China's air travel over the coming two decades, projected from income and population alone and counted on Chinese airlines only. Re-run on the curve fitted to all years, the model ran below China's pre-COVID growth.
Conditions and method
Status: projection (income-only central projection from income per person and population; supply-side expansion, network formation and low-cost entry are not modelled).
The baseline is an income-only projection from the 2024 passenger level. The 5th and 95th percentile lines are a 90% band around it that takes its income path to 2030 and the 2024 level as given, so it covers part of the uncertainty only and cannot be added to the scenario spread. Re-run with actual income, on the curve fitted to all years, the method projected less traffic than was flown in every window of 5 years or more ending by 2019, and its band held the outcome in 18 of 30. Outside the United States the bands held 55% of those 5-year windows, 25% of 10-year and none of 15-year.
The chart counts passengers a year, not passenger-kilometres. For China the count is CAAC (the Civil Aviation Administration of China) passengers on Chinese airlines only; domestic includes Hong Kong, Macao and Taiwan routes. Other markets count passengers on their own national definitions.
The baseline grows about 0.99% a year. The fitted curve’s median income elasticity averages 0.412 against 1.509 on the long-run ladder, where the result for China is inconclusive, and most of that gap is already there in 2024.
Baseline: the IMF growth forecast to 2030, then income growth rising or falling toward one assumed long-run rate, about 1.8% a year, which has no external source. Optimistic and pessimistic: income growth to 2030 is 1.4 and 0.6 times the baseline’s, on the UN high and low population variants; by 2044 most of the gap between them comes from the income path (population is about 42%). Structural deceleration: from 2031 income growth converges to 1.2% a year. Disruption: income growth is 3 percentage points lower in 2025 and 2026, less so to 2029, so income stays about 10% below the baseline.
Two refit tests set the method against holding trips per person flat: it did no better one year ahead and was closer several years ahead in most markets. The second, on data to 2014, was 25% low here in 2019 and 22% low in 2024. A third, across the pandemic, starts from a predicted level, so its error counts the curve’s gap to the data and across markets is several times the other two. Neither refit behind them passed the project’s own sampling screen, so their figures are less firm than the projection’s. These tests reach 10 years at most; the projection runs 20.
The projection uses passenger data through 2024 and the IMF’s October 2025 growth forecast, which runs to 2030. Later observations exist for some markets and are not used. Income growth per person is IMF real GDP growth minus UN medium-variant population growth. The IMF path embeds its own population projection; the two differ most for Canada, Australia and the United Kingdom. The international share of traffic is held inside its historical range. Short-history markets borrow their curve’s shape from the others. China’s fitted curve is the flattest of the 9 pooled markets.