Platform competition and adoption heterogeneity: How dockless bike sharing shapes urban air quality

• 成果类型: Article; Early Access
• 署名作者: Liu, Qianqian; Jiang, Shenyang; Qiu, Liangfei; Huo, Baofeng
• 署名单位: Tianjin University; Hong Kong Polytechnic University; State University System of Florida; University of Florida; Zhejiang University
• 刊物名称: PRODUCTION AND OPERATIONS MANAGEMENT
• ISSN/ISSBN: 1059-1478
• DOI: 10.1177/10591478261472202
• 发表日期: 2026
• 关键词: ride-hailing services; big data; TECHNOLOGY; INVESTMENT; capacity; IMPACT; matter; CHINA

摘要

Dockless bike sharing has been widely recognized for its potential environmental benefits by bridging the last-mile gap and encouraging a shift toward greener modes of travel. However, these benefits depend on large-scale bike adoption rather than mere service presence. Unfortunately, this adoption is often undermined by limited bike supply or unfavorable operating contexts. Leveraging the staggered rollout of Ofo and Mobike across Chinese cities, we employ a quasi-experimental stacked difference-in-differences design to identify how air quality changes following the entry of dockless bike sharing. The twin-platform setting enables two complementary analyses. First, treating the two platforms as a unified provider, we find that bike introduction significantly improves urban air quality by 3.82%, corresponding to a 3.13-point reduction in the Air Quality Index (AQI). These results remain robust across extensive alternative specifications and validation checks. Second, distinguishing between the two platforms reveals a supply-side mechanism. Dual-platform entry reduces AQI by 6.47 points, which is more than twice the baseline estimate. In contrast, single-platform entry has no significant impact. We also examine the contextual determinants that may influence bike-riding demand. Heterogeneity analyses indicate larger air quality improvements in cities with more domestic migrants, larger well-educated populations, more severe traffic congestion, denser public transit networks, and stronger bike-lane infrastructure, but smaller improvements where terrain is steep or rainfall is frequent. Overall, our study moves beyond average effects by showing how platform market structure and local demand conditions shape the realized environmental benefits of dockless bike sharing.