![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
Abstract
Focusing on conventional taxis and e-hailing, this paper discusses the technology, job and mobility choices of a conventional occupational group – taxi drivers – faced with an algorithm-enabled mode of mobility. Based on six-month ethnographic fieldwork in Xi’an, China, it shows that taxi drivers generally prefer taxis to e-hailing. Because the e-hailing algorithm treats each driver independently, drivers’ spatio-temporal skills become marginalised and taxi drivers are no longer able to maintain a regular spatio-temporal arrangement that facilitates their community nodes as they do in taxi-driving. Their preference for taxis is a response to the potential threat to their community and social values imposed by algorithm-enabled mobilities. The paper emphasises how workers’ response to algorithmic digital automation are centred around and operationalised by spatio-temporal mobility. It also shows that the impacts of new mobilities are distributed unevenly across groups with different socio-economic backgrounds and life experiences, in this case vis-à-vis the privatisation and urbanisation of Chinese society.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Acknowledgement
I wish to extend heartfelt thanks to this journal’s editorial board represented by the coeditor, Professor Peter Adey, and to gratefully acknowledge the feedback and comments from two anonymous Mobilities referees, whose valuable comments and suggestions contributed significantly to improving this article. I wish to also thank the organizers of the special issue (as well as the corresponding workshop held in Vancouver in November, 2019), Joshua Barker, Sheri Gibbings and Bronwyn Frey for providing a great forum for exchanging ideas among scholars studying the platform economy all over the world. I would like to acknowledge all the supportive feedbacks from other authors in this special issue. In addition, I thank Shuang L. Frost, Alex Rosenblat, Gonçalo Santos and Naubahar Sharif for their valuable suggestions during the early stages of the development of this article.
Notes
1 I have used pseudonyms for the names of all informants and WeChat groups.
2 The exchange rate during my fieldwork was 1 USD = 6.7 RMB. The data about income difference was gathered by myself during fieldwork. To ensure the accuracy of the income difference, besides asking taxi and e-hailing drivers’ their own monthly income, I also asked them about their knowledge of the monthly income of their counterparts. Their answers indicate that they generally know the monthly income of their counterparts and know the income difference between the two jobs (RMB 1,000–1,500).
3 Here, ‘e-hailing’ refers to private car-based e-hailing.
4 Similarly, other works urge mobilities to be studied with reference to the social and affective relations embedded in them (Salazar Citation2010; Sheller Citation2004).
5 ‘Mobility’ refers to the status of mobility but not necessarily the highly mobile, flexible or uncertain status. In other words, both high mobility, flexibility or uncertainty and immobility, regularity or stability are depictions of mobility. As Hannam, Sheller and Urry (Citation2006) pointed out, mobilities cannot be described without moorings. Immobile nodes and platforms are the prerequisite of certain patterns (especially the more stable patterns) of mobility.
6 As will be shown below, taxi-driving also has this tendency, though not as much as e-hailing.
7 “I interviewed.” Among the taxi drivers I spent time with, 90 are males and 10 are females. Among the e-hailing drivers I interviewed, 55 are males and 5 are females. Both groups are highly masculine.
8 Some owners operate several cabs, and some do not drive. Therefore, employed drivers predominate among taxi drivers.
9 E-hailing’s rise in China is built not only on its flexibility and convenience enabled by digital technologies, but also on social factors, including the mass automobile ownership, increase of a freelance labour force, and the increasing precarity of taxi-driving (Xing Citation2021).
10 In and before 2018, such ambiguity significantly shaped taxi drivers’ preference for taxis. Without clear regulations, transportation administrators in Xi’an could fine e-hailing drivers or turn a blind eye by chance. Taxi drivers averred such uncertainty. After 2018, taxi drivers worried significantly less because they, as full-time drivers, could simply obtain a licence. In contrast, licencing became a concern for part-time drivers because it was not worthwhile for them to bear the financial and time cost of licencing.
11 In some but infrequent cases, owners want to become Didi drivers, so they recruit another employed driver for the shift run.
12 All the former taxi drivers are full-time Didi drivers.
13 At the early stage e-hailing companies subsidised drivers heavily during the price war. Once the e-hailing market stabilised (in Didi’s case, starting in 2018), they significantly reduced subsidies and rates for drivers.
14 Notably, such networks usually do not completely diminish, at least in short-term. Many taxi drivers know Didi drivers’ working conditions via the connections with their old colleagues who switched to Didi.
15 In 2016, Didi’s rate (approximately RMB 3.5 per kilometre with occasional surge pricing) was much higher than the taxi rate (RMB 1.5 per kilometre) in Xi’an. In 2018, Didi’s rate decreased to approximately RMB 1.4 per kilometre. Nevertheless, daily private car rentals in the e-hailing business (approximately RMB 100 per day) were still lower than the amount of fenziqian in the taxi business (RMB 160), meaning that e-hailing was still more profitable.
16 Similarly, Chen and Sun (Citation2020) study food delivery workers’ mundane and opportunistic tactics in reconstructing their temporality.
17 The difficulties for drivers to return to taxis vary. As for cars, those who use their own cars for Didi can easily switch back; those who bought new cars need to sell the cars or turn them into self-use; those renting cars usually have contract renewable every 3, 6 or 12 months and can switch when contracts end – those with longer renewal periods suffer. Among my 13 informants, all 3 were using their own cars (including Mr. Wang), 1 out of 3 bought new cars, and 6 out of 7 were renting cars. After deciding to switch, drivers contact their taxi driver or owner friends to get work opportunities, which takes less than one month. Most drivers did not have significant difficulties readapting into taxi-driving because most have not been away for very long and still retain spatio-temporal skills and some connections with old colleagues.
18 As shown by other articles in this issue, e-hailing drivers also started to form work-based communities as e-hailing gradually matured, yet taxi drivers in my case tend to prefer their existing communities to those (possibly) rebuilt after another round of deprivation.