTA room occupancy is shown for July 2026 (second preliminary) and September 2025 (final), and our OTA-listed-inventory estimate for September 2026. Because the columns cover different months, please read across with the understanding that “both the definition and the month differ.” On the OTA-inventory basis for September, the highest was Aomori at 95.5%, followed by Aichi 90.4%, Hokkaido 90.3% and Tokyo 89.8%, and the lowest was Kagoshima at 71.4%. For JTA July, Tokyo 75.3%, Osaka 72.6% and Fukuoka 72.4% led, with the national figure at 61.0%.
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated; September 2026; 47 prefectures, 15,460 properties / 807,290 rooms) / Japan Tourism Agency, Overnight Travel Statistics Survey, July 2026 second preliminary
Open the 47-prefecture data table (JTA Jul 2026 and Sep 2025 / OTA-inventory Sep 2026)
| Prefecture | JTA Jul 2026 (2nd prelim.) | JTA Sep 2025 (final) | OTA-inventory estimate Sep 2026 | Our coverage (Sep) |
|---|---|---|---|---|
| Hokkaido | 69.4% | 67.4% | 90.3% | 793 properties / 54,434 rooms |
| Aomori | 63.4% | 69.4% | 95.5% | 105 properties / 7,378 rooms |
| Iwate | 58.0% | 57.9% | 85.8% | 186 properties / 9,717 rooms |
| Miyagi | 63.8% | 63.5% | 87.9% | 210 properties / 13,896 rooms |
| Akita | 52.5% | 54.7% | 86.4% | 104 properties / 4,792 rooms |
| Yamagata | 50.0% | 53.2% | 87.4% | 187 properties / 6,462 rooms |
| Fukushima | 51.3% | 52.1% | 85.4% | 270 properties / 12,571 rooms |
| Ibaraki | 54.6% | 55.6% | 78.3% | 201 properties / 9,784 rooms |
| Tochigi | 54.7% | 56.9% | 81.2% | 409 properties / 14,893 rooms |
| Gunma | 48.4% | 47.8% | 81.4% | 357 properties / 14,059 rooms |
| Saitama | 64.7% | 65.3% | 84.7% | 131 properties / 7,966 rooms |
| Chiba | 63.1% | 60.2% | 83.5% | 478 properties / 24,629 rooms |
| Tokyo | 75.3% | 76.5% | 89.8% | 861 properties / 105,602 rooms |
| Kanagawa | 66.4% | 61.7% | 84.2% | 480 properties / 31,978 rooms |
| Niigata | 46.8% | 48.7% | 85.9% | 285 properties / 13,519 rooms |
| Toyama | 61.9% | 61.3% | 83.5% | 117 properties / 7,709 rooms |
| Ishikawa | 58.7% | 58.5% | 84.2% | 258 properties / 14,879 rooms |
| Fukui | 52.5% | 46.3% | 84.5% | 181 properties / 5,519 rooms |
| Yamanashi | 47.8% | 46.5% | 82.2% | 405 properties / 10,192 rooms |
| Nagano | 43.3% | 42.8% | 88.4% | 718 properties / 18,876 rooms |
| Gifu | 53.4% | 56.9% | 88.1% | 302 properties / 11,179 rooms |
| Shizuoka | 57.3% | 55.0% | 83.0% | 885 properties / 32,395 rooms |
| Aichi | 62.9% | 70.0% | 90.4% | 368 properties / 34,587 rooms |
| Mie | 52.0% | 53.8% | 83.1% | 284 properties / 14,149 rooms |
| Shiga | 57.1% | 56.9% | 85.7% | 164 properties / 7,220 rooms |
| Kyoto | 63.3% | 70.1% | 84.0% | 785 properties / 33,397 rooms |
| Osaka | 72.6% | 83.4% | 80.4% | 436 properties / 60,604 rooms |
| Hyogo | 61.0% | 61.1% | 86.3% | 563 properties / 20,179 rooms |
| Nara | 50.6% | 50.6% | 82.5% | 139 properties / 5,083 rooms |
| Wakayama | 53.3% | 48.2% | 79.0% | 243 properties / 8,222 rooms |
| Tottori | 47.3% | 52.9% | 85.5% | 125 properties / 5,371 rooms |
| Shimane | 52.4% | 56.2% | 85.2% | 199 properties / 6,484 rooms |
| Okayama | 56.0% | 56.4% | 85.1% | 173 properties / 8,682 rooms |
| Hiroshima | 61.3% | 66.5% | 88.5% | 274 properties / 15,375 rooms |
| Yamaguchi | 54.8% | 55.9% | 82.0% | 132 properties / 6,652 rooms |
| Tokushima | 50.1% | 50.4% | 82.4% | 92 properties / 3,709 rooms |
| Kagawa | 55.9% | 58.0% | 80.1% | 198 properties / 5,798 rooms |
| Ehime | 59.9% | 56.6% | 82.8% | 179 properties / 8,629 rooms |
| Kochi | 56.6% | 52.2% | 86.9% | 117 properties / 5,127 rooms |
| Fukuoka | 72.4% | 69.9% | 88.3% | 428 properties / 29,820 rooms |
| Saga | 55.1% | 50.5% | 80.3% | 135 properties / 4,137 rooms |
| Nagasaki | 52.4% | 52.5% | 77.2% | 249 properties / 11,398 rooms |
| Kumamoto | 49.1% | 59.3% | 88.0% | 361 properties / 13,957 rooms |
| Oita | 46.4% | 50.0% | 78.8% | 454 properties / 10,841 rooms |
| Miyazaki | 54.8% | 48.8% | 78.4% | 149 properties / 6,550 rooms |
| Kagoshima | 49.5% | 49.6% | 71.4% | 368 properties / 11,785 rooms |
| Okinawa | 61.8% | 62.6% | 86.7% | 922 properties / 37,105 rooms |
Source: Japan Tourism Agency, Overnight Travel Statistics Survey, time-series tables (4-1 and former 4-2) / compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated; based on listings on major booking sites). Property and room counts are the mode within the month.
High values in regional prefectures need careful reading. Aomori’s 95.5% comes from a small base of 105 properties / 7,378 rooms in a month when holidays and leisure demand absorbed almost all listed inventory; on the JTA basis, Aomori was 63.4% in July 2026 and 69.4% in the September 2025 final figures, only slightly above the national level (61.0% and 62.8%). The smaller the base, the more the OTA-inventory value swings, so it is premature to conclude that supply is tight from a high level alone. By contrast, prefectures with large room counts — Tokyo (861 properties / 105,602 rooms), Osaka (436 properties / 60,604 rooms) and Hokkaido (793 properties / 54,434 rooms) — average out day-to-day fluctuations, so their monthly values are relatively more reliable.
Naha, Hakata, Five Osaka Wards and Two Yokohama Wards — Occupancy by City and Ward
When searching by city name — “Yokohama occupancy,” “Osaka occupancy” — one wants values a level finer than the prefecture. Tallying our OTA-listed-inventory estimate by city and ward, September 2026 came in at Naha City 87.8% (181 properties / 15,455 rooms), Fukuoka City Hakata Ward 91.8% (151 properties / 13,492 rooms), Osaka City’s Kita Ward 84.2%, Chuo Ward 82.9%, Yodogawa Ward 80.0%, Naniwa Ward 77.1% and Tennoji Ward 77.0%, and Yokohama City’s Nishi Ward 85.9% and Naka Ward 80.2%. The JTA also publishes room occupancy for major municipalities in its second preliminary release; for July 2026 these were Naha 78.0%, Osaka City 75.7%, Yokohama City 69.7% and Fukuoka City 82.8%. City values exceed their prefecture values (Okinawa 61.8%, Osaka 72.6%, Kanagawa 66.4%, Fukuoka 72.4%) by 3–16 points because the ryokan, simple lodgings and suburban properties in each prefecture drop out and the population shifts toward urban limited-service hotels — here too, the difference in population matters more than the definition.
| City / ward | OTA-inventory estimate Sep 2026 | Same, Aug 2026 | Same, Jul 2026 | JTA by municipality Jul 2026 (2nd prelim.) | Our coverage (Sep) |
|---|---|---|---|---|---|
| Naha City (Okinawa) | 87.8% | 86.0% | 90.2% | 78.0% (Naha City) | 181 properties / 15,455 rooms |
| Fukuoka City, Hakata Ward | 91.8% | 93.0% | 93.1% | 82.8% (Fukuoka City) | 151 properties / 13,492 rooms |
| Osaka City, Kita Ward | 84.2% | 84.1% | 85.2% | 75.7% (Osaka City) | 39 properties / 8,102 rooms |
| Osaka City, Chuo Ward | 82.9% | 82.2% | 83.8% | 135 properties / 19,731 rooms | |
| Osaka City, Yodogawa Ward | 80.0% | 79.5% | 76.7% | 17 properties / 2,795 rooms | |
| Osaka City, Naniwa Ward | 77.1% | 79.7% | 79.9% | 45 properties / 7,837 rooms | |
| Osaka City, Tennoji Ward | 77.0% | 73.8% | 74.9% | 17 properties / 1,467 rooms | |
| Yokohama City, Nishi Ward | 85.9% | 86.3% | 87.3% | 69.7% (Yokohama City) | 11 properties / 2,878 rooms |
| Yokohama City, Naka Ward | 80.2% | 80.7% | 78.0% | 50 properties / 9,972 rooms |
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated; based on listings on major booking sites. Yokohama Nishi Ward is the average of 18 observed days in August 2026 and 26 days in July; all others cover every day of the month) / Japan Tourism Agency, Overnight Travel Statistics Survey, July 2026 second preliminary, Reference Table 12 (room occupancy by major municipality). Yokohama’s Kohoku and Kanagawa wards are not shown because the number of covered properties falls below our threshold.
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated; September 2026)
By ward, even within Osaka City there is a 7-point spread between Kita Ward (84.2%) and Tennoji Ward (77.0%). Kita Ward, centered on Umeda, has many large hotels (39 properties / 8,102 rooms) and captures both business and leisure demand, while Tennoji Ward has a small base (17 properties / 1,467 rooms), so daily swings show up in the monthly average. Naniwa Ward (around Namba and Shin-Imamiya) eased for three straight months — 79.9% in July, 79.7% in August, 77.1% in September — suggesting that supply built up during the Expo period has entered an absorption phase. In Yokohama, Nishi Ward (Yokohama Station and Minato Mirai) was 85.9% versus 80.2% for Naka Ward (Kannai, Yamashita-cho and Chinatown), both above the JTA figure of 69.7% for Yokohama City as a whole. Naha moved from 90.2% in July to 86.0% in August and 87.8% in September, tracking almost the same level as Okinawa Prefecture overall (86.7% in September).
Luxury Hotel Occupancy — the ¥100,000+ Listed-Price Band and REIT-Disclosed Figures
For the question “what is luxury hotel occupancy?”, JTA statistics have no price-band breakdown, so the answer has to come from two proxy indicators. The first is our OTA-listed-inventory estimate narrowed to properties with listed prices of ¥100,000 or more per night (2 guests per room, tax included). In September 2026 it was 83.9% in Okinawa (34 properties / 197 rooms), 89.1% in Kanagawa (23 properties / 441 rooms) and 83.7% in Kyoto (22 properties / 565 rooms), versus 89.3%, 85.6% and 83.2% respectively in August. For Osaka, Fukuoka and Tokyo, observations in this price band were not complete across the month, so no value is shown. With only a few hundred rooms the figures are volatile, but Okinawa’s high-price band fell just over 5 points from 89.3% in August, while the prefecture overall rose from 85.5% to 86.7% — indicating that demand at the upper end of resorts settled first after the summer holidays.
The second is REIT-disclosed figures. Japan Hotel REIT’s Okinawa region in August 2026 (its group of upper-tier resort hotels) posted occupancy of 82.5%, ADR of ¥47,112 and RevPAR of ¥38,853 — an ADR about twice that of the REIT’s 29-hotel total (occupancy 84.1%, ADR ¥23,337). Fusaki Beach Resort Hotel & Villas (フサキビーチリゾートホテル＆ヴィラズ) on Ishigaki Island, disclosed by Invincible Investment Corporation, recorded occupancy of 84.8% and ADR of ¥86,729 in the same month, and Art Hotel Ishigakijima (アートホテル石垣島) 87.4% with ADR of ¥27,544. Hoshino Resorts REIT (星野リゾート・リート投資法人, 3287) posted portfolio-wide occupancy of 81.4% and ADR of ¥25,004 in August, down 1.9 points in occupancy and 4.6% in ADR year on year. The going rate for high-end resorts in August is the low-to-mid 80s %, a few points below business-hotel-centric portfolios (87.3% for Invincible’s 101 domestic hotels), but with ADR 1.4 to 4.8 times higher.
| Indicator | Coverage | Aug 2026 | Sep 2026 | Notes |
|---|---|---|---|---|
| OTA-inventory estimate (listed price ¥100,000+) | Okinawa | 89.3% | 83.9% | 34 properties / 197 rooms |
| Same as above | Kanagawa | 85.6% | 89.1% | 23 properties / 441 rooms |
| Same as above | Kyoto | 83.2% | 83.7% | 22 properties / 565 rooms. August covers 25 observed days |
| REIT-disclosed | Japan Hotel REIT, Okinawa region | 82.5% | Not yet disclosed | ADR ¥47,112, RevPAR ¥38,853 |
| REIT-disclosed | Invincible: Fusaki Beach Resort Hotel & Villas (Ishigaki Island) | 84.8% | Not yet disclosed | ADR ¥86,729, RevPAR ¥73,563 |
| REIT-disclosed | Hoshino Resorts REIT, full portfolio | 81.4% | Not yet disclosed | ADR ¥25,004; occupancy −1.9pt YoY |
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (OTA-listed inventory basis, estimated; listed prices for 2 guests per room, tax included) / compiled by HotelBank Editorial Team from the August 2026 monthly operating results of Japan Hotel REIT Investment Corporation, Invincible Investment Corporation and Hoshino Resorts REIT
RevPAR Benchmarks from Occupancy × Estimated Settled ADR
Occupancy alone does not show the revenue picture. Multiplying by the estimated settled ADR for the same September 2026 (area median, tax-excluded equivalent) gives RevPAR benchmarks of ¥10,400 for Okinawa (86.7% × ADR ¥12,000), ¥7,500 for Osaka (80.4% × ¥9,300), ¥11,900 for Kanagawa (84.2% × ¥14,200) and ¥10,200 for Fukuoka (88.3% × ¥11,600). Note that because OTA-listed-inventory occupancy runs higher than actual occupancy, these benchmarks also sit toward the upper bound. Multiplying the JTA July 2026 room occupancy rate by the same month’s ADR instead gives Okinawa ¥8,500 (61.8% × ¥13,700), Osaka ¥6,400 (72.6% × ¥8,800), Kanagawa ¥8,500 (66.4% × ¥12,800) and Fukuoka ¥7,900 (72.4% × ¥10,900) — 15–29% lower. The same rule applies: use the latter for business plans and the former for recent relative comparisons.
| Area | OTA-inventory occupancy Sep 2026 | Estimated settled ADR Sep 2026 | RevPAR benchmark (Sep, OTA-inventory basis) | JTA occupancy Jul 2026 | Estimated settled ADR Jul 2026 | RevPAR benchmark (Jul, JTA basis) |
|---|---|---|---|---|---|---|
| Okinawa | 86.7% | ¥12,040 N=515 | ¥10,439 | 61.8% | ¥13,722 N=508 | ¥8,480 |
| Osaka | 80.4% | ¥9,310 N=655 | ¥7,485 | 72.6% | ¥8,761 N=647 | ¥6,360 |
| Kanagawa | 84.2% | ¥14,154 N=579 | ¥11,918 | 66.4% | ¥12,784 N=576 | ¥8,489 |
| Fukuoka | 88.3% | ¥11,594 N=477 | ¥10,238 | 72.4% | ¥10,873 N=473 | ¥7,872 |
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (occupancy on OTA-listed inventory basis, estimated; ADR is estimated settled ADR, area median, tax-excluded equivalent; N is the number of properties covered by the estimated settled ADR) / Japan Tourism Agency, Overnight Travel Statistics Survey, July 2026 second preliminary. RevPAR benchmarks are the simple product of occupancy × ADR and are estimates calculated for this article.
Source: Compiled by HotelBank Editorial Team from MetroEngines Research data (September 2026, OTA-inventory occupancy × estimated settled ADR)
The same calculation works at the city and ward level. Estimated settled ADR for September 2026 was ¥9,700 for Naha City (N=144), ¥14,200 for Hakata Ward (N=151), ¥10,400 for Osaka Kita Ward (N=74), ¥10,600 for Chuo Ward (N=193), ¥23,300 for Yokohama Nishi Ward (N=17) and ¥12,400 for Naka Ward (N=51); multiplying by occupancy gives RevPAR benchmarks of ¥8,500 for Naha, ¥13,100 for Hakata Ward, ¥8,700 for Osaka Kita Ward, ¥8,800 for Chuo Ward, ¥20,100 for Yokohama Nishi Ward and ¥10,000 for Naka Ward. Hakata Ward combines high occupancy of 91.8% with an ADR more than 20% above the prefecture median, giving it the thickest RevPAR among the four cities’ central districts, excluding Yokohama Nishi Ward with its small coverage. In Yokohama Nishi Ward, large hotels in Minato Mirai push up ADR; while the small coverage of 17 properties calls for a discount, it outperforms Kanagawa Prefecture overall on both occupancy and rate.
How to Look Up the Occupancy of Any City Yourself
For cities not covered here, or when you need values for a more recent month, here is the procedure. Because the right source depends on the purpose, the shortcut is to decide first “what the occupancy figure is for.”
| Step | What to do | Where to look |
|---|---|---|
| 1 | Check the official prefecture value. Sheet 4-1 of the JTA’s time-series xlsx lists monthly room occupancy by prefecture and property type. The latest month is the second preliminary; earlier years are final. | JTA Overnight Travel Statistics Survey / Guide to reading the statistics |
| 2 | Check the official city value. Reference Table 12 of the second preliminary release gives room occupancy for major municipalities such as Naha, Osaka, Yokohama, Fukuoka and Kyoto. There are no official ward-level values. | JTA monthly statistical tables (second preliminary) xlsx |
| 3 | If you need the latest month, ward-level or segment-level data, look at the OTA-listed-inventory estimate. Nationally, the gap with the JTA figure was 22.7 points in July 2026, and the reasons break down into four factors. | July 2026: 60.8% official vs 83.5% OTA-inventory, decomposed |
| 4 | To gauge room for growth in supply and demand, compare occupancy with “guest-nights per room” across the 47 prefectures. | June 2026 occupancy of 56.2% and guest-nights per room |
| 5 | For results by operator brand or comparisons alongside ADR and RevPAR, use REIT monthly operating results. Some REITs provide regional breakdowns. | REIT August results: Osaka at 93.5% occupancy with ADR −29.8% |
JTA values from Steps 1 and 2 can be cited as “levels,” while the OTA-listed-inventory values from Step 3 are for “relative comparison and change.” When placing the two side by side in the same sentence, stating the definition and target month in each column header, as in this article’s tables, keeps readers from confusing them.
Summary
Estimated occupancy on an OTA-listed-inventory basis for September 2026 was 86.7% in Okinawa, 80.4% in Osaka, 84.2% in Kanagawa and 88.3% in Fukuoka, with a national room-weighted average of 85.4%. At the city/ward level, Hakata Ward (91.8%), Naha City (87.8%), Yokohama Nishi Ward (85.9%) and Osaka Kita Ward (84.2%) led, while Osaka’s Naniwa and Tennoji wards stayed in the 77% range. On the JTA room occupancy rate, the latest prefecture-level month is July 2026 (second preliminary), at Okinawa 61.8%, Osaka 72.6%, Kanagawa 66.4% and Fukuoka 72.4%, with city values of Naha 78.0%, Osaka City 75.7%, Yokohama City 69.7% and Fukuoka City 82.8%. The gap between the two reflects differences in definition and population; read JTA figures for levels, OTA-inventory figures for relative comparison and change, and REIT-disclosed figures for brand-level results, according to purpose. For the luxury tier, the ¥100,000+ listed-price band shows Okinawa 83.9% and Kanagawa 89.1%, and REIT-disclosed upper-tier resorts at 82–85% — the going rate from summer to early autumn 2026. JTA prefecture-level values for August are expected in the second preliminary release at the end of October, and September values at the end of November; once published, the tables in this article can be updated in the same columns and read again.
Note on the data: Occupancy in this article (OTA-listed inventory basis, estimated) is an estimate based on how inventory sold on OTAs is absorbed and differs from a property’s actual overall occupancy. Because rooms not listed on OTAs are treated as sold, it runs higher than actual occupancy. All values are monthly, not full-year occupancy. JTA figures for July 2026 are second preliminary and for August 2026 first preliminary, and may be revised in the final figures. JTA prefecture-level and municipality-level values for August 2026 onward had not been published at the time of writing. ADR is an estimate (median error about 7% against REIT-disclosed results), and RevPAR is a benchmark computed as the simple product of occupancy × ADR.
Frequently Asked Questions
Q. What was hotel occupancy in Okinawa in September 2026?
A. 86.7% on MetroEngines Research’s OTA-listed-inventory estimate (922 properties / 37,105 rooms, monthly value). The JTA room occupancy rate for the latest prefecture-level month, July 2026, was 61.8%, and the September 2025 final figure was 62.6%. Because the definitions differ, the two cannot be compared directly.
Q. Where can I find occupancy for Osaka and Yokohama?
A. Room occupancy for Osaka City and Yokohama City appears in Reference Table 12 of the JTA’s second preliminary release (July 2026: Osaka City 75.7%, Yokohama City 69.7%). Since there are no official ward-level values, this article shows OTA-listed-inventory estimates for five Osaka wards and two Yokohama wards (September 2026, e.g., Osaka Kita Ward 84.2% and Yokohama Nishi Ward 85.9%).
Q. How can I find luxury hotel occupancy?
A. JTA statistics have no price-band breakdown. The practical options are the OTA-listed-inventory estimate narrowed to properties listed at ¥100,000 or more (September 2026: Okinawa 83.9%, Kanagawa 89.1%) or the monthly occupancy of upper-tier resorts disclosed by listed hotel REITs (August 2026: Japan Hotel REIT Okinawa region 82.5%, etc.).
Q. Why does OTA-inventory occupancy run higher than the JTA figure?
A. The main reasons are that rooms not listed on OTAs are counted as sold, that coverage skews toward properties with high OTA listing ratios, and that the JTA population includes lower-occupancy property types such as ryokan and simple lodgings. It is best used for relative comparisons within the same definition and for month-to-month changes, not for levels.
Related Articles
- Japan Hotel Occupancy July 2026: 60.8% Official vs 83.5% OTA Estimate
- Overnight Travel Statistics Survey Guide: 61.6% OCC, Prelim vs Final (occupancy is the final figure for January–December 2025)
- June 2026 Occupancy 56.2%: Guest-Nights per Room in 47 Prefectures
- JHR Aug 2026 by Region: Osaka ADR -29.8% at 93.5% Occ, Okinawa -9.9%
- REIT vs OTA Occupancy: How to Use a Gap of Up to 10.1pt (July 2026)
- Japan Lodging Stats: Preliminary Figures Move 3.4%, Occupancy 1.0pt
- Okinawa 176 Resorts: Obon vs Early Sep, 3.0pt Gap at 30 Days Out
- Silver Week 2026: Sep 20 Filled First, Final Night Sep 23 at 78.5% OCC (stays on September 23, 2026)
References and Sources
■ Data sources
MetroEngines Research (estimated occupancy on an OTA-listed-inventory basis and estimated settled ADR, July–September 2026); Japan Tourism Agency, Overnight Travel Statistics Survey (time-series tables; July 2026 second preliminary; August 2026 first preliminary; 2025 final figures); August 2026 monthly operating results of Japan Hotel REIT Investment Corporation, Invincible Investment Corporation, Ichigo Hotel REIT Investment Corporation and Hoshino Resorts REIT.
■ Estimation assumptions
Occupancy (OTA-inventory basis) is a monthly average calculated from the remaining-inventory trajectory from 45 days before the stay date to the stay date, for properties listing at least 30% of total rooms on major booking sites. Prefecture and national values are room-weighted. Osaka City and Yokohama City were tallied by ward because no city-level aggregation is available. Estimated settled ADR is the area median (tax-excluded equivalent). RevPAR benchmarks are the simple product of occupancy × ADR and are not actual results.
■ Limitations
The OTA-inventory basis counts rooms not listed on OTAs as sold, so it runs higher than the JTA room occupancy rate. JTA figures for July 2026 are second preliminary and for August first preliminary, and may be revised in the final figures. Prefectures and wards with few covered properties (Aomori, Osaka Tennoji Ward, Yokohama Nishi Ward, etc.) and the ¥100,000+ listed-price band (a few hundred rooms) fluctuate widely from month to month. REIT-disclosed figures are based on each REIT’s definitions and holdings.
- Japan Tourism Agency press release, “Overnight Travel Statistics Survey (August 2026 first preliminary, July 2026 second preliminary)” (September 30, 2026)
- Japan Tourism Agency, Overnight Travel Statistics Survey statistics page — time-series tables (4-1 room occupancy by prefecture and property type; former 4-2 by month), July 2026 second preliminary statistical tables (Reference Table 12, room occupancy by major municipality), August 2026 first preliminary statistical tables (Table 5)
- Japan Hotel REIT Investment Corporation, “Monthly Operating Results, August 2026” (29 hotels with variable rent etc., by region)
- Invincible Investment Corporation, “Monthly Operating Results, August 2026” (September 25, 2026)
- Ichigo Hotel REIT Investment Corporation, “Monthly Operating Results, August 2026” (September 25, 2026)
- Hoshino Resorts REIT IR Library, “Monthly Operating Status, August 2026”
- MetroEngines Research — occupancy (OTA-listed inventory basis, estimated; July–September 2026; 47 prefectures and by city/ward; based on listings on major booking sites), estimated settled ADR (July and September 2026, area median)