中国网约车服务对道路交通事故的影响研究

doi:10.16265/j.cnki.issn1003-3033.2024.02.0277

摘要/Abstract

摘要：

为深入探究网约车服务是否直接影响道路交通事故,以及是否通过改变公交车和私家车的使用量而间接影响道路交通事故。基于2005—2017年我国114个城市的面板数据,采用多期双重差分法(DID)和逐步法,研究网约车服务对道路交通事故的影响机制。结果表明:网约车服务进入城市导致道路交通事故增加5.6%,这一结论在进行平行趋势和倾向得分匹配(PSM)-DID等稳健性检验后仍然成立;网约车服务显著降低约19.3%公交使用量,从而间接增加0.71%道路交通事故,说明公交使用量发挥中介效应;网约车服务显著增加5%私家车使用量,从而间接增加0.14%道路交通事故,说明私家车使用量发挥中介效应。公交使用量的减少是网约车服务间接增加道路交通事故的重要原因。政府部门应当加强网约车服务的监管,同时,鼓励网约车服务与公共交通协同发展。

关键词: 网约车服务, 道路交通事故, 公交, 中介效应, 多期双重差分法(DID)

Abstract:

To deeply explore whether ride-hailing services directly affect road traffic crashes and whether they indirectly affect road traffic crashes by changing the usage of buses and private cars, based on panel data from 114 cities in China from 2005-2017, the multi-period DID method and the causal steps method are used to estimate the impact mechanism of ride-hailing services on road traffic crashes. The results show that the entry of ride-hailing services increases road traffic crashes by 5.6%, and this conclusion still holds after performing robustness tests such as parallel trend and Propensity Score Matching-DID. Ride-hailing services significantly reduce public transit usage by about 19.3%, thus indirectly increasing road traffic crashes by 0.71%, which indicates that public transit usage plays a mediating effect. Ride-hailing services significantly increase private car usage by 5%, thus indirectly increasing road traffic crashes by 0.14%, which indicates that private car usage plays a mediating effect. The decrease in public transit usage is an important reason why ride-hailing services indirectly increase road traffic crashes. Government departments should strengthen the regulation of ride-hailing services, while encouraging the synergistic development of ride-hailing services and public transit.

Key words: ride-hailing services, road traffic crashes, public transit, mediating effect, multi-period difference-in-differences (DID)

中图分类号:

X951交通运输安全

王卓伦, 林岩. 中国网约车服务对道路交通事故的影响研究[J]. 中国安全科学学报, 2024, 34(2): 209-216.

WANG Zhuolun, LIN Yan. Research on impact of ride-hailing services on road traffic crashes in China[J]. China Safety Science Journal, 2024, 34(2): 209-216.

图/表 6

表1

变量的描述性统计

变量类别	变量定义	平均值	最小值	最大值	标准差
被解释变量	交通事故死亡人数ln(D_it)	4.971	1.386	7.461	0.903
解释变量	滴滴是否进入 $R i t$	0.132	0.000	1.000	0.339
中介变量	公交客运量ln(B_it)/×10⁴	9.156	1.974	13.612	1.546
中介变量	私家车保有量ln(E_it)/辆	11.816	7.844	15.452	1.376
控制变量	常住人口ln(P_it)/×10⁴	5.823	3.054	8.031	0.766
	国内旅游人数ln(T_it)×10⁴	7.173	2.827	10.487	1.189
	人均国内生产总值ln(G_it)/元	10.338	8.232	12.320	0.772
	人均工资ln(W_it)/元	10.427	9.252	11.813	0.480
	失业率U_it/%	3.474	1.100	7.600	0.780
	人均道路面积ln(A_it)/m²	2.579	0.438	4.321	0.528
	降雨量ln(F_it)/mm	6.340	3.616	7.971	0.910
	医院数ln(H_it)/个	4.485	2.639	6.613	0.800

表1

表2

网约车服务影响道路交通事故的回归结果

变量	$l n (D i t)$
$R i t$	0.070 0^* (0.036 8)	0.055 8^* (0.033 7)
$l n (P i t)$	—	0.471 0^*** (0.148 0)
$l n (T i t)$	—	0.050 0^* (0.027 5)
$l n (G i t)$	—	-0.112 9^** (0.051 7)
$l n (W i t)$	—	-0.007 9 (0.116 0)
$U i t$	—	-5.416 5^*** (1.721 1)
$l n (A i t)$	—	0.029 5 (0.043 8)
$l n (F i t)$	—	-0.009 0 (0.033 6)
$l n (H i t)$	—	0.052 3^* (0.031 0)
常数项	5.308 9^*** (0.022 0)	3.394 8^** (1.437 7)
时间固定效应	是	是
城市固定效应	是	是
观测值	1 482	1 482
R²	0.342 6	0.375 7

表2

图1

表3

PSM-DID稳健性检验结果

变量	$l n (D i t)$
$R i t$	0.079 8^** (0.038 0)	0.067 3^* (0.037 7)
常数项	5.308 4^*** (0.022 2)	3.525 0^** (1.477 2)
控制变量	否	是
时间固定效应	是	是
城市固定效应	是	是
观测值	1 45 4	1 45 4
R²	0.345 3	0.378 5

表3

表4

公交客运量和私家车保有量的中介效应结果

变量	列1	列2	列3	列4
变量	$l n (B i t)$	$l n (D i t)$	$l n (E i t)$	$l n (D i t)$
$R i t$	-0.192 6^*** (0.054 4)	0.048 7^* (0.024 7)	0.050 0^* (0.029 1)	0.054 3^* (0.032 3)
$l n (B i t)$	—	-0.036 8^* (0.019 1)	—	—
$l n (E i t)$	—	—	—	0.028 6^* (0.016 3)
常数项	5.336 3^** (2.470 1)	3.5913^** (1.421 6)	3.2189^** (1.537 7)	3.3083^** (1.453 4)
控制变量	是	是	是	是
时间固定效应	是	是	是	是
城市固定效应	是	是	是	是
观测值	1 48 2	1 48 2	1 48 2	1 48 2
R²	0.445 5	0.396 3	0.602 3	0.479 6

表4

表5

Sobel检验和Bootstrap检验

中介变量	Sobel-Goodman	Z值	P值	Bootstrap 抽样次数	95%置信区间
中介变量	Sobel-Goodman	Z值	P值	Bootstrap 抽样次数	下限	上限
$l n (B i t)$	Sobel	-1.937	0.052 7	500	-0.057 7	-0.002 4
	Goodman-1	-1.931	0.053 4	1 000	-0.058 3	-0.001 7
	Goodman-2	-1.943	0.052 0	1 500	-0.056 6	-0.003 4
$l n (E i t)$	Sobel	3.987	0.000 1	500	0.034 1	0.081 9
	Goodman-1	3.974	0.000 1	1 000	0.033 0	0.083 1
	Goodman-2	4.001	0.000 1	1 500	0.034 1	0.082 0

表5

参考文献 22

[1]	每日经济新闻. 2021城市交通出行报告[EB/OL]. [2023-06-15]. https://new.qq.com/rain/a/20210418A09YLI00.
[2]	中国互联网络信息中心. 第49次中国互联网络发展状况统计报告[EB/OL]. [2023-06-14]. https://www.cnnic.cn/n4/2022/0401/c88-1131.html.
[3]	ERHARDT G D, HOQUE J M, GOYAL V, et al. Why has public transit ridership declined in the United States?[J]. Transportation Research Part A: Policy and Practice, 2022, 161: 68-87. doi: 10.1016/j.tra.2022.04.006
[4]	BANSAL P, SINHA A, DUA R, et al. Eliciting preferences of TNC users and drivers: evidence from the United States[J]. Travel Behaviour and Society, 2020, 20:225-236. doi: 10.1016/j.tbs.2020.04.002
[5]	左盛丹. 深圳网约车交通事故率高于出租车[EB/OL]. [2023-06-15]. https://www.chinanews.com.cn/auto/2017/06-02/8239874.shtml.
[6]	DILLS K, MULHOLLAND E. Ride-sharing, fatal crashes, and crime[J]. Southern Economic Journal, 2018, 84(4): 965-991. doi: 10.1002/soej.v84.4
[7]	GREENWOOD N, WATTAL S. Show me the way to go home: an empirical investigation of ride-sharing and alcohol related motor vehicle homicide[J]. Mis Quarterly, 2017, 41(1): 163-187. doi: 10.25300/MISQ
[8]	MORRISON N, MEHRANBOD C, KWIZERA M, et al. Ridesharing and motor vehicle crashes: a spatial ecological case-crossover study of trip-level data[J]. Injury Prevention, 2021, 27: 118-123. doi: 10.1136/injuryprev-2020-043644 pmid: 32253258
[9]	BARRIOS J M, HOCHBERG Y V, YI H. The cost of convenience: ridesharing and traffic fatalities[J]. Journal of Operations Management, 2023, 69(5): 823-855. doi: 10.1002/joom.v69.5
[10]	KIRK S, CAVALLI N, BRAZIL N. The implications of ridehailing for risky driving and road accident injuries and fatalities[J]. Social Science & Medicine, 2020, 250(4): DOI: 10.1016/j.socscimed.2020.112793.
[11]	艾媒网. 网约车行业数据分析:2019年中国网约车客运量占出租车客运量的37.1%[EB/OL]. [2023-06-15]. https://www.iimedia.cn/c1061/71939.html.
[12]	GUO Yue, XIN Fu, BARNES J, et al. Opportunities or threats: the rise of online collaborative consumption (OCC) and its impact on new car sales[J]. Electronic Commerce Research and Applications, 2018, 29: 133-141. doi: 10.1016/j.elerap.2018.04.005
[13]	清华大学社科院. 2021年中国一线城市出行平台调研报告[EB/OL]. [2023-06-15]. http://www.199it.com/archives/1246525.html.
[14]	TURNER D, ALLSOP E. Road casualties and public transport fares in London[J]. Accident Analysis & Prevention, 1986, 18(2):147-156. doi: 10.1016/0001-4575(86)90059-X
[15]	宝驾租车. 国内首份重点城市“留守汽车”数据发布[EB/OL]. [2023-06-16]. http://www.12gang.com/article_10634.html.
[16]	LI Xin, WU Liyu, YANG Xianfeng. Exploring the impact of social economic variables on traffic safety performance in Hong Kong: a time series analysis[J]. Safety Science, 2018, 109: 67-75. doi: 10.1016/j.ssci.2018.05.010
[17]	温忠麟, 叶宝娟. 中介效应分析:方法和模型发展[J]. 心理科学进展, 2014, 22(5):731-745. doi: 10.3724/SP.J.1042.2014.00731
	WEN Zhonglin, YE Baojuan. Analyses of mediating effects: the development of methods and models[J] Advances in Psychological Science, 2014, 22(5): 731-745. doi: 10.3724/SP.J.1042.2014.00731
[18]	YAO Di, XU Liqun, ZHANG Chunqin, et al. Revisiting the interactions between bus service quality, car ownership and mode use: a case study in Changzhou, China[J]. Transportation Research Part A: Policy and Practice, 2021, 154: 329-344. doi: 10.1016/j.tra.2021.10.017
[19]	MACKINNON P, LOCKWOOD M, WILLIAMS J. Confidence limits for the indirect effect: distribution of the product and resampling methods[J]. Multivariate Behavioral Research, 2004, 39(1): 99-128. doi: 10.1207/s15327906mbr3901_4 pmid: 20157642
[20]	李杰, 曾叙砜, 孙领, 等. 全国道路交通安全水平的时空布局演化研究[J]. 中国安全科学学报, 2021, 31(12):136-143. doi: 10.16265/j.cnki.issn1003-3033.2021.12.018
	LI Jie, ZENG Xufeng, SUN Ling, et al. Research on temporal and spatial evolution of road safety in China[J]. China Safety Science Journal, 2021, 31(12): 136-143. doi: 10.16265/j.cnki.issn1003-3033.2021.12.018
[21]	路庆昌, 崔欣, 徐标, 等. 公交接驳场景下轨道交通网络脆弱性研究[J]. 中国安全科学学报, 2021, 31(8):141-146. doi: 10.16265/j.cnki.issn1003-3033.2021.08.020
	LU Qingchang, CUI Xin, XU Biao, et al. Vulnerability research of rail transit network under bus connection scenarios[J]. China Safety Science Journal, 2021, 31(8): 141-146. doi: 10.16265/j.cnki.issn1003-3033.2021.08.020
[22]	朱彤, 秦丹, 魏雯, 等. 基于机器学习的公交驾驶员事故风险识别及影响因素研究[J]. 中国安全科学学报, 2023, 33(2):23-30. doi: 10.16265/j.cnki.issn1003-3033.2023.02.0034
	ZHU Tong, QIN Dan, WEI Wen, et al. Research on accident risk identification and influencing factors of bus drivers based on machine learning[J]. China Safety Science Journal, 2023, 33(2): 23-30. doi: 10.16265/j.cnki.issn1003-3033.2023.02.0034