VOICE AND REPRESENTATION OF PLATFORM DRIVERS IN VIETNAM Do Quynh Chi, Research Centre for Employment Relations(ERC)

Voice and Representation of Platform Drivers in Vietnam July 2022

i Table of Contents ACKNOWLEDGEMENTS.................................................................................................................................... ii ABBREVIATIONS................................................................................................................................................ iii INTRODUCTION................................................................................................................................................ 1 CHAPTER 1: WORK PLATFORMS – AN OVERVIEW............................................................................................. 3 CHAPTER 2: DEFINING THE RELATIONSHIP OF PLATFORMS AND DRIVERS........................................................ 6 2.1. Platform drivers and riders in Vietnam: Who are they?......................................................................... 6 2.2. Defining the platform-driver relationship............................................................................................. 7 Legal relationship................................................................................................................................. 7 Economic relationship.......................................................................................................................... 9 Technological relationship.................................................................................................................... 10 2.3. Channels for communication and grievance-handling......................................................................... 11 CHAPTER 3: LOYALTY, EXIT OR VOICE? PLATFORM WORKERS’ ORGANISING STRATEGIES................................. 12 3.1. Platform drivers’ responses to dissatisfaction........................................................................................ 12 3.2. Building labour agency........................................................................................................................ 14 3.3. Collective bargaining by riots............................................................................................................... 15 3.4. Prospects for organising gig workers.................................................................................................. 16 CONCLUSION.................................................................................................................................................. 18 REFERENCES..................................................................................................................................................... 19

ii ACKNOWLEDGEMENTS This research would not have been possible without the funding and consistent support from Friedrich-Ebert-Stiftung (FES) in Vietnam. We would also like to warmly thank members of the Vietnam Labour Research Network(VLRN) for contributing to the design and data collection for the study and the experts, lawyers, labour administrators and union officials who provided their insights. Finally, our deepest thanks go to the platform drivers and riders who overcame their fears to share their important stories and thoughts with us.

iii ABBREVIATIONS CB DOLISA ERC FES FM FOL GDP HCMC ILO IT MOLISA USD VGCL VLRN VND Collective bargaining Department of Labour, Invalids and Social Affairs Research Center for Employment Relations Friedrich-Ebert-Stiftung Female Federation of Labour Gross Domestic Product Ho Chi Minh City International Labour Organization Information Technology Ministry of Labour, Invalids and Social Affairs United States Dollar Vietnam General Confederation of Labour Vietnam Labour Research Network Vietnam Dong

IV Introduction This page is intentionally left blank. Voice and representation of platform drivers in Vietnam

INTRODUCTION 1 INTRODUCTION The way in which we have this flexible model means that there are certain ways in which you can protect yourself. So things like collective bargaining and other things, because of the flexible nature of our work, because you can come on and off the platform, the purpose that collective bargaining was originally structured for doesn’t necessarily hold. Amit Singh, Uber Head of Work-Policy and Research 1 Flexibility, as argued by this top researcher of Uber – one of the largest firms in the global gig economy, is the reason platform workers can protect themselves (choosing to stay or leave the platform) without the need of traditional industrial relations institutions, such as through collective bargaining. Is it true, as argued by the Uber manager, that platform workers can protect themselves and that the usual mechanisms of collective voice, representation, and collective bargaining have become outdated in the gig economy? This research, conducted by ERC and members of the VLRN and funded by Friedrich Ebert Stiftung Vietnam, sets out to answer these questions through examining the case study of platform drivers in Vietnam. Gig and platform-based work is one type of nonstandard work facilitated through technology and digital markets, on-demand. The number of digital labour platforms, both online web-based and locationbased, has grown rapidly over the world over the past decade. Online web-based platforms(microtask, freelance and competitive programming) and locationbased platforms in the taxi and delivery sector, globally, saw at least 777 active platforms operating in January 2021(ILO 2021). The number of platforms in the delivery sector was the highest(383), followed by online web-based platforms(283), and the taxi sector(106). Estimates suggest that at least seven million platform workers worldwide perform work worth approximately USD5 billion annually outsourced through applications or platforms. 2 While platform economy work and workers currently comprise a small share of employment and the overall labour force 3 , this form of work is growing exponentially. 4 The benefits and costs of gig and platform work for employers, workers and society remain highly contested. Advocates contend that digital labour platforms can economically benefit socially marginalised groups including the unemployed, geographically isolated, and refugees(De Stefano, 2016; Byrne and Waters, 2015). For firms, gig work combines technological innovation with various contractual relationships that can reduce transaction and labour costs, provide‘numerical flexibility’ in the face of fluctuating demand, and increase competitiveness(Peck and Theodore, 2012; ILO, 2016). However, like non-standard employment more broadly, work content and arrangements in the gig economy are diverse. Despite the possible benefits, jobs in the gig economy can also be structured in ways that can negatively impact workers(unpredictable scheduling, inconsistent earnings, unreliable long-term employment prospects) and firms(unfair competition, lower productivity and absenteeism)(Ibid, see also: Peck and Theodore, 2012; ILO, 2016; De Stefano, 2016). In Vietnam, the emergence of work platforms started in 2014 with the entrance of Grab and Uber, the two ridehailing platforms. Since then, despite work platforms mushrooming in Vietnam, the ride-hailing ones(which have since extended to more than 20 additional services) remain the most developed and engage the largest number of workers. By the end of 2019, the five biggest vehicle-sharing platforms(Grab, GoViet, FastGo, Be, and MyGo) collectively had 170,000(car) drivers and 5 400,000(motorbike) riders. 6 Platform workers are those who are self-employed and responsible for their own social protection and fulfilling their own tax obligations. Their employment relationship with the platform companies, in many countries including Vietnam, remains largely ambiguous. As such, they can be defined as informal workers or independent contractors, which means they are excluded from forming and joining unions and engaging in collective bargaining. Despite these challenges, the platform drivers and riders in Vietnam have managed to mobilise for(informal) collective actions which, in many cases, went beyond one platform or locality. For instance, in 2018, thousands of Grab drivers and riders organised collective actions 1 ILO Facebook Event on 21 November 2017(Link: https://www.ilo.org/global/about-the-ilo/multimedia/video/video-interviews/WCMS_601901/ lang--en/index.htm). 2 Kuek, S.C., Paradi-Guilford, C., Fayomi, T., Imaizumi, S., Ipeirotis, P.(2015). The Global opportunity in online outsourcing, Washington DC, World Bank. 3 In 2018, 36 percent of US workers joined the gig economy through either their primary or secondary jobs. This rate is expected to rise to more than 50 percent by 2023. This figure in Europe was less than 10 percent(Pendell 2018; Choi 2019). 4 Berg, Janine. 5 https://www.tienphong.vn/xa-hoi/so-phan-hang-van-lai-xe-cong-nghe-di-ve-dau-1618303.tpo 6 https://laodong.vn/kinh-te/hang-tram-ngan-xe-om-cong-nghe-se-vuong-khau-tru-trong-thue-thu-nhap-752598.ldo Voice and representation of platform drivers in Vietnam

2 INTRODUCTION in the capital Hanoi, the central city of Danang and the southern metropolis of Ho Chi Minh City at the same time. These collective actions were used to bargain collectively with platform companies with regards to a profit-sharing formula between the platform and workers. Since 2014, these spontaneous and informal collective actions, repeated at different scales almost every time the platform companies adjusted their profit-sharing formulas and bonus systems, showed that these so-called independent contractors may not be fully content with the‘flexibility’ system provided by the platforms. Instead, this action amplified their need for a voice, representation and collective bargaining to protect their own interests. This research, therefore, will examine the needs, prospects and strategies employed by platform drivers in voicing their opinions, organising and bargaining collectively with platform companies. In particular, the study also investigates the relationships between the drivers/riders and technology companies, the structural and institutional challenges for platform drivers in protecting and advancing their interests as well as examining the possible models for workers’ voice and representation. Methodology Survey of Platform Drivers and Riders in Hanoi and Ho Chi Minh city FM: 10.6% Hanoi: 62% 274 HCMC: 38% the drivers and riders randomly at departure points and between rides for interviews. The identities of all respondents is kept confidential. The total survey population is 355 people, comprised of 81 car-drivers and 274 bike-riders. With the estimated population of drivers and riders in the top five ridehailing platforms(Grab, GoViet, FastGo, Be, and MyGo) being 170,000 and 400,000 respectively 7 , the sample for the car-drivers has a confidence level of 90 percent and an error margin of 0.09, while the sample of bikeriders reached a confidence level of 95 percent and an error margin of 0.06. As Grab accounts for a 70 percent market share, the same percentage of sample sizes was allocated to it. Among bike-riders, 10.6 percent are women. However, no female car-driver was reached by the survey. There are no statistics on female participation in these ride-hailing platforms. However, according to a survey by Grab in 2020 covering more than 7,000 partner drivers and riders, women accounted for only 1.4 percent of the platform partner population(Grab 2020). Based on the survey sample, the research team selected 10 drivers and riders for in-depth interviews. These drivers and riders were either the de facto leaders of strikes or were active in organising and representing the platform workers’ voices. The interviews focused on the impacts of the COVID-19 pandemic and their perspectives of channels to voice workers’ needs inside and outside of the platforms. In-depth interviews were also conducted with the head of platform rider syndicates(‘ nghiep doan xe om cong nghe’) in Ho Chi Minh City, the management of a major platform service provider and labour administrators from MOLISA and this southern city’s DOLISA. M: 100% Hanoi: 66.7% 81 HCMC: 33.3% The research is primarily based on a survey of platform car-drivers and bike-riders in Hanoi and Ho Chi Minh City, the two biggest local markets for ride-hailing services. While the drivers are all engaged in ridehailing services, the bike-riders work in ride-hailing and delivery. The survey was conducted between December 2020 and February 2021. Interviewers approached 7 Estimations based on the platforms’ public information by the end of 2019. Voice and representation of platform drivers in Vietnam

CHAPTER 1: Work platforms- an overview 3 CHAPTER 1: WORK PLATFORMS – AN OVERVIEW The digital economy is widely touted as a driving force for growth in upcoming decades. According to a report by Google, Temasek and Bain& Company, Southeast Asia’s digital economy has grown more than three-fold between 2015 and 2019 to reach USD100 billion. It is expected to grow to USD300 billion by 2025 and account for 8.5 percent of the region’s GDP. Vietnam’s digital economy is similarly growing at a fast pace of about 40 percent each year and is the second fastest growing one in the region, behind only Indonesia. Contributing 5 percent of the country’s GDP with a gross merchandise value of USD12 billion in 2019, Vietnam’s digital economy is expected to almost quadruple in size to reach USD43 billion by 2025(Google, Temasek and Bain 2021). The growth of the digital economy has created opportunities for new business models and services to emerge, most prominent of which are the digital platforms. These platforms are spaces for information exchanges that connect supply and demand. They act as an intermediary between individuals, and make it possible to hire workers for specific tasks for fixed terms. These platforms change companies by reducing transaction costs, making unprofitable business activities viable, challenging the market structure, and transforming labour relations. In Vietnam, the growth of the‘gig economy’ started in 2014 with the entry of ride-hailing platforms such as Grab and Uber, which operate in several markets across the region. Over the past few years, digital platforms have become more varied and encompassed many other sectors such as e-commerce, food/parcel delivery, freelance tasking, and healthtech/telemedicine. The original ride-hailing platforms like Grab and Gojek(or Go Viet) have also expanded their services to include food/parcel/grocery delivery, housecleaning, personal beauty and hairstyling services, entertainment ticket-selling, and telemedicine. In Vietnam the ride-hailing platforms, as the focus of this research, have been expanding at an annual rate of 40 percent since 2014. The value of this market is expected to reach USD2 billion by 2025. After Uber’s exit from Vietnam in 2018, Grab became the biggest ride-hailing app, accounting for a 73 percent market share. Be and Go-Viet have 16 and 10 percent of market values, respectively. 8 Business Model Work platforms present themselves as IT enterprises that act as intermediaries to facilitate the encounter between the users and the service providers rather than providing a specific service, such as transportation or delivery. In case of the ride-hailing apps, the drivers/riders are defined as ‘ business partners’ not‘ employees’. The business partners are supposed to enjoy optimum flexibility of working(or not) with the platform companies by simply turning on or off their apps. The apps also claim to provide fair and unbiased treatment of the drivers/riders as the allocation of work and payment are managed by the algorithms(De Stefano 2016). 8 https://thanhnien.vn/cong-nghe/nong-thi-truong-ung-dung-dat-xe-cong-nghe-nam-2020-1174231.html Voice and representation of platform drivers in Vietnam

4 CHAPTER 1: Work platforms- an overview At a closer look, the algorithmic management does not necessarily support the drivers/riders’ flexibility. The algorithmic management has five essential characteristics:(i) the continuous monitoring of the behaviour of workers,(ii) the constant evaluation of workers through assessments by customers,(iii) decisions are made and implemented without human interventions,(iv) workers interact through the application, thereby hindering the possibility of workers being able to communicate with each other and to personally appeal certain decisions and(v) a certain lack of transparency about how the algorithm works, due to both competitive commercial practices and the very adaptability of the algorithm(Berg et al., 2018; Möhlmann and Zalmanson, 2017). 9 In addition, a worker’s rating, usually based on algorithms that combine the evaluations of customers and the data collected from monitoring by the platform itself, involves considerable work-related consequences. Generally, an insufficient score can have a negative impact on the possibility of gaining access to more attractive or profitable job offers(reserved for workers who are best positioned in this scoring system), and it can even lead to the temporary or definitive deactivation of a worker(De Stefano, 2016). Moreover, the fact that not accepting work often causes a decrease in the rating of workers who provide services through platforms(with the risks that such a decrease implies) brings into question the true flexibility of this type of employment. Employment Relationship Because of the non-traditional business model of platform companies, existing national and international legal frameworks have not been fully applied to determine whether the employment relationship between the platforms and their‘partners’ exists. At the international level, ILO Recommendation 198(Art. 13) specifies two groups of indicators of the existence of an employment relationship. Those indicators might include: (a) the fact that the work is carried out according to the instructions and under the control of another party; involves the integration of the worker in the organisation of the enterprise; is performed solely or mainly for the benefit of another person; must be carried out personally by the worker; is carried out within specific working hours or at a workplace specified or agreed by the party requesting the work; is of a particular duration and has a certain continuity; requires the worker’s availability; or involves the provision of tools, materials and machinery by the party requesting the work; (b) periodic payment of remuneration to the worker; the fact that such remuneration constitutes the worker’s sole or principal source of income; provision of payment in kind, such as food, lodging or transport; recognition of entitlements such as weekly rest and annual holidays; payment by the party requesting the work for travel undertaken by the worker in order to carry out the work; or absence of financial risk for the worker. The nature of platform work does not necessarily satisfy all these indicators. For instance, the work does not require specific work hours or a workplace, nor does it specify a particular duration or continuity because platform workers, in theory, can self-determine when they stop working. However, the recommendation also urges member countries to allow‘ a broad range of means for determining the existence of an employment relationshi p’; and provide‘ for a legal presumption that an employment relationship exists where one or more relevant indicators is present’. Due to the growing protests of platform workers and their unions during the past five years, there were several court rulings, mainly in Europe in early 2021, to acknowledge the employment relationship between work platforms and their‘ partners’, including: • Britain: in February 2021, the UK Supreme Court ruled Uber drivers should be classified as workers. 10 • The Netherlands : in February 2021, the Amsterdam Court of Justice ruled that couriers from the Deliveroo meal delivery service are entitled to an employment contract. 11 • Italy: in February 2021, the Italian Justice Department forced the main delivery companies to regularise 60,000 riders as salaried workers. 12 • European Union : on February 24, 2021, the European Commission launched a first phase 9 https://www.ilo.org/wcmsp5/groups/public/---americas/---ro-lima/---ilo-buenos_aires/documents/publication/wcms_765150.pdf 10 https://www.supremecourt.uk/cases/docs/uksc-2019-0029-press-summary.pdf 11 https://myprivacy.dpgmedia.nl/consent?siteKey=PUBX2BuuZfEPJ6vF&callbackUrl=https%3a%2f%2fwww.volkskrant.nl%2fprivacy-wall% 2faccept%3fredirectUri%3d%252fnieuws-achtergrond%252fdeliveroo-koeriers-hebben-recht-op-een-contract-sluit-het-net-zich-rondde-platformeconomie%257eb5f357b3%252f%253futm_campaign%253dshared_earned%2526utm_medium%253dsocial%2526utm_ source%253dtwitter 12 https://www.ilsole24ore.com/art/deliveroo-glovo-uber-eats-e-foodora-sotto-scacco-pm-milanesi-sanzioni-733-milioni-ADzTFqLB Voice and representation of platform drivers in Vietnam

CHAPTER 1: Work platforms- an overview 5 consultation of social partners that w ould run for six weeks. 13 • Spain: in late February 2021, the Ministry of Labour of Spain committed to giving a maximum of three months to platforms to incorporate distributors as salaried employees. 14 In other countries, governments have stepped up efforts to extend social security benefits to gig workers. For example: • India ’s Finance Minister Nirmala Sitharaman proposed in the annual budget to extend social security benefits to gig workers. 15 • Australia ’s Labour Party proposed a plan of reforms to improve job security and provide minimum pay to those in insecure work, gig workers included. 16 • The French Minister of Labour, Elisabeth Borne, entrusted three personalities with a mission on the social and legal status of the selfemployed working for web platforms( Uber , Uber Eats or Deliveroo ). 17 • The Danish Chamber of Commerce and 3F Transport has concluded a nationwide collective agreement for food delivery services . 18 In 2019, the ILO’s Global Commission on the Future of Work called for the‘development of an international governance system for digital labour platforms(and their clients) to respect certain minimum rights and protections’(ILO 2019, p. 13). It could set minimum standards as well as develop the infrastructure necessary for facilitating payments to social security systems, and it could also establish a representative board to adjudicate disputes between platforms, clients and workers. Impacts of COVID-19 on ride-hailing platforms The ride-hailing platforms across the world were hit hard by the COVID-19 pandemic due to the application of lock-down measures. The number of trips in the United States for Uber has fallen to 70 percent of the pre-pandemic level(Hawkins, 2021). In Southeast Asia, Grab laid-off around 360 employees or almost 5 percent of its headcount, after slashing discretionary spending in the first half of 2020. 19 Grab also reported that its ride-hailing drivers saw their incomes decrease by about a double-digit percentage in April 2020, compared to October 2019. 20 However, by September 2020, its turnover recovered by 95 percent of the pre-pandemic level with the surge of food delivery services. Grab, Uber and Lyft observed an influx of new driver requests mainly from those who were laid-off, furloughed or small merchants who lost income during the pandemic. 21 In Vietnam, platform drivers and riders claimed that their income in 2020 declined by 30-50 percent compared to that in 2019(Interviews with platform drivers, December 2020-March 2021). At the same time, the ride-hailing platforms interviewed also reported a notable increase of new driver requests during 2020. However, the platforms have intentionally slowed the recruitment process as“ the market demand is low so we want to ensure that the existing drivers have sufficient orders”(Interview with the representative of Grab Vietnam, March 2021). 13 https://ec.europa.eu/commission/presscorner/detail/en/qanda_21_656 14 https://wageindicator.org/labour-laws/platformeconomy/rights-of-platform-workers 15 https://indianexpress.com/article/india/budget-2021-esic-other-social-security-safety-nets-to-cover-gig-economy-workers-7170725/ 16 https://www.theguardian.com/australia-news/2021/feb/10/anthony-albanese-promises-more-gig-workers-portable-annual-sick-and-longservice-leave 17 https://objectifaquitaine.latribune.fr/business/2021-01-27/travailleurs-independants-la-feuille-de-route-de-la-mission-sur-les-plateformesnumeriques-871610.html 18 https://fagbladet3f.dk/artikel/danes-can-now-order-takeaways-clean-conscience#:~:text=%E2%80%8B%E2%80%8B%E2%80%8B%E2% 80%8B%E2%80%8B,first%20to%20sign%20the%20agreement. 19 https://www.reuters.com/article/grab-results-idINKBN2770YV 20 https://techcrunch.com/2020/05/28/how-grab-adapted-after-covid-19-hit-its-ride-hailing-business/ 21 Ibid. Voice and representation of platform drivers in Vietnam

6 CHAPTER 2: Defining the relationship of platforms and drivers CHAPTER 2: DEFINING THE RELATIONSHIP OF PLATFORMS AND DRIVERS 2.1. Platform drivers and riders in Vietnam: Who are they? A online voluntary survey carried out by Grab Vietnam in 2020 described three types of drivers based on their working hours:(i) the professional driver-partner – those working on a full-time basis(192 hours per month or more),(ii) casual driver-partners – those working on a part-time basis(fewer than 192 hours per month, but above zero hours per month) and(iii) ad-hoc driverpartners – those working on irregular, ad-hoc basis(at times, zero hours in a given month). With the sample of more than 7,000 partners, this survey found that the vast majority of driver-partners worked on a part-time, casual basis(84.5 percent of the sample). These drivers are described as younger(under 26 years old) with a tertiary education and have other employment outside of the gig economy(Grab 2020). Our survey conducted between December 2020 and February 2021 painted a different demographic picture of platform workers. First, younger workers(18-25 years old) accounted for only one-fifth of the sample. More than half of the platform workers fell into the age group of 26-40(Figure 1). This finding is in line with a global survey by ILO in 2017, which showed that more than 60 percent of crowd workers were aged 26-40 years(ILO 2017). Figure 1: Distribution of platform workers by age groups (%) Source: Survey of platform riders and drivers in Vietnam (n=355) 70.0 60.0 58.0 50.0 40.0 % 30.0 20.6 21.1 20.0 10.0 0.3 0.0 18-25 26-40 41-60>60 Ages Second, 84.5 percent of platform workers claimed their work in the gig economy was their only and fulltime job at the time of survey. The average working hours of these full-time drivers/riders for the platforms was 75.6 hours per week , nearly double the standard 48 hours per week adopted by the Grab survey. 22 The working hours include time spent on active driving and waiting for jobs. Depending on the allocation of orders by the apps, the drivers/riders may have to wait a few minutes or a few hours between rides. Third, the survey found a strong influx of workers from the formal and semi-formal, traditional sectors into the gig economy, especially during the pandemic year of 2020. As much as 62.5 percent of platform drivers used to work for enterprises and household businesses. 23 More than one-third of these workers lost their previous jobs due to COVID-19, while the remainder hoped to generate more income from platform work. The selfemployed and students make up the second and third biggest groups in the sample. Figure 2: Prior occupations of platform drivers(%) Source: Survey of platform riders and drivers in Vietnam (n=355) PRIOR OCCUPATIONS 11.5 Students 62.5 Employees in companies/household business 22.5 Self-employed 2.0 The urge for drivers to make regular and liveable incomes accounted for more than 70 percent of the surveyed platform drivers with one to two children(Figure 3). Also, 59 percent of drivers were migrants in Hanoi and Ho Chi Minh City, living in rented accommodation (Figure 4). 22 In accordance with the 2008 Transportation Law(Art. 65), a taxi driver is not allowed to drive continuously for more than four hours and not more than 10 hours per day. 23 The researchers were not able to check whether the household businesses(such as restaurants, small garment workshops) were registered. Therefore, we refer to these household businesses as‘semi-formal’. Voice and representation of platform drivers in Vietnam

0.00.0 181-285-25 262-460-40 414-610-60 AgAegses >6>060 SelSfe-elfm-epmlopyloedyed 2.20.0 TraTdraitdioitnioanl ablikbei-kreid-reidrsers CHAPTER 2: Defining the relationship of platforms and drivers 7 Figure 3: Family status of platform workers Figure 4: Housing status of platform workers 0.208.%28% 7.074.%04% SinSgilnegplearpeanrtent SinSgilnegle 41.4411.%41% 28.2485.%45% MaMrraierrdiewditwhiotuhtout chicldhrilednren 21.2619.%69% MaMrraierrdiewditwhit0h101 chicldhild 1.113.%13% MaMrraierrdiewditwhit0h202 chicldhrilednren MaMrraierrdiewditwhit0h303 chicldhrilednren 35%35% 3%3% 3%3% 59%59% MiMgraignrta,notn, ornenrtent MiMgraignrta,notw, onwn houhsoeuse LocLaolc, aoln, ornenrtent LocLaolc, aolw, onwhnouhsoeuse Source: Survey of platform riders and drivers in Vietnam(n=355) 2.2. Defining the platform-driver relationship The platform-driver relationship is a non-traditional, non-standard one compared to the labour-management relationship in the traditional economic sector. In this research, we will explore the platform-driver relationship in three aspects: legal, economic and technological. Legal relationship As a rider or driver applying to work for a ride-hailing app, s/he will sign a business partnership contract(‘ hop dong hop tac kinh doanh’) which is regulated by Art. 28 of the 2014 Investment Law and Art. 504 of the 2015 Civil Code. In accordance with these contracts, the driver/rider is regarded as a‘ partner’ of the platform. However, among the 355 drivers and riders interviewed, none kept a copy of the contract with their platforms. Some claimed there were no such contracts, while others said they were asked to sign electronic contracts at the time of registration, but they received no(virtual or hard) copies. Regardless of the type of contract signed between the two parties, the 2019 Labour Code specifies three criteria for defining the existence of labour relations: (i) working conditions, rights and responsibilities of each party in the transaction,(ii) direct payment from the employer to the employee and(iii) under the management and monitoring of the employer(Art. 3.1). The relationship between the ride-hailing apps and the drivers clearly meets the first criterion: the regulations for platform driver-partners, such as usage of their own vehicles, driving licenses, uniforms, as well as the rights and responsibilities of each party are specified in the business partnership contract. But, the platform does not pay the driver-partners directly, which has made it difficult to determine the existence of an employment relationship. For full-time platform drivers, the income from ridehailing apps constitutes their principal source of income. Their work is also carried out according to instructions and under the control of the apps. However, platform drivers have no such entitlements as periodic rest and holidays or social insurance. In theory, the drivers can turn off the apps when they do not want to work and are not obliged to carry out work within specific working hours required by the other party. The prolonged debate in Vietnam about the legal status of ride-hailing platforms(in comparison with traditional transportation service providers) led to the promulgation of governmental Decree 10/2020/NDCP on auto transport business and conditions for auto transport business attached with Circular No.12/2020/ TT-BGTVT in 2020 by the Ministry of Transport. Decree 10 opens up two options for ride-hailing platforms. First, in order to provide transportation services in the context of controlling prices and coordinating drivers, these platforms can register themselves as transportation service providers, which also means they may need to treat their drivers as employees, rather than contractors and business partners.‘Be’, a Vietnamese ride-hailing and delivery platform, is among the few platforms registering as a service provider. 24 However, at the time of writing, Be riders and drivers have not been recognised as employees. Second, the platforms can register as transportationconnection software providers(‘ don vi cung cap phan mem ho tro ket noi van tai’), which means they are not directly involved in managing vehicles and drivers nor 24 https://vietnambiz.vn/khang-dinh-la-doanh-nghiep-van-tai-ceo-be-group-cho-rang-tat-ca-cong-ty-van-tai-deu-co-the-ap-dung-duoc-congnghe-40-113615.htm https://www.baogiaothong.vn/ung-dung-goi-xe-viet-dau-tien-nhan-la-doanh-nghiep-van-tai-d464412.html Voice and representation of platform drivers in Vietnam

8 CHAPTER 2: Defining the relationship of platforms and drivers fixing the fees(Decree 10, Art. 35). At the moment, Fast Go is the only ride-hailing and delivery platform that offers fixed rates of fees for users and drivers, while others adjust fees unilaterally. 25 This means the ride-hailing platforms do not fully meet the criteria to register as software providers only. There is also an important difference between cardrivers and bike-riders in terms of legal status. Decree 10 requires that car-drivers must register either with a transportation service provider or a transportation cooperative to provide ride-hailing services, while this requirement does not apply to bike-riders. This means there are two types of car-drivers working across the platforms:(i) drivers of transportation companies(such as traditional taxi companies) who also receive orders from platforms that partner with their companies(for example, Grab has GrabTaxi, Fast Go has Fast Taxi). In this case, the drivers are primarily employees of taxi companies,(ii) self-employed drivers registering as members of transportation cooperatives, but only working for the platforms. In the latter case, drivers have to pay a registration fee and annual fees for cooperatives to get a license for providing ride-hailing services, although cooperatives do not have actual relationships with them. The official perspective of whether or not platform drivers/riders are employees is expressed in Report No. 45/BC-BTP to the Prime Minister by the Ministry of Justice(dated March 17, 2021):“Drivers for platforms such as Grab are not paid by Grab but receive transportation fees from service users(consumers) through the platform after subtracting the service fee for the platform. On the other hand, more importantly, drivers for platforms such as Grab are not subject to Grab’s management, operation and supervision as a traditional worker, drivers have the right to turn off/ on the app to participate in providing services when needed, have the right to refuse to pick up or even cancel the car after receiving. Grab’s Code of Conduct is considered an agreement between Grab and drivers as drivers must agree to this Code of Conduct when participating in the platform[.] Grab also calls drivers who join its platform‘driver partners’, i.e. consider them business partners and not workers. As analyzed above, depending on the specific case as prescribed by law, the nature of the contract relationship between the service provider and the connected platforms is still a commercial contract(be it a brokerage contract or an authorisation contract). 26 In other words, the Ministry of Justice stated four factors for determining that there is no employment relationship between platform drivers and platform companies. Firstly, the platform driver is not the one paid by the platform company, but receives transportation fees from consumers. Secondly, they are not subject to the management, operation and supervision of the platform company, manifesting in the fact that drivers have the right to actively, that is discretion to turn off/on the application, declining a work assignment. Thirdly, the platform company’s code of conduct that app-based drivers must comply with is not imposed by the platform company unilaterally, but instead is an agreement between the two parties. Finally, app-based drivers are seen by platform companies as‘business partners’ rather than workers. The main basis of these arguments is Article 3(1), Labour Code 2019, which defines EE as“an employee who works for an employer as agreed upon, paid and subject to the management, administration and supervision of the employer.” The VGCL has not made an official statement on the issue. However, some VGCL leaders and experts have publicly expressed their views towards viewing appbased drivers as EE. In a recent statement, VGCL Vice Chairman Ngo Duy Hieu expressed his support for the view that app-based drivers were employees and their relationship with platform companies 27 was an employment relationship. Explaining this view, he said: “[Regardless of] the fact that platform companies always consider drivers as partners, but[the] daily work of drivers-workers must comply with their instructions on the division of‘rides’, on handling labour discipline, as well as on safety, uniforms, hours clearly show the employment relationship at the enterprise.” 28 With the same view, another VGCL expert – Le Dinh Quang – also emphasized: “This is clearly not a cooperation contract, but a relationship between the employer and the employee. Grab is taking advantage of strict labour regulations to give drivers the form of cooperation contracts.” 29 Quang also referenced Article 13 of the 2019 Labour Code to show that it was not important how a driver is defined by the platform itself, but that between the driver and company there was an agreement on paid employment. In our survey, drivers were asked to define their relationship with the platforms. Most drivers defined themselves as‘business partners’; yet, a much higher rate of bike-riders(40.15 percent) identified as being in an‘employment relationship’ with platforms than 25 https://www.brandsvietnam.com/18154-Cuoc-chien-ve-gia-cuoc-giua-cac-dich-vu-goi-xe-cong-nghe 26 Department of Justice, p. 16. This report consulted with numerous agencies, including the Ministry of Labour, Invalids and Social Affairs. 27 Interview with PTL(Zoom, 25/3/2021); Interview with LĐQ(TPHCM, 6/4/2021). See also: https://laodong.vn/cong-doan/tong-ldldvn-se-nghiencuu-ky-ve-kien-nghi-doi-voi-nld-la-xe-om-cong-nghe-861872.ldo 28 https://tuoitre.vn/goi-tai-xe-cong-nghe-la-doi-tac-cach-goi-luon-lach-20210322222647495.htm. 29 https://zingnews.vn/tai-xe-la-lao-dong-cua-grab-post1164432.html. Voice and representation of platform drivers in Vietnam

CHAPTER 2: Defining the relationship of platforms and drivers 9 the car-drivers(Figure 5). In the strike against Grab by Second, the rate for each trip is completely determined bike-riders in December 2020, one of the key demands by the platforms, based on the traffic density and of platform workers was that the company should sign availability of drivers in the area, among other factors. labour contracts and recognise them as employees. 30 According to the drivers, the rates were sometimes fixed so low they were unable to cover fuel expenses. Figure 5: What is your relationship with the ride-hailing However, they could not decline such orders out of fear platform company?(%) 31 of being disciplined. Three continuous refusals of orders result in deactivation of the driver’s app for one day. Figure 5: Wh S a o t ur is ce y : o S u u r rv r e e y la o t f io p n la s t h f i o p rm wi d th riv t e h r e s a ri n d d e r i h d a e i r l s in in g p V l ie a t t n fo am rm company? (Car-drivers= 81; bike-riders= 274) Third, the platforms flexibly and unilaterally adjust Car-drivers Bike-riders the bonus target and rate for drivers. Each platform provides a bonus system, whereby drivers get a cash 91.36 83.58 gift or priority in order allocation if a targeted number of trips or bonus points is reached. However, the bonus target is frequently increased whereas the gift value has reduced over the past few years, resulting in a number 40.15 22.22 43.21 31.39 of strikes by the apps’ drivers(see Table 1, Chapter 3). As the manager of a platform company argued in such a strike in 2016:“ the company holds the right to flexibly adjust the bonus system in line with our competitive strategy. The driver-partners have all the right to decide upon their working time or whether or not to continue Business partnership Employment Using platform only, with our platform”. 33 relationship no relationship Finally, the platforms unilaterally adopt complicated Economic relationship scoring systems to determine the allocation of orders to partners, which according to the drivers and riders, Role of platform Th c e o b m u p si a n n e i s e s s m in o det l e o r f m ri i d n e in -h g a t i h li e ng in p c l o a m tfo e r a m n s d is w t o o rk se in rv g e condit w io e n r s e o n f o p t la n t e f c o e r s m sarily fair and transparent. The score of as d in r t i e v r e m rs e a d n ia d ri r e i s de f r o s r (R u a se n r k s in a g nd of d im riv p e o r r s ta t n o ce n 1 c o 5 u ) nter, while driver-partners must prepare and maintain their a driver-partner is based on users’ satisfaction ratings and the number of trips completed within a month. vehicles, pay for any registration and transportation For instance, a car-driver will achieve the highest fees as well a I s nc f o u m e e l on O th th e e i r r w o o w rk n in . g c S o t n il d l, iti t o h n e s platform ranking(Gold) if s/he completes at least 450 trips in the companies play a significant role in determining the previous month, which means that s/he will be among income of drivers because of fo 4 u .0 r 3 reasons: the most 4 p .1 ri 3 oritised in allocation of orders. However, if 3.95 3.96 3.75 3.76 the next month, the driver is not able to complete the First, the platform companies unilaterally fixes the sufficient number of trips, its ranking will drop and its profit 3 -s .2 h 2 aring rate with driver-partners. According to priority in ord 3 e . r 14 allocation will be affected accordingly. the survey 2 e .6 d 9 drivers, before the profit-sharing ra 2 t . e 76 was In other words, this system encourages drivers to work 59 2.56 adjusted, they would re 2 c .5 e 4 ive an ema 2 i . l 5 f 9 rom the platform f 2 r . e 4 q 2 uently with a full-time commitment rather than company informing of the new rate. Yet, they were not taking into account their seniority and time served with able to provide feedback or negotiate the changes. A a company as per a traditional human resources system. bike-rider 32 told us: Drivers who have made long-term commitments to platforms find this system unfair:“ The old drivers like us have worked hard for years and brought profits to Full-time We all received an email from the company, stating that the sharing rate is going to increase. Of course we disagreed because the sharing rate has beco P m ar e t-ti t m o e o high, G m ra u b ch highe G r o V th ie a t n that in Be other [ride-hailing platform] companies. But we could not give feedback. No one is going to listen. We may be deactivated from the app. the company, but we receive no incentives at all. If I get sick for a week, my ranking will drop and I will not get as many orders as before and my income will be Bike r a id ff e e rs cted” C . ar T d h r e iv r e e rs are also rumours among drivers/riders about manipulation of the order-allocation system of a platform:“ A few years ago, in a forum I heard that if a rider pays VND3 million per month, he will get as many orders as he wants”. 34 30 The bike-riders on strike sent their petition to the Hanoi Federation of Labour. This strike will be discussed further in page 28. 31 Note that the respondents can choose more than one option in this question. 32 All riders and drivers participating in the survey and in-depth interviews requested anonymity. 33 http://tamlongvang.laodong.com.vn/kinh-te/tai-xe-uber-dinh-cong-phan-doi-chinh-sach-moi-tai-anh-hay-tai-a-510501.bld 34 We were not able to verify this claim. According to the interviewed riders, the operators who took bribes from riders were removed by the platform company. Voice and representation of platform drivers in Vietnam

Car-drivers 91.36 10 CHAPTER 2: Defining the relationship of platforms and drivers 83.58 Bike-riders 40.15 43.21 In our survey, the drivers/riders were asked to rank time workers o 3 n 1. p 39 latforms is much higher than partthe importance of platform companies in determ 2 in 2 i . n 2 g 2 time ones and that of car-drivers is also higher than their income and other working conditions(work bike-riders. Comparing the three biggest ride-hailing hours, work safety, work security). As shown in Figure platforms – Grab, GoViet and Be – it is interesting 6, the workers found that platform companies play that the economic dependence is highest with GoViet, a significantly more decisive role in determining their followed by Grab and the lowest with Be – the only Business partnership Employment Using platform only, income(3.84/5) than other work environments(2.59/5) platform that has registered as a transportation service relationship no relationship they experience. The economic dependence of full- provider. Figure 6: Role of platform companies in determining the income and working conditions of platform drivers and riders (Ranking of importance 1-5) Figure 6: Role of platform companies in determining the income and working conditions of platform *Ranking from the least important of 1 to most important of 5 drivers and riders (Ranking of importance 1-5) Source: Survey of platform drivers and riders Income Other working conditions 3.84 2.59 3.95 2.56 3.22 2.69 3.96 2.54 4.03 2.59 3.75 2.76 3.76 2.42 4.13 3.14 Total Full-time Part-time Grab Go Viet Be Bike riders Car drivers Technological relationship The supervision of work in the ride-hailing platforms is divided between the platform company and users. The users’ rating plays an important role in determining the drivers’ performance, access to order allocation, and even employment in the platform. A bike-rider explained:“ If I get a five-star rating for six months continuously, I will get many incentives like bonus points, better orders. If I get to just below five, the number of orders drops immediately. In the worst case, if a rider raised his voice with the client and the client complains via the app, that rider may risk being removed from the app permanently”. Although in principle, the drivers/riders have the right to reject orders suggested to them via the apps, they are subject to discipline if they do so repeatedly. For instance, three continuous rejections or more than 15 percent of trips rejected in one day by the driver may result in temporary suspension of access to the app. The drivers’ smartphones are also scrutinised by the app. Some drivers reported that they got warning messages from the platform if they download any other apps. Algorithmic management is supposed to be automatic and fair. However, the algorithms are designed in such a way that the number of drivers/riders reaching the higher rankings and bonus points is under control. The drivers and riders often find the apps prolong the time required to reach the incentive target:“ I’m supposed to finish 20 trips for a day to get to the bonus point. I started working from 6am and I completed 18 trips by 2pm. But then I did not get any more orders until 6pm. I had to stay alert in the street, waiting for hours. Sometimes I became too tired and called it a day before I got enough orders for a bonus” – said a bike rider in Ho Chi Minh City. This situation was fairly common in other countries reported Uma Rani, a senior economist at the ILO 35 : 35 https://www.bangkokpost.com/business/1980967/labour-strikes-against-the-algorithm Voice and representation of platform drivers in Vietnam

CHAPTER 2: Defining the relationship of platforms and drivers 11 This is something we have seen across the world. Supposedly the drivers are told that if they reach a target of 10 customers within a certain amount of hours or days they will receive a bonus of a particular amount, but what happens is they reach the eighth customer easily, the ninth one takes a lot of time and then they never reach the last. 2.3. Channels for communication and grievancehandling The ride-hailing and delivery platforms in this survey commonly operate four main channels for communicating with and receiving grievances from their driver-partners: 1. Hotlines via the apps: drivers can call platforms’ support centres via the apps. The workers at these call centres are provided with basic information about the companies’ rules and regulations. More complicated queries must be noted and forwarded to other departments. 2. Facebook fanpages: these fanpages offer a forum for drivers, but it is not necessary a support centre as the administrators do not normally respond to questions posted on these pages. 3. Offline gatherings: platform companies organise periodic offline gatherings with driver-partners and respond to their questions. In 2020, these gatherings were organised online instead due to the pandemic. 4. Team leaders: platforms appoint experienced drivers to act as team leaders, each covering several hundred drivers in a certain geographical area. The team leaders are responsible for communicating the platform’s policies to the members and receiving grievances, if any, from the drivers. As much as 87 percent of platform workers in the survey are aware of the grievance channels provided by platform companies. However, as reflected in interviews with platform companies and drivers/riders, these channels are intended to inform the drivers/riders about the companies’ policies rather than offering a gateway for two-way dialogue and grievance-handling. Even in the case of team leaders, the airing and documenting of workers’ opinions is not expected. The manager of a platform company explained:“ The team leaders are experienced riders who understand our policies and why we adopt these policies. Their role is to transfer this understanding to the team members”. 36 In other words, the team leaders’ responsibility is to persuade the riders to accept the company’s policies rather than representing the riders’ voice. 36 Interview conducted by the researcher, March 2021. Voice and representation of platform drivers in Vietnam

12 CHAPTER 3: Loyalty, exit or voice? Platform workers’ organising strategies CHAPTER 3: LOYALTY, EXIT OR VOICE? PLATFORM WORKERS’ ORGANISING STRATEGIES 3.1. Platform drivers’ responses to dissatisfaction “The hotlines via the app are not good. They give Hirschman(1970) categorises three categories of employees’ responses to job dissatisfaction, namely:(i) Loyalty – sticking to the firm, accepting the situation and hoping for improvement,(ii) Voice – attempts to change the situation by a range of actions from me automatic answers. So what we would do is to file our complaints over the app and fix a date when we would meet with the platform people in their office. Then, they were prepared and they would offer us a thorough response.” appeals to protests and(iii) Exit – voluntary separation or turnover from the job. Although these categories emerged from and have been mainly applied to workers in traditional sectors, they provide a useful framework “I raised my feedback over the app that I rejected the order because there were more passengers than allowed but the platform did not respond.” for analysing the responses of platform drivers to technology companies. “If we make complaints, we may get blacklisted. Our orders will be reduced. If we strike, they[the As shown in Figure 7, the overall satisfaction rate of the surveyed workers is not high – standing at 2.91 out of five. Car-drivers were more satisfied with the platforms than bike-riders(3.17 vs. 2.84). Among the platforms] will monitor our registration numbers and we may get deactivated permanently. I know some riders who posted complaints on social media and their apps were deactivated.” three biggest platforms, Go-Viet earned the highest satisfaction rate among its peers, while Grab had the lowest. Figure 7 “ : S W at e isfa c c o ti m on p ra la te in s o e f d pla a t b fo o rm ut dr t iv h e e rs/ n ri e d w ers [ w b it o h n th u e s i ] r p p la o tf l o ic rm y s b (m ut ax.: 5.0) they[the platform representatives] replied if you do n 4 ot accept it, you can quit! We are the underdogs, Figure 7: Satisfaction rates of platform drivers/riders with 3 w .5 e can not do 3.1 a 7 nything.” 3.34 3.21 their platforms(max.: 5.0) 3 2.91 2.84 2.66 Figure 7: Satisfaction rates of platform drivers/riders with their platforms(max.: 5 I . t 0) is interesting to note that while nearly 40 percent 2.5 Source: Survey of Platform Drivers(n=355) of drivers chose loyalty, meaning they accepted the 4 curre 2 nt conditions and continued work, around 20 3.5 3.17 3.34 3.21 3 2.91 2.84 2.66 2.5 perc 1. e 5 nt would exit the company and 25 percent decid 1 ed to take collective action to push for changes (Fig 0 u .5 re 8). Collective actions have proven to be the most effective way for drivers to push platforms into 2 0 dialogue. To A ta s l show Car n in Fi B g ik u e re 9, Gr 3 ab 8 pe G rc o e Vi n et t of d Be rivers 1.5 who ever raised grievances found that the platforms would only Source: Survey of Platform Drivers (n=355) respond to them upon the leverage of 1 collective actions. 0.5 0 Total Car Bike Grab Go Viet Be Figure 8: What would you do if you had a grievance(%)? Figure 8: What would you do if you had a grievance(%)? (n=354) 37 Source: Survey of Platform Drivers (n=355) 60.00 The satisfaction rates tended to climb when the 53.95 51.41 platform companies were willing to respond properl 5 y 0.00 to Fig d u r r i e v 8 e : r W s’ h q at u w e o ri u e l s d y a o n u d do g i r f i y e o v u a h n a c d e a s. gr I i n eva fa nc c e t, (% m )? ore tha 4 n 0.00 39.27 half of drivers/riders claimed they would turn to official platform communication channels first to rais 3 e 0.00 24.86 60.00 20.00 their grievances before resorting to other measures 19.21 (Figure 8). However, am 53 o .9 n 5 g drivers 51 w .41 ho raised their 50.00 grievances, only 18 percent received prompt an 1 d 0.00 satisfactory responses from the platforms(Figure 9). 40.00 39.27 In the meantime, 22 percent got no response and 18 0.00 percent received automatic or general replies: Accept Quit for another Discuss with peers platform Using the platform's Collective actions 30.00 24.86 grievance channels 20.00 19.21 37 Note that in this question, the respondents could choose more than one option. 10.00 0.00 Accept Quit for another Discuss with peers Using the Collective actions platform platform's grievance channels Voice and representation of platform drivers in Vietnam

CHAPTER 3: Loyalty, exit or voice? Platform workers’ organising strategies 13 Figure 9: How does the platform company respond to your queries/grievances?(n=277) 18% 18% 4% 22% 38% No response Threats to drivers who voice up Unsatisfactory response Only respond upon collective actions Immediate and thorough response In contrast to the supposedly non-traditional business model of work platforms in which driver-partners are granted the flexibility to turn off the app if they are not happy with the conditions, as this survey shows, only 20 percent would choose to do so. More importantly, a quarter of surveyed drivers would opt to stay and negotiate with the platforms to uphold their working conditions. In these cases, they have turned to collective action, just like workers in traditional sectors have done for centuries, to make their voices heard. Across the world, platform drivers have resorted to various forms of collective action, such as strikes(turning off their apps), demonstrations, neglection(for instance, by faking orders on apps), and legal action(by filing cases at court, as in the case of Uber workers in the United Kingdom). Strikes have been the most common: in India alone by September 2020, there had been 115 strike actions taken against local players in the ridehailing and food delivery sectors. In Indonesia, there had been 25 strikes against the two biggest ride-hailing platforms, Grab and Gojek. 38 Table 1: Strikes by ride-hailing and delivery platform workers in Vietnam Source: Review of online Vietnamese newspapers(last updated 20 April 2021) No. Time of strike Platform Number of drivers on strike Causes of strikes Outcomes 1 22-01-2016 Uber Hanoi N/A New trip target to reach bonus point. No change 2 14-08-2017 Grab Hanoi N/A Increase of extraction rate for bikeriders by 5 percent. No change 3 Jan. 2018 Grab A few hundred Increase of extraction rate from 20% to 23.6 percent. Keep the 20% rate 4 18-07-2019 GoViet A few hundred New bonus system making it harder for riders to reach target. No change Grab ceased VAT collection 5 27-08-2019 Grab HCMC A few hundred Grab collect value-added tax(VAT) for for riders and riders, but not in a transparent way. returned the collected taxes to riders 6 11-11-2019 Be Hanoi N/A Be reduced the rate of bonus(based on daily turnover) from 15 to 10 percent. No change 7 25-11-2019 GoViet Hanoi N/A Go Viet reduced the daily bonus rate from VND20,000 to VND14,000. No change 8 12-08-2020 Now A few hundreds Now applies a new set of bonus targets, which makes delivery riders work longer hours(at least 29 orders per day). No change 38 https://www.bangkokpost.com/business/1980967/labour-strikes-against-the-algorithm Voice and representation of platform drivers in Vietnam

14 CHAPTER 3: Loyalty, exit or voice? Platform workers’ organising strategies No. Time of strike Platform 9 17-08-2020 Be 10 Grab in & 7-9/12/2020 Hanoi and 11 HCMC 12 1113/12/2020 Be Hanoi Number of drivers on strike Causes of strikes Outcomes The profit-sharing formula is set high. The new bonus system makes it harder A hundred for riders to Fi r g e ur a e c 9 h : H a ow b d o oe n s u th s e p pl o at i f n or t m . c F o o m r pany re N sp o ond change instance: a rider t m o y u o s u t r q f u in er i i s es h /g 4 rie 0 va t n r c i e p s? s (n t = o 277) reach the highest bonus point. 40.00 37.55 35.00 Decree 126/2020/ND-CP on VAT for 30.00 technology-based transportation companies(3-10 percent). Grab 25.00 22.38 increased prices by 6 percent, resulting A few thousands 20.00 in higher extraction rates for drivers 18 N .41 o change 18.05 15.00 (28.3 to 32.8 percent) and riders(23.6 10.00 to 28.3 percent) leaving the remaining burden on drivers/riders. 5.00 3.61 A few hundreds 0.00 No response Only respond upon Threats to drivers collective actions who voice up Be reduced the bonus rate by 30 percent without informing its riders. Be stopped the Immediate and Unsatisfactory thor n ou e g w h bonus response response programme and adopted the previous one. In Vietnam, since 2014, there have been reportedly 13 strikes by ride-hailing and delivery platform workers (see Table 1). However, although the collective actions can force platform companies to hold serious dialogues with the drivers/riders, they were not effective in helping workers to improve their working conditions. Among the 13 reported strikes between 2016 and 2020, in only three cases(or 23 percent of the strikes) did the platforms concede and meet the requests of the drivers/riders. 3.2. Building labour agency The most important difference between workers in the gig economy and those in the traditional sectors is their isolation from other co-workers. Each has their own routes and works on different hours. While bike riders can socialise when they gather around a pick-up area, car-drivers rarely meet up physically with other drivers. In traditional sectors, Vietnamese workers have managed to stage collective actions because they can mobilise via networks built based on their employment, origins, gender, community and identity of class(Tran 2014; Clarke et al. 2007). This survey found that platform drivers and riders have also built their own networks, mostly online, to mobilise for collective action. Figure 10: How do you define your relationship with other platform drivers?(%) Figure 10: How do you define your relationship with other platform drivers?(%) 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00 69.58 Independent 75.49 71.27 61.13 Colleagues In the same boat, supportive of one another Shared interests, need to unite The foundation for any labour network is the workers’ perception of shared interests(Herod 2001). The surveyed drivers/riders have mixed perceptions of their relationships with other co-workers. While nearly 70 percent of drivers/riders believed they were independent of one another, like freelancers, as much as 71 percent found they were in the same position and thus, should support one another. More importantly, 61.1 percent of platform drivers felt they shared the same interests and thus, should unite to protect their own rights and interests(Figure 10). As they work on digital platforms, 74.8 percent of drivers use online forums(on social media) to build their networks. There are hundreds of social media groups set up by drivers independently with membership ranging from a few dozen to tens of thousands. The groups are established either based on geographical Voice and representation of platform drivers in Vietnam

CHAPTER 3: Loyalty, exit or voice? Platform workers’ organising strategies 15 areas of members(a specific district or city), their origins (migrants originating from the same province), and gender(groups of women bike-riders in Hanoi). The organisation of these informal groups varies: most of groups have no clear leadership and rarely coordinate for any common activity, while a small number of groups have informal leaders and provide coordinated support to members. I am a member of a bike-rider group on Zalo. The current group leader is brother M. They advise on how to prevent fake orders, cheaters or which routes are dangerous” - A woman bike-rider in Hanoi As the survey found, the biggest benefits for platform drivers in joining the online forums included socialisation and learning from experiences in handling problems at work. For nearly 80 percent of the drivers, the online forums also provided the foundation for mobilisation to fight for their interests(Figure 12). Figure 11: How did you get to know the online forums?(%)(n=291) Figure 11: How did you get to know the online forums? Figure 11: How did you get to know the online forums?(%)(n=291) 39 (%)(n=291) 60.00 50.00 60.00 40.00 50.00 48.11 48.11 51.20 51.20 40.55 30.00 40.00 40.55 20.00 30.00 15.12 10.00 20.00 7.90 15.12 0.00 10.00 Suggested by the platform Recommended by peers Internet 7.90 Recommended by social groups Others 0.00 Suggested by the Recommended by Internet Recommended by Others platform peers social groups Figure 12: Benefits of joining the online forums?(%)(n=248) Figure 12: Benefits of joining the online forums?(%) Figure (n 12 = : 2 Be 4 ne 8 fi ) t 4 s 0 of joining 94.8 the online forums?(% 9 ) 2 ( . n 7 =248) 79.8 94.8 92.7 59.7 69.4 79.8 69.4 59.7 Sharing orders Sharing orders Socialising Socialising Experiences in handling work problems Discussing platform policies Mobilise to protect mutual interests Experiences in handling work problems Discussing platform policies Mobilise to protect mutual interests Similar to(informal) networks of workers in traditional sectors, coordination among platform drivers is spontaneous, loose and fragmented. The leadership, if any, only emerges during the time of collective actions and is not sustained. One of the leaders of the strike by Grab bike-riders in Hanoi in December 2020 admitted: The Facebook and Zalo groups are mainly for the riders to pour out their stress. There is no formal organisation. I joined these groups in the hope to protect the interests of the riders. I tried to mobilise the riders to raise our[collective] petitions to the Federation of Labour, but only a few riders were willing to join. Everything was spontaneous and we did not get the sufficient impact.” - One leader of the 2020 strike by Grab bike-riders, interviewed in March 2021 3.3. Collective bargaining by riots As long as platform driver-partners are not recognised as employees of ride-hailing apps, they do not have the right to collective bargaining. That is not to say that in the case of car-drivers who are registered with transportation cooperatives, their official employers are the cooperatives rather than the ride-hailing apps. On the other hand, the platforms have consistently adopted the approach of‘take it or leave it’ when it comes to their policies. According to a platform, they inform drivers/riders three days before application of a new policy so partners have time to consider to‘accept it or leave the app’. However, as discussed earlier, drivers/ riders have consistently launched strikes to‘bargain by riots’ with the platforms. A typical platform driver strike starts with calls on online forums to collectively switch off apps on a specific date. Then a group of a few hundred to a few thousand riders/drivers will gather in front of the platform company’s office. Several representatives of drivers/riders meet up with company management to present their demands. The platform management will explain its new policy again and persuade the drivers/riders to accept. Otherwise, it will promise to consider the partners’ demands. To put extra pressure on the platforms, the drivers/riders also use other tactics such as: posting fake orders on the apps(to crowd out real orders), organising a parade of hundreds of drivers/riders in the streets to catch the attention of the media and authorities, or blocking the entrance to the platform company’s office building. The police’s presence during these strikes was only to maintain public order. Between 2014 and 2019, the bargaining focused mainly on economic issues such as the profit-sharing formula between the platform and partners or bonus targets and rates. Since 2019, bargaining issues have extended 39 Note that in this question, the respondents could choose more than one option. 40 Ibid. Voice and representation of platform drivers in Vietnam

16 CHAPTER 3: Loyalty, exit or voice? Platform workers’ organising strategies to non-economic topics including the interpretation of VAT regulations for platform companies(Decree 126/2020/ND-CP) and the establishment of a employment relationship between platform and drivers/ riders(see Picture 1 for an example). Although the success rate of spontaneous bargaining efforts is low (23 percent), the incidence of strikes has increased over the past two years. Particularly, in the second half of 2020, despite the COVID-19 pandemic, there were five strikes by thousands of ride-hailing and delivery bikeriders. Apparently, unless an effective formal channel for dialogue and negotiation is provided, wildcat strikes will continue to be the way for platform workers to bargain with platforms in Vietnam. Picture 1: Petition by Grab bike-riders to the FOLs in Hanoi and HCMC demanding to sign labour contracts with all riders with more than one month’s experience with the platform and the provision of social benefits 3.4. Prospects for organising gig workers Unions around the world have made various attempts to organise non-standard workers in the gig economy, ranging from legal efforts to establish an employment relationship between gig companies and workers to form union-like agencies such as guilds, work councils and worker centres(Box 1). Box 1: Unions’ Initiatives to Organise Gig Workers Union-affiliated guilds 41 Union-affiliated guilds represent an avenue for unions to form relationships with gig and platform-based workers. When conditions change and formal union recognition becomes an option, the unions can mobilise members into a formal organising drive. An example is the Independent Drivers Guild(IDG, or‘the Guild’), an affiliate of the International Association of Machinists and Aerospace Workers(IAM). IDG asserts that it represents 50,000 New York City Taxi and Limousine Commission-Licensed Uber drivers(Independent Drivers Guild, 2017). IDG was formed to help achieve wide-reaching industry reforms and create opportunities for dialogue between Uber drivers and the corporation. Work councils In Austria, Foodora app-based delivery workers recently joined together to form a work council with the support of Vida, the Austrian union representing workers in the transport and services sector. Worker representatives in such councils enjoy a wide variety of rights from information, consultation and participation, to special consultation rights in staff and economic matters, as well as to collective bargaining of work agreements. Worker Centres Over the past two decades, worker centres have emerged as a new type of institution advocating for worker rights, mainly in the United States. Operating independently and often within a limited geographical scope, they provide social services and labour resources to wage earners in a variety of sectors, especially in sectors where non-standard forms of employment predominate and in industries where workers face barriers to formal unionisation. Cooperatives SMart, a Belgium-based cooperative operating throughout Europe, has used its cooperatives 41 Guilds have existed for hundreds of years as an avenue for people to pursue mutual purpose. Occupational guilds, commonplace in preindustrial Europe, were organised by craft. Craftspeople, artisans, service providers, and manufacturers would join guilds for the purpose of mutual aid. These member-driven associations were economically important, serving not only as platform for expressing collective voice but also in securing market access for members and helping to formalise and professionalise work. That they have reemerged concurrently with the growth of non-standard employment links to the historical fact that guild membership was reserved for artisans seeking to protect and advance their interests in a context pre-dating the employer-employee relationship. Voice and representation of platform drivers in Vietnam

CHAPTER 3: Loyalty, exit or voice? Platform workers’ organising strategies 17 to create an employer where none existed, People’s Committee asking for clarification of Decree providing workers who would otherwise be 126/2020/ND-CP. 42 classified as self-employed with the security of a formal employment relationship. The cooperative will mediate between workers and their clients, so that clients become the clients of SMart, and workers become the employees of SMart. In accordance with the 2013 VGCL Charter, the syndicates do not have the right to represent their members in collective bargaining. As the platform companies are not recognised as employers of platform workers, the syndicates limit their mandate to transferring members’ demands to the local authority Legal Strategies rather than representing them in negotiations with the platforms. The New York Taxi Worker Alliance(NYTWA), discussed more extensively in the section on When asked how to better protect drivers’ rights and organising, has also helped workers bring interests, nearly 80 percent of drivers and riders called multiple cases against Uber. Initially, NYTWA for the selection of their own representatives to hold assisted two Uber drivers with filing a successful formal dialogues with platform companies on upcoming unemployment claim against the company policy changes. Almost 77 percent of drivers also asked (Rivoli, 2016). This ruling may create subsequent for more effective channels to provide feedback on opportunities for workers to gain employment- platform policies as well as work conditions, with 65.5 related protections including un F e ig m u p r l e oy 1 m 3 e : n W t, hat are t p h e e rc n e e n c t e w s a s n a t r i y ng ch to an es g t e a s bl t is o h p u r n o io te n c s t (F b ig e u tt re er 1 d 3) r . ivers’ rights and int guaranteed minimum wage, and other social protection measures. In the United Kingdom, Figure 13: What are the necessary changes to protect GMB- the union for professional drivers, was better drivers’ rights and interests?(n=351) 43 successful in bringing forth, to date, one of the largest cases regarding worker misclassification against Uber. GMB argued that despite Uber’s classification of drivers as independent Establishing unions 65.53 contractors, a more appropriate classification for drivers would be the United Kingdom’s‘worker’ status. The ruling provided 30,000 drivers across the United Kingdom access to basic employment provisions including holiday pay, minimum wage, and breaks(GMB, 2016) Representatives of drivers to dialogue with platforms 79.49 In Vietnam, organising platform bike-riders into syndicates(‘nghiep doan’) was piloted in Ho Chi Minh City since 2020. The syndicate is an occupation-based union model in the informal sector. The VGCL has established syndicates of housemaids, lottery sellers, garbage collectors, fishermen, and traditional bike-riders for the past decade. Starting with the first syndicate established in District 7 in July 2020 with 30 members, now there are 12 syndicates for platform bike-riders in Ho Chi Minh City, as reported by a representative of the HCMC Federation of Labour, in March 2021. The leaders of these syndicates were appointed by regional federations of labour. The current mandate of the syndicates is limited to providing benefits, such as New Year’s gifts or sickness benefits. During the strike by Grab riders in December 2020, the syndicates collected riders’ opinions and sent a petition to the HCMC Effective feedback channels 76.92 None of the surveyed drivers were aware of differences between enterprise unions and syndicates for selfemployed workers. What is more important for the surveyed respondents is to have an official channel of representation to voice their demands and get heard. The organiser of the bike-rider strike against Grab in Hanoi said:“ The last strike was mainly spontaneous. We made calls on Facebook, but only a small number of drivers responded. We also sent petitions to the unions[Federation of Labour of Hanoi], but it was no use because nothing was official”. 44 42 Interview with the HCMC Federation of Labour, March 2021. 43 Note that for this question, the respondents could choose more than one option. 44 Interview conducted by the researcher in March 2021. Voice and representation of platform drivers in Vietnam

18 CONCLUSION CONCLUSION The sharing economy is a new and growing business model. The regulatory frameworks of numerous countries around the world has been constantly challenged, re-interpreted and revised to fit with this new form of work. In Vietnam, whether or not the engagement between the platforms and their drivers is defined as an‘employment relationship’ remains highly controversial among State agencies. There have been calls by the VGCL and labour experts to classify the platform drivers as‘employees’ of the technology companies. 45 However, labour administrators(MOLISA and DOLISAs) described this relationship as‘business partnership’ and defined the platform drivers as‘selfemployed’ due to their flexibility of working or leaving the platforms(Interviews with representatives from MOLISA and DOLISAs in Hanoi and HCMC). However, as demonstrated in this study, the drivers – especially those who rely on the platforms as their principal source of income and full-time employment, do not necessarily enjoy a lot of flexibility. The algorithmic management that provides incentives for prolonged working hours and sanctions for rejections of work tends to tie the drivers tightly to the platforms rather than offering them more choices. The study also reaffirms that despite the new business model, the needs of the platform workers are not different from those in traditional sectors: the need to raise their voices, get their voices heard and negotiate with the platforms before any substantial changes that affect their work conditions and income are implemented. Before the question of legal classification is settled, there are a number of things the platforms and relevant authorities can do to improve the relationship between the platforms and their drivers as well as protect both parties’ interests: • The platform companies should improve their communication channels to allow for effective and regular feedback from drivers/riders. Instead of the ‘take it or leave it’ approach which has proven to be unsustainable, the platforms should allow for two-way dialogues with their partners. • There should be a categorisation of drivers/ riders who work full-time for the platforms and those who do not. More benefits and protection should be given to those who work full-time for a minimum period of time(for example: in the United Kingdom, the platform drivers who work full-time for at least one month are entitled to social and health insurance). • Instead of using traditional organising tactics, the trade unions should use internet-based channels to recruit membership from platforms for the syndicates. • The newly-established syndicates should establish official connections with the local platform management and the network of team leaders to facilitate communication and consultations. The union-affiliated guild could be a good model for these syndicates to adapt to because the best way to‘win the hearts’ of potential members is to be able to represent their voices in dialogues with the platforms. • In the case of car-drivers who are registered with cooperatives, the experience of SMart in Belgium is that the cooperatives can act as the employers of drivers and deal with the platforms as clients of the cooperatives. In that way, the sharing rate, bonus, allocation of orders and other working conditions will be negotiated between the cooperatives and platforms. In this way, the drivers will become employees and are entitled to the employees’ rights and benefits. The surest thing about gig and platform work is that it will continue to evolve. The trends and technological innovations are shaping, and will continue to shape the future of work. Workplace models that encourage platform work are impacting workers now and we should expect these trends to continue. As the ILO Director-General recently pointed out,“it is fundamentally important that we confront these challenges from the conviction that the future of work is not decided for us in advance. It is a future that we must make according to the values and the preferences that we choose as societies and through the policies that we design and implement”(ILO, 2017). 45 See this article, for instance: https://zingnews.vn/tai-xe-la-lao-dong-cua-grab-post1164432.html Voice and representation of platform drivers in Vietnam

REFERENCES 19 REFERENCES Choi, Gisan(January 2019).“Global Gig Economy Status and Implications”. International Economy Focus(in Korean) Clarke, S., Lee, C.-H.,& Chi, D. Q.(2007). From Rights to Interests: The Challenge of Industrial Relations in Vietnam. Journal of Industrial Relations, 49(4), 545– 568. https://doi.org/10.1177/0022185607080321 De Stefano, V. 2016.“The rise of the‘just-in-time workforce’: On demand work, crowdwork, and labor protection in the‘gig-economy’”, Working Paper 71. Conditions of Work and Employment Series (Geneva, ILO). Forbath, W. and Rogers, B. 2017.“A new type of labor law for a new type of worker”, New York Times, 4 September. https://www.nytimes.com/2017/09/04/ opinion/labor-laws-workers.html?emc=eta1 Google, Temasek and Bain. 2021.“E-conomy SEA 2021: Soaring 20s – The Digital SEA Decade”. Online report. Link: https://www.bain.com/globalassets/ noindex/2021/e_conomy_sea_2021_report.pdf(last accessed May 4, 2022) GMB. 2016.“GMB wins monumental victory in employment case against Uber.” GMB experts in the world of work: Newsroom. October 28. http://www. gmbwalesandsouthwest.org.uk/news/gmbwinsmonumental-victory-in-employment-case-against-uber Grab. 2020. The gig economy in Vietnam: Characteristics and recommendations for the future. Unpublished report Hawkins, A. 2021. Uber and Lyft are getting less unprofitable, but COVID-19 is still a drag on their business. The Verge(February 11, 2021). Available at: https://www.theverge.com/2021/2/11/22277043/uberlyft-earnings-q4-2020-profit-loss-covid Herod, A. 2001. Labor Internationalism and the Contradictions of Globalization: Or, Why the Local is Sometimes Still Important in a Global Economy. Antipode 33(3): 407-426 International Labour Office(ILO). 2017. Remarks by ILO Director-General Guy Ryder at the launch of the ILO Global Commission on the Future of Work. http://www.ilo.org/global/about-the-ilo/how-the-iloworks/ilo-directorgeneral/statements-and-speeches/ WCMS_570882/lang--en/index.htm International Labour Office(ILO). 2018. Digital labour platforms and the future of work: Towards decent work in the online world. Geneva: ILO. Available at: https://www.ilo.org/global/publications/books/WCMS_ 645337/lang--en/index.htm International Labour Office(ILO). 2019. Global Commission on the Future of Work: Working for a Brighter Future. Geneva: ILO. International Labour Office(ILO). 2021. World Employment and Social Outlook 2021: The role of digital labour platforms in transforming the world of work. ILO Flagship Report. Geneva: ILO, 2021. Available at: https:// www.ilo.org/wcmsp5/groups/public/---dgreports/--dcomm/---publ/documents/publication/wcms_771749. pdf Johnston, H. and Land-Kazlauskas, C.(2019). Organizing On-Demand: Representation, Voice, and Collective Bargaining in the Gig Economy. Conditions of Work and Employment Series No. 94. Geneva: ILO Pendell, Ryan; Mcfeely, Shane(August 16, 2018).“What Workplace Leaders Can Learn From the Real Gig Economy”. Gallup. Tran, A.(2014). Ties that Bind: Cultural Identity, Class, and Law in Vietnam’s Labor Resistance. Cornell University Press, Ithaca, NY Voice and representation of platform drivers in Vietnam

About the author: Do Quynh Chi is co-founder and director of the Research Centre for Employment Relations(ERC), a Vietnamese social enterprise conducting research on labour issues. She has a doctorate degree in industrial relations from the University of Sydney and her current research is focused on labour relations in manufacturing industries, rural migration, labour law and global supply chains. The views expressed in this publication are not necessarily those of Friedrich-Ebert-Stiftung. Imprint © 2022 Friedrich-Ebert-Stiftung Vietnam Office 7 Ba Huyen Thanh Quan, Ba Dinh Hanoi, Vietnam IPo Box 44 Responsible: Axel Blaschke| Resident Representative t:+84 24 3845 5108 vietnam.fes.de Vietnam To order publication: mail@fes-vietnam.org Commercial use of all media published by FriedrichEbert-Stiftung(FES) is not permitted without the written consent of FES. Friedrich-Ebert-Stiftung(FES) is the oldest political foundation in Germany. Founded in 1925, FES is named after Friedrich Ebert, the first democratically elected president of Germany. FES opened its Vietnam Office in Hanoi in 1990. It was one of the first international nonprofit organisations to work in Vietnam. As a political foundation, it works on the principles of promoting social justice and political participation. The Vietnamese partner organizations of FES are central actors in the fields of politics, economic development, climate change, energy, and environmental policies. asia.fes.de