The estimated settled ADR of city hotels in Tokyo (N=109 properties) fell to ¥21,265 in August 2026, down 20.6% from ¥26,769 (N=107) a year earlier. Business hotels (N=915 properties) also slipped to ¥11,554, 8.9% below last year’s ¥12,689 (N=900), making this a summer in which both categories fell below the prior year on a finalized-month basis. Adding the just-finalized September, however, the picture splits: business hotels recovered to ¥13,364 (+2.2% YoY), while city hotels stood at ¥24,650 (−10.3%) and remain below last year. This article overlays three years of finalized monthly values since 2024, examines where the summer decline came from and how far it has recovered, and lays out—from an operator’s perspective—how to reflect this in October–December pricing.

Scope: Tokyo city hotels N=109 properties and business hotels N=915 properties (properties included in the estimated settled ADR calculation as of August 2026). The price metric in this article is estimated settled ADR (the settled price level estimated from OTA and other sales data, tax-excluded equivalent); occupancy is an estimate based on OTA-listed inventory. Definitions are provided at the end of the article. Data as of: October 18, 2026.

Above last year in January–April, then a reversal into summer — the decline is concentrated in the summer

Lining up the nine finalized months of 2026 against the same months a year earlier shows that the decline was not year-round but concentrated in the summer. City hotels posted ¥26,419 in January (+8.3% YoY), ¥30,282 in March (+5.8%) and ¥31,241 in April (+0.5%), staying at or above last year’s level through spring. Once May dipped below at −3.8%, however, the decline widened month after month—−12.8% in June, −17.1% in July and −20.6% in August—and September, at −10.3%, still fell short of last year.

Business hotels, on the other hand, followed a different shape. They stayed above last year by +0.4% to +5.7% from January to May, and although June–August turned negative at −4.7%, −5.3% and −8.9%, the declines were less than half those of city hotels. Then in September, at ¥13,364, they exceeded last year’s ¥13,072 by 2.2%, exiting the summer decline within a single month. In sum, Tokyo’s summer rate adjustment followed the pattern of “common to both categories, but deeper and slower to recover for city hotels.” Neighboring Kanagawa, based on finalized August values, saw declines of only −3.3% for city hotels and −4.8% for business hotels, so the depth of the drop among Tokyo’s city hotels stands out even within the Greater Tokyo area.

Source: Compiled by HotelBank Editorial Team from MetroEngines Research. All values are for finalized (past) months; N is the number of properties included in each month’s calculation.

Source: Compiled by HotelBank Editorial Team from MetroEngines Research

Chart 1 overlays city hotels’ estimated settled ADR for 2024–2026. While the 2025 line ran above 2024 in every month, the 2026 line peaked at ¥31,241 in April and then fell, in June–August, below even the 2024 line. The dotted line from October onward is an estimate based on current sales and should be viewed as a different yardstick from finalized values (discussed below).

August fell below even 2024 — a “payback from a high prior year” alone cannot explain it

The next question is whether the 2026 summer decline can be dismissed as a “payback from an overly high summer 2025.” As Table 2 shows, city hotels’ August rose +13.0% from ¥23,695 in 2024 to ¥26,769 in 2025, and in 2026 came in at ¥21,265—−20.6% YoY and −10.3% versus 2024. It not only gave back the prior year’s gain but fell further below the level of two years ago. Business hotels’ August likewise grew +6.4% from ¥11,924 in 2024 to 2025, and then in 2026, at ¥11,554, came in −3.1% versus 2024, slightly below two years earlier.

The picture does not change when looking at the three-month June–August average. City hotels moved from ¥23,332 in 2024 to ¥26,422 in 2025 and ¥21,965 in 2026, leaving summer 2026 −5.9% below two years earlier. Business hotels went ¥11,822 → ¥12,718 → ¥11,917, +0.8% versus 2024—roughly back to the level of two years ago. In other words, for Tokyo’s city hotels specifically, summer rates can be read as having fallen beyond the range of a mere “payback.”

Source: Compiled by HotelBank Editorial Team from MetroEngines Research. N for August of each year: City 107 / 107 / 109 properties; Business 908 / 900 / 915 properties.

Source: Compiled by HotelBank Editorial Team from MetroEngines Research

Chart 2 is the same year-over-year overlay for business hotels. The 2026 line set its high for the year at ¥17,802 in April, fell to ¥11,890 in June, bottomed at ¥11,554 in August and rebounded to ¥13,364 in September. The gap with the 2025 line opening in summer and closing in September matches the YoY trend in Table 1.

Behind this is a slowdown on the demand side. In the Japan Tourism Agency’s Overnight Travel Statistics Survey (July 2026, second preliminary figures), Tokyo’s total guest nights were 7,865,630, down 10.7% YoY, of which Japanese guests accounted for 3,511,180 (−14.8%) and foreign guests 4,354,450 (−7.0%)—both domestic and inbound demand fell below the prior year. Nationwide, total guest nights in August 2026 were 62.68 million (−6.7%) and foreign guest nights 12.64 million (−10.3%), while the Japan National Tourism Organization (JNTO) reported 3,098,900 international visitor arrivals in August, down 9.6% YoY. With foreign guests making up 55.4% of Tokyo’s July guest nights—more than half of lodging demand—the market is structured such that a slowdown in inbound travel feeds directly into rates. For the supply side, the other factor alongside demand, our breakdown of Tokyo’s 68 new hotels and 2,055 rooms opening in 2026 by category tracks new openings by category together with the turn to negative estimated settled ADR.

Occupancy stayed high while rates were conceded — estimated OCC alongside same-month ADR

So was the rate decline an adjustment made to protect occupancy? Table 3 lines up estimated OCC for June–September (OTA-listed inventory basis, monthly average) with same-month ADR YoY. Estimated OCC for city hotels (53 properties, 17,172 rooms) held high at 89.8% in June, 89.5% in July, 87.2% in August and 90.0% in September, and business hotels (534 properties, 78,154 rooms) likewise moved from 90.9% in June to 88.8% in August and 90.3% in September. Because estimated OCC is derived from how listed inventory is absorbed and tends to run higher than actual room occupancy, the levels themselves should be interpreted with some latitude—but at the very least, there is no sign that a large share of listed inventory went unsold over the summer.

What this suggests is that Tokyo’s summer followed the pattern of “rooms got filled, but rate was conceded to fill them.” High occupancy appears to have been maintained alongside rate adjustments of −20.6% YoY for city hotels and −8.9% for business hotels, and it can be said that most properties did not choose to hold prices and let occupancy fall. Note that because fewer than two years of estimated OCC observations have accumulated, no YoY comparison is made; here we only confirm levels within the same year.

Source: Compiled by HotelBank Editorial Team from MetroEngines Research. Estimated OCC is the monthly average of daily values for a fixed cohort of properties (the same set within each month) that list at least 30% of their total rooms on major booking sites. Cohort property and room counts are intra-month averages.

Published REIT results pointed in the same direction. As Table 4 shows, August 2026 ADR YoY was −8.8% for Japan Hotel REIT Investment Corporation (ジャパン・ホテル・リート投資法人), −3.2% for Invincible Investment Corporation (インヴィンシブル投資法人), −2.7% for Ichigo Hotel REIT Investment Corporation (いちごホテルリート投資法人) and −4.6% for Hoshino Resorts REIT (星野リゾート・リート投資法人)—all four below the prior year—while the YoY change in room occupancy stayed within −0.1 to −1.9pt. The combination of holding occupancy while rates fall was thus observed in nationally diversified portfolios as well. However, REIT figures are actual results for holdings spread across the country, and the −20.6% for Tokyo city hotels is far deeper. It is natural to see this as an adjustment specific to Tokyo’s upper-tier city hotels layered on top. Why REIT actuals and Tokyo’s estimated settled ADR diverge in sign and magnitude from month to month is explained in Hotel REIT July 2026 Results vs Tokyo Settled ADR: Why Signs Diverge.

Source: Monthly operating results published by each investment corporation (linked in table). Portfolios are located nationwide; these are not Tokyo figures.

YoY by month — business hotels broke through in September, city hotels did not

Source: Compiled by HotelBank Editorial Team from MetroEngines Research

Chart 3 plots the YoY changes from Table 1 as bars. Both categories sit on the positive side for January–April and the negative side for June–August, with only business hotels returning above the zero line in September. Looking at the ratio of city to business hotel rates, city hotels were 2.11x business hotels in August 2025 and 2.10x in September, versus 1.84x in both August and September 2026. Given that April was nearly level at 1.81x in 2025 and 1.75x in 2026, the erosion of city hotels’ premium from summer into autumn can be read as a defining feature of Tokyo in 2026.

Current estimates for October–December — treat them as a separate yardstick from finalized values

When thinking about autumn pricing, it is important first to recognize that values from October onward are different in nature. The October–December figures in Table 5 are all estimates based on current sales and will shift as sales progress. City hotels’ October, at ¥33,876, sits +37.4% above the finalized September value of ¥24,650, but there is a seam in the calculation basis between finalized values and current estimates, so this gap cannot be read as “rates rebounding sharply in October.” A simple comparison with the prior year’s finalized values (October 2025: City ¥31,970, Business ¥15,348) should likewise wait until the month is finalized.

Source: Compiled by HotelBank Editorial Team from MetroEngines Research. October–December are estimates based on current sales and may change once the month is finalized. YoY against prior-year finalized values is not calculated.

What remains useful even so is the seasonal shape of finalized values. For Tokyo’s city hotels, October came in well above August in both 2024 and 2025 (2024: ¥23,695 → ¥27,220, +14.9%; 2025: ¥26,769 → ¥31,970, +19.4%), and business hotels moved in the same direction (+11.9% in 2024, +21.0% in 2025). November 2025 was also the month in which city hotels set their high for the year at ¥32,756 and business hotels their second-half high at ¥16,540. Carrying the summer’s lowered levels straight into autumn risks voluntarily giving up the seasonal upside that is available.

Table 6 puts numbers on the yardsticks for setting autumn price bands. It simply applies the YoY rates confirmed in this article to the finalized values for October and November 2025. For city hotels, if September’s −10.3% continues, October would be ¥28,677; if it reverts to the summer June–August average of −16.9%, ¥26,567—a gap of roughly ¥3,300–¥5,400 versus the flat-YoY ¥31,970. For business hotels, if September’s +2.2% continues, ¥15,686, slightly above the flat-YoY ¥15,348. This is not a forecast of where results will land but a calculation to check which assumption a property’s own settings are built on.

Source: Compiled by HotelBank Editorial Team from MetroEngines Research. YoY assumptions: City September −10.3%, Jun–Aug avg. −16.9%; Business September +2.2%, Jun–Aug avg. −6.3% (Tables 1 and 2). Current estimates use a different calculation basis from finalized values, so their gap with the scenario values should not be read as YoY.

For revenue managers running city and business hotels in Tokyo — implications and action plan

Summary — three yardsticks for reading the summer decline

In Tokyo’s summer of 2026, estimated settled ADR fell on a finalized-month basis by −20.6% YoY in August for city hotels and −8.9% for business hotels, while estimated OCC held at high levels in both categories—best summarized as “a summer that protected occupancy by conceding rate.” In September, business hotels recovered to +2.2%, while city hotels lagged at −10.3%. There are three yardsticks for reading this market. First, use “finalized-month YoY” to compare your property’s decline with the market’s; second, use “comparison with two years ago” to distinguish a payback from a structural decline (city hotels’ August was −10.3% versus 2024); and third, “treat current estimates separately from finalized values.” Re-applying these three in the same order once October and November values are finalized will serve as a re-check of pricing from autumn into winter.

Frequently Asked Questions

About the Data

References and Sources

■ Data sources

Estimated settled ADR and estimated OCC are aggregations by MetroEngines Research (Tokyo city hotels N=104–110 properties, business hotels N=883–924 properties, monthly from January 2024 to December 2026). For the demand side, we used the Japan Tourism Agency’s Overnight Travel Statistics Survey and JNTO’s International Visitor Arrivals; for actual-results comparison, we used the monthly operating results of four listed hotel REITs.

■ Calculation assumptions

YoY is calculated only between finalized months; current estimates for October–December are not used in YoY calculations. The autumn price band scenarios in Table 6 are a mechanical calculation that applies the YoY rates confirmed in this article (September actual, June–August average, flat) to the finalized values for October and November 2025, and are not forecasts of where results will land.

■ Limitations and caveats

Estimated settled ADR is an estimate derived from sales data, with a median error of 6.6% versus published operating results. Estimated OCC is based on OTA-listed inventory and therefore runs higher than actual room occupancy. REIT figures are actual results for nationwide portfolios and cannot be compared directly with Tokyo values. The Japan Tourism Agency changed its stratification criteria starting with January 2026 data.