Where will Japan’s inbound numbers land in 2026? Foreign visitor arrivals for July 2026, published by the Japan National Tourism Organization (JNTO) on 19 August 2026, came to 3,442,100 — a record high for the month of July. Year on year, however, that is just +0.1%, essentially flat, and the January-July cumulative total sits 1.7% below the same period a year earlier. Rather than asserting that “2026 will come in at X million,” this article sets out how far through the year we are, what monthly pace the remaining five months would require, and how many room-nights of demand that headcount converts into — as a calculation anyone can reproduce from public statistics alone. Every figure can be refreshed each month from JNTO’s release and the Japan Tourism Agency’s Overnight Travel Statistics Survey.

1. The full-year arithmetic – 57.5% done, and the pace the last five months require

Building a full-year outlook takes nothing more than subtraction and division. Add up what is already confirmed, then set an assumption for the remaining period. That is the whole exercise.

Foreign visitor arrivals for January-July 2026 came to 24,527,200: 3,597,881 in January, 3,466,848 in February, 3,619,159 in March, 3,692,364 in April, 3,560,256 in May, 3,148,600 in June and 3,442,100 in July (July being an estimate). Against 24,955,693 over the same period a year earlier, that is -1.7%.

Next, fix the denominator. Full-year 2025 arrivals were 42,683,301, of which 17,727,608 fell in August-December — a seasonal structure in which 41.5% of the year is concentrated into those five months. January-July 2026’s 24.527 million therefore represents 57.5% progress against the full-year 2025 total.

That much is observed. What remains is entirely a question of what to assume for August-December. Setting a band of +/-5% year on year gives the three scenarios below.

Source: Japan National Tourism Organization (JNTO), “Foreign Visitor Arrivals Statistics.” Compiled by the HotelBank Editorial Team

Table 1 | Full-year landing scenarios and the monthly average pace August-December would require

Full-year landing	Aug-Dec total	Required monthly average	Year on year	What it represents
41.37 million	16.841 million	3.368 million	-5.0%	Soft scenario
41.40 million	16.873 million	3.375 million	-4.8%	JTB forecast (published 8 January 2026)
42.00 million	17.473 million	3.495 million	-1.4%	Round-number threshold
42.25 million	17.728 million	3.546 million	+/-0.0%	Flat scenario (base case)
42.68 million	18.156 million	3.631 million	+2.4%	Level with 2025 (the line for a new record)
43.14 million	18.614 million	3.723 million	+5.0%	Upside scenario
44.00 million	19.473 million	3.895 million	+9.8%	Strong upside (would need a real tailwind)

Source: Japan National Tourism Organization (JNTO), “Foreign Visitor Arrivals Statistics”; JTB Corp., “2026 (January-December) Inbound Travel Market Trend Forecast” (8 January 2026). Compiled by the HotelBank Editorial Team

Using the table is simple. Each month, when JNTO publishes the single-month figure, check which “required monthly average” line that month’s actual cleared. As August, September and October accumulate, the range narrows quickly; by the October release covering September, the landing point is broadly settled.

For reference, JTB Corp.’s outlook published on 8 January 2026 forecasts 41.40 million for the full year (97.2% of the prior year) and spending of ¥9.64 trillion (100.6%). The shape of it: softer demand from China and Hong Kong leaves headcount marginally below the prior year, while growth in the longer-staying Western and Australian markets lifts spending above it. Progress as of end-July sits broadly consistent with that forecast.

The distance to the government’s target is worth noting as well. The 5th Basic Plan for Promoting a Tourism-Oriented Country, approved by the Cabinet on 27 March 2026, sets out 60 million foreign visitors and ¥15 trillion in travel spending by 2030. If 2026 lands flat at 42.25 million, reaching that would require 9.2% annual growth over the following four years. On the soft 41.37 million landing, the requirement is 9.7% a year. (Both growth rates are HotelBank Editorial Team calculations.)

2. Contribution by market – who filled China’s -3.2 million, and by how much

The January-July 2026 total is -1.7%, but that number is the residue of market-level moves cancelling each other out. Broken apart, the picture changes.

China first. The January-July cumulative total was 2,486,500, down 56.3% year on year — a negative contribution of 3,206,000 visitors in absolute terms. July alone came to 428,200 (-56.1%), still running at less than half the prior-year level.

The gains, meanwhile, are substantial. Of the 23 major markets, all 22 other than China grew, adding 2,687,000 visitors between them — enough to fill 83.8% of China’s decline. South Korea (6,569,800, +20.3%, +1,108,000) and Taiwan (4,736,000, +21.8%, +847,000) led the way, and those two markets alone offset 61% of the China shortfall. In July alone, Taiwan reached 763,800, clearing 700,000 in a single month for the first time, while South Korea’s 894,700 (+31.9%) was a record high for the month of July.

The result: excluding China, the January-July total was 22,040,700, up 14.4% year on year. In other words, 2026 is not a “contracting market.” It is a test of whether +14.4% combined growth across 22 markets can fully absorb the adjustment in one very large one. Whether you read the full-year outlook with that structure in mind makes a considerable difference.

Source: Japan National Tourism Organization (JNTO), “Foreign Visitor Arrivals Statistics.” Compiled by the HotelBank Editorial Team (January-July 2026 cumulative; July is an estimate)

Table 2 | Cumulative January-July 2026 by market and contribution (top 10 markets plus China)

Market	Jan-Jul 2026	Jan-Jul 2025	Change (visitors)	Year on year
South Korea	6,569,800	5,462,088	+1,107,712	+20.3%
Taiwan	4,736,000	3,888,980	+847,020	+21.8%
United States	2,107,900	1,978,506	+129,394	+6.5%
Hong Kong	1,571,000	1,447,126	+123,874	+8.6%
Malaysia	395,100	342,078	+53,022	+15.5%
Indonesia	418,700	373,572	+45,128	+12.1%
India	230,600	189,420	+41,180	+21.7%
Singapore	407,400	369,457	+37,943	+10.3%
Canada	419,000	387,497	+31,503	+8.1%
Vietnam	448,700	418,173	+30,527	+7.3%
China	2,486,500	5,692,537	-3,206,037	-56.3%
Total	24,527,200	24,955,693	-428,493	-1.7%

Prior-year figures are derived from JNTO’s published cumulative totals and year-on-year growth rates. Source: Japan National Tourism Organization (JNTO), “Foreign Visitor Arrivals Statistics.” Compiled by the HotelBank Editorial Team

The operational point is that a change in market mix changes the character of lodging demand. South Korea and Taiwan are short-haul, high-frequency, short-stay markets with a high share of repeat visitors. The room types, stay lengths and booking lead times they need differ from the group and shopping-led demand that came from China. Even with total headcount flat, the booking profile a property sees is not the same one it saw a year ago. Notably, city-level measurements taken while Chinese demand was running at roughly half the prior year still showed ADR rising in Tokyo and Kyoto — substitution demand can hold rates up even as one source market halves.

3. Converting arrivals into guest nights – 3.95 nights per visitor in 2026

Foreign visitor arrivals are a count of entrants, not a metric that maps directly onto accommodation revenue. Converting it into lodging demand requires a coefficient: how many nights per visitor actually land in accommodation. That is obtained by dividing foreign guest nights from the Japan Tourism Agency’s Overnight Travel Statistics Survey by JNTO’s arrivals figure.

Foreign guest nights for January-July 2026 totalled 96,817,900 (January-June are second preliminary values, July is a first preliminary value). Divided by the 24,527,200 arrivals over the same period, that gives 3.95 nights per visitor. By month: 3.57 in January, 4.05 in February, 3.95 in March, 4.16 in April, 4.07 in May, 3.88 in June and 3.94 in July — peaking in April and otherwise moving within a 3.6-4.2 night range.

That coefficient is lower than a year ago. Across 2025 it was 179.92 million foreign guest nights divided by 42,683,301 arrivals, or 4.22 nights per visitor — a decline of 0.27 nights.

Reading that entire gap as a change in underlying behaviour would be premature, however. From the January 2026 survey onward the Japan Tourism Agency changed the stratification variable from “number of employees” to “number of rooms,” so a simple year-on-year comparison may carry some effect from that revision. On top of that, the mix shift works in the same direction: longer-staying group demand from China fell away while shorter-staying South Korea and Taiwan grew as a share. The market rotation described in section 2 most likely shows up directly in this coefficient.

For a full-year conversion, then, the sensible approach is a band centred on 3.95 nights (measured in 2026) with 4.22 nights (the 2025 actual) as the upper bound. Applying 3.95 nights to the flat scenario of 42.25 million gives 167 million guest nights; at 4.22 nights it gives 178 million.

Source: Japan National Tourism Organization (JNTO), “Foreign Visitor Arrivals Statistics”; Japan Tourism Agency, “Overnight Travel Statistics Survey.” Compiled by the HotelBank Editorial Team

4. Converting to room demand – what share of Japan’s 1.69 million rooms inbound takes

Turning guest nights into room demand takes one more conversion: how many guests occupy each room.

The July 2026 figures (first preliminary) from the Overnight Travel Statistics Survey let that coefficient be calculated directly. Guest nights that month were 54,671,130 against 31,930,050 occupied room-nights, giving 1.71 guests per room. June (second preliminary) was 45,815,140 guest nights over 28,826,670 occupied room-nights, or 1.59 — more guests per room in leisure-season July, as expected.

The same statistics also allow national room stock to be worked back. Dividing July’s 31,930,050 occupied room-nights by the 60.8% room occupancy rate (July 2026, published first preliminary value from the Japan Tourism Agency’s Overnight Travel Statistics Survey) and by 31 days gives a national total of roughly 1.69 million rooms. The same calculation on June (28,826,670 occupied room-nights, a 57.2% occupancy rate from the same survey’s second preliminary values, 30 days) gives roughly 1.68 million — near-identical across the two months.

For reference, of the roughly 168,000 domestic properties MetroEngines Research tracks, about 27,000 with roughly 1.26 million rooms have operation confirmed through OTA and similar listings. That is about 75% of the 1.69 million on the Japan Tourism Agency basis. The gap covers ryokan, minshuku and simple lodgings that do not list on OTAs, plus properties working mainly on group and corporate contracts.

Combining these gives the share of room demand attributable to inbound. July 2026’s 13,575,980 foreign guest nights divided by 1.71 guests per room comes to roughly 7.93 million room-nights — 24.8% of that month’s 31,930,050 occupied room-nights, and 15.1% of total supplied room-nights (1.69 million rooms x 31 days = 52.52 million). The same calculation for June gives 26.7% of occupied room-nights and 15.3% of total supply: a stable level.

Table 3 | Full-year foreign room demand – scenario x nights-per-visitor sensitivity (converted at 1.65 guests per room)

Full-year landing	3.95 nights/visitor	4.10 nights/visitor	4.22 nights/visitor
Soft: 41.37 million	99.0 million room-nights (16.1%)	102.8 million room-nights (16.7%)	105.8 million room-nights (17.2%)
Flat: 42.25 million	101.2 million room-nights (16.4%)	105.0 million room-nights (17.0%)	108.1 million room-nights (17.5%)
Upside: 43.14 million	103.3 million room-nights (16.7%)	107.2 million room-nights (17.4%)	110.3 million room-nights (17.9%)

Figures in brackets are the share of total supplied room-nights nationwide (1.69 million rooms x 365 days = 616.85 million room-nights). At 1.60 guests per room the share rises by about 0.5pt; at 1.80 it falls by about 1.4pt. Source: Japan National Tourism Organization (JNTO); Japan Tourism Agency, “Overnight Travel Statistics Survey.” Compiled by the HotelBank Editorial Team

Source: Japan Tourism Agency, “Overnight Travel Statistics Survey” (June 2026 second preliminary, July 2026 first preliminary). Compiled by the HotelBank Editorial Team

To summarise: inbound consumes roughly one sixth of national room supply, or about a quarter if you count only occupied rooms. What makes that 16% figure useful is that whether the year lands at 41.37 million or 43.14 million, the national-average share of room demand moves only within 16.1-16.7%. At the national macro level, a 1.77 million difference in the landing point is worth less than 1pt of occupancy.

Put differently, for any individual property which areas and facility types the demand lands in matters more than where the full-year total settles. The next section looks at that distribution.

5. Where the demand lands – regional areas at 35.8%, and the gap to the 130 million target

Breaking the 83,241,920 foreign guest nights recorded in January-June 2026 down by prefecture shows sharp concentration at the top. Tokyo took 27,041,700 (32.5%), Osaka 11,158,570 (13.4%) and Kyoto 7,573,550 (9.1%) — 55.0% across those three alone. Hokkaido follows with 6,273,340 (7.5%), then Fukuoka with 4,136,630 (5.0%) and Okinawa with 3,945,380 (4.7%).

The three major metropolitan areas — the eight prefectures of Tokyo, Kanagawa, Chiba, Saitama, Aichi, Osaka, Kyoto and Hyogo — total 53,469,060 guest nights, or 64.2%. The remaining 39 prefectures, the “regional areas,” account for 29,772,860, or 35.8%. Across 2025 the regional figure was 60,562,510 guest nights at 33.7%, so the regional share has edged up.

This is where the distance to the government’s target becomes the issue. The 5th Basic Plan for Promoting a Tourism-Oriented Country sets a 2030 goal of 130 million foreign guest nights in regional areas, bringing them to a 1:1 ratio with the three major metropolitan areas. From the 60.56 million recorded in 2025, that requires a 2.15x expansion — around 21% annual growth over four years. Holding nights per visitor constant, regional areas and the three major metropolitan areas would together come to 260 million guest nights, which is close to 60 million visitors multiplied by 4.3 nights each. Read that way, the internal consistency of the government’s target assumes nights per visitor holds at around 4.3.

Which is why the drop to 3.95 nights described in section 3 means more than a rounding difference. Even if headcount reaches 60 million, nights per visitor stuck at 3.95 would produce 237 million guest nights — about 9% short of the 260 million target. The headcount target and the guest-night target will not necessarily be met together, and that is a point anyone planning supply should register. The mismatch between regional room stock and regional lodging share, and the growth rate the 130 million target implies, are analysed in detail in Regional Japan: 61.7% of Rooms, 33.7% of Inbound Nights, 16.5% Growth.

Source: Japan Tourism Agency, “Overnight Travel Statistics Survey” trend workbook (January-June 2026). Compiled by the HotelBank Editorial Team

It is worth sizing the room demand that 60 million visitors at 4.0 nights each in 2030 would create. Dividing 240 million guest nights by 1.65 guests per room gives roughly 145 million room-nights — 44 million room-nights more than the 2026 flat scenario of about 101.2 million. Absorbing that at a 60% room occupancy rate (an assumed full-year average) would take roughly 200,000 additional rooms; at 70%, roughly 170,000. Against the current national stock of about 1.69 million rooms, that is an increase of more than 10%.

6. Flat headcount, moving rates – summer 2026 from the property side

Everything so far has concerned demand volume as measured by public statistics. It is worth closing on how that shows up in prices at the property level, because headcount and rate do not necessarily move in the same direction.

MetroEngines Research data puts estimated settled ADR for July 2026 in Tokyo at approximately ¥12,600 (N=1,136 properties), down 4.8% from around ¥13,300 a year earlier. Kyoto came in at approximately ¥12,200 (N=604), down 12.8% year on year, and Osaka at approximately ¥8,800 (N=647), down 30.7%. Much of the Osaka gap reflects how far Expo 2025 Osaka, Kansai — held from April to October 2025 — lifted the prior-year base.

Even in a month that set a July record for arrivals, area-level rates can sit below the prior year. That is precisely why tracking headcount alone opens a gap against what properties actually feel in revenue. Macro headcount, guest nights and area-level rates each carry different information. As section 4 showed, the national-average demand share barely moves between landing scenarios, so differences between properties are driven far more by area and facility-type mix — and by what happened in the base year. How far growth in arrivals actually passes through into rates is a separate question, and one that resolves differently by facility type and area: in past periods the gains have concentrated in regional ryokan and resorts rather than in the large urban markets.

7. Updating it each month – just swap in three numbers

Every calculation in this article can be brought up to date each month by swapping in the three published figures below. No special data is required.

Table 4 | How to update the progress dashboard

What to update	Source	Release timing	How to use it
Single-month foreign visitor arrivals	JNTO, “Foreign Visitor Arrivals”	Mid to late the following month	Add to the cumulative total and compare against the “required monthly average” in Table 1
Foreign guest nights	Japan Tourism Agency, “Overnight Travel Statistics Survey”	About two months in arrears (first preliminary)	Divide by arrivals to refresh nights per visitor
Occupied room-nights and room occupancy rate	Same source (Tables 3 and 5)	Same as above	Guest nights / occupied room-nights refreshes guests per room; occupied room-nights / occupancy rate / days refreshes total room count

Source: Japan National Tourism Organization (JNTO); Japan Tourism Agency, “Overnight Travel Statistics Survey.” Compiled by the HotelBank Editorial Team

Two cautions. First, the Japan Tourism Agency’s statistics come as first preliminary, second preliminary and final values, and are revised after the fact — reading a several-month moving average is more stable than reacting to small single-month moves. Second, because the stratification variable changed with the January 2026 survey, comparisons against 2025 and earlier may carry some effect from that revision; it is worth footnoting whenever year-on-year figures are discussed.

Conclusion – how to read Japan’s 2026 inbound numbers

The facts as of end-July 2026 come down to these five points.

Table 5 | Summary of the key questions as of end-July 2026

Question	The answer as of end-July 2026
Progress	24.527 million cumulative January-July, -1.7% year on year; 57.5% of the full-year 2025 total.
Landing range	41.37-43.14 million if August-December runs +/-5% year on year. Matching 2025’s 42.68 million would need +2.4% over August-December.
Structure	China is down 3.206 million. Gains of 2.687 million across the other 22 markets offset 83.8% of that. Excluding China, +14.4%.
Lodging conversion	3.95 nights per visitor (4.22 across 2025). A flat landing gives 167 million guest nights for the year.
Room demand	Roughly 101 million room-nights, about 16% of the 617 million room-nights of national supply. A 1.77 million swing in the landing point moves the national-average share by less than 1pt.

In aggregate, 2026 looks likely to land at roughly the same level as 2025. The composition, though, is moving on every axis: market mix, length of stay and geographic distribution. Being able to work out for yourself how many room-nights a given headcount becomes, and where those land, is more useful in day-to-day operations and supply planning than guessing the headline number. All three coefficients used here — nights per visitor, guests per room and total room count — can be refreshed every month from published Japan Tourism Agency figures.

Related reading

• Japan Inbound July 2026: 3.44M Arrivals, 41.7-42.8M Full-Year Range (the monthly news coverage of the July figures)
• Regional Japan: 61.7% of Rooms, 33.7% of Inbound Nights, 16.5% Growth
• Inbound Stay Length Varies 2.0x: Tokyo 2.32 vs Yamanashi 1.16 Nights
• City Hotels: 45.9% of Guest Nights Are Foreign – Ryokan +3.7%
• Only 6 of 47 Prefectures Peak in Autumn: Japan’s Inbound Share Map