Japan welcomed 3,618,900 international visitors in March 2026, an all-time high for the month of March. Seven markets — Indonesia, Vietnam, the United States, Canada, the United Kingdom, Germany and the Nordic region — each posted their single best month on record. Yet foreign guest nights in the same month fell 3.6% year on year. And when you look at room rates, the increases showed up not in the major cities but in regional Japan. This article traces the path inbound demand actually takes before it reaches room rates, using ADR data across all 47 prefectures.
Metric Definitions Used in This Article
- ADR (average daily rate): an estimated settled rate (tax-excluded equivalent) derived by applying category-specific correction factors to each property’s publicly listed lowest-plan level (double occupancy, per-room, tax-included). Cross-checked against property-level results disclosed by listed hotel REITs (91 properties, most recent three months), the median error is approximately 7%. These are estimates and differ from each property’s actual transacted rates and accounting figures. Area-level ADR is the median across the properties covered (i.e. the level of a typical property in that area).
- ADR YoY: the rate of change in estimated settled ADR for March 2026 versus March 2025. Comparisons between prefectures use each prefecture’s median.
- Regional Japan: following the Japan Tourism Agency definition, the 38 prefectures outside the three major metropolitan areas (Tokyo, Kanagawa, Chiba, Saitama, Aichi, Osaka, Kyoto, Hyogo and Nara).
- Data source: MetroEngines Research (March 2026, 47 prefectures, N=17,127 properties)
- — 3.619 million international visitors, a record for March. Yet foreign guest nights fell 3.6% year on year: the two volume indicators pointed in opposite directions.
- — Estimated settled ADR across the 47 prefectures rose a median +5.1% year on year, with 41 prefectures positive. But the gains spread through regional Japan, not the big cities.
- — The correlation between growth in foreign guest nights and growth in ADR was r=0.16 (n=8) — effectively none. Visitor counts do not predict ADR.
- — The transmission runs through property category. Resorts +7.9% and ryokan +7.2% versus business hotels +4.0% and city hotels +4.9%: stay-oriented categories grew roughly twice as fast.
- — Regional Japan’s share reached 34.6% (from 32.3% in March 2025). Whether that tailwind converts into rate depends on product design; it does not happen automatically.
Record arrivals, falling guest nights — a contradictory starting point
The first thing to register is that two statistics moved in opposite directions in March 2026. According to Japan National Tourism Organization (JNTO) estimates, international arrivals reached 3,618,900, up 3.5% year on year, bringing the January–March cumulative total to 10.68 million. Meanwhile, the Japan Tourism Agency’s first preliminary figures from its Accommodation Survey put foreign guest nights at 14.28 million for March 2026, down 3.6% year on year.
More people entered the country, yet fewer nights were stayed. What produces that gap is a change in the mix of source markets. Short-haul markets grew: South Korea reached 795,600 (+15% YoY) and Taiwan 653,300 (+24.9% YoY). China, by contrast, fell 55.9% to 291,600. Because the Chinese market includes segments that stay comparatively long, a rise in headcount does not necessarily translate into a rise in guest nights. For how large those length-of-stay differences become at the prefecture level, see Inbound Stay Length Varies 2.0x: Tokyo 2.32 vs Yamanashi 1.16 Nights.
In other words, March was less a month in which inbound demand grew in volume than one in which its composition changed. How that change transmitted into room rates is the subject of the rest of this article.
Seven record markets — five of them Western
The line-up of the seven markets JNTO reported at single-month record highs captures that shift. Setting aside Indonesia and Vietnam, the remaining five — the United States, Canada, the United Kingdom, Germany and the Nordic region — are all Western markets. Reporting attributes the surge to the overlap of the Easter holidays with the late-March cherry blossom season, alongside increased air capacity.
Source: Japan National Tourism Organization (JNTO), “International Visitor Arrivals (March 2026 estimates)”; compiled by the HotelBank Editorial Team
Western markets generally stay longer and are more likely to travel on multi-destination itineraries. The longer the stay, the more dispersed the destinations, and the less likely the trip is to be confined to Tokyo and Osaka. Long-haul markets also rose broadly: Mexico +69.7%, Spain +36.3% and Italy +23.6%. Expansion in long-haul markets tends to come with a wider geographic spread of accommodation demand.
Regional Japan’s share hits 34.6% — dispersion is visible in the numbers
In the Japan Tourism Agency’s tally for March 2026, the three major metropolitan areas accounted for 9.34 million foreign guest nights (-6.8% YoY, a 65.4% share) and regional Japan for 4.94 million (+3.2% YoY, a 34.6% share). With the metropolitan areas down and the regions up, regional Japan’s share rose 2.3 points from 32.3% in March 2025.
Source: Japan Tourism Agency, “Accommodation Survey”; compiled by the HotelBank Editorial Team (2019 and 2024 are full-year figures; March 2025 and March 2026 are monthly figures)
Worth noting here is that the regional dispersion rate had fallen sharply at one point, from 37.3% for full-year 2019 to 30.9% for full-year 2024. In the post-COVID recovery phase, inbound demand concentrated first in the large cities and the regions were left behind. The 34.6% recorded in March 2026 can be read as a sign that this concentration has begun to loosen. That said, it is a single month’s figure and includes the seasonal effect of the cherry blossom season, so it should be interpreted with that discount in mind.
By prefecture, the strongest growth in foreign guest nights came from regional prefectures: Tottori +78.5%, Ibaraki +57.4%, Hyogo +56.7%, Nagasaki +54.4% and Fukushima +51.2%. The major tourism destinations declined together: Tokyo -9.3%, Kyoto -13.3% and Osaka -3.6%.
The 47-prefecture ADR YoY map — gains spread across western and northern Japan
So did that dispersion reach ADR? Plotting the 47-prefecture data compiled by MetroEngines Research (March 2026, N=17,127 properties) on a map gives the following picture.
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
Forty-one of the 47 prefectures posted ADR above the year-earlier month, with a median of +5.1%. The largest gain was Okinawa at +16.5% (¥11,700), followed by Hyogo at +13.9% (¥14,600) and Aichi at +12.2% (¥9,200). At the bottom, Osaka was the only large decline at -7.6% (¥9,600).
| Prefecture | ADR, March 2026 | YoY | Properties covered |
|---|---|---|---|
| Okinawa | ¥11,736 | +16.5% | 493 |
| Hyogo | ¥14,605 | +13.9% | 565 |
| Aichi | ¥9,207 | +12.2% | 511 |
| Mie | ¥12,774 | +11.5% | 363 |
| Wakayama | ¥9,411 | +10.1% | 226 |
| Tokushima | ¥6,347 | +10.1% | 143 |
| Fukuoka | ¥11,708 | +9.6% | 472 |
| Miyagi | ¥8,860 | +9.5% | 291 |
| … (31 mid-ranked prefectures omitted) … | |||
| Miyazaki | ¥6,690 | +0.6% | 169 |
| Akita | ¥7,348 | +0.5% | 169 |
| Tochigi | ¥9,979 | -0.8% | 411 |
| Toyama | ¥6,866 | -0.9% | 165 |
| Saga | ¥12,712 | -1.8% | 155 |
| Tottori | ¥8,309 | -2.2% | 152 |
| Kumamoto | ¥9,993 | -2.6% | 394 |
| Osaka | ¥9,646 | -7.6% | 646 |
Source: MetroEngines Research; compiled by the HotelBank Editorial Team (47 prefectures, N=17,127 properties)
Osaka’s decline reads most naturally as the payback from Expo 2025 Osaka, Kansai. By March 2025, demand ahead of the Expo opening was already being priced in, so any comparison against that level tends to produce a negative figure. Indeed, Osaka’s ADR YoY widened month by month to -22.1% in April 2026, -32.3% in June and -30.7% in July, confirming that the comparison against Expo-period levels is doing the work. On that basis, reading Osaka’s numbers as weakness in inbound demand would be a mistake.
Growth in guest nights and growth in ADR do not move together in the short run
Here is the core test of this article. We place the Japan Tourism Agency’s year-on-year change in foreign guest nights alongside our own year-on-year ADR change for the same eight prefectures.
Sources: foreign guest nights — Japan Tourism Agency, “Accommodation Survey”; ADR — MetroEngines Research; compiled by the HotelBank Editorial Team
The correlation coefficient between the two is just 0.16 (n=8) — essentially no relationship. Tottori, where foreign guest nights rose 78.5%, saw ADR at -2.2%; Nagasaki, up 54.4%, managed only +1.2%. Conversely, Kyoto, where foreign guest nights fell 13.3%, posted ADR +7.2%, and Tokyo, down 9.3%, still rose +4.8%.
The result looks counterintuitive, but the structure explains it. First, in many regional prefectures foreign guests are only a fraction of total guests, and the smaller the base, the larger the growth rate. Tottori’s +78.5% is likely a limited increase in absolute terms. Second, ADR is a composite of several factors — domestic demand, supply volume, property mix — and is not a function of inbound alone. Third, price responds to how tight inventory is: in areas already running high occupancy, a small increment moves price, while in areas with slack, added demand struggles to move it. The tendency for inbound volume to sit loosely against supply-demand indicators is not unique to this article; Inbound Share vs Weekday Gap: −0.13 Across Japan’s 47 Prefectures likewise finds a 47-prefecture correlation of only −0.13.
Put differently, tracking the volume indicator of visitor arrivals will not explain how your own property’s ADR moves. The transmission runs somewhere else.
The transmission runs through property category — stay-oriented types grew roughly twice as fast
So where does it run? Aggregating the same March 2026 data by property category brings the outline into focus.
Source: MetroEngines Research; compiled by the HotelBank Editorial Team (aggregated from prefecture-level median ADR within each category)
Resort hotels posted a median ADR YoY of +7.9% and ryokan +7.2%, well ahead of business hotels at +4.0% and city hotels at +4.9%. The absolute ADR levels differ as well: resort hotels at ¥15,300 and ryokan at ¥13,200 are roughly double the ¥7,200 of business hotels.
That gap is consistent with the composition of the seven record-setting markets. Travellers from the five Western markets, who tend to travel long and tour multiple destinations, are less likely to string together consecutive nights in an urban business hotel and more likely to build itineraries that include onsen and resort destinations. Stay-oriented properties carry a higher rate per night and a higher share of plans that include meals, so an additional unit of demand has a larger effect on rate. On how much inbound each category is absorbing, City Hotels: 45.9% of Guest Nights Are Foreign — Ryokan +3.7% breaks down inbound share by category, and reading the two together gives a fuller view of the pathway.
The transmission, then, does not follow “visitor arrivals → big-city occupancy → big-city ADR.” It follows “a change in guest mix → demand shifting to stay-oriented and regionally located categories → ADR gains in those categories.” Regional Japan’s share rising to 34.6% and the resort and ryokan categories growing at roughly twice the pace are two cross-sections of the same phenomenon.
Inside the five inbound hub prefectures — they cannot be treated as one
Finally, consider the monthly path of the five prefectures usually described as having high inbound exposure.
Source: MetroEngines Research; compiled by the HotelBank Editorial Team
From January to March 2026, Fukuoka and Hokkaido posted double-digit or near double-digit year-on-year gains: Fukuoka +25.7% in January and +9.6% in March, Hokkaido +26.5% in January and +8.9% in March. Tokyo was steady in a +4.8% to +9.7% range, while Kyoto rose only modestly, from +1.1% to +7.2%. Osaka alone sits in negative territory on the Expo payback.
Grouping these five prefectures as “inbound hubs” and treating them as moving together does not work. Fukuoka and Hokkaido are benefiting from the tailwind of added direct flights from short-haul Asian markets; Tokyo’s depth of supply is keeping price increases gradual. Kyoto is already at a high ADR level (¥20,300, the highest in the country), which makes gains harder to show up as a growth rate.
Conclusion — shift the metric you watch from visitor counts to guest mix
The March 2026 data points to three things.
First, foreign guest nights can fall even when international arrivals hit a record. A change in the mix of source markets pulls headcount and guest nights apart. Second, at the prefecture level the correlation between growth in foreign guest nights and growth in ADR is close to zero (r=0.16, n=8), so a volume indicator cannot be used to predict price. Third, where transmission is actually observable is along the category axis: resort hotels +7.9% and ryokan +7.2% against business hotels +4.0% and city hotels +4.9%, with stay-oriented categories growing roughly twice as fast.
The implication for your own pricing strategy is straightforward. Rather than asking whether visitor numbers in your area are rising, ask whether your property is positioned to receive the change in guest mix — that sits much closer to how ADR moves. In a phase where long-stay, touring Western guests are increasing, stay-oriented product design — support for consecutive nights, plans that include meals, larger rooms — is likely to feed directly into rate. Conversely, in areas anchored to short-stay short-haul markets, the same increase in demand is less likely to pass through into rate.
A regional dispersion rate of 34.6% is a tailwind for regional properties, and at the same time a reminder that whether that tailwind converts into rate depends on product design. Tottori, where guest nights rose 78.5% and ADR did not move, is evidence that the conversion does not happen on its own.
A note on the data: the ADR figures in this article are estimates; cross-checked against property-level results disclosed by listed hotel REITs, the median error is approximately 7%. The Japan Tourism Agency’s March 2026 figures are first preliminary values and may be revised in the final release. The correlation analysis rests on a limited sample of n=8 and does not represent the relationship across all 47 prefectures.
Related reading
- City Hotels: 45.9% of Guest Nights Are Foreign — Ryokan +3.7%
- Inbound Share vs Weekday Gap: −0.13 Across Japan’s 47 Prefectures
- Inbound Stay Length Varies 2.0x: Tokyo 2.32 vs Yamanashi 1.16 Nights
- Regional Japan: 61.7% of Rooms, 33.7% of Inbound Nights, 16.5% Growth
- US Now Tops Japan Inbound at 13.5% — Prefecture Host Map 2026
- Japan Hotel ADR Polarization 2026: 57.6pt Prefecture×Category Gap
- Lodging Is 37% of Japan’s ¥2.51T Inbound Spend: ¥10,446 vs Real ADR
References and sources
■ Data sources
Three series are used: estimated settled ADR across the 47 prefectures compiled by MetroEngines Research (January 2025 – July 2026; N=17,127 properties as of March 2026); Japan National Tourism Organization (JNTO), “International Visitor Arrivals (March 2026 estimates)”; and Japan Tourism Agency, “Accommodation Survey (March 2026, first preliminary release)”.
■ Calculation assumptions
ADR is an estimated settled rate (tax-excluded equivalent) derived by applying category-specific correction factors to each property’s publicly listed lowest-plan level (double occupancy, per-room, tax-included). Area-level ADR uses the median of the properties covered, and year-on-year change is measured against March 2025. Category-level aggregates re-aggregate prefecture-level median ADR by category. Regional Japan follows the Japan Tourism Agency definition: the 38 prefectures outside the nine that make up the three major metropolitan areas.
■ Limitations and caveats
ADR figures are estimates; cross-checked against property-level results disclosed by listed hotel REITs (91 properties, most recent three months), the median error is approximately 7%. The Japan Tourism Agency’s March 2026 figures are first preliminary values and may be revised in the final release. The correlation analysis rests on a limited sample of n=8 and does not represent the relationship across all 47 prefectures. March 2026 is a single month that includes the seasonal effect of the cherry blossom season, and Osaka’s decline reflects payback against the 2025 Expo period.
- Japan National Tourism Organization (JNTO), “International Visitor Arrivals (March 2026 estimates)”
- Japan Tourism Agency, “Accommodation Survey (February 2026 second preliminary release; March 2026 first preliminary release)”
- Yamatogokoro.jp, “March 2026 inbound guest nights down 4% to 14.28 million: metropolitan areas fall while Tottori, Ibaraki and other regions grow”
- Yamatogokoro.jp, “3.61 million inbound visitors in March 2026, cumulative total passes 10 million”
- Travel Voice, “International visitors to Japan reach 3.62 million in March 2026”
- Kanko Keizai Shimbun, “Foreign guest nights reached 160 million in 2024 — a record high, with stark regional differences”
- MetroEngines Research — estimated settled ADR across the 47 prefectures (January 2025 – July 2026; N=17,127 properties as of March 2026)
