How China Counts Foreign Visitors: Making Sense of the 2026 Inbound Numbers


Last updated: November 2026

China publishes four different inbound numbers, and all four are correct. The 499 million figure is every border crossing in both directions. Foreign arrivals run about 30.6 million. The 154.5 million inbound visitors for 2025 include Hong Kong, Macau and Taiwan residents. Match the definition to your question before you quote any of them.

What does the 499 million border-crossing figure actually count?

On September 10, 2026, the National Immigration Administration (NIA) reported that border inspectors cleared 499 million entries and exits between January and August, up 8.7 percent year on year. The breakdown matters more than the total:

  • Mainland Chinese residents: 242 million
  • Hong Kong, Macau and Taiwan residents: 196 million
  • Foreign nationals: 61.287 million

Mainland residents and Hong Kong, Macau and Taiwan compatriots therefore account for roughly 88 percent of the headline. It is a traffic count, not a visitor count. Each person is logged on the way out and again on the way back, so a single round trip contributes two units. Nothing in the series distinguishes a holiday from a commute, a factory inspection from a family visit, or a first-time visitor from a Shenzhen resident crossing weekly.

Its one genuine use is capacity: these are the people who will queue, take a seat, or occupy a berth. If you are timing a transfer at a specific port, watch this series.

Why is the foreign arrival count only about half of the foreign crossing count?

Because the same passport is counted twice per round trip. The third quarter of 2026 offers a clean check: NIA logged 23.272 million foreign crossings and 11.601 million foreign arrivals in the same three months, a 49.8 percent split.

Apply that to the eight-month figure and the 61.287 million foreign crossings imply about 30.6 million foreign arrivals. A second route arrives at the same place. NIA said 23.72 million of those arrivals were visa-free, and that this represented 77.6 percent of all inbound foreigners. Divide 23.72 million by 0.776 and you get 30.6 million again.

The agreement is reassuring, and also where the confusion starts: 30.6 million is still not a tourist count. It includes foreign residents of China returning from business trips, airline and maritime crew, border-area residents crossing on local permits, transit passengers who clear immigration, and one consultant arriving four times on four separate visa-free trips.

How does the tourism ministry define an inbound visitor?

The Ministry of Culture and Tourism (MCT) and the National Bureau of Statistics work from a separate rulebook, the Statistical Survey System for Culture and Tourism. Its definitions are narrower and more explicit than a border count.

Term Definition Unit
Visitor (游客) Away from usual country or place for 12 months or less, main purpose not remunerated locally One trip
Inbound visitor (入境游客) Foreigners plus Hong Kong, Macau and Taiwan compatriots visiting the mainland One entry
Inbound overnight visitor (入境过夜游客) Inbound visitors who spend at least one night in a tourism accommodation establishment One entry with a room night
Inbound day visitor (入境一日游游客) Inbound visitors who do not stay overnight in such an establishment One entry without a room night

Three details do most of the work. First, the test is country of usual residence, not nationality, so a Chinese passport holder living in Toronto who tours China counts in the foreign-market statistics. Second, the overnight definition only recognizes nights in tourism accommodation, which keeps stays in private homes out of hotel-demand data. Third, eight groups are excluded outright: senior officials on official visits, diplomats, residents of more than one year including students and journalists, transit passengers who never clear immigration, border residents, compatriots resettling, settled foreigners, and returning Chinese nationals.

That exclusion list is the filter. It strips out commuters and crew, and it pushes long-stay residents into the domestic series instead.

Why are most of China's inbound visitors from Hong Kong, Macau and Taiwan?

In 2025 the MCT recorded 154.5 million inbound visitors, up 17.1 percent. Of those, 119.32 million were Hong Kong, Macau and Taiwan residents and 35.17 million were foreign nationals. The compatriot share was 77.2 percent.

This is geography, not a quirk of accounting. A Macau resident shopping in Zhuhai, a Hong Kong student crossing at Lo Wu, a family visiting relatives in Fujian: each crossing counts as one inbound visit. The same border that makes day trips ordinary makes the national total large.

For anyone selling travel, the national figure is a weak proxy for hotel demand and a strong one for cross-border traffic; city data is closer to what gets sold. Shanghai received 5.31 million inbound visitors in the first half of 2026, of whom 5.02 million stayed overnight, a 94.5 percent share. That is a market you can pitch a six-night itinerary to. The national 77 percent compatriot share is not.

Which number should you cite, and which should you never mix?

If the question is Cite Latest reference point
How large is the inbound market? MCT and NBS inbound visitors 154.5 million in 2025, 22.8 percent foreign
How fast is foreign demand growing? NIA foreign arrivals About 30.6 million in January to August 2026
How much visa friction is left? NIA visa-free share of foreign arrivals 77.6 percent in January to August, 77.5 percent in Q3
How many room nights can I sell? Inbound overnight visitors, city level Shanghai at 94.5 percent overnight in H1 2026
What will the year end at? China Tourism Academy forecast 163 million visitors and 36.66 million foreigners for 2026

The 2026 trend lines agree across publishers. NIA reported foreign crossings up 19.5 percent in January to August and inbound foreigners up 15.6 percent in the third quarter, with visa-free arrivals growing faster at 26.8 and 24.1 percent. The MCT put first-half inbound foreigners at 22.914 million, up 20.4 percent, close to 80 percent of them visa-free. The China Tourism Academy forecasts 163 million inbound visitors for the year, including 36.66 million foreigners.

The pattern is consistent: the visa-free channel is growing faster than the total, so the gap between the two headline series should widen rather than close.

How do you sanity-check an inbound statistic before quoting it?

Five questions catch almost every error.

  1. Who is in the numerator? If Chinese residents are included, it is a travel figure, not an inbound one.
  2. One direction or two? Halve anything labeled entries and exits before comparing it with an arrival count.
  3. Are compatriots included? Otherwise a 154.5 million figure reads as 154.5 million foreigners, overstating that market by more than four times.
  4. Which period? Eight-month and full-year numbers circulate in the same autumn news cycle.
  5. Is it a forecast? The 163 million for 2026 is a projection published in March, not a result.

For itinerary work the payoff is concrete. If a client asks whether China is crowded, answer from port-level crossings and city overnight figures, not the national total. If a client asks whether this is the moment to book, answer with the visa-free share, because that is the variable that changed. The same holds for leisure, MICE and incentive programs and study tours: demand sits in the filtered series, capacity pressure in the border series.

Frequently asked questions

Is 499 million the number of tourists who visited China?

No. It is every entry and exit by every nationality in both directions from January to August 2026, including Chinese residents returning home. The comparable figure for a full year is 154.5 million inbound visitors, of whom 35.17 million were foreigners.

Why do different outlets report different inbound numbers for the same year?

Because they quote different series. A border count, a foreign arrival count, an inbound visitor count and an overnight count can legitimately differ by a factor of four or more. The publisher and the period tell you which one you are reading.

Do business travelers and students count as inbound visitors?

Yes. Business travel, meetings, study, medical treatment and family visits all count, provided the trip is not remunerated locally and lasts under 12 months. That is why corporate delegations and study tours sit in the same series as holidays, and why that series cannot isolate leisure alone.

How many foreigners actually visited China in 2025?

On the tourism definition, 35.17 million, up 30.6 percent. NIA's visa-free arrivals alone came to 30.08 million, which shows how much of the growth ran through the visa channel and why the two series belong side by side rather than subtracted.

Do the definitions change in 2027?

The four definitions are unchanged. What shifts is the policy underneath them: most unilateral visa-free arrangements run to December 31, 2026, with Russia extended to 2027 and Brunei carrying no end date. Any 2027 proposal should name the arrangement it relies on.

---

For how these flows land in a single holiday period, see Golden Week 2026 inbound numbers. For the source-market side, read Southeast Asia's outbound surge; for how advisory language moves demand, see travel advisories and China demand. Our China Travel Insights section carries the full series, and the 2026 budget breakdown turns these volumes into per-trip costs. Tailor-made itineraries, inbound leisure groups and business delegations all price from the same capacity data.