A Shock Effect of Ride-hailing Services on Using Traditional Taxis in Urban Areas
-
摘要: 网约车服务的迅速发展对包括传统出租车在内的城市出行方式产生了重大冲击。针对传统出租车使用的冲击问题，收集中国33个地级及以上行政级别城市2010—2016年的平衡面板数据，采用双重差分法（DID）进行量化，并进行动态效应分析、稳健性检验、城市规模等级异质性和城市区位异质性分析。研究结果表明，网约车服务对传统出租车的使用产生显著的负向冲击，使传统出租车使用平均减少了25.46%；随着时间的推移，网约车服务对传统出租车使用的负向冲击呈现先加强后变弱的规律；城市规模异质性分析发现网约车服务致使超大城市中传统出租车使用减少了28.68%，致使大城市中传统出租车使用减少了22.12%；城市区位异质性分析发现网约车服务使东部城市传统出租车使用减少了27.96%，使西部城市传统出租车使用减少了21.2%。Abstract: The rapid development of ride-hailing services significantly affects urban travel modes, including traditional taxis. An important issue is how it affects the use of traditional taxis. The balanced panel data of 33 cities in China from 2010 to 2016 are collected and quantified by difference-in-differences(DID)method. This paper also performs dynamic effect analysis, robustness test, and analysis on the city scale and urban location heterogeneities. The results show that ride-hailing services reduce the use of traditional taxis by an average of 25.46%. The negative impact of ride-hailing services initially strengthens and then weakens over time. Analysis of the city scale heterogeneity shows that ride-hailing services reduce the use of traditional taxis by 28.68% in megacities and 22.12% in the metropolis. Analysis of the city location heterogeneity of shows that ride-hailing services reduce the use of traditional taxis by 27.96% in eastern cities and 21.2% in western cities.
-
Key words:
- urban transportation /
- ride-hailing services /
- shock effect /
- difference-in-differences method /
- taxi
-
表 1 各相关变量的定义和描述性统计
Table 1. Definition and descriptive statistics of related variables
类别 名称 含义 最大值 最小值 均值 标准差 因变量 lntr 传统出租车客运量对数/万人* 11.260 2 6.095 8 8.329 34 1.365 3 自变量 treat × time 网约车服务对传统出租车使用的平均效应 1 0 0.065 9 0.248 8 控制变量 lnnt 出租车数量对数/veh* 11.134 3 4.718 4 7.007 7 1.576 9 lnnpc 私家车数量对数/veh* 15.325 8 10.004 2 12.255 5 1.170 8 lnptr 公交客运量对数/(万人·次)* 13.152 7 5.214 9 8.949 1.737 4 mr 地铁客运量/(亿人·次） 37 0 3.065 9 8.036 9 lnpd 人口密度对数/(人/km2)* 7.732 8 4.127 1 5.821 5 0.883 3 lnne 城镇单位就业人数对数/万人* 2.864 5 1.484 5 6.674 5 0.693 1 lnawe 职工平均工资对数/元* 11.717 9 10.013 1 10.784 0.329 5 lnnmp 城市移动电话数对数/万户* 8.312 8 3.663 5 5.706 2 1.055 8 
下载: 导出CSV
表 2 网约车服务对传统出租车使用的平均效应
Table 2. Average effect of the ride-hailing service on using traditional taxis
因变量: 传统出租车 客运量对数/万人 ⑴ (2) DID (timextreat) -0.301 7*** -0.254 6** (0.0882) (0.100 4) lnnt 0.195 3* (0.112 4) lnnpc -0.081 7*** (0.019 3) lnptr -0.167 3*** (0.051 1) mr -0.068*** (0.009 6) lnpd 0.097 5* (0.049 5) lnne 0.275 8* (0.135 8) lnawe 0.363 8 (0.458 3) lnnmp 0.422 5* (0.202 6) 常数项 8.136*** 10.225 3** (0.051 9) (4.132 3) 城市固定效应 YES YES 时间固定效应 YES YES 聚类到城市 YES YES R2 0.683 4 0.719 1 观测点个数 231 231 *表示p < 0.1；**表示p < 0.05；***表示p < 0.01；括号内为稳健标准误差。
下载: 导出CSV
表 3 网约车服务对传统出租车使用的动态效应
Table 3. Dynamic effect of the ride-hailing service on using traditional taxis
因变量: 传统出租车 客运量对数/万人 (1) (2) time1 x treat -0.351 6*** -0.279*** (0.088 24) (0.054 5) time2 x treat -0.328 4*** -0.259 9** (0.077 3) (0.100 7) 控制变量 NO YES 城市固定效应 YES YES 时间固定效应 YES YES 聚类到城市 YES YES 常数项 8.229 4*** 10.091 5** (0.052 2) (3.776 7) R2 0.680 5 0.758 6 观测点数量 231 231
下载: 导出CSV
表 4 平行趋势检验
Table 4. Test for the parallel trend
因变量: 传统出租车 客运量对数/万人 ⑴ (2) time-1 x treat -0.133 5 -0.101 9 (0.156 6) (0.067 1) time-2x treat -0.031 7 -0.031 4 (0.074 6) (0.064 5) time-3x treat -0.062 -0.048 4 (0.068 3) (0.077 1) 控制变量 NO YES 城市固定效应 YES YES 时间固定效应 YES YES 聚类到城市 YES YES 常数项 8.136 6.994 5 (9.231 7) (10.472 1) R2 0.689 7 0.722 61 观测点数量 231 231
下载: 导出CSV
表 5 基于PSM-DID的稳健性检验
Table 5. Robustness test based on PSM-DID
因变量: 传统出租车 客运量对数/万人 (1) (2) DID (time x treat) -0.368 4*** -0.281*** (0.077 3) (0.075) 控制变量 NO YES 城市固定效应 YES YES 时间固定效应 YES YES 聚类到城市 YES YES 常数项 8.268 5*** 10.223 3*** (0.050 5) (4.703 6) R2 0.675 0.771 1 观测点数量 133 133
下载: 导出CSV
表 6 控制变量均滞后1期
Table 6. Control variables lagging by one period
因变量: 传统出租车 客运量对数/万人 DID(timeXtreat) -0.250 6* (0.127) 控制变量均滞后1期 YES 城市固定效应 YES 时间固定效应 YES 聚类到城市 YES 常数项 10.202** (4.301 3) R2 0.750 5 观测点数量 198
下载: 导出CSV
表 7 基于城市规模异质性的分析
Table 7. Analysis of urban scale heterogeneity
因变量: 传统出租车 客运量对数/万人 超大城市 大城市 DID(time×treat) -0.286 8** -0.221 2* (0.114 8) (0.108 6) 控制变量 YES YES 城市固定效应 YES YES 时间固定效应 YES YES 聚类到城市 YES YES 常数项 9.142 7** 6.943 4* (4.348 1) (3.895 5) R2 0.733 4 0.678 6 观测点数量 210 203
下载: 导出CSV
表 8 城市区位异质性分析
Table 8. Analysis of urban location heterogeneity
因变量: 传统出租车 客运量对数/万人 东部城市 西部城市 DID(time×treat) -0.279 6** -0.212 0** (0.112 7) (0.093) 控制变量 YES YES 城市固定效应 YES YES 时间固定效应 YES YES 聚类到城市 YES YES 常数项 -17.253 7*** 12.600 3*** (5.527 8) (4.499 8) R2 0.535 3 0.532 5 观测点数量 217 196
下载: 导出CSV
-
[1] YOUNG M, FARBER S. The who, why, and when of Uber and other ride-hailing trips: An examination of a large sample household travel survey[J]. Transportation Research Part A: Policy and Practice, 2019, 119: 383-392. doi: 10.1016/j.tra.2018.11.018 [2] 中国互联网络信息中心. 第43次《中国互联网络发展状况统计报告》(全文)[R/OL]. (2019-02-28)[2020-10-18]. www. cac. gov. cn/2019-02/28/c_1124175677. htm.China Internet Network Information Center. The 43rd statisti cal report on the development of internet in China (full text)[R/OL]. (2019-02-28)[2020-10-18]. www. cac. gov. cn/2019-02/28/c_1124175677. htm. (in Chinese) [3] 陈文强, 顾玉磊, 吴群琪, 等. 出租汽车行业管理制度变迁、政策演变与效应评析[J]. 公路交通科技, 2020, 37(4): 148-158. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK202004020.htmCHEN Wenqiang, GU Yulei, WU Qunqi, et al. Analysis on reg ulatory regime change, policy evolution and effect of taxi indus try[J]. Journal of Highway and Transportation Research and De velopment, 2020, 37(4): 148-158. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK202004020.htm [4] 唐立, 邹彤, 罗霞, 等. 基于混合Logit模型的网约车选择行为研究[J]. 交通运输系统工程与信息, 2018, 18(1): 108-114. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201801017.htmTANG Li, ZOU Tong, LUO Xia, et al. Choice behavior of taxi-hailing based on mixed-logit model[J]. Journal of Transpor tation Systems Engineering and Information Technology, 2018, 18(1): 108-114. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201801017.htm [5] NIE Y. How can the taxi industry survive the tide of ridesourcing? Evidence from Shenzhen, China[J]. Transportation Re search Part C: Emerging Technologies, 2017(79): 242-256. http://www.sciencedirect.com/science/article/pii/S0968090X17301018 [6] KIM K, BAEK C, LEE J D. Creative destruction of the sharing economy in action: The case of Uber[J]. Transportation Re search Part A: Policy and Practice, 2018(110): 118-127. http://www.sciencedirect.com/science/article/pii/S0965856417304627 [7] WARD J W, MICHALEK J J, AZEVEDO I L, et al. Effects of on-demand ridesourcing on vehicle ownership, fuel consump tion, vehicle miles traveled, and emissions per capita in U.S. States[J]. Transportation Research Part C: Emerging Technologies, 2019(108): 289-301. http://www.sciencedirect.com/science/article/pii/S0968090X18316449 [8] 宋绪扬, 陈力诚, 肖烨. 微观情况下网约车对交通能耗与排放的影响研究[J]. 交通节能与环保, 2017, 13(6): 46-50. doi: 10.3969/j.issn.1673-6478.2017.06.011SONG Xuyang, CHEN Licheng, XIAO Ye. Study on the impact of network car on traffic energy consumption and emissions un der microscopic conditions[J]. Energy Conservation & Enviro mental Protection in Transportation, 2017, 13(6): 46-50. (in Chinese) doi: 10.3969/j.issn.1673-6478.2017.06.011 [9] RAYLE L, DAI D, CHAN N, et al, Just a better taxi A sur vey-based comparison of taxis, transit, and ride-sourcing servic es in San Francisco[J]. Transport Policy, 2016(45): 168-178. [10] 钟军, 林岩, 杭宇. 中国城市网约车服务对公交使用量的影响[J]. 交通运输系统工程与信息, 2020, 20(5): 234-239. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202005034.htmZHONG Jun, LIN Yan, HANG Yu. Impact of ride-hailing ser vices on use of public transport in China's cities[J]. Journal of Transportation Systems Engineering and Information Technol ogy, 2020, 20(5): 234-239. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202005034.htm [11] BERGER T, CHEN C, FREY C B. Drivers of disruption?Esti mating the Uber effect[J]. European Economic Review, 2018 (110): 197-210. http://www.sciencedirect.com/science/article/pii/S0014292118300849 [12] CONTRERAS S D, PAZ A. The effects of ride-hailing compa nies on the taxicab industry in Las Vegas, Nevada[J]. Trans portation Research Part A: Policy and Practice, 2018(115): 63-70. http://www.researchgate.net/publication/321310692_The_effects_of_ride-hailing_companies_on_the_taxicab_industry_in_Las_Vegas_Nevada [13] GREENWOOD B N, WATTAL S. Show me the way to go home: An empirical investigation of ride-sharing and alcohol related motor vehicle fatalities[J]. MIS Q, 2017, 41(1): 163-187. doi: 10.25300/MISQ/2017/41.1.08 [14] ZHONG J, LIN Y, YANG S. The Impact of ride-hailing servic es on private car use in urban areas: an examination in Chi nese cities[J]. Journal of Advanced Transportation, 2020(7): 1-15. http://www.researchgate.net/publication/347138161_The_Impact_of_Ride-Hailing_Services_on_Private_Car_Use_in_Urban_Areas_An_Examination_in_Chinese_Cities/download [15] PATERNOSTER R, SIMPSON S. Sanction threats and ap peals to morality: Testing a rational choice model of corporate crime[J]. Law & Society Review, 1996, 30(3): 549-583. [16] CARSON V, DUGAN L, YANG S. A comprehensive applica tion of rational choice theory: How costs imposed by, and ben efits derived from, the U.S. federal government affect inci dents perpetrated by the radical eco-movement[J]. Journal of Quantitative Criminology, 2019, 36(3): 701-724. doi: 10.1007/s10940-019-09427-8 [17] 张昱, 孙岩, 刘学敏. 基于TPB的北京市居民低碳通勤选择机制研究[J]. 北京师范大学学报(自然科学版), 2020, 56 (6): 831-837. https://www.cnki.com.cn/Article/CJFDTOTAL-BSDZ202006012.htmZHANG Yu, SUN Yan, LIU Xuemin. Theory of planned be havior applied to choice of low carbon commuting of Beijing residents[J]. Journal of Beijing Normal University(Natural Sci ence), 2020, 56(6): 831-837. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BSDZ202006012.htm [18] LONG X, ZHANG L, WANG Y, et al. Dynamic evolution of traveler's bounded-rational route choice behavior[J]. Journal of Advanced Transportation, 2018(3): 1-12. http://www.researchgate.net/publication/329421728_Dynamic_Evolution_of_Traveler's_Bounded-Rational_Route_Choice_Behavior [19] 参考消息. 滴滴并购优步中国引发市场对网约车行业垄断忧虑[R/OL]. (2016-08-04)[2020-10-18]. www. cankaoxiaoxi. com/finance/20160804/1256899. shtml.CANKAOXIAOXI. Didi's acquisition of Uber in China has raised market concerns about ride-hailing industry monopoly [R/OL]. (2016-08-04)[2020-10-18]. www. cankaoxiaoxi. com/finance/20160804/1256899. shtml. (in Chinese) [20] 速途研究院. 2017年网约车市场研究报告[R/OL]. (2018-03-33) [2020-10-18]. www. sootoo. com/content/675157. shtmlSootoo. com. 2017 Ride-hailing market research report[R/OL]. (2018-03-33)[2020-10-18]. www. sootoo. com/content/675157. shtml(in Chinese) [21] TIRACHINI A. Ride-hailing, travel behaviour and sustainable mobility: an international review[J]. Transportation, 2020, 47 (3): 2011-2047. doi: 10.1007/s11116-019-10070-2 [22] 海外网. 优步在中国年亏10亿美元[R/OL]. (2016-02-20)[2020-10-18]. http://news.haiwainet.cn/n/2016/0220/c3541839-29655205. html.WWW. HAIWAINET. CN. Uber loses $1 billion a year in China[R/OL]. (2016-02-20)[2020-10-18]. http://news.haiwainet.cn/n/2016/0220/c3541839-29655205. html.(in Chinese) [23] 腾讯QQ大数据. 2016年全国城市年轻指数报告[R/OL]. (2016-05-04)[2020-10-18]. http://www.199it.com/archives/469119.html.TENCENT QQ BIG DATA. 2016 National Urban Youth Index Report [R/OL]. (2016-05-04)[2020-10-18]. http://www.199it.com/archives/469119.html.(in Chinese) -
点击查看大图
表(8)
计量
- 文章访问数: 400
- HTML全文浏览量: 182
- PDF下载量: 24
- 被引次数: 0