Answers are grounded in this paper and cite their sources.
Study estimates causal effects of travel time on airfare in Chengdu's multi-airport system, indicating significant price sensitivity disparities.
As multi-airport systems continue to form in large cities, the impact of airport ground accessibility on passenger travel decisions and price sensitivity is becoming increasingly prominent. This study takes the Chengdu Shuangliu and Tianfu (TFU) international airport system as a case study, utilizing airfare data from the start of TFU’s operations through 2024. By applying the causal forest model in conjunction with Double/Debiased Machine Learning (DML), the research aims to estimate the Conditional Average Treatment Effect (CATE) of travel time changes on airfare under the DML and causal forest framework. The findings reveal significant regional disparities in price sensitivity. For driving, the areas with the highest price sensitivity are concentrated in the western and southwestern regions farthest from the airports. Among these, Pujiang and Qionglai show CATE values as high as −17.84 and −16.02, respectively. This signifies that for every 10-min increase in driving time to an airport, residents in these areas are expected to purchase air tickets that are ∼16–18 RMB cheaper as they require substantial price discounts to compensate for high ground transportation costs. In contrast, the center of price sensitivity for public transport unexpectedly shifts to the eastern and northern regions with inadequate rail transit coverage, such as Qingbaijiang (CATE of −8.54) and Jintang (CATE of −8.19). This study explores the potential costs arising from transportation inconvenience, offering preliminary insights that could inform the design of transportation infrastructure planning, refined airline pricing strategies, and specialized airport connection services.
No takes yet. Share an insight, caveat, or question.
Luo et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: