BEHIND THE MYTH OF ‘BUSINESS PARTNERSHIPS’ VULNERABLE APP-BASED DRIVERS IN URGENT NEED OF LEGAL PROTECTION IN VIET NAM DO HAI HA ET AL. INFORMATION AND COMMUNICATIONS PUBLISHING HOUSE BEHIND THE MYTH OF ‘BUSINESS PARTNERSHIPS’ VULNERABLE APP-BASED DRIVERS IN URGENT NEED OF LEGAL PROTECTION IN VIET NAM JUNE 2024 2 LIST OF AUTHORS DO Hai Ha: Research associate, Asian Law Centre, Melbourne Law School, University of Melbourne Lecturer, Ho Chi Minh City University of Law Toan LE: Senior lecturer, Swinburne Law School, School of Business, Law and Entrepreneurship, Swinburne University HUYNH Thi Ngoc Tuyet: Retired senior research fellow, Southern Institute of Social Sciences NGUYEN Thi Minh Chau: Executive Director, Centre for Environment and Development, Southern Institute of Social Sciences DINH Thi Chien: Lecturer, Ho Chi Minh City University of Law LUONG Minh Son: Lecturer, Ho Chi Minh City University of Law NGUYEN Quoc Thang: Lecturer, University of Labour and Social Affairs II 3 CONTENTS LIST OF ABBREVIATIONS................................................................................................... 5 LIST OF FIGURES................................................................................................................ 6 LIST OF TABLES................................................................................................................. 7 FOREWORD........................................................................................................................ 8 EXECUTIVE SUMMARY....................................................................................................... 9 INTRODUCTION.................................................................................................................. 10 Starting Point of the Research.................................................................................................10 Research Aims and Focus........................................................................................................11 Research Methods and Data Collection...................................................................................12 Structure of the Report...........................................................................................................13 PART I................................................................................................................................ 14 The Platform Economy and Platform Work: A Snapshot..........................................................15 Platform Work and the Rise of App-based Ride-hailing and Delivery Services in Viet Nam.......17 PART II............................................................................................................................... 20 Low and Vulnerable Incomes..................................................................................................21 Long Working Hours with Limited Rest...................................................................................25 Lack of Safety at Work............................................................................................................28 Algorithmic Control and Unfair Management..........................................................................34 Algorithmic Control................................................................................................................31 Unfair Management...............................................................................................................34 Low Job Security.....................................................................................................................36 Limited Social Protection.........................................................................................................37 Weak Voice and Representation..............................................................................................40 Summative Remarks................................................................................................................43 PART III.............................................................................................................................. 45 Unclear Status of Platform Motorbike Drivers under Vietnamese Law......................................46 Problems with the Employee Test and Its Application to Platform Drivers................................49 Working under an Agreement.................................................................................................49 Receiving Wages.....................................................................................................................51 Working under Management, Direction, and Supervision.......................................................53 Summative Remarks................................................................................................................55 The Impact of Legal Uncertainties on the Vulnerability of Platform Motorbike Drivers.............56 4 PART IV.............................................................................................................................. 59 Clarifying Existing Criteria to Identify Employment Relationships..............................................60 Receipt of Wages....................................................................................................................60 Subordination......................................................................................................................... 60 Administrative Guidance as the Main Means..........................................................................63 Reforming the Legal Framework for Identifying Employment Relationships.............................63 The Shift towards More Flexible Work Arrangements and Regulatory Responses around the World...............................................................................................................................64 Recommendation 198 as a Generic Guide..............................................................................64 Introducing a Rebuttable Presumption of Employment Status.................................................70 Platform Workers as Third-Category Workers?........................................................................73 Enacting Platform-specific Protection......................................................................................75 Summary and Recommendations............................................................................................78 CONCLUSION..................................................................................................................... 80 ENDNOTES......................................................................................................................... 82 REFERENCES...................................................................................................................... 90 Articles, Books, Papers, and Reports........................................................................................90 Code of Interviews..................................................................................................................104 Legal Instruments....................................................................................................................106 Others..................................................................................................................................... 107 5 LIST OF ABBREVIATIONS CEACR COC DFS EU GPS HCMC ILO MOJ MOLISA MPI OECD OHS USD VGCL VND WRO Committee of Experts on the Application of Conventions and Recommendations Code of Conduct Digital Future Society European Union Global positioning system Hồ Chí Minh City International Labour Organisation Ministry of Justice Ministry of Labour, War Invalids, and Social Affairs Ministry of Planning and Investment Organisation for Economic Cooperation and Development Occupational health and safety United States Dollar Viet Nam General Confederation of Labour Viet Nam Dong 1 Worker representative organisation 6 LIST OF FIGURES Figure 1. Parties involved in app-based ride-hailing and delivery work..............................................19 Figure 2. Factors impacting app-based motorbike drivers’ incomes..................................................25 Figure 3. Drivers’ assessment of platforms’ safety measures.............................................................30 Figure 4. Platform indicators for customer feedback and ratings......................................................33 Figure 5. Drivers’ rating of the effectiveness of periodic meetings with grab....................................41 Figure 6. Drivers’ assessment of platforms’ openness to dialogue....................................................41 7 LIST OF TABLES Table 1. App-based motorbike drivers’ average monthly incomes....................................................21 Table 2. Full-time app-based motorbike drivers’ average monthly incomes per platform..................22 Table 3. Full-time app-based motorbike drivers’ average working hours in a month.........................26 Table 4. Drivers’ normal rest places..................................................................................................27 Table 5. Sources of danger reported by drivers................................................................................29 Table 6. Health issues reported by drivers.......................................................................................31 Table 7. Accident insurance provided by platforms..........................................................................38 Table 8. Healthcare insurance provided by platforms.......................................................................39 Table 9. Platforms’ hierarchical powers over drivers........................................................................62 8 FOREWORD This report highlights the precarious working conditions and vulnerability of app-based motorbike drivers in Viet Nam and their linkage with an outdated labour law framework. It not only pinpoints the problems, but also considers solutions to enhancing legal protection for platform drivers, taking into consideration regulatory innovations around the world and the unique context of Viet Nam. Data collection was undertaken between 2019 and 2023. This consisted of a quantitative survey of workers from four major ride-hailing and delivery platforms, conducted in Ho Chi Minh City(HCMC) between February and March 2021. From May 2019 to August 2023, the researchers also interviewed and re-interviewed 50 workers and 15 other actors, including company managers, union cadres, state officials, and labour law experts. Moreover, data for this research was drawn from observations made by the researchers and surveyors during their conversations with workers and participation in workers’ Facebook groups. Finally, the research analysed numerous documents including-among other thingsplatforms’ contracts, policies, and documents; statutory laws; administrative regulations; court judgments; government papers and reports; academic works; and newspaper articles. These documents were collected until October 2023. The authors would like to express their deep appreciation to the Friedrich-Ebert-Stiftung(FES) for funding this major project. We are particularly indebted to Axel Blaschke, Timo Rinke, Nguyen Thi Ha Giang, Nguyen Thanh Thuy, and Pham Hung Son for their patience, support, and valuable advice throughout the project. Our gratitude also goes to the people involved in data collection, especially the interviewees and surveyors. We appreciate Dr. Do Quynh Chi, Dr. Pham Sy Thanh, and participants in the webinar on Labour Research in Digital Platforms in Viet Nam organised by the Southern Institute of Social Sciences, FES, and the Viet Nam Labour Research Network in October 2021 for their insightful feedback on our early work. Last, but not least, special thanks go to Laurence Newman and Simon Drought for their proofreading. All errors are ours. 9 EXECUTIVE SUMMARY The rapid rise of digital labour platforms in Viet Nam, especially since the arrival of Grab and Uber in 2014, has impacted the lives of hundreds of thousands of workers(if not more). So far, policy and media discourse in the Southeast Asian country has largely emphasised the benefits of these platforms for workers-namely, job creation, high pay, autonomy, and flexible working. There has been limited discussion of the negative aspects of platform work, especially in policy and scholarly forums in Viet Nam. However, growing strikes-including multiple regional strikes taking place in late 2020-by motorbike ride-hailing and delivery platform workers(hereafter referred to as“platform drivers” or“app-based drivers”) in this country indicate that this work may be structured in ways that have negative impacts on workers and that the much-touted benefits of digital labour platforms are more speculative than evidence-based. To provide an evidence-based foundation for policymakers and stakeholders to determine the impacts of this fast-growing business model on workers, this study evaluates the work, pay, and conditions of platform workers in Viet Nam with a particular focus on the ride-hailing and delivery sector and motorbike drivers. It reveals that platform drivers in Viet Nam are highly vulnerable in a multitude of ways. These range from receiving low, unstable pay and working in harsh, unsafe conditions to lacking basic social protections, suffering from insecure jobs and managerial abuses, and having a weak voice and representation in determining pay and conditions. Moreover, their autonomy and flexibility at work is substantially restricted in reality due to their heavy financial dependence on platform work and the algorithmic and traditional controls exerted by platform operators. The dominant narrative that platform workers are well paid, and that the job allows autonomy and flexibility, is a myth for a majority of app-based motorbike drivers. The study further reveals that the existing labour law’s ambiguity regarding the status of platform drivers as employees has played a role in their vulnerability. It has enabled digital platforms to classify app-based motorbike drivers as independent contractors to avoid ensuring that these drivers enjoy the labour rights to which employees are generally entitled. These loopholes have also permitted ride-hailing and delivery platforms to maintain unfair contractual arrangements with motorbike drivers and have prevented drivers from seeking to redress their precarious situation. The acute vulnerability of platform drivers and its close linkage with their vague legal status underlines the need to strengthen legal protection for these workers. This study calls for greater clarity of the criteria used to determine the existence of an employment relationship under Viet Nam’s Labour Code, particularly the‘receipt-of-wages’ and‘subordination’ criteria. Furthermore, it suggests that the legal framework’s distinction between employed and self-employed workers be reformed substantially, taking account of ILO Recommendation 198 and legal developments in other jurisdictions. Another possible reform option, separately or in conjunction with the clarification of the scope of the labour law, is introducing a rebuttable presumption of employment status in favour of platform drivers, in particular, or platform workers in general. Lastly, the Vietnamese government could enact legislation that specifically deals with these workers and grants them basic collective labour rights and greater protection, including in respect of algorithmic management and occupational health and safety. 10 INTRODUCTION Starting Point of the Research “Industry 4.0 Era: Motorbike Taxi Drivers’ Salaries are from Three to Four Times Higher than Public Servants’ Salaries.”(Hà Kiều, 2019),“Driving GrabBike to Buy a Mercedes Car: Pipe Dream or Feasible Goal?”(Tiểu Phượng, 2020), Attentiongrabbing headlines like these can easily be found in Vietnamese newspapers, including those with the biggest readership. They create a strongly positive impression about a newly-emerging occupation in this Southeast Asian country: Appbased motorbike driving( xe ôm công nghệ). 2 Frequently, it is portrayed in the media as a desirable job with high pay and stability(Hà, 2018). Known as‘partners’( đối tác) of digital ride-hailing and/or delivery platforms, appbased motorbike drivers are hailed as their own boss, with the freedom to determine their work schedule(Hà, 2018). Explicitly or not, these drivers appear in the media as beneficiaries of business innovations and technological advancements (Hà, 2018). The emergence of app-based motorbike drivers is a result of the explosion of platformbased ride-hailing and delivery services in Viet Nam, especially since the arrival of Grab and Uber in 2014(Trương Quốc Lâm et al, 2021). Departing from traditional transportation services, these platforms give customers quick access to transportation on an on-demand basis through software applications(apps) installed on smartphones. These apps offer a range of utilities for customers, including booking, price and time estimation, remote access to products and suppliers, real-time tracking of trips and drivers, electronic payment, and driver ratings. As elsewhere, these innovative transportation services have quickly become popular in Viet Nam. It is estimated that the market value of the app-based ride-hailing and food delivery sector in this country increased eight times from 0.2 billion USD in 2015 to 1.6 billion USD in 2020, and is expected to reach seven billion USD in 2025 (Google, Temasek and Bain& Company, 2020). There are no official statistics on the workforce of this quickly expanding sector. In 2021, Grab, Be, and Gojek-the three biggest platforms at that time-were estimated to have nearly 500,000 drivers(Viễn Thông, 2021). However, Grab, Be, and Xanh SM claimed to have roughly 690,000 motorbike drivers(Nhật Minh, 2024). Of note, this number does not include car drivers or several other platforms, including Gojek. Looking more broadly, the boom of app-based ride-hailing and delivery services in Viet Nam is part of the rapid growth of economic activities conducted on digital platforms, often known as the‘platform economy,’ 3 across the world (Richard Sheposh, 2020; ILO, 2021). These include digital labour platforms that match ‘supply and demand for paid work,’(Cyrille Schwellnus, Assaf Geva, Mathilde Pak, and Rafael Veiel, 2019; ILO, 2021) such as the ridehailing and delivery apps mentioned above. The work performed through these apps is usually called‘platform work’(ILO, 2021). Similarly, those engaging with such platforms to match themselves with customers and provide services for money are commonly referred to as‘platform workers’(Nicola Duell, 2020). The growth of platform work is considered‘one of the most important transformations in the world of work during the past decade.’(Janine Berg et al , 2018). The rise of platform work has triggered extensive discussion around the world on its costs and benefits. On the one hand, it is praised as a new source of employment and a new form of work that offers freedom and flexibility to workers (DFS, 2019; Workbank, 2016). On the other, platform work is criticised as another form of ‘precarious work’ and a means for businesses to evade labour standards and regulations(Andrew Stewart and Jim Stanford, 2017). Although the impact of platform work remains contested, the emergent literature suggests that it generates 11 both opportunities and challenges for workers and societies(ILO, 2021). Notwithstanding this, the discourse around platform work in Viet Nam has been relatively one-sided. Only since app-based motorbike drivers began to organise prominent collective protests, such as multiple regional strikes regarding a new tax policy for motorbike ridehailing and delivery services in late 2020, has local media increased its coverage of the negative side of platform work(Mai Hà, 2020) while often retaining an optimistic view(Nguyệt Anh, 2022). Particularly, policy and scholarly discourse tends to look at platform work through a favourable lens, emphasising job creation, high incomes, flexibility, and the autonomy of work(MPI, 2020). Exhibiting such an upbeat view on the labour impact of platform business models, a government report remarks: [The platform economy] facilitates the transformation of working arrangements from‘master-servant’ relationships to‘partnerships’ between workers and enterprises. This in turn results in transformation in the[nature] of workers’ wages/incomes(from wages[and] remunerations to shared profits) which together with increased productivity will enable a quick rise in workers’ earnings....[The platform economy] transforms employment models in a way favourable to workers (MPI, 2020). Research Aims and Focus This research challenges the mainstream view in Viet Nam about platform work, focusing on the ride-hailing and delivery sector. This has been one of the fastest-growing sectors in the booming platform economy(Trương et al, 2021; ILO, 2022). Therefore, it provides a good case to examine the general situation of platform workers in this Southeast Asian state. Significantly, the optimistic view that prevails in local discourse on platform work has been built largely on a mistaken perception of app-based drivers(MPI, 2020). While assessing the situation of app-based drivers, this study concentrates chiefly on motorbike drivers. Although motorbike drivers account for the majority of the workforce in the ride-hailing and delivery sectors(Truong et all, 2021; Nguyễn Thành Nhân, 2018), regulatory discussion in Viet Nam has paid more attention to car drivers( Phạm Hoài Huấn, 2021). Meanwhile, unlike in the case of app-based car driving, no specific regulatory measure has been adopted to address issues related to app-based motorbike driving. This would, arguably, render app-based motorbike drivers more vulnerable than app-based car drivers, at least from a legal perspective. Drawing on empirical data, the study argues that platform drivers in Viet Nam are highly vulnerable in several ways. These include low and precarious incomes; harsh, unsafe working conditions; managerial abuses and insecure jobs; a lack of basic social protections; and a weak voice and representation in determining pay and conditions. Furthermore, as a result of their heavy financial dependence on platform work and the algorithmic and traditional controls exerted by platform companies, the apparent freedom and flexibility of platform workers are substantially diminished in practice. Simply put, the narrative about app-based drivers as business partners of platform companies with desirable earnings and flexible work schedules is, essentially, a myth. Moreover, this research reveals that the vulnerability of app-based drivers is closely related to the failure to clearly determine whether or not these workers are employees under Vietnamese labour law. This uncertainty has allowed ridehailing and delivery platforms to classify their workers as independent contractors-excluding them from the coverage of labour and social security laws. Even worse, the absence of 12 legal protection has created opportunities for these platforms to structure contractual work arrangements in a way that is remarkably unfair to app-based drivers and inhibits their ability to redress the situation. In view of that, the study calls for enhanced legal protection for app-based drivers in Viet Nam. It suggests that the criteria to determine the existence of employment relationships under the 2019 Labour Code be better defined, especially in terms of whether the employee receives wages and whether he or she works under the ‘management, direction,[and] supervision’ of the employer. Furthermore, the legal framework for determining the existence of employment relationships should be thoroughly revised to address the increasing mismatch between traditional approaches to classifying workers and rapid changes in the world of work, drawing upon ILO Recommendation No. 198 on Employment Relationships and legal developments in other jurisdictions. Apart from that, the Vietnamese government could also introduce a rebuttable presumption of employment status for appbased drivers or platform workers, in general, and adopt regulations specifically dealing with these workers. Such regulations would establish minimum standards for platform workers, whether or not they are legally considered employees, at least in relation to algorithmic management, occupational health and safety, and basic collective labour rights and social protections. Research Methods and Data Collection This study adopts a mixed-method approach to data collection-both quantitative and qualitative. The first source of data is a survey of app-based motorbike drivers undertaken in HCMC, Viet Nam’s most important economic hub, where the platform economy is the most advanced and dynamic. The survey took place between February and March 2021, with drivers selected from the following platforms: Grab: Entering Viet Nam in 2014, the Singaporebased‘super app’ quickly expanded from(car and motorbike) ride-hailing to other services, including parcel delivery, food delivery, and grocery delivery, cashless payment, and hotel bookings. The platform became a dominant player in the local ride-hailing sector after its merger with Uber, with 73 percent market share in 2019(Viễn Thông, 2020). Holding 37 percent market share in 2020 and 45 percent in 2023, it is also the leading competitor in the food delivery market(Nhật Minh, 2020; Công Trung, 2023). Grab claimed to have 190,000 drivers in 2020 (Hoàng Nam, 2020). More recently, in 2024, it is estimated to have around 300,000 motorbike drivers(Nhật Minh, 2020). Be: Launched in 2018, the home-grown app is owned by Be Group. Be provides both ridehailing and delivery services, including car and motorbike rides, package delivery, food delivery, and grocery shopping. From 2019 to 2023, the platform was the second-biggest player in the ride-hailing and delivery sector, with 16 percent market share and a reported workforce of 100,000 drivers in 2020(Viễn Thông, 2020). Like Grab, Be is said to have about 300,000 motorbike drivers in 2024(Nhật, 2024). Now: Originally a Vietnamese start up, Now has operated as a food delivery app since 2015. Following its acquisition by Singaporean tech conglomerate Sea Group in 2017, the app was re-branded as ShopeeFood in 2021. Now/ ShopeeFood is the second-biggest player in food delivery services, with 34 percent and 41 percent market share in 2020 and 2023, respectively (Nhật, 2024). At the time of the survey, the platform was reported to have at least 20,000 motorbike drivers in 2020(Thiên Trang, 2020). Four years later, its workforce has increased 2.5 times to 50,000 drivers(Lam Le and Nhung Nguyen, 2024). Baemin: The South Korean-owned app entered the Viet Nam market in 2019, focusing on food delivery services. A year later, it became the third-most popular food delivery app accounting 13 for 16 percent market share, claiming tens of thousands of drivers in mid-2020(Nhat, 2020). Baemin, however, ceased its operation in Viet Nam in December 2023(Công Trung, 2023). The total number of drivers(aged from 18 to 63) responding to the survey was 203. 4 As Grab held a dominant position in the ride-hailing and delivery market, 99 respondents were recruited from this platform. The rest of the sample was spread roughly evenly among the other three platforms to ensure the statistical significance of the data. App-based drivers were randomly approached at their regular parking places, such as pavements outside apartment buildings and shopping malls and spaces under bridges. A snowballing technique was also used to identify additional participants. Each participant was invited to complete a 60-minute questionnaire with the assistance of a surveyor. Upon the completion of the questionnaire, the respondent was paid 150,000 VND(6 USD) to compensate for lost working time. Another major source of data in this study is in-depth interviews conducted from 2019 to 2023. Fifty app-based motorbike drivers were interviewed, including three women. We also interviewed 15 other actors, including current and former managers of platform operators, government officials, union cadres at local and national levels, and lawyers. Interviews were semi-structured, typically lasting one hour, with some up to three hours. Worker interviewees were recruited and compensated like survey respondents. Meanwhile, other interviewees were identified from the network of the researchers, and were not paid for their participation. Interview and survey data has been de-identified and pseudonymised to protect anonymity and confidentiality. Apart from surveys and interviews, data in this research was also drawn from observations. While assisting worker participants to complete the questionnaire, surveyors were also observing and taking note of their gestures, attitudes, and expressions in response to the questionnaire as well as the surrounding environment(including fellow workers). Additionally, the researchers participated in Facebook groups of platform drivers to observe how they interacted with each other and other actors; how they perceived pay and conditions; and how they relied on social media to voice concerns, mobilise support, and seek changes. We focused particularly on the collective protests organised by app-based motorbike drivers in late 2020 with respect to the introduction of a new tax policy for the ridehailing sector. Lastly, the study elicits data from documentary and online research. Company websites were an important resource to understand the background, policies, procedures, and practices of the surveyed platforms. We also obtained standard agreements between platforms and their drivers from lawyers and workers. In addition, we collected and analysed-among other thingsstatutory laws, administrative regulations, court judgments, government papers and reports, academic works, and newspaper articles. Structure of the Report The report contains four main parts. Part One explains major concepts and provides a general background to the platform economy and platform work, with some focus on Viet Nam and its ever-growing ride-hailing and delivery sectors. Drawing from survey and interview data, Part Two highlights the vulnerabilities of appbased motorbike drivers in Viet Nam, specifically in respect of their incomes, working time, labour safety, labour management, job security, social protection, and voice and representation. Part Three highlights the failure of the current legal framework to deliver a clear answer on the employee status of app-based drivers and its contribution to their vulnerabilities. Part Four considers a range of regulatory solutions to enhance legal protection for these workers, taking into consideration local needs and contexts as well as international and foreign experiences. PART I THE PLATFORM ECONOMY, PLATFORM WORK, AND THE RISE OF APP-BASED RIDE-HAILING AND DELIVERY SERVICES IN VIETNAM 15 THE PLATFORM ECONOMY, PLATFORM WORK, AND THE RISE OF APP-BASED RIDE-HAILING AND DELIVERY SERVICES IN VIETNAM The Platform Economy and Platform Work: A Snapshot The increasing popularity of the Internet; the spread of smartphones, computers, and servers; and several advancements in digital technologies, like cloud computing and artificial intelligence, have led to a rapid rise of the platform economy in the past decade(Richard Sheposh, 2020; ILO, 2021). Nowadays, digital platforms have penetrated virtually all economic sectors, ranging from transportation and deliveries to tourism, hospitality, finance, and professional services-to name a few(ILO, 2022). As evidence of this, a study conducted in 2018 identified 242 platform companies with a market value of 7.176 trillion USD, a 67 percent increase from the 4.304 trillion USD reported by a survey conducted two years earlier(Stephanie Hottenhuis et al, 2018; Peter Evans and Annabelle Gawer, 2016). By 2022, five of the ten biggest companies in the world were part of this fast-growing platform economy (ILO, 2022). Briefly, a digital platform is‘a digital service that facilitates interactions between two or more distinct but interdependent sets of users (whether firms or individuals) who interact through the service via the Internet.’(OECD, 2019). Platform-based business represents a departure from traditional linear business in which value is created at every stage of a supply chain, like sourcing, manufacturing, distribution, marketing, and after-sales services(ILO, 2022). By bringing different users(customers, service providers, producers, suppliers, and advertisers) together, online platforms generate value by facilitating exchanges or transactions or through fostering innovation among these actors(Evans and Gawer, 2017). Online platforms can be used to facilitate the exchange of goods, services, or software(ILO, 2021). This report focuses only on digital labour platforms: Those which‘match workers on one side of the market to customers(final consumers or businesses) on the other side on a per-service (gig) basis.’(Schwellnus et al, 2019). Despite the absence of comprehensive statistics, there are several indicators of the quick proliferation of digital labour platforms around the world. According to one estimate, the annual revenue of these platforms was no less than 50 billion USD in 2019(Mark Graham et al, 2019). From 2010 to 2020, the number of digital labour platforms rose more than five times(from 142 to 777)(ILO, 2021). In several countries, platform workers were estimated to range from 0.3 percent to 22 percent of the adult population(de Groen et al, 2018). Although these workers remain relatively small in number, they are expected to continue increasing quickly and to profoundly reshape the world of work in the near future(DFS, 2019). Labour platforms are regularly categorised into two groups. The first group contains online webbased platforms, also known as crowd work, global, or online-to-online platforms, where work is outsourced to a geographically dispersed crowd and executed virtually(ILO, 2018; Berg et al, 2018). These platforms mediate various services, including-among other things-data processing, translation, graphic design, and medical advice(ILO, 2021). The second contains location-based platforms, also known as onlocation, in situ, or online-to-offline platforms, where work is distributed to individuals through software apps and performed in a geographically specified location(Berg et al, 2018; DFS, 2019; ILO, 2021). On-location platforms normally involve more“manual” tasks, such as food delivery, taxi services, domestic work, care provision, and home services(ILO, 2021). Notwithstanding the divergences mentioned above, location- and web-based platforms share BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 16 important similarities. Both rely on the Internet and digital technologies to connect the supply and demand for paid work(Valerio De Stefano, 2016). In both cases, work is disaggregated into short-term tasks(gigs) which are outsourced to a‘just-in-time’ workforce and compensated on a ‘pay-as-you-go’ basis whilst workers are required to supply their own capital equipment(ILO, 2021; Meijerink, Jansen and Daskalova, 2021). Unlike traditional employment relationships, platform work-whether on-location or virtualinvolves at least three parties, with platforms mediating transactions between workers, customers, and sometimes suppliers(like grocery shops or restaurants)(Meijerink, Jansen and Daskalova, 2017). In these arrangements, platform workers are normally classified as freelancers, independent contractors, or selfemployed workers(ILO, 2021; Meijerink, Jansen and Daskalova, 2021). As digital labour platforms are spreading worldwide, they generate opportunities for capital, labour, and society(ILO, 2021). These platforms have enabled businesses to expand market reach; access a larger labour force; and improve efficiency, productivity, and profitability (ILO, 2021). In addition, they have increased convenience, transparency, and the quality of service for consumers(Hottenhuis et al, 2016). For workers, the main benefit of platform work is job creation, particularly for marginalised groups like disabled, young, and migrant workers and stay-at-home mothers, thanks to flexible working arrangements and low entry barriers(Annabelle Gawer, 2021). Despite this, at the same time, the boom in digital labour platforms has produced significant challenges. One of these is to maintain fair competition, as platform operators may utilise their substantial marketing power to create dependence in enterprises associated with them, particularly small and medium-sized enterprises (ILO, 2022; Hottenhuis et al, 2018). Additionally, platform businesses may have an unfair advantage over traditional businesses due to the lack of regulation of platform-based business models, including in connection with labour issues(ILO, 2022; ILO, 2021). They may also improperly collect and utilise users’ data, and violate their privacy and other rights(Hottenhuis et al, 2018). Further, platform work can be a source of informal labour, causing difficulties for the collection of taxes and social contributions (ILO, 2022). Significantly, digital labour platforms present major challenges for workers to achieve decent work. Critics have pointed to several problems associated with platform work, such as job insecurity, low wages, lack of social protection, harsh working conditions, exclusion from basic labour rights, and the‘demutualisation of risks’ (a shift of risks and responsibilities mainly to individual workers)(De Stefano, 2022; Gawer, 2021). These problems are closely related to the prevalence of informal and non-standard work arrangements which excludes platform workers from coverage of the labour law and social protection(DFS, 2019; ILO, 2021).To make matters worse, platform workers often suffer from a serious imbalance in bargaining power which is inherent in the structure of most platforms and face significant practical and legal impediments to collective organisation(DFS, 2019; Gawer, 2021). For these reasons, digital labour platforms have been criticised for creating ‘conditions of precarity and vulnerability’ among platform workers(Mohammad Amir Anwar and Mark Graham, 2021), opening space for‘a severe commodification of work’(De Stefano, 2019). There has been an increasing call that platform work be seen and addressed as a part of broader problems of the changing world of work: The informalisation of labour and the spreading of atypical forms of employment(De Stefano, 2022; DFS, 2019). Despite the ongoing debate over the costs and benefits of platform work, a growing consensus has arisen that the regulation of platform work must balance different interests(De Stefano, 2019; Hottenhuis et al, 2018). On the one hand, regulatory measures should foster business and technological innovations by platform operators BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 17 (Hottenhuis et al, 2018). On the other, they also need to address the concerns of workers and other(traditional) businesses in respect of decent work(Hottenhuis et al, 2018). Though still nascent and fragmentary, regulatory innovations have been variously initiated at both national and transnational levels to achieve this two-fold objective(ILO, 2021). Platform Work and the rise of app-based Ride-hailing and delivery services in Viet Nam As in other parts of the world, platform work has expanded at an exponential rate in Viet Nam, starting with ride-hailing services(Trương et al, 2021). Ride-hailing is an on-demand or prearranged service in which passengers seeking a ride are connected with drivers of personal vehicles through a mobile app operated by a digital platform(ILO, 2021; Susan Shaheen, Adam Cohen, and Ismail Zohdy, 2016). The app provides passengers with updates at every stage, including estimated waiting time, trip fares, and ride duration, and enables them to track their trips and drivers in real time(ILO, 2021). It is not the drivers but the platform that determines trip fares and the percentage that it takes from the total fare of each trip as commission(Andrés Fiebauln and Alejandro Tirachini, 2020). For example, Grab and Be’s commission rates for motorbike ride-hailing services were set at 25.93 percent and 33 percent in 2023, respectively (Grab, 2023; Hoàng Thuỳ, 2023). The first ride-hailing platform-Easy Taxi, a popular taxi platform originally from Brazil-came to Viet Nam in 2013. However, the ride-hailing sector began to explode only after the entrance of Grab and Uber in 2014(Trương et al, 2021). Very quickly, these platforms overtook traditional taxi businesses and became the biggest taxi service operators. In mid-2017, they had around 25,000 car taxi drivers in HCMC. To put that in perspective, Mai Linh and Vinasun-two traditional taxi companies that once dominated BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 18 the local market-had only about 16,000 car taxi drivers(Nguyễn Nhật Tiến, 2017). These platforms have also mediated motorbike taxi services, a favourite transportation option for local people, which had previously been occupied by individual drivers(Trương et al, 2021). There were reportedly more than 400,000 app-based motorbike drivers in 2019(Thế Lâm, 2019), more than double the number of app-based car drivers (Anh Trọng, 2020a). Following the merger of Uber and Grab in Southeast Asia, Grab became a dominant player in Viet Nam’s ride-hailing market, holding a market share of 73 percent in 2019. However, Grab and Uber have not been the sole players. Several other ride-hailing platforms have emerged since the mid-2010s(Trương et al, 2021), with major players including Be and Gojekthe Indonesia-based app that has been Grab’s major competitor in Southeast Asia. By 2021, these two competitors had gained 18 percent and 19 percent market share, respectively, as the Singaporean app’s market share declined to 60 percent(Việt Hưng, 2023). However, Xanh SM-a new platform backed by Vingroup, one of Viet Nam’s largest private conglomerates-has recently become Grab’s biggest rival in the ride-hailing market. Shortly after its incorporation in March 2023, this platform occupied second position in the market, accounting for 18.17 percent by the end of 2023(Khánh Vy, 2024). Grab retained its leading position with 58.68 percent while Be and Gojek held 9.21percent and 5.87 percent, respectively. Like Grab, these and other platforms offer both car and motorbike ride-hailing services (Trương et al, 2021). App-based delivery services have also grown rapidly, including during the Covid-19 pandemic (Google, Temasek and Bain& Company, 2020). Several ride-hailing platforms-like Grab, Be, and Gojek-have expanded their services to include parcel, food, and grocery delivery(Trương et al, 2021), using one app for all ride-hailing and delivery services. There are also platforms that specialise in delivery services, like Now and Baemin-which focus on food delivery-and Ahamove-which provides a range of freight delivery and logistics services. Unlike ride-hailing services, app-based delivery services involve four parties, with platforms mediating transactions between customers, drivers, and business clients(e.g., restaurants and grocery shops) through an app or website (ILO, 2021). They enable business clients to access a broader pool of customers whilst, at the same time, permitting customers to approach various products without leaving their physical location(ILO, 2021). Delivery platforms make profits primarily by taking a fixed percentage from delivery fees charged to customers and, in the case of food and grocery delivery, a percentage-based commission from each order (Shaheen, Cohen, and Zohdy, 2016). The multiparty relationships of ride-hailing and delivery platform work are illustrated in Figure 1 below. This section has provided a snapshot of the platform economy, platform work, and the rise of app-based ride-hailing and delivery services in Viet Nam. The next part will look in particular at the pay, conditions, and representation of platform motorbike drivers in Viet Nam. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 19 Figure 1. Parties Involved in App-based Ride-hailing and Delivery Work Note: The broken lines that connect the supplier to other actors indicate that the supplier merely exists in certain cases, such as food and grocery deliveries, adapted from Figure 2 in Duggan et al (2017) Customer • End-user of ridehailing/delivery services • Generates demand on platforms by initially requesting services • Abides by service terms and conditions, including service charges, set out by the platform Driver • ‘Driver-partner’ of the platform who works flexibly on a gig basis • Executes assigned tasks in the same geographic location as customers in exchange for income • Abides by pay and conditions set out by the platform Platform Company • Provider of ri-de-hailing/delivery services through an intermediary digital platform/app, like Grab, Be, Now, and Baemin • Assigns tasks to the drivers via apps on an as-needed basis • Controls drivers via algorithms and decides pay and conditions • Determines terms and conditions to be met by customers and suppliers, including trip fares, delivery fees, and commission rates Supplier • An additional‘partner’ of the platform in services such as food and grocery deliveries • Supplies goods to be delivered to customers • Abides by supply terms and conditions, including commissions paid for each order, set out by the platform PART II THE VULNERABILITY OF PLATFORM MOTORBIKE DRIVERS IN VIETNAM 21 THE VULNERABILITY OF PLATFORM MOTORBIKE DRIVERS IN VIETNAM As discussed, high-earning jobs, autonomy in working conditions, and flexibility in time commitments are the commonly assumed benefits of platform work in Viet Nam. Drawing from quantitative and qualitative data, this section presents a very different story that features the vulnerability of app-based motorbike drivers in HCMC. With low and unstable incomes, these drivers often work in harsh and dangerous conditions while constantly facing unfair management, job insecurity, and limited social protection. Compounding this is their lack of voice and representation in setting pay and conditions. Part II also shows that the apparent freedom and flexibility of platform motorbike drivers is highly restricted in practice due to their dependence on platform work to earn a liveable income and that these workers are subjected to algorithmic and traditional controls exerted by platform operators. The claim that platform workers are well paid and the job allows autonomy and flexibility is, therefore, a myth for most app-based drivers. Low and Vulnerable Incomes Our survey reveals that, contrary to the dominant narrative about platform work, app-based motorbike drivers in HCMC suffer from low pay. In particular, it shows that if a driver worked on a full-time basis, i.e., 48 hours per week or more, and for a full 30-day month, 5 his or her average income would be 9,290,344 VND(380 USD) after deducting essential work-related expenses, such as petrol costs and mobile phone charges. For part-time drivers-namely, those working less than 48 hours each week-their average monthly income would be 4,794,792 VND(196 USD). Table 1. App-based Motorbike Drivers’ Average Monthly Incomes(N=203) Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February& March, 2021 Type of Driver Full-time Part-time Number of Drivers 154 49 Average Monthly Income Before Costs(VND) 11,073,312 6,024,375 Average Monthly Income After Costs(VND) 9,290,344 4,794,792 Of the four platforms surveyed, full-time drivers from Baemin had the highest average monthly income, followed by those from Now and Be, respectively. Full-time Grab drivers had the lowest average monthly incomes. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 22 Table 2. Full-time App-based Motorbike Drivers’ Average Monthly Incomes Per Platform Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February&March, 2021. Platform Baemin Be Grab Now Number of Drivers 24 26 75 29 Average Monthly Income before Costs(VND) 13,016,667 11,619,231 9,915,200 11,970,690 Average Monthly Income after Costs(VND) 11,508,125 9,480,192 8,109,640 10,338,276 These results suggest that app-based motorbike driving is not a well-paid job. According to the Global Living Wage Coalition, the living wage for HCMC in 2020 was 7,446,294 VND(304 USD), excluding overtime pay(Global Living Wage Coalition, 2020). Meanwhile, the survey reveals that platform motorbike drivers had to work, on average, 11.1 hours per day and 28.03 days per month to earn 9,290,340 VND. In other words, they had to work about 103.1 hours overtime every month on top of a regular 48-hour working week. This goes well beyond the 40hour monthly limit on overtime under the labour law, not to mention its overtime cap of 300 hours per year( Labour Code 2019 art 107). If an app-based motorbike driver merely worked eight hours a day and 26 days a month-i.e., within standard working hours-our calculation suggests that the driver would earn only 6,210,672 VND (253 USD). That figure is 16.6 percent below the liveable wage cited above, though still higher than the minimum wage of 4,420,000 VND (180 USD) that applied to most areas in HCMC in 2020-2021. 6 If we take into consideration the fact that appbased drivers do not enjoy several benefits of conventional employees, their situation is even less desireable. Between 2020 and 2021, an employee would normally be entitled to social, health, and unemployment insurance contributions from his or her employer. Together, these would be equivalent to 21.5 percent of his or her base salary plus allowances. 7 If such contributions were added to the salary of a trained employee, his or her monthly income would be, at a minimum, 5,746,221 VND(235 USD). That would be 4.64 percent below the avergage monthly income of appbased motorbike drivers working a standard 48hour working week. Put differently, when social security contributions are counted, the difference between the average income of app-based motorbike drivers and the minimum income of trained employees was insignificant. Let us then count paid leave. Under the 2019 Labour Code, employees are typically entitled to at least 23 days of paid leave for public holidays and vacations every year. 8 Assuming that an appbased driver working standard working time took 23 days off for holidays and recreational purposes in a year, his/her average monthly income would fall to 5,752,834 VND(235 USD) because he/she is not entitled to paid leave. This would be 22.7 percent lower than the 2020 living wage rate for HCMC(7,446,294 VND/month) and virtually the same as the monthly income of a trained employee earning the legal minimum rate plus social contributions made by his or her employer (5,746,221 VND). Moreover, this has not taken into account several other employee benefits from which platform drivers are excluded. These include not only occasional paid leave-like BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 23 leave for marriage, compassionate reasons, and work-related accidents-but also various social security entitlements, such as those relating to sickness, maternity, work-related accidents, and retirement( Law on Social Insurance, 2014; Law on Employment, 2013; Law on Health Insurance (Revised), 2014). Unlike app-based drivers, conventional employees also receive overtime pay at premium rates, salary allowances, and shift meals-to name a few(Joe Buckley, 2022). Suffice to say that, when employees’ entitlements are counted, app-based motorbike drivers are not paid more, but less than trained employees receiving a minimum pay rate. The current situation of platform drivers may be even worse due to the fast-paced growth of the ridehailing workforce which has been driven, in part, by the shrinkage of manufacturing factories(Lê Tuyết, 2023). As well as being low, the incomes of app-based drivers are also vulnerable in various ways. Our survey indicates that the main income source of such workers was service charges-i.e., trip fares or delivery fees, usually plus additional fees, 9 paid by customers-after deducting a percentage as commission to platforms 10 Specifically, these charges accounted for, on average, 79.41 percent of their gross earnings from platform work each month. However, these earnings could be easily affected owing to policy changes unilaterally made by platform operators. From 2014 to 2020, Grab raised the percentage deducted from service charges three times, rising from 15 percent to over 27 percent. 11 Describing how these increases impacted his income, one Grab motorbike driver said: Previously, the deduction rate was 15 percent. Then it was increased to 20 percent. Now, it is nearly 30 percent. We, drivers, can’t bear[the new rate]. After subtracting all costs, we only have half of our earnings left. Working on streets every day, whether it is rainy or sunny,[but] we earn just enough for daily needs. However, we have no choice but to sell our health to survive (Hiếu Đam, 2020). Bonuses paid by platform companies are another important source of earnings for app-based motorbike drivers. These bonuses are designed to incentivise drivers to take more rides and orders and to ensure customer satisfaction(See Baemin, 2020; Be, 2022d, Grab, 2020a). Our survey indicates that these bonuses amounted to, on average, 17.36 percent of the gross earnings of an app-based motorbike driver each month. This income stream was also vulnerable to changes in platform policies. Be, Now, and Grab frequently revise their bonus policies, making it increasingly difficult for workers to achieve company rewards(Joe Buckley, 2020). One motorbike driver explained how changes to bonus formulas affected his earnings: It has become nearly impossible for us [i.e., drivers] to achieve bonus levels set out by the company since it introduced a new bonus policy recently. Therefore, our daily incomes have decreased. Previously, the platform rewarded 280,000 VND for every 700 gems [i.e., bonus points]. This formula has changed. To earn the same bonus, we now have to obtain 850 gems. And the gems given for each trip have reduced to 15. 12 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 24 Bonuses are also dependent on customer ratings. And these are, to a considerable degree, beyond the control of app-based drivers. Interviewed workers consistently noted that they received low ratings for unfair or unclear reasons, but had limited chances to appeal them. 13 Low customer ratings might also diminish drivers’ chances of receiving bookings and, thus, their actual earnings-a fact noted by both platform managers and workers. 14 In the worst cases, customer complaints led to a temporary locking of drivers’ apps pending investigation, causing significant lost income. 15 One driver revealed how his income was vulnerable to customer ratings: I rode carefully, but another motorbike rider didn’t. He hit my passenger’s leg. The passenger left feedback that I rode carelessly.[On another occasion] I rode into an alley. The alley was stuck. The passenger commented that I wasn’t good at direction. In fact, I know that area very well. I rode into the alley to avoid traffic jams, but I was unlucky that time.... My rating was five-star, but it has decreased because of such ratings. This has not affected my bonuses yet. However, it reduced the number of bookings allocated to me. Say, for example, I often earned 50,000 VND for every hour. But, after receiving one or two low ratings, I could earn only 100,000 VND in four hours.... It is very unfair. 16 Moreover, the incomes of app-based drivers are vulnerable to changing market conditions. Those in our survey frequently commented on the growing number of platform workers and its relation to their decreasing earnings. 17 One worker lamented:“Our incomes have declined because there are more workers in the job each passing day. Meanwhile, the number of customers has not changed. For this reason, our income will continue to decline further.” 18 The Covid-19 pandemic also provides an example of how app-based motorbike drivers’ incomes are precarious and variable depending on market conditions. Their earnings declined considerably when strict social distancing was in place due to reduced mobility needs(Bình Minh, 2020). One platform driver reflected on his hardship during this time: Normally, I had 20-25 rides per day. With such a number of rides, I could receive some bonus and make 300,000400,000 VND per day. Now, I have only four to five rides and earn 100,000200,000 VND in a day. After subtracting percentage-based commissions, petrol costs, and maintenance expenses, I have only 30,000-50,000 VND left.[That is] not enough to live. 19 The surge in petrol prices in 2022 offers another illustration of how app-based drivers are vulnerable in unfavourable market conditions. As ride-hailing and delivery platforms did not immediately raise service rates, their drivers’ take-home incomes were severely impacted(Thế Lâm, 2022). The following figure illustrates how app-based motorbike drivers’ incomes can be vulnerable to factors beyond their control. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 25 Figure 2. Factors Impacting App-based Motorbike Drivers’ Incomes Source: Designed by the authors. Note: The arrows in blue represent impact relations. Driver’s income after costs Gross income Service charges Platform bonuses Customer tips Deductions Platform commissions Work-related costs Platform policies Customer rating and feedback Market conditions To sum up, app-based motorbike drivers in HCMC are not well-paid as has been widely presumed. Although the average monthly income of fulltime drivers in our survey was higher than the living wage they had to work, on average, 103.1 hours of overtime every month to achieve it. If an app-based driver merely worked 48 hours a week, his/her income would fall well below the living wage. Moreover, if we also take regular paid leave and social security contributions into account, the average income of drivers would be almost equal to the legal minimum income of trained employees. This has not counted various entitlements under labour and social security laws from which app-based drivers are excluded. Making matters worse, their incomes are highly unstable and vulnerable to the behaviour of other market actors and changing market conditions. As demonstrated, this results largely from the fact that platform drivers have limited say on their pay. Long Working Hours with Limited Rest App-based motorbike drivers in HCMC regularly work long shifts. In total, 154 respondents (over 75 percent of those surveyed) noted that they usually worked at least 48 hours a week. Their average number of working hours in one day was 11.1 and their average number of working days in one 30-day month was 28.03. Put another way, full-time drivers, on average, worked 311.1 hours in a 30-day month; 103.1 hours more than the standard working hours. As discussed, this far exceeds overtime caps under the 2019 Labour Code: 40 hours per month and 300 hours per year. Of the four platforms, Be had the longest average working day. Grab and Baemin came next, followed by Now. Unlike conventional employees, app-based drivers are, however, not paid at premium rates(which range from 150 percent to 300 percent, depending on working days). 20 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 26 Table 3. Full-time App-based Motorbike Drivers’ Average Working Hours in a Month Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February&March, 2021. Platform Average Working Hours Per Month Working Hours Exceeding the Monthly Overtime Cap Baemin 301.5 93.5 Be 344.4 136.4 Grab 308.7 100.7 Now 296.4 88.4 Overall 311.1 103.1 One major reason that explains why platform motorbike drivers often work long hours is that they could not earn a liveable wage if they only worked 48 hours a week. One veteran app-based driver remarked: belief, which was confirmed to be true by a platform manager, that if they regularly worked long hours, the platform would allocate a stable number of tasks to them. 24 One driver explained: Several drivers work up to 15 hours a day. Why? Why don’t they just work eight or ten hours?... In fact, they don’t want to work 15 hours. They do so just because they need a liveable income. 21 Furthermore, platform companies strategically deploy incentive-based tools to encourage drivers to work long hours. Bonuses are one example. For drivers to receive a bonus from platforms, they are required to complete a prescribed number of rides and/or orders(See Baemin Bonus Policy). Since these bonuses are a major component of drivers’ earnings, several accepted long working hours. 22 To make matters worse, as many realised, drivers often had to wait longer than usual to receive new tasks when they were about to achieve a bonus level. 23 Working long hours is also considered by platform drivers as a way to secure stable work and income. Interviews with drivers reveal a Grab has a policy on how to distribute work and income to drivers. For example, if I usually work twelve hours a day, I will usually receive enough work to earn about one million VND a day.... There is no explicit regulation about this. But it is a well-known, implicit norm. 25 App-based motorbike drivers also regularly work at night. Of the 203 surveyed drivers, 47(23.2 percent) reported that they often drove between 10:00PM and 6:00AM. Driving at night increases health risks and the probability of traffic accidents, robberies, and sexual harassment for drivers. Nevertheless, these workers are not compensated at a premium rate, unlike employees who are paid at least 130 percent of the normal wage rate( Labour Code 2019, art 98). Be, Grab, and Now(but not Baemin) drivers were merely entitled to a fixed nightly fee(usually 10,000 VND per booking) which might be partly deducted by platforms(Be, 2022f). BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 27 While working long hours, app-based motorbike drivers rarely take a good break. As they are under pressure to complete as many bookings as possible and secure a liveable income, many keep their apps on during breaks. 26 This, and the on-demand nature of ride-hailing work, prevents workers from having scheduled and non-disruptive break periods. One female worker said: I drive all day and do not turn the app off. I eat while awaiting rides. On one occasion, it took me four times to finish a bánh mì [Vietnamese roll].... Every time buying a meal, I put and eat it in a lunchbox. I don’t eat it on a plate because I may have a ride before finishing the meal. I often skip meals because of work.... Sometimes, I am too tired to eat. I just drink a glass of orange juice. 27 Short breaks usually take place in outdoor spaces like pavements, lawns, and parks, followed by public indoor spaces, such as cafés, eateries, and shopping malls(see Table 4). Table 4. Drivers’ Normal Rest Places Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February&March, 2021. Note: 17 respondents did not reply to this question. Place Outdoor space Public indoor space Home Number of Drivers Reporting 126 88 68 Percentage 67.7% 47.3% 36.6% As indicated above, app-based motorbike drivers working on a full-time basis took, on average, two days-off every 30-day month. This figure excluded occasional leave, such as leave for sickness, broken vehicles, public holidays, or vacations. It is far from meeting the weekly dayoff standard. Survey responses also show that a majority of drivers(84.7 percent) usually worked on public holidays. In that case, they were not paid triple like employees, but received a fixed additional fee from customers, ranging from 5,000 to 15,000 VND, which might be deducted by platforms(Ng Hải, 2021). Only 100 out of the 203 respondents stated that they arranged recreational leave beyond regular days-off each month. The respondents also reported that, on average, they took 3.73 and 1.9 days off each year for sickness and vehicle repair and maintenance, respectively. All of these days off and leave were unpaid. This was a major contributor to app-based drivers’ reluctance to take leave, even when they were unwell. The following statement from a platform driver reflected this: BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 28 Lack of Safety at Work I had a fever yesterday. But I still went to work. The[platform] company does not prohibit me from taking a day off. But I would not be paid for that day. I would not be entitled to 50 percent of the daily wage or something like employees of[other] companies. 28 Danger is, alas, everywhere on the road. The line between life and death of a driver is very delicate. Accidents. Robberies. Being attacked. Even killed. Anything can happen.... Do you know why we[drivers] still work? Money. We need money to survive. 29 To conclude, app-based motorbike drivers normally work under harsh conditions which feature long working hours with limited rest. As already seen, the full-time drivers in our survey had to work, on average, 11.1 hours a day for 28.03 days a month. This far exceeds the legal caps on overtime that protect workers’ health. In addition, the app-based drivers in our survey did not have sufficient rest, whether in the form of short breaks, periodic days-off, public holidays, recreational leave, or even sick leave. Nor were they entitled to any kind of paid leave like conventional employees. These extreme conditions derive substantially from how ridehailing and delivery platform work is structured and regulated. As illustrated, the low-paying nature of such work, its gig- and incentive-based pay system, the on-demand nature of the service, and the non-application of labour standards on working time have led platform workers to prioritise working and de-prioritise rest times. This, in turn, demonstrates that ride-hailing work does not provide motorbike drivers with freedom and flexibility as platform operators often promise. The above statement from an app-based motorbike driver arguably captures most of his occupational hazards. Consistent with this, survey responses show that app-based drivers were exposed to serious safety risks. In fact, 128 of the 203 respondents(62.6 percent) reported that they encountered at least one dangerous incident during work. As shown in Table 5 below, traffic accidents were the most common source of danger. Other frequently cited incidents included assault or the threat of assault by other drivers; sexual harassment by customers; and assault or the threat of assault by customers. Only 38.4 percent of surveyed workers stated that they had never faced a dangerous situation. These results accord with periodic media coverage of tragic outcomes of app-based motorbike drivers getting into traffic accidents or suffering from robberies or physical violence, including severe injuries and deaths. 30 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 29 Table 5. Sources of Danger Reported by Drivers(N=128) Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February&March, 2021. Source of Danger Traffic accident Assault or threatened assault by other drivers Assault or threatened assault by customers Assault or threatened assault by other people Sexual harassment by customers Sexual harassment by other people Other dangerous circumstances Number of Workers Reporting 92 69 22 11 30 5 5 Percentage 71.9% 53.9% 17.2% 8.6% 23.4% 3.9% 3.9% These statistics only reflect the situation of male drivers. In-depth interviews with female drivers suggest that they constantly face high risks of assault and harassment while having limited means to protect themselves. One part-time female driver stated: I’ve never been abused. But some of my female friends were abused when driving at night. One even encountered a robber who put a knife to her throat. She didn’t resist, letting him take her mobile phone and motorbike away. However, I’ve encountered passengers who flirted with me. Some even solicited for sex. Normally, I didn’t reply. [I] just kept driving. I just said:“Please stop talking. I have to concentrate on driving; otherwise, we will fall off the motorbike.” 31 Another female driver who worked full time encountered more serious forms of sexual harassment. She recalled: I was abused several times. I was touched. My buttock was patted. Some passengers even hugged my body. Others suggested me sleeping with them for money. Or sent text messages and invited me to meet at hotels. One time, when I was still a new driver, a passenger suddenly hugged me while I was driving him to the airport. I was nervous. Uncomfortable. But I worried that if I refused to drive him, he might demand me to pay compensation for his missed flight. I was inexperienced then. Fortunately, I saw a Grab male driver while stopped at a traffic light. I asked him to take over the passenger. 32 Only one of the three female drivers interviewed said that she had never suffered from sexual harassment. She worked in food delivery only and, therefore, did not frequently meet with customers. 33 Of note, harassed platform drivers, whether male or female, are rarely reported in newspapers, indicating low public awareness of the problem. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 30 Platform companies have introduced measures to manage the risks facing platform drivers. All of the surveyed platforms have enacted safety rules and required their drivers to comply(Baemin, 2019). In addition, they have provided training to workers in traffic laws, first aid, self-defence, and management of assault and harassment. 34 Ride-hailing platforms also operate emergency communication channels, provide drivers with safety alerts on apps, and offer safety tips. 35 Nevertheless, only a small proportion(26.9 percent) of surveyed drivers believed that such measures were‘highly effective’(see Figure 3 below). Over half(57.2 percent) rated their effectiveness as‘moderate’ while the rest(15.9 percent) rated their effectiveness as‘low.’ Figure 3. Drivers’ Assessment of Platforms’ Safety Measures(N=201) Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February&March, 2021. Effectiveness We[app-based drivers] are often insulted, dispelled, and threatened by [traditional] motorbike drivers at coach stations. We are very frustrated.... However, the company has no policy to protect us. It only recommends us to contact the police.... If we fight back, we will be disadvantaged. If Grab knows this, our app will be deactivated. Therefore, we only leave quietly when facing such a situation. 36 App-based motorbike driving also involves significant health risks. Of the 198 workers who responded to the survey, 53.5 percent believed that their health had been adversely impacted after taking this job. Respiratory diseases are the health problem most cited by respondents, followed by joint pains, digestive problems, and urology and cardiovascular issues. 15.9% 26.9% 57.2% High Average Low Qualitative interviews provide examples of how platform companies have failed to address the safety concerns of their workers. One platform driver said: BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 31 Table 6. Health Issues Reported by Drivers(N=198) Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February and March, 2021. Health Issues Respiratory diseases Joint pains Digestive problems Urologic diseases Cardiovascular issues Other problems Number of Workers Reporting 29 84 32 12 06 22 Percentage 14.2% 41.4% 15.8% 4.0% 2.9% 10.8% However, ride-hailing and delivery platforms pay little attention to their workers’ health risks. Almost two-thirds(64.3 percent) of respondents said that their platforms had no measures to protect workers’ health. Of the four surveyed platforms, Be appeared to be the best at looking after workers’ health, with 62.1 percent of respondents from this platform reporting the existence of health protection measures. Grab came next(40 percent), followed by Now (25.7 percent) and Baemin(13.5 percent). The main measure implemented by platforms was providing health advice and information to workers, noted by 24.5 percent of respondents. Only 6.1 percent reported the existence of company training. Platforms also purchased health and work-related accident insurance for some of their workers. These insurance policies will be discussed later. In the final analysis, app-based motorbike driving is a job with high risks to workers’ safety, health, and dignity. Although platform companies have introduced some measures to reduce such risks, these measures remain modest and ineffective in the eyes of most drivers. Algorithmic Control and Unfair Management Algorithmic Control Though they classify app-based motorbike drivers as‘partners,’ i.e., independent contractors, delivery and ride-hailing platforms exercise tight control over their drivers. As we shall see, this control is often exerted remotely through algorithmic management. As Duggan et al explained, an‘algorithm is a computational formula that autonomously makes decisions based on statistical models or decision rules without explicit human intervention.’(Duggan et al, 2017). Digital labour platforms rely heavily on algorithms to match workers with customers and oversee the performance of platform workers(Alessandro Gandini, 2018). Algorithmic management, a term first coined by Lee et al in 2015(Lee et al, 2015), refers to‘a system of control where self-learning algorithms are given the responsibility for making and executing decisions affecting labour, thereby limiting human involvement and oversight of the labour process.’(Duggan et al, 2017). BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 32 Conversations with workers and a former platform manager reveal that platform operators rely on algorithms to closely monitor the labour process of app-based motorbike drivers. As soon as drivers turn on their app, algorithmic management determines how their work is assigned. In particular, an algorithm will allocate trips and couriers to drivers based on-among other thingsperformance metrics like time spent working, acceptance, cancellation and completion rates, customer ratings, and movement directions(as compared with customers’ directions). 37 Utilising algorithms, platforms also implement strategies to encourage drivers to accept assigned tasks. All surveyed platforms imposed on their drivers minimum performance rates, such as minimum acceptance, cancellation, completion, or ‘working time’ rates. 38 Failure to meet these rates triggers automatic, immediate sanctions including temporary or permanent deactivation of drivers’ apps. 39 For example, if a Be driver repeatedly fails to meet the minimum acceptance, cancellation, or completion rates prescribed by the platform, he or she would be at risk of temporary(three or seven days) or permanent deactivation 40 . Therefore, app-based drivers are under constant pressure to accept work assigned by their platforms. 41 This is well reflected in the following statements of app-based drivers: I don’t cancel any bookings. I accept all bookings, whether they are good or bad. If I don’t accept the tasks assigned to me, the company will consider me a selective driver.... They think that I am not hardworking enough. They will distribute fewer rides to me[and] will give fewer bonuses. 42 I drive every day, including Saturdays and Sundays.[I am] only off when I am very ill. If I am ill, but it’s not too bad, I will still work. Otherwise, I won’t achieve performance targets. 43 On occasion, the app-based drivers surveyed were even forced to perform work assigned by their platforms. Grab drivers, for example, reported that their app sometimes accepted allocated bookings automatically, i.e., without allowing them to give their consent. 44 Algorithms also allow platforms to identify regions and timeframes(e.g., peak hours or weekends) with high customer demand and offer higher service charges or bonus rates to entice drivers to work. 45 In the same way, app-based drivers are encouraged to accept less preferable tasks, such as short rides, long-distance or multiorder deliveries, and high-price food orders. 46 Hence, platforms do not simply match drivers with requestors of services. Through algorithms, they tightly control the assignment of work to drivers. The freedom of app-based motorbike drivers to determine their own work schedule is, therefore, essentially theoretical. In reality, they are compelled to accept most-if not all-of the tasks assigned to them by their platforms in order to maintain a stable job with liveable income. If it exists, their freedom is arguably confined mainly to the choice of the start and end-rather than the length-of their working time. Aside from dictating work assignments, algorithms also exercise the HR function of performance management. Deploying a global positioning system(GPS) and the apps installed in drivers’ smartphones, platforms track the location of drivers and their movement in real time and verify their identity as well as activities. 47 Consequently, they can discover, warn against, and penalise workers’ breaches of company policies, such as having their accounts used by other drivers, working for other ride-hailing platforms at the same time, cancelling bookings on the app but BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 33 still accepting passengers, and exceeding speed limits, etc. 48 To ensure productivity, platforms also set high-performance expectations for drivers like acceptance, completion, and cancellation rates, and revenue targets(Be, 2022c). If they fail to meet these expectations, app-based drivers would not only be disciplined 49 , but also automatically disqualified from bonus payments and other benefits like accident and health insurance coverage. 50 Customer rating is another important tool for platforms to manage the work performance of motorbike drivers. Since low ratings or negative feedback from customers triggers algorithmic (automatic) sanctions, such as a reduction of assigned tasks, exclusion from bonus schemes, or app deactivation, app-based drivers have no choice but to enhance customer satisfaction. 51 Nevertheless, customer ratings do not only press platform drivers to maintain high-quality services. All of the platforms surveyed have issued rules and guidelines to which their drivers must adhere when rendering their services, including but not limited to those relating to customer service and communication, safety and professionalism, and hygiene and vehicle conditions( Baemin COC; Be COC; Grab COC; Now COC). By suggesting that customers assess drivers on such issues(see Figure 4 below), platforms also exercise indirect, but effective, oversight over the performance of app-based drivers and their compliance with work rules and instructions. Figure 4. Platform Indicators for Customer Feedback and Ratings Source: Drawn from platforms’ instructions for customer ratings on Baemin, Be, Grab, and Now/ ShopeeFood apps. Algorithmic management, therefore, functions like a‘virtual automated manager’(Duggan et al, 2017) that assigns work to app-based drivers, sets performance expectations, evaluates their performance, supervises their compliance with work rules and instructions, and disciplines those with poor performance or misconduct. Using algorithmic management, platform operators exert tight control over the labour process of their driver-partners no less-if not more-than the traditional control of employers over employees. Drivers repeatedly commented on this: I have to hail Grab[laugh]. It has created mechanisms that make drivers always anxious, pressing them to comply with company policy and change their behaviour. If drivers cannot afford this pressure and change themselves, they have no way but to quit and look for another job. 52 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 34 I have never committed any breach because I always remind myself: “Compliance! Otherwise, I will pay for my breaches right away.” 53 Although platforms rely extensively on algorithms, these are not their only means of control over app-based motorbike drivers. Platforms have also utilised traditional methods to manage labour. Grab, for example, has team leaders who provide instructions to drivers and supervise their compliance with company policy. 54 Also, Grab drivers widely reported the existence of ‘undercover inspectors’ from the company who went around taking pictures of drivers’ infringements and testifying if drivers committed dishonest acts, such as cancelling bookings on the app while still taking passengers. 55 In short, despite labelling app-based drivers as independent contractors, ride-hailing and delivery platforms deployed both algorithmic and traditional means to closely monitor the labour process of these drivers. Unfair Management Working under the tight control of their platforms, app-based motorbike drivers suffer heavily from unfair management practices. Over half(57.2 percent) of the 203 surveyed drivers claimed that the allocation of bookings by platforms was unjust while 49 percent complained that the process was not transparent. These results are understandable, considering that platform companies do not disclose algorithms determining work assignments to drivers. Although platform drivers might recognise some factors impacting their receipt of rides and orders, they never fully understand how the distribution of bookings is done. 56 This lack of transparency has caused many to question the fairness of the booking allocation system. 57 One driver, for example, said that: To receive several rides, you have to work for long hours. Then your app will be identified as a hardworking app and you will receive more rides. You had better drive around rather than wait for rides in one place. Drivers with higher customer ratings also get more bookings.... However, it does not always work that way. In many cases, one driver receives a ride before another despite the fact that both wait in the same place, have the same ratings, and work the same hours. This system is not always fair! 58 Others went further, saying that platform administrative staff improperly manipulated booking allocations. 59 Additionally, several drivers reported that it would usually take more time to receive a new booking when they were about to achieve a bonus level. 60 An interview with a former manager of two platforms confirmed the existence of these underhand practices. 61 Significantly, the ride-hailing and delivery platforms in our survey maintained a disciplinary regime which was overly harsh, arbitrary, and disrespectful of due process standards. Interviews with drivers reveal the extensive use of monetary fines. 62 Particularly, if a driver-partner was deemed to have violated the platform’s code of conduct, his or her account would be easily deactivated temporarily or permanently, causing a loss of income or even job loss for the driver. 63 As noted, such deactivation could also be imposed pending investigation. Even if the driver subsequently succeeded in proving his or her non-infringement, he or she would not receive any compensation for the loss of income. 64 Platform drivers repeatedly commented on the severe impact of deactivations. 65 One said: BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 35 They[the platform] pay me. So, they have the right to manage me. I am fine with that. If I commit a breach, they will sanction me. It’s fine, too. But they deactivated my app. That was too harsh! It affected my family’s bread and butter(chén cơm, manh áo). I was very unhappy, and I told them about this. But nothing has changed at all. 66 It should be noted that the extensive use of severe disciplinary sanctions by ride-hailing platforms departs considerably from basic labour standards. Under the 2019 Labour Code, monetary fines are strictly prohibited( Labour Code 2019 art 127(2)). Temporary suspension of employment cannot be implemented as a disciplinary sanction, but only to foster a disciplinary investigation(where the continued working of the employee under investigation may impede that investigation) ( Labour Code 2019 art 128). During such a suspension, the employee is entitled to 50 percent of his or her salary. 67 Meanwhile, if the employer fails to prove his or her wrongdoing, the employee will receive their full wage upon the expiration of the suspension. 68 Arguably, the exclusion of app-based motorbike drivers has rendered them highly vulnerable to severe disciplinary actions. Another problem with disciplinary processes on ride-hailing and delivery platforms is the lack of transparency, a problem reported by 61 of the 203 survey respondents(30 percent). Notwithstanding their frequent use, monetary fines are not provided for in any code of conduct of the platforms investigated( Baemin COC; Be COC; Grab COC; Now COC). App-based drivers knew neither how a disciplinary decision was made nor by whom. 69 Not merely untransparent, disciplinary actions were also arbitrary and unpredictable. According to interviewed drivers, platform companies were not always consistent in enforcing codes of conduct and determining sanctions. 70 Due process was largely disregarded. Disciplinary action was not always sufficiently substantiated by reasons and evidence, a fact not only reported by app-based drivers, 71 but also acknowledged by a former platform manager. 72 The following statement of a Grab driver exemplifies this unfair practice: My app was deactivated twice. In one case, it was due to the complaint of a customer.... I didn’t hit or crash into anyone. But the company[Grab] insisted that I did it.... They based[their decision] only on the statement of the customer to punish me. They didn’t listen to me. 73 Meanwhile, 61.9 percent of the 203 surveyed workers claimed that they had no or limited chance to explain their circumstances. While providing a long list of concrete offences and corresponding sanctions applicable to driverpartners, platforms’ codes of conduct offer no mechanism, such as a disciplinary meeting, that enables their drivers to explain and to receive an explanation( Baemin COC; Be COC; Grab COC; Now COC). One platform driver expressed discontent about this: Disciplinary action is not reasonable at all. Why? If the company wants to discipline a driver, they must sit down and talk with the driver.... They cannot just send a notice to the driver while he/she is on the street. They must also provide sufficient evidence. 74 There is no guarantee of workers’ right to counsel( Baemin COC; Be COC; Grab COC; Now COC). Quite often, surveyed platform drivers were only provided with a short, automated notice of their infringement and the imposed penalty. 75 Workers’ right to appeal was essentially BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 36 disregarded. Almost two-thirds(62.4 percent) of survey respondents noted that their platforms failed to maintain an effective mechanism to handle drivers’ appeals. More particularly, drivers regularly criticised their platforms for failing to reply to their appeals, to exhibit empathy towards drivers, and to give regard to their circumstances. 76 Confirming this, a former platform manager admitted: My company has tens of thousands of drivers. Consequently, we can only resolve a limited number of drivers’ appeals. Normally, we only acknowledge rather than completely resolve their appeals. We try to address important complaints from drivers, such as why their apps are deactivated. For less important complaints, we just reply that the decision can’t be changed because it has been made on the basis of the code of conduct. 77 The app-based drivers in our study were especially dissatisfied with the customer rating system. One reason is that it was not always explained when they received low ratings. 78 In particular, drivers widely commented on their limited capability to reply to customer feedback, and repeatedly complained that their platforms“did not respect drivers opinions”,“merely listened to customers”, and“privileged customers over drivers”. 79 As indicated by a veteran app-based driver, this does not necessarily mean that drivers entirely reject the customer rating system; rather, they “only want it to be fairer and give more regard to drivers’ opinions and circumstances.” 80 In conclusion, contrary to the widely circulated narrative about the freedom of platform workers, app-based motorbike drivers are tightly controlled by ride-hailing and delivery companies, especially through algorithmic management. Furthermore, they constantly suffer from unfair management practices. As illustrated, the platform companies in our study not only failed to ensure transparent and equitable allocation of work. They also maintained a disciplinary system that featured undue harshness, arbitrariness, a disregard of due process, and a customer rating system that lacked transparency and fundamentally ignored the voice of workers. Low Job Security Aside from facing constant managerial abuses, the app-based motorbike drivers in our study also had a low level of job security. Of the four platforms examined, Baemin, Bee, and Now only offered fixed-term contracts to their drivers, 81 typically ranging from one to two years, with possible renewal subject to the discretion of the platform companies. 82 Although Grab often signed indefinite-term contracts with drivers, it retained the right to terminate such contracts at any time by providing three-days of notice. 83 Simply put, there was no contractual guarantee of a permanent job for platform drivers. Furthermore, platform companies maintain a broad discretion to unilaterally terminate contracts with their drivers. Similar to Grab, Gofast-Now’s owner-reserves the right to terminate a driver simply by notifying him or her three days in advance 84 . Additionally, all of the examined platforms retained, whether in contract or company rules, the right to terminate their drivers without notice in various circumstances, including where the drivers were deemed to violate the Code of Conduct issued by the platforms. 85 Even worse, platform drivers are not entitled to any benefits upon the termination of their contracts because they are not classified as employees. 86 App-based drivers are, therefore, highly vulnerable to job loss, particularly owing to the virtually unconstrained disciplinary power of platform companies. Conversations with drivers suggest that many realised not only the insecurity of their jobs, but also the necessity of legal intervention in this area, including the protection of the labour law. 87 This is particularly evident in the following statement from an informal leader of platform drivers: BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 37 As partners, we[drivers] are considered to be on an equal footing with Grab. Each of us has an agreement with the company. But this agreement is subject to Grab’s discretion. The agreement says that Grab can unilaterally change its terms[and] can terminate it whenever they want.... We have no say in these agreements. No rights. No entitlements. Just obligations. If we breach any of these obligations, the company may deactivate our app. We’ll be kicked to the street then, helpless and with anything in hands.... We are different from employees. If employees lose jobs, they will receive unemployment allowances and retraining for new jobs. We have nothing.... We want the state to intervene to protect us[,] to compel Grab to sign employment agreements with us[and] to ensure that we have more rights in our agreements with Grab. 88 Limited Social Protection As illustrated above, app-based motorbike drivers experience a high level of unemployment and health and safety risks. Despite this, they have received very limited social protection. Since platform companies categorise their drivers as independent contractors, they have refused to pay social insurance, health insurance, and unemployment insurance contributions for these drivers. The consequence is that platform drivers have been excluded from several benefits of state-run social security programmes, like sickness and maternity payments, work-related injury and occupational disease allowances and subsidies, medical expense reimbursement, unemployment allowances, and retirement pensions-to name only the most important( Law on Social Insurance 2014; Law on Employment 2013; Law on Health Insurance(Revised) 2014). Platform companies have sponsored some insurance policies for their drivers. In 2017, for instance, Grab became the first platform to purchase accident insurance for its drivers. This provided reimbursement of medical expenses and compensation in case of permanent disability or accidental death(Grab, 2014). Be, Now(ShopeeFood), and Baemin introduced similar insurance schemes in 2019, 2020, and 2021, respectively. Varying between platforms, the basic terms and conditions of these insurance policies are summarised in Table 7 below. In addition, Be, Baemin, and ShopeeFood initiated healthcare insurance schemes for drivers in 2019, 2021, and 2022. As shown in Table 8 below, these schemes do not apply to all-but only a group of-drivers on each platform, and vary considerably from one platform to another. Grab has not provided any insurance like this for its platform drivers. Instead, it has worked with insurance companies to offer additional insurance products, such as enhanced accident insurance and health insurance, to drivers(Grab, 2020b; Grab, 2022d). Clearly, the insurance schemes purchased by platform companies far from adequately compensate for their workers’ exclusion from state-run social security. Contrary to a widely circulated presumption that platform workers are not interested in social security rights(Kim Dung, 2022; Bùi Xuân Hải and Hà Thị Thanh Bình, 2020), qualitative interviews suggest that app-based drivers are concerned about their lack of access to social protection. 89 One, for example, remarked:“Our job is very precarious because we are not covered by insurance. If we have bad luck, we must manage it alone. There is no assistance for us. Nobody helps app-based motorbike drivers.” 90 In summary, while they face a high possibility of job loss and high occupational risks, platform motorbike drivers have been largely excluded from state-run social security schemes. Although platform companies have maintained their own insurance policies, these policies remain inadequate and fall short of the level of protection guaranteed by state-run social security schemes, worsening the already vulnerable situation of platform drivers. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 38 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ Table 7. Accident Insurance Provided by Platforms Source: Collected on digital apps by the authors. Platform Baemin (Baemin, 2022) Be ( Be, 2020) Grab (Grab, 2022) Now(ShopeeFood,2022)(ShopeeFood) Benefits Insured Drivers Covered Accidents Medical Expenses Permanent Disability/Death Compensation Allowance Drivers with operating accounts who have performed at least one assigned task within 60 days prior to the accident “Loyal BeDrivers,” i.e., drivers achieving certain prescribed rates and requirements regarding customer rating and feedback, booking acceptance, booking cancellation, monthly revenue, and training participation (Be, 2020; Be, 2022) Accidents taking place while the driver is performing an assigned task Accidents taking place in any of the following circumstances: (i) the driver has turned the app off;(ii) the driver is turning the app on and waiting for a booking; or (iii) the driver has accepted a booking and is waiting for a customer or performing a ride or order Up to Up to 100,000,000 N/A 20,000,000 VND VND per case per case Up to 1,000,000 or 3,000,000 or 6,000,000 VND per year 91 Up to 30,000,000 or 60,000,000 or 120,000,000 VND per case 92 Daily allowance of 100,000 or 200,000 VND for days of hospitalisation for no more than 30 days per case, 60 days per year 93 Daily allowance of 100,000 or 200,000 VND for days of income loss for no more than 30 days per case, 60 days per year Any driver Accidents taking place while the driver has turned the app on Up to Up to 113,636,364 N/A 45,454,545 VND VND per case per case Drivers who have (i) worked for at least 30 days; (ii) worked at least 20 days within the last 30 days, and; (iii) completed at least 30 orders within the last 30 days Any accident Some expenses up to 20,000,000 VND per year Up to 100,000,000 VND per case Daily allowance of 100,000 VND for days of medical treatment or hospitalisation for no more than 60 days per year Table 8. Healthcare Insurance Provided by Platforms Source: Collected on digital apps by the authors. Benefits Platform Insured Drivers Medical Expenses Permanent Disability/Death Compensation Allowance Baemin (Beamin, 2021 Drivers with high work performance Up to 20,000,000 Up to 100,000,000 N/A approved by Baemin VND per case VND per case Bee (Be, 2020) ShopeeFood(ShopeeFood, 2022) “Loyal BeDrivers” ranked as Pro or Pro+-namely, drivers achieving very high rates and requirements regarding customer rating and feedback, booking acceptance and cancelation, monthly revenue, and training participation (Be, 2020; Be, 2022) Up to 1,000,000 or 3,000,000 or 6,000,000 VND per year 94 Up to 30,000,000 or 60,000,000 or 120,000,000 VND per case 95 Daily allowance of 100,000 or 200,000 VND for days of hospitalisation for no more than 30 days per case, 60 days per year 96 Daily allowance of 100,000 or 200,000 VND for days of income loss for no more than 30 days per case, 60 days per year HUB drivers-i.e., drivers registered to work on a regular basis (ShopeeFood, 2022)-who have worked for the platform for at least 30 days, including at least 12 days within the last 30 days, and completed at least 30 orders within the last 30 days Some expenses Up to 100,000,000 Daily allowance of 100,000 VND for days of up to 20,000,000 VND per case medical treatment or hospitalisation for no VND per year more than 60 days per year 39 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 40 Weak Voice and Representation Added to the vulnerability of platform motorbike drivers is their lack of representation and marginal voice in the determination of pay and conditions. None of the drivers at Baemin, Bee, Grab, or Now have a trade union( công đoàn). 97 Nor are they organised into‘worker representative organisations’( tổ chức của người lao động tại doanh nghiệp,“WROs”) 98 . As explained by union cadres, platform drivers can form neither trade unions nor WROs within their enterprises because they have not been recognised as employees. 99 Arguably, the absence of the implementation of regulations has also prevented workers, including app-based drivers, from establishing WROs in accordance with the 2019 Labour Code. Since 2020, the HCMC Federation of Labour( Liên đoàn Lao động thành phố Hồ Chí Minh) and its district subsidiaries have attempted to organise app-based motorbike drivers into‘syndicates’ ( nghiệp đoàn). 100 Unlike trade unions, these syndicates are not established at the workplace, but structured along administrative units: Wards ( phường, xã) and districts( quận, huyện). 101 This means that the syndicate of a particular ward or district would consist of drivers from various platforms. 102 The underlying reason for organising app-based motorbike drivers into syndicates rather than trade unions is that these drivers are not considered employees. 103 It remains to be seen whether these newlyestablished syndicates for app-based motorbike drivers( nghiệp đoàn xe ôm công nghệ) will become effective representatives for labour interests. Qualitative interviews with platform drivers in 2020-2021 indicate their broad unfamiliarity with these organisations. 104 Until 2021, the HCMC Federation of Labour and its syndicates focused chiefly on the extension of their organisations and only started to provide some welfare to their members. 105 Little was done to discuss or negotiate with platform companies over remuneration and working conditions for app-based motorbike drivers. 106 Not only do they lack effective representation, appbased motorbike drivers also struggle to express their voice collectively. All four of the platforms scrutinised here have maintained channels for communication with drivers, including receipt of and replies to their concerns like chat boxes, call centres, emails, Q&A webpages, and Facebook pages. 107 These channels are, however, designed primarily for communication between platforms and individual-rather than collective-workers. Further to this, their actual impact has been substantially limited. Drivers consistently noted slow responses, or even silence, from platform operators and their frequent disregard of drivers’ opinions. 108 The following statement from an app-based driver exemplifies this: I called the centre, complaining that it was too difficult to acquire gems under the new policy and that my income had been affected. I called just to express my discontent. I didn’t think it would be of any help. Call centre operators have a default response:‘We acknowledge your opinion. We will forward it to management.’ That’s all! I never know if they forwarded my opinion to any managers or they just paid lip service. Nobody has ever informed me of how my feedback was handled. 109 Of the investigated platforms, Grab appears to be the only one to have introduced an official mechanism that permits some form of dialogue with a labour collective. The platform convened quarterly meetings with drivers which contain Q&A sessions for drivers to express their opinions. 110 However, our survey and interview data indicate that the actual impact of these meetings was limited. Of the 101 driverrespondents from Grab, fifteen(14.85 percent) were unaware of this mechanism. Meanwhile, 39 drivers(38.61 percent) rated its effectiveness as‘low’ and 26(25.74 percent) chose‘average.’ BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 41 Only fourteen(13.86 percent) drivers ranked it as‘high.’ Interviews with drivers further revealed the lack of genuine dialogue at these quarterly meetings. As one driver noted,“all speakers at the meetings were Grab’s people. It is difficult for other people to speak” 111 In the same vein, another interviewee recalled his experience at a meeting organised by Grab as follows: I attended one meeting. The company explained newly promulgated regulations to drivers. No driver complained about anything. All hailed Grab. They said: the company was so good, the company created income for workers, and so on and so forth. 112 Figure 5. Drivers’ Rating of the Effectiveness of Periodic Meetings with Grab(N=101) Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February& March, 2021. E ff ec � veness 6.93% 14.85% 13.86% 38.61% 25.74% High Average Low Unrated Unaware The app-based motorbike drivers we spoke to regularly reported the reluctance of their platforms to have a dialogue with drivers. Replying to our question about their platforms’ openness to dialogue with workers, only 29.1 percent of the 203 surveyed drivers said‘yes’ while 36.5 percent stated that their platforms accepted dialogue only in the event that drivers were taking collective action. Around one-quarter(27.8 percent) claimed that their platforms were not open to dialogue at all, while the rest(7.4 percent) chose‘unknown.’ At the same time, interviewed drivers regularly complained about platform companies’ failure to listen and give regard to their opinions. 113 One bitterly commented:“The company never seeks our opinions. In their eyes, drivers are smaller than a nail( không bằng cái đinh).” 114 Figure 6. Drivers’ Assessment of Platforms’ Openness to Dialogue(N=203) Source: Data of the survey of app-based motorbike drivers undertaken in HCMC, carried out on February& March, 2021. Openness 15.7% 55. 27% 59. 29% 74. 37% Yes Only in case of collec � ve ac � on No Unknown In the absence of an official representative organisation, app-based motorbike drivers have spontaneously developed self-help groups. 115 Lacking a shared physical workplace, these groups rely heavily on social media platforms, such as Facebook and Zalo, to maintain relationships between members. 116 Some of the groups observed in our study were rather active, as were their informal leaders. They developed Facebook pages for app-based motorbike drivers with membership in the tens of thousands( CLB Be Bike Sài Gòn, 2022). These pages have become an important way for platform drivers to build their collective identity and solidarity, to develop a shared sense of injustice, and to mobilise collective efforts to defend their rights. 117 Based on our observations, active drivers’ groups and their informal leaders appeared to BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 42 play an important role in organising collective action and communicating drivers’ concerns to platform companies, state authorities, trade unions, and state media. 118 Quite often, such collective action involved typical strike action in which platform drivers collectively turned off their apps. 119 But it also took many other forms, such as mass exits; collective petitions; sabotage (like mass cancellation of accepted bookings, including self-bookings); motorbike marches in public places; and picketing outside the offices of platform companies or state media agencies (Fieldnote, 2022). Notably, app-based drivers have regularly organised media campaigns to support their protests. Apart from proactively sharing their stories with state media, they also extensively utilised Facebook pages to show their discontent, make collective claims, exert pressure on platform companies, and draw support from the state and society. 120 Briefly speaking, appbased drivers use social media platforms not only to coordinate collective action, but also as a space for their protests. Facing the increasing activism of drivers, platform companies have resorted to oppressive and manipulative measures. Interviews with platform drivers reveal that those involved in collective protests, especially informal leaders of active groups, were subject to retaliation, including having their apps permanently deactivated. 121 All COCs of the four investigated platforms contain broadly-drafted provisions that allow easy punishment of drivers engaging in collective action, such as picketing, mass refusal or cancellation of bookings, mobilisation for collective action, and circulation of information detrimental to the platforms( Baemin COC ss 5, 7; Baemin Driver Agreement s 3.2; Be COC ss 2, 21-22, 24, 28; Grab COC ss 14.1, 14.2; Now COC ss 2.4, 2.5, 3.3; Now Driver Agreement s 5(d)). Furthermore, platform operators closely followed the Facebook groups of motorbike drivers and infiltrated them with a view to manipulating drivers’ discussions. For Grab, it also invited informal leaders of drivers to co-opt their groups with the platform. 122 In addition, it began to organise drivers into groups with leaders appointed by the company. 123 Although these and co-opted groups might provide a channel for two-way dialogue, they were deployed by Grab mainly as a means to foster top-down communication and contain the activism of drivers. 124 Platform companies have been relatively successful in neutralising the activism of motorbike drivers and its effects. Despite their increasing activism, app-based drivers achieved little success in objecting to the adjustment of deduction rates and bonus formulas by platform companies that adversely affected their takehome earnings. 125 The protest of Grab motorbike drivers in January 2018 was a rare case, as it pressed the platform to revoke an increase of the deduction rate from 20 percent to 23.6 percent. 126 However, this success was shortlived. Grab subsequently raised the rate to 27.2 percent in December 2020, irrespective of strong protests by its drivers. 127 Arguably, the most visible consequence of the increasing activism of app-based drivers was the growing attention of other actors, including state authorities, trade unions, state media, legal experts, and labour scholars, to their situation. 128 This has triggered emergent policy and scholarly discussion on how to regulate platform work, 129 the outcome of which remains to be seen. In a nutshell, platform motorbike drivers have had severe difficulty in organising and expressing their voices collectively. Lacking effective representation and genuine mechanisms for dialogue with platform operators, they have relied principally on informal networks and ‘wildcat’ collective action to defend their rights. This activism was, however, largely unsuccessful and significantly weakened by repressive and manipulative measures implemented by platform companies. Nonetheless, the increasing activism of app-based motorbike drivers caused other actors, including regulatory bodies and stateaffiliated unions, to pay attention to their issues, leading to an emergent debate about the regulation of digital labour platforms. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 43 Summative Remarks The analysis above has shown that the narrative that platform workers are well paid, and that the job allows autonomy and flexibility, is essentially a myth. Demonstrating the vulnerability of platform motorbike drivers in HCMC, it finds that: ▪ App-based motorbike drivers are poorly paid. The average monthly income of the surveyed full-time drivers was 9,290,344 VND. Although this was higher than the liveable wage for HCMC in 2020(7,446,294 VND per month), these drivers had to work, on average, 103.1 hours on top of the standard working time to achieve it. If an app-based driver worked 48 hours a week, his or her average earnings would be only 6,210,672 VND a month, 16.6 percent lower than the living wage-though still higher than the local minimum wage of 4,420,000 VND. This has not taken into account the fact that platform drivers are excluded from several benefits of conventional employees, including: Social security contributions by employers; paid leave; and social security entitlements in relation to sickness, maternity, workrelated accidents, occupational diseases, unemployment, and retirement- to name a few. If regular paid leave and social security contributions are calculated in this comparison, the average income of appbased drivers working 48-hour standard working time would be only 5,752,834 VND per month, almost equal to the minimum income of a trained employee and 22.7 percent lower than the living wage. ▪ Platform motorbike drivers regularly suffer from unstable incomes. Their take-home earnings are highly vulnerable to the behaviour of other market actors, especially platform companies and their customers, and the fluctuations of market conditions. ▪ App-based motorbike drivers often work under risky conditions characterised by long working hours with insufficient rest. On average, full-time drivers participating in the survey worked 103.1 hours overtime every month on top of a regular 48-hour working week. This went well beyond legal caps on overtime that protect workers’ health. Platform drivers did not have sufficient rest, whether in the form of short breaks, periodic days-off, public holidays, recreational leave, or even sick leave. Neither were they entitled to any kind of paid leave. ▪ Although platform motorbike drivers are promised discretion in determining working time, this discretion is substantially curtailed in reality due to the low-paying nature of the job, its gig- and incentive-based pay system, the on-demand nature of ride-hailing and delivery services, and the exclusion of platform drivers from labour law entitlements. ▪ App-based motorbike drivers work in dangerous conditions, with high risks to their safety, health, and dignity. Nevertheless, platform companies provide their workers with little help, leaving most- if not all- risks on the shoulders of app-based drivers. ▪ Contrary to the widely-circulated narrative about the freedom of platform workers, appbased motorbike drivers are tightly controlled by platform operators, particularly through algorithmic management. Significantly, they constantly suffer from unfair management practices. Not only do platform companies fail to ensure a transparent and equitable allocation of work, but they also maintain an overtly harsh and arbitrary disciplinary regime that essentially disregards due process and a customer rating system that falls short of transparency and pays little regard to workers’ voices. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 44 ▪ Platform motorbike drivers face a high level of job insecurity. Having no chance to secure a permanent job, they also face the risk of termination by platform companies at any time after a very short, or even without any, notification. In all cases, platform drivers are not eligible to any benefits upon the termination of their contracts. ▪ Being classified as independent contractors, app-based motorbike drivers are excluded from state-run social security schemes. Although platform companies maintain their own accident and/or health insurance programmes, these programmes far from adequately compensate for the benefits that app-based drivers would have received if they had participated in state-run social security schemes. ▪ App-based motorbike drivers have had major difficulties organising and expressing their voices collectively. Lacking effective representative organisations and mechanisms for genuine dialogue with platform companies, they have turned to informal networks and spontaneous collective action to exhibit discontent and push back against deteriorating pay and conditions. While triggering attention from state authorities, trade unions, and society at large, this informal activism ran into systematic repressive and manipulative responses from platform companies and appears to have achieved only moderate success. This section has revealed that ride-hailing work is not a desirable job for motorbike drivers who are exposed to substantial vulnerabilities. The next section will illustrate that this is closely related to the ambiguous status of these drivers under Vietnamese labour law. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 45 PART III LEGAL UNCERTAINTY AND ITS IMPACT ON THE VULNERABILITY OF PLATFORM MOTORBIKE DRIVERS IN VIETNAM 46 LEGAL UNCERTAINTY AND ITS IMPACT ON THE VULNERABILITY OF PLATFORM MOTORBIKE DRIVERS IN VIETNAM As evidenced in Part II, app-based motorbike drivers in HCMC face highly precarious conditions to earn a living. In this third part of the report, we argue that these circumstances derive substantially from the ambiguous legal status of such drivers. To advance this argument, Part III first shows how major regulators have failed to deliver a clear answer on whether platform drivers are regarded as employees. It then demonstrates that this failure has its roots in problems relating to the employee test under the 2019 Labour Code and its application to platform drivers. The last section reveals that legal uncertainties have contributed to and reinforced the vulnerability of platform drivers by excluding them from the protection of labour and social security laws, permiting platform companies to maintain unfair contractual arrangements and creating difficulties for those seeking to redress the situation. Unclear status of platform motorbike drivers under Vietnamese Law On-demand ride-hailing and delivery services exploded in Viet Nam from the mid-2010s. However, discussion of the regulation of platform work here only emerged in 2018. 130 At that time, debate was confined mainly to experts involved in the drafting of the 2019 Labour Code(Nguyễn Trang, 2018; Hoàng Mạnh, 2018). The issue has since attracted increasing attention from state authorities, 131 labour unions, 132 legal practitioners (Phạm, 2021; Lạc Duy, 2020) and scholars, 133 especially following escalating conflicts between app-based drivers and their platforms. 134 More recently, platform motorbike drivers also called for legal protection for themselves. 135 So far, the discourse on regulating platform work in Viet Nam has focused principally on ride-hailing and delivery services, particularly in respect of whether app-based(car and motorbike) drivers are employees under the labour law. If these drivers are considered employees, they would enjoy various rights under labour and social security laws, such as minimum wage protection; reasonable limitation of working hours; rest breaks and paid leave; and social, health, and unemployment insurance coverage- to name just a few. Despite the importance of this question, Vietnamese regulators have yet to provide a clear and decisive answer to the legal status of app-based drivers. The Ministry of Labour, War Invalids, and Social Affairs(MOLISA) is an important institution responsible for the interpretation and enforcement of labour law. However, it has not clarified the legal status of app-based drivers. The Ministry of Justice(MOJ) is one state agency to have expressed a view on the legal nature of work arrangements between workers and platforms. In Report 45/BC-BTP, submitted by the MOJ to the Prime Minister on 17 March 2021, it stated: BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 47 [Workers] of ride-hailing platforms, such as Grab, are not paid by Grab, but receive transportation fees from end-users(customers) after subtracting service fees for their use of the platform. More importantly,[Grab] workers are not subject to Grab’s management, direction, and supervision as traditional employees.[They] have the right to turn off/on the app to participate in providing services when they want to do so, have the right to refuse an allocated trip or even cancel it after acceptance. Grab’s Code of Conduct can be considered an agreement between Grab and its workers because the workers agree to this Code of Conduct upon joining the platform[.] Grab calls workers participating in its platform‘worker partners,’ meaning that the workers are business partners rather than employees,[The] contractual arrangement between the[worker] and the platform is a commercial contract(be it a brokerage contract or an authorisation contract) (MOJ, 2021). Put differently, MOJ rejects the employee status of app-based drivers on the basis of three arguments. Firstly, it argues that these workers do not receive‘wages’ from platform entities, but transportation fees from customers. Secondly, it is contended that platform workers do not work under the‘management, direction, [and] supervision’ of platform companies, as exemplified by their discretion in turning on/ off the app and accepting or refusing bookings. The MOJ argues that the code of conduct with which platform workers must comply is not imposed unilaterally by platforms, but as per agreement between the two parties. Thirdly, the ministry reasons that platform workers are called ‘business partners’ by their platforms, rather than ‘employees.’ These arguments rest on Article 3(1) of the 2019 Labour Code, which defines an‘employee’ as a person who works under an agreement, is paid by an employer, and is subject to its management, direction, and supervision. The MOJ’s view appears to reflect mainstream opinion within MOLISA, though divergent perspectives do exist within the latter. 136 The MOJ’s position is, nonetheless, not shared by the courts, at least in the case of Vinasun vs Grab. In this case, Vinasun Corporation-a traditional taxi company-sued Grab for damages caused by alleged breaches of legal requirements regarding providing transportation services. 137 At the centre of the case were questions about whether Grab provided taxi transportation services and, if so, whether it had violated related legal requirements. 138 In addressing these questions, the courts of first instance and appeal both touched on the legal status of GrabCar workers. 139 The appellate court, the Superior People’s Court( Toà án nhân dân cấp cao) in HCMC, reasoned: [Grab] has carried out activities of a transportation service provider, including: storing and managing workers’[personnel] records; receiving customers’ bookings; organising communication between workers and customers; allocating bookings and deciding their journeys; determining prices upon the completion of bookings; receiving money[i.e., taxi fares] directly from customers; conducting promotional programmes and customer care; and handling customers’ feedback, including deducting workers’ remuneration if they displease customers. As such, it is believed that[Grab] has used the platform to put[worker] partners under its management. 140 [Emphasis added] BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 48 Moreover, the court of appeal held that Grab had violated legal requirements regarding car transportation services, including‘compliance with labour law[and] payment of social[and] health insurance contributions’ for workers. 141 In short, the court confirmed the employee status of GrabCar drivers. Vinasun vs Grab only concerns car ride-hailing services. The courts did not examine the cases of motorbike drivers or Grab’s delivery services. Unlike car transportation services, motorbike transportation services are largely unregulated. 142 There is no explicit requirement that transportation service providers comply with labour law in their relationships with motorbike(ride-hailing/ delivery) drivers. In particular, Vietnamese courts have never adopted the doctrine of stare decisis as in common law jurisdictions. 143 Therefore, it remains to be seen to what extent the viewpoint of the Superior People’s Court in HCMC on the status of GrabCar drivers will influence other courts and state agencies. The case at least illustrates that different state authorities-such as the MOJ and Superior People’s Court in HCMCsee platform work arrangements and their legal nature differently. Similar to MOLISA, the Viet Nam General Confederation of Labour(VGCL) has not expressed its formal position regarding the employee status of app-based drivers, let alone platform workers in general. The VGCL has, nonetheless, directed its Department of Labour Relations, Department of Propaganda and Education, and the Institute for Workers and Labour Unions to research the issue. 144 From time to time, senior VGCL officers have spoken in favour of app-based drivers. In an interview in 2021, Ngọ Duy Hiểu, VGCL’s Deputy President, strongly argued for these workers to be treated as employees(Trần, 2021): [Notwithstanding that] the owners of ride-hailing platforms always consider workers their partners, the fact that the workers have to obey directions of the platforms in respect of booking allocation[and] disciplinary measures as well as[labour] safety, [work] uniforms,[working] time clearly indicates[the existence of] employment relationships in enterprises. 145 In the same vein, Lê Đình Quảng, Deputy Head of the Labour Relations Department of the VGCL, stressed: Undoubtedly, this[contractual arrangement between Grab and its workers] is not a partnership contract, but a relationship between an employer and an employee. Grab has taken advantage of the inadequacy of labour law to put workers under the so-called partnership contract(Văn, 2020). Citing Article 13 of the 2019 Labour Code, Mr. Quảng further contended that the key issue was not how platform workers were categorised by platforms; rather, it was the existence of an agreement having features of an employment contract that mattered. 146 These statements, however, merely reflect the personal standpoints of VGCL officials, not the official view of their organisation. 147 As with MOLISA, there are differing opinions within VGCL. 148 Nevertheless, it is worth noting that our interviews with labour and union officials from 2019 to 2021 suggest increasing support for recognising app-based drivers as employees. 149 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 49 In summary, authorities have yet to clarify the legal status of app-based drivers. While refraining from stating their positions publicly, they have continued to discuss the issue internally with divergent opinions. The following section will illustrate that this reluctance stems from shortcomings relating to legal criteria to identify an employee under the 2019 Labour Code and their application to app-based drivers. Problems with the Employee Test and Its Application to Platform drivers The divergence between Vietnamese authorities and their reluctance to confirm or reject the employment status of ride-hailing work does not simply reflect unfamiliarity with this new type of work. As demonstrated below, it has been underpinned by a longstanding problem of labour law in Viet Nam: The lack of clear, established, and consistent criteria to determine the existence of an employment relationship. No less importantly, regulators have often adopted rigid, formalistic approaches to such determinations and failed to fully understand how platform(ride-hailing and delivery) work is performed and remunerated. The governing scope of the labour law has never been clearly defined in Viet Nam. The sole indicator implied in the 1994 Labour Code, the first labour code of the market reform era, is the existence of an‘employment contract’ ( hợp đồng lao động)( Labour Code 1994 arts 2, 6; Pham Cong Bay, 2010). Without further elaboration, this indicator was usually understood by administrative authorities and labour inspectorates as the existence of an agreement formally named as an‘employment contract’(Do Hai Ha, 2016a). The courts appeared to be less rigid and formalistic in their identification of employment contracts, as they considered a diverse range of factors, including: The subordination of the worker to managerial powers; supply of work tools, equipment, and facilities; remuneration of the worker; contents of the underlying contract and its title; the duration of work; working time; etc(Do Hai Ha, 2016a; Pham, 2010). However, they remained far from establishing a clear and standardised set of criteria, leading to inconsistent and unpredictable rulings in cases of employment status(Do Hai Ha, 2016a; Pham, 2010). To provide more clarity, the 2012 Labour Code introduced two indicators for an employment relationship-namely, that the worker‘is paid wages, and is subject to the management and direction of the employer’-whilst retaining the requirement of an employment contract( Labour Code 2012 art 3(1)). In practice, the notion that employees must have a contract titled ‘employment contract’ remained significant, paving the way for the increased use of sham employment arrangements(Do Hai Ha, 2016a). To tackle this problem, the 2019 Labour Code no longer requires that workers have an employment contract, only that they work under an agreement( Labour Code 2019 art 3(1)). Additionally, it states that if such an agreement has a name other than‘employment contract,’ ‘but... has contents relating to paid work, wages, and the management, direction,[and] supervision of one party, it shall be considered an employment contract.’( Labour Code 2019 art 13(1)). Simply put, it is not the title, but the contents of an agreement that determines the employment status of the underlying work arrangement. Notwithstanding the above-mentioned reforms, there remain problems with the criteria used to identify employment relationships. As we will see when we examine each of these below, this has caused difficulties in determining the legal status of platform motorbike drivers. Working under an Agreement The first indicator of an employment relationship under the 2019 Labour Code is that the worker works under an agreement. Baemin, Be, Grab, and Now drivers all satisfy this indicator because they have to sign an agreement with their platforms. According to these agreements, they are obliged to perform the tasks(rides or food/grocery/parcel delivery orders) that they receive through the BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 50 mobile apps. None of the investigated platforms calls such agreements employment contracts. Instead, they have used titles such as‘Cooperation Agreement’(Grab),‘Cooperation Contract’(Be), ‘Business Cooperation Contract’(Baemin), and ‘Independent Contractor Agreement’(Now). As discussed, these titles are yet to be considered a legal basis to deny platform drivers employee status. It should be noted that app-based drivers are not referred to as an‘employee,’ but a‘partner,’ a‘party to a partnership’ or an‘independent contractor’ in their agreements with platforms. Baemin and Now’s agreements go even further, explicitly stating that no employment relationship exists between platform and driver 150 . Though rejecting the relevance of the title of an agreement, Article 13(1) of the 2019 Labour Code emphasises the importance of its contents to the identification of an employment relationship. This means that the implicit/ explicit non-recognition of the employee status of platform drivers in their agreements with platforms would be taken into account. In fact, the MOJ has relied on this to argue against the employee status of platform drivers. Interviews reveal that such a view also exists in other state agencies and even union organisations. 151 In addition, proponents of this view often reason that the state should respect parties’ freedom to opt out of a contract of employment and refrain from intervening in market transactions. 152 This line of argument has two problems. The first is the regular dependence of Vietnamese regulators on the contract’s wording to determine the existence of an employment relationship. This approach departs markedly from the principle of primacy of facts set out in Paragraph 9 of Recommendation 198 of the ILO, which suggests that such determination‘be guided primarily by the facts relating to the performance of work and the remuneration of the worker’ instead of the contractual characterisation of the relationship. This principle has been widely recognised in other jurisdictions(ILO, 2007; ILO, 2013). The reliance on formal contractual arrangements does not merely represent a departure from internationally recognised legal principles. It is also inconsistent with Article 124 of Viet Nam’s Civil Code, which states that:‘Where parties enter into a civil transaction falsely for the purpose of concealing another transaction, such false transaction shall be invalid; notwithstanding, the transaction which is concealed shall be valid...’ The idea underlying this provision is that the legal nature of a transaction be assessed in light of the actual intent of the parties which, arguably, has to be inferred from the actual performance of the transaction more than its formal wording. Vietnamese courts have, from time to time, based their decisions on an employment relationship on the performance of work and the remuneration of the worker(Do, 2016a; Pham, 2010). Vinasun vs Grab is a case in point. Nevertheless, the courts have never consistently relied on factual circumstances and upheld their primacy over formal agreements(Do, 2016a; Pham, 2010). Quite often, they have based their rulings on employment status on contractual terms and titles(Do, 2016a; Pham, 2010). The rejection of the employee status of platform drivers on the basis of their contractual classification, therefore, indicates the continued adherence of regulators to formalistic approaches in deciding the employment status of workers. Article 13(1) of the 2019 Labour Code correctly rejects the significance of contract titles. But it does not represent a shift towards a factual approach to the identification of employment relationships. Rather, the new provision has reinforced the longstanding formalistic approach by pointing to the importance of contractual terms. The call to respect the freedom of platform companies and their drivers to characterise their relationships as an independent contractor relationship is also problematic. It disregards the unequal bargaining position between the two parties which, essentially, prevents platform drivers from having substantive negotiations with their companies. 153 The primary-though BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 51 not only-reason for the existence of labour law is to redress such imbalances of power by setting out labour standards and guaranteeing freedom of association(Richard Mitchell and Jill Murray, 2002). Such protections would be easily neutralised if the employers who dominate contract drafting and negotiation have absolute freedom to classify their workers as independent contractors in formal contracts. In conclusion, while app-based motorbike drivers meet the indicator of‘working under an agreement,’ many-though not all-regulators contend that these drivers have, explicitly or not, opted out of an employment relationship under their agreements with platform companies. Underlying this contention is the persistence of the dependence on formal contractual arrangements to determine the existence of employment relationships. Although Article 13(1) of the 2019 Labour Code has ruled out the relevance of contract titles to such determination, at the same time, it has reinforced this formalistic approach by emphasising the importance of contractual terms. Receiving wages Pursuant to Article 3(1) of the 2019 Labour Code, the second indicator of employee status is that the worker‘is paid wages’( được trả lương) for his/her work. Clearly, platform motorbike drivers receive incomes for the performance of work arising in connection with their agreements with platform companies. The question that then arises is whether such incomes are legally seen as‘wages.’ The 2019 Labour Code offers a few hints to answer this question. Article 90(1) reads: ‘Wage means the amount of money which the employer pays to the employee pursuant to the agreement in order for the latter to undertake work.’ In other words, payment to a worker would be considered a‘wage’ if it is paid by a party for whom the worker has performed work. The MOJ has opined that platform drivers are not employees of their platforms because the former are paid by customers rather than the latter(MOJ, 2021). This argument, however, disregards the fact that these workers regularly receive bonuses directly from platform companies. Representing the second-largest component(17.36 percent) of the monthly incomes of the surveyed workers, 154 these bonuses are rewarded to appbased drivers only if they meet performance and disciplinary criteria set out by platform companies ( Baemin Bonus Policy; Be Bonus Policy; Grab Bonus Policy; Now Bonus Policy). Calculated in accordance with formulas set out by platforms, these payments are clearly made by platform companies rather than end users. In addition, although the main income of an app-based driver-service charges after deducting commissions for his or her platform-seems to come from customers, the latter have no say in the determination of such earnings. The platform operator decides how each booking is charged (trip fare, delivery fee and, if any, additional fees and discount rate) and how much the driver can take from such revenues. 155 Here, the situation of the platform driver is similar to that of a traditional taxi driver who is remunerated on a revenue-sharing basis(the latter is usually recognised as an employee). Both drivers collect service charges from customers and retain a part of such revenues as remuneration for their work. In both cases, the rates of service charges and the sharing percentages are set by the platform and taxi companies and are accepted by the drivers on a‘take it or leave it’ basis. For this and the other reason stated above, app-based drivers cannot be seen as paid by customers, but by the platform company. Having established that app-based drivers receive payments from platform companies, the next question is whether these payments are made for work that the former undertake for the latter. Those rejecting the employee status of such workers reason that these are revenues shared between two business partners, rather than wages paid by platforms, 156 meaning that they do not work for the platforms, but for themselves. This reasoning fails to account for the bonuses that platforms pay their drivers. It is hard to consider these rewards revenue sharing in a partnership because they depend substantially BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 52 on the extent to which drivers adhere to platform policies and regulations, such as codes of conduct (i.e., disciplinary rules); performance targets like acceptance, completion, or cancellation rates; and customer service policies( Baemin Bonus Policy; Bee Bonus Policy; Grab Bonus Policy; Now Bonus Policy). Meanwhile, the revenue of a partnership would be distributed principally on the basis of the proportionate contribution of each party to co-business activities and its corresponding results( Civil Code 2015 art 504(1); Investment Law 2020 art 28(1)(c)). Such earnings are contribution- and outcome-based whereas the bonuses of platform drivers are rewarded on a performance basis. The latter are, therefore, more proximate to employee remuneration than revenue sharing in partnership arrangements. Significantly, the sharing of service charges between an app-based driver and his or her platform, in itself, is not sufficient to indicate a partnership. A genuine partnership does not merely contain revenue or profit sharing. It also involves the sharing of ownership, business operations, and liabilities arising thereof( Civil Code 2015 arts 504-509; Investment Law 2020 arts 27, 28). Such sharing is not clear between app-based drivers and platform companies, especially when it comes to business operations. Deciding how ride-hailing and delivery services are charged and rendered to customers, distributing tasks to drivers, and closely monitoring their performance, platform companies have, arguably, exercised dominant control over the operation of these services. These relationships cannot be seen as partnerships, regardless of the existence of revenue/profit sharing. Moreover, in defining‘wages’, Article 90(1) of the 2019 Labour Code neither implies nor excludes any particular form of payment. Conventional employment does involve outcome-based compensations, such as piece wages or profitsharing bonuses( Labour Code 2019 arts 90; 95; 96; 104). In addition, Article 13(1) of the 2019 Labour Code refers not only to‘wages’ ( tiền lương), but also‘remuneration’( tiền công) as an element of employment agreements. Despite the absence of an explanation of the latter, this indicates that the Code contemplates a broad and flexible conception of employee payment. Notably, although Vietnamese courts were initially of the view that wages had to take the form of regular, periodic payments, they have subsequently accepted other forms of remuneration, including profit sharing(Pham, 2010). It should be emphasised that ILO Recommendation 198 does not consider income or profit-sharing as grounds to deny the employee status of workers. It merely suggests that some forms of payment, like periodic payment or remuneration in kind, be regarded as indicators of employment relationships( Recommendation 198[13]; ILO, 2013). What’s more, the ILO has adopted a broad and flexible conception of‘wages’ which disregards their designation and calculation. Article 1 of Convention 95 concerning the Protection of Wages says: In this Convention, the term wages mean remuneration or income, however designated or calculated, capable of being expressed in terms of money and fixed by mutual agreement or by national laws or regulations, which are payable in virtue of a written or unwritten contract of employment by an employer to an employed person for work done or to be done or for services rendered or to be rendered.[Original emphasis] To sum up, there are good grounds to say that app-based motorbike drivers are not paid by end-users, but platform operators, and that their earnings are not essentially different from, but substantially comparable to, employee remuneration. Despite this, the existence of a narrow, rigid conception of wages-which excludes revenue/profit-sharing remunerationhas produced opposing opinions among and within regulators. Such views also have their roots in failures to fully understand how appBEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 53 based drivers are actually remunerated. In part, these failures reflect the regulators’ struggle to comprehend the distinctive features of ride-hailing work compared with traditional employment, namely: Its multi-party involvement; disaggregation into short-term tasks; and‘payas-you-go’ compensation. Working under Management, Direction, and Supervision The third legal criterion to determine the existence of an employment relationship is that one party is subject to the‘management, direction, and supervision’ of the other party 157 . Equivalent to the‘subordination’ criterion recognised in national labour laws across the world(Giuseppe Casale, 2011), this was not recognised in Vietnamese labour legislation until the adoption of the 2012 Labour Code 158 . Vietnamese courts have, nonetheless, applied it in their rulings about employee status for decades(Pham, 2010). Notwithstanding its official recognition, the criterion of‘working under management, direction, and supervision’ remains to be clarified. MOLISA, the institution mainly responsible for guiding the implementation of the Labour Code, has never elaborated the meaning of this criterion. Nor has it provided guidance on how to testify if a worker is subject to‘management, direction, and supervision.’ Whilst labour administrations remain silent, court rulings have shed some light on the newlyintroduced criterion. To assess the subordination of a worker to his or her principal, the courts have looked at various aspects of working arrangements, including the assignment of tasks, instruction of working methods, organisation of work, work and rest times, control over work performance, and setting of performance targets. However, they have failed to develop well-defined and uniform standards for such assessment and have not always resorted to the subordination test to determine the existence of employment relationships. Normally, the courts give weight to this indicator only if there is no written contract or the contract is unclear. Vinasun vs Grab is a rare case in which the court prioritised the subordination factor over contractual characterisation. As discussed, the impact of this case remains to be seen. The uncertainties over the criterion of‘working under management, direction, and supervision’ have impacted its application to app-based drivers. Divergences have emerged within and between regulators as well as the VGCL with respect to whether app-based drivers are subordinate employees. For those arguing against the subordinate nature of these workers, the main concern is the discretion of app-based drivers in turning on or off platform apps and in accepting, declining, or cancelling allocated tasks. 159 In their opinion, these are indicative of the autonomy-rather than subordination-of appbased drivers. This contention, however, overlooks the fact that the apparent freedom of app-based drivers is substantially restricted in practice. All of the investigated platforms require some degree of work commitment from drivers, the absence of which would subject them to disciplinary action, such as suspension or termination of the contract. For instance, if a GrabBike worker fails to meet the acceptance or cancellation rate prescribed by the platform, his or her account would be temporarily or permanently deactivated depending on how many times the failure was committed 160 . Baemin, Bee, and Now all maintain similar regulations 161 . Furthermore, as discussed above, app-based drivers must achieve high performance metrics, for example, in relation to acceptance, completion, and cancellation rates; working time; and revenues to receive regular bookings and bonuses from platforms. Failure to do so does not necessarily trigger disciplinary action. Nevertheless, since receiving regular bookings and bonuses is essential for drivers to earn a liveable income, they usually choose to work long hours and accept the tasks assigned by platforms. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 54 Put differently, the theoretical freedom of appbased drivers to choose tasks and decide their working time is largely curbed by platform companies by two means: Disciplinary measures and economic incentives. The latter’s impact varies according to workers’ financial dependency on their platforms. Workers not overly-reliant on ride-hailing work for their income-such as those working part-time, with multiple sources of income, and/or with no economic dependents-would normally enjoy more discretion in determining their working time. 162 Meanwhile, for those depending heavily on such work-like those working full-time, with little to no income from other sources, or with economic dependents-such discretion is substantially limited. 163 One veteran ride-hailing platform worker realised this and commented: This is not an additional job, but a job that requires high[work] commitment. It is not a[legally] imposed commitment, but our[workers’] own commitment to earn a living. At the beginning, we were said to work only when we had free time. But, if we do so,[we] would not have sufficient income. If a worker works every day [and] ten hours per day, he/she will be allocated many orders because Grab favours such workers. 164 Accordingly, the actual autonomy of a platform driver is very limited, especially when he or she depends largely on platform work for a liveable income. In this case, his or her autonomy, if any, merely exists to the extent that he or she can decide the start and end-rather than the length-of his or her working time. This limited flexibility alone is inadequate to reject employee status. In addition, a flexible work schedule has never been a unique feature of independent contractor relationships, but also legally exists in employment relationships. 165 Another argument which has been advanced by the MOJ to reject the subordinate nature of app-based drivers is that these workers are not forced to comply with the codes of conduct, i.e., disciplinary rules, issued by their platforms, but have voluntarily agreed to do so(MOJ, 2021). Put another way, the MOJ contends that the managerial power of the platform is not an inherent feature of the relationship because it stems from mutual consent between the two parties. The problem with this argument is that the 2019 Labour Code does not, explicitly or implicitly, contemplate such an understanding of subordination. In fact, there are no major differences between the code of conduct applicable to a platform driver and the work rules ( nội quy lao động) applicable to a conventional employee. Both are unilaterally enacted by the company and serve as the legal foundation for its managerial power. Both have a binding force upon a worker only if there exists an agreement between the two parties. In both cases, the company retains its right to revise rules at any time while, for the worker, the agreement to be bound by such rules is only a take-it-or-leave-it clause. Meanwhile, those arguing for app-based drivers to be recognised as subordinate employees have failed to present a consistently reasoned argument. To support their position, they have cited various different reasons, with limited explanation, including, for example: ▪ The platform company retains personnel records of the driver; ▪ The platform company assigns bookings to the driver, and decides the price and journey for each booking; ▪ The platform company controls communication between the driver and clients; ▪ The driver has to wear a platform uniform; ▪ The driver has to follow company policy on working time; ▪ The driver has to comply with safety measures prescribed by the platform, and; BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 55 ▪ The platform company has power to discipline the driver if the latter violates its code of conduct. 166 The lack of consistency and reasoning has rendered the argument for the subordinate nature of app-based drivers less convincing. 167 This, alongside the existence of dissenting opinions, explains why regulators have been hesitant to give a decisive answer on the status of such workers. As shown above, the problem stems partially from the struggle of regulators to comprehend new aspects of ride-hailing work. Compared with the traditional control of an employer over an employee, the control of a platform company over app-based drivers is less visible. This is because the latter is often exercised in a remote, indirect manner and hidden behind algorithmic management, incentive-based schemes, and strategically structured contractual arrangements. Failure to understand these features has led several regulators to deny the subordinate nature of app-based drivers. The absence of a well-defined and uniform standard or method to evaluate the subordination of a worker has made the determination of the employee status of app-based drivers more troublesome. This is reflected in the failure of some regulators to advance a consistent, reasoned argument to substantiate the subordinate nature of such workers. It can also be seen in the conflicting ways that regulators view the freedom of platform drivers in choosing their work and time schedules. Another illustration is the contention that the managerial power of a platform company is different from that of an employer because the former derives from the worker’s consent. As discussed, this contention represents a misunderstanding of the‘subordination’ concept contemplated in labour law. Moreover, as with the first two criteria, the criterion of‘working under management, direction, and supervision’ has been frequently interpreted in a rigid, formalistic manner. As already seen, several regulators have relied mainly on the flexible work schedule of appbased drivers, without considering other facts, to argue against the subordinate nature of their work relationships. This exhibits a very strict view of the subordinate relationship between an employee and an employer which subjects the former to direct, total control by the latter. The reliance on the worker’s consent to be bound by the platform’s code of conduct to reject his/her subordination, meanwhile, indicates a formalistic approach which places more emphasis on contractual wording than factual circumstances. In a nutshell, the existence of conflicting views about the subordination of app-based drivers to their platforms has contributed to the reluctance of regulators to clarify the employee status of these workers. In part, the problem reflects the failure of regulators to realise the control that platform companies have over their workers hidden behind algorithmic management, incentive-based monitoring, and complicated contractual arrangements. To a large extent, it also stems from the lack of an established, consistent standard or method to identify a subordinate worker, the prevalence of a strict perception of subordinate relationships, and the frequent dependence on contractual wording to distinguish between dependent and autonomous workers. Summative remarks This discussion reveals the three factors contributing to the neglect of Vietnamese authorities to clarify the status of platform drivers in Viet Nam. Firstly, no coherent or consistent understanding or method has been established to assess two indicators of an employment relationship, namely: The receipt of wages by the worker and his/her subordination to another party. This has resulted in divergent opinions among regulators about platform drivers, discouraging them from giving a conclusive answer on the status of these workers. Secondly, regulators have regularly employed rigid, formalistic approaches to determining the existence of employment relationships, including relying extensively on formal agreements instead of factual circumstances, adopting a BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 56 narrow conception of wages, and maintaining a strict view about subordinate relationships. Such rigidity and formalism not only deviate from international standards and practices, but also lack a sound basis in Vietnamese law. Explicating the existence of such rigid, formalistic approaches requires further research. Nevertheless, the above analysis suggests that, in applying such approaches, regulators have failed to pay adequate attention to the traditional goal of labour law: Redressing the lack of bargaining power of workers and protecting vulnerable workers. 168 Last, but not least, regulators have failed to comprehensively capture how platform drivers perform their work and how they are remunerated by platform companies. In part, this failure reflects their struggle to unravel novel aspects of this newly-emergent type of work, such as multi-party involvement, gig-based work,‘pay-as-you-go’ remuneration, algorithmic management, incentive-based monitoring, and complex contractual arrangements. The impact of legal uncertainties on the vulnerability of platform motorbike drivers The uncertainty over the legal status of appbased mototorbike drivers has contributed to and reinforced their vulnerability in at least three ways. First and foremost, it has enabled platform companies to categorise these drivers as independent contractors and, in doing so, exclude them from labour rights and social protection that apply to employees. As demonstrated in Part Two, this exclusion plays an important role in the vulnerability of platform motorbike drivers. Take the low and unstable incomes of app-based motorbike drivers, for example. As analysed above, the average income of drivers in our survey working a standard 48-hour working week was lower than the legal minimum income of a trained employee, if statutory benefits under labour and social security legislation were included in the latter’s income. This would never have happened if platform drivers enjoyed minimum wage protection, paid leave, and social security contributions. Further to this, they would have been entitled to overtime and night-shift pay like conventional employees. Recognised as employees, their incomes would also have been less vulnerable to policy changes made by platforms because, in that event, they would have been consulted or their consent sought on changes that might impact their remuneration ( Labour Code 2019 arts 33, 91, 98). In a similar vein, app-based motorbike drivers’ long working hours, insufficient rest, and lack of labour safety is closely linked with the deprivation of their basic labour rights, such as reasonable limitations on working hours, provision of paid leave, and safe and healthy working conditions. Classified as independent contractors, they are also deprived of protection from managerial abuses and job insecurity under labour law. For the same reason, they have been excluded from state-run social security schemes. This has made them more vulnerable to work and health-related risks, missed work, lost income and job loss and, accordingly, pressed them to work harder to safeguard against such uncertainties. Secondly, the lack of legal protection of platform motorbike drivers-the direct consequence of their undetermined legal status-has allowed platform companies to leverage a superior bargaining position to establish unfair contractual arrangements. Baemin, Be, Grab, and Gofast (Now’s owner) all exercise total control over the drafting and negotiation of agreements with drivers, leaving the latter to‘take it or leave it.’ 169 Unsurprisingly, these agreements give platform companies substantial advantages over platform drivers. Normally, they require that the driver supply, or otherwise pay for, working equipment and devices(other than the software app), ensure their operability and legality, and bear all operating costs(except for those incurred in the development and operation of the digital platform. 170 Additionally, he or she has to assume all risks, damages, and liabilities arising during his or her performance of work. 171 By dominating the contract drafting and negotiation process, platform companies have shifted most major risks and responsibilities to motorbike drivers. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 57 While setting out detailed and burdensome obligations for the worker, Baemin, Be, Grab, and Now’s agreements with drivers also grant these platforms enormous power, discretion, and flexibility. In doing so, they have increased the vulnerability of platform drivers. All these agreements state that the platform company retains the right to change commission rates and bonus policies during the course of the contract without consulting or obtaining consent from the worker. 172 It is also stated that the former has the right to determine and revise, at any time, the methods, standards, and conditions for the services to be rendered by the latter(despite his or her characterisation as an independent service provider under the contract). 173 At the same time, the platform company-while exempted from the legal responsibilities of an employer-has the power to direct and manage the worker, to enact and freely revise the rules by which he or she must abide during his or her performance of work, and to discipline him or her for violating such rules. 174 Meanwhile, platform drivers are provided with no protection from arbitrary policies and decisions, whether in their contracts or company rules. The unfair nature of the contractual arrangement between platform companies and their drivers is well captured in the following statement from a veteran app-based driver:“We have no say in these agreements. No rights. No entitlements. Just obligations.” 175 Last, but not least, the unclear legal status of platform drivers has impeded attempts to address their vulnerable situation. As discussed, this has precluded them from organising into trade unions or WROs. Although the HCMC Federation of Labour has established syndicates for platform drivers, these syndicates do not have the right to represent their members in bargaining, consultation, and other dialogical processes with platform companies. 176 Nor do they have the right to represent workers in dispute resolution. This has, arguably, circumscribed the capacity of platform drivers in defending and furthering their occupational interests. Several driverrespondents acknowledged the problem of lacking representation. 177 One GrabBiker stated: I really want to have a representative organisation. All workers want this. [However,] as motorbike drivers(xe ôm) we cannot have such an organisation. We don’t have a trade union. That is why we have to bear miserable conditions. If there is a trade union, all of us will support it. 178 The vagueness of the employee status of platform drivers has also discouraged them from seeking redress for their situation. One GrabBike worker, for instance, said that:“If I sue the platform company, nobody will help me. I am only a “partner.” If Grab no longer wants my services, it can release me easily.” 179 Likewise, another commented:“Making complaints would change nothing. Grab has specified in the contracts[with drivers] that its decisions are always right. How can we speak our opinions?” 180 In other words, the perception of a weak legal position has made platform drivers reluctant to fight for their rights. Significantly, as the question about the employee status of app-based motorbike drivers remains unresolved, labour authorities and trade unions have hesitated to engage with and support collective action by such drivers, 181 unlike the wildcat strikes undertaken by factory workers (Do Hai Ha, 2016b; Lee Chang-Hee, 2006). Interviews with labour officials and union cadres indicated that they were unsure whether to intervene in emergent conflicts between platform companies and app-based drivers and, if so, how such conflicts should be handled. 182 In addition, since the right to strike is only recognised for employees, 183 there is no clear legal basis for strike action by platform workers. Nor are there sound legal grounds for their demands in strike incidents. This has reduced the sympathy of labour authorities and trade unions towards strike action by app-based drivers and their demands. 184 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 58 To conclude, the ambiguous legal status of platform drivers is an important contributor to their vulnerability. As these workers are not yet recognised as employees, they have been excluded from basic labour rights and social protections. Not only does this pave the way for platform companies to maintain unfair contractual arrangements, this legal ambiguity also hinders motorbike drivers from effectively fighting for more decent work. Part III has underlined how the unclear status of platform motorbike drivers has shaped their vulnerability. It has also shown that the problem stems from the lack of established, well-defined parameters to determine the existence of employment relationships; the frequent reliance of regulators on a rigid, formalistic approach to such determination; and their difficulty comprehending novel features in the way that platform drivers perform work and receive remuneration. The next section of the report considers possible solutions and reforms that can enhance legal protection for platform drivers with a view to safeguarding and providing a more equitable work environment. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ PART IV ENHANCING LEGAL PROTECTION FOR PLATFORM MOTORBIKE DRIVERS IN VIET NAM: OPTIONS AND RECOMMENDATIONS 60 ENHANCING LEGAL PROTECTION FOR PLATFORM MOTORBIKE DRIVERS IN VIET NAM: OPTIONS AND RECOMMENDATIONS The vulnerability of platform motorbike drivers, underpinned by the close relation to the uncertainty regarding their employee status and its consequences, including their de facto deprivation of labour and social security rights, underscores the urgent need to provide legal protection for these workers in Viet Nam. In light of the local situation and legal developments in other jurisdictions and international laws, this section discusses four potential solutions, namely: (i) Clarification of the existing legal criteria to determine the existence of an employment relationship;(ii) reform of the legal framework to that end;(iii) the introduction of a rebuttable legal presumption of employment status for platform workers, and;(iv) enactment of specific protections for platform workers. These solutions are neither exhaustive nor exclusive, and can be implemented at the same time or one after the other. Clarifying existing criteria to Identify employment relationships As we have seen, the vagueness of legal criteria to identify an employment relationship has resulted in divergent opinions among and within regulators, discouraging them from confirming the employment status of platform drivers. Properly enhancing the clarity of these criteria would reduce legal uncertainty regarding their employment status and increase their chances of accessing labour law and social protection. Receipt of wages The first criterion that requires elaboration is the worker’s receipt of wages. As shown above, different opinions have arisen about the legal nature of platform drivers’ earnings, including a rigid view that such earnings are not wage payment because they take the form of revenue sharing. This can be addressed by making it clear that employee remuneration does not need to include or exclude any particular form. That is to say, the employee is unnecessarily paid on a regular, periodic basis. Nor is he/she excluded from receiving shared profit or revenue. As previously discussed, such a broad conception of wages is well supported by the 2019 Labour Code and aligned with the‘wage’ conception enshrined in ILO instruments, like Convention 95 and Recommendation 198. It should be mentioned that such clarification does not deny the relevance of how a worker is remunerated to the determination of his/her employment status. This is an important factor for such determination under Recommendation 198 and national laws across the world ( Recommendation 198[13]; ILO, 2013). Indicators that an employment relationship has been established include where the worker:(i) Is remunerated on a periodic basis;(ii) receives remuneration that constitutes his or her sole or principal source of income, or;(iii) is provided with payment in kind( Recommendation 198 [13]; ILO, 2013). What must be emphasised here is that the absence of such facts does not necessarily point towards an independent contractor relationship, especially where other facts indicate the contrary. Subordination Guidance should be provided to elaborate the meaning of the‘subordination’ factor or-using the words of Article 3(1) of the 2019 Labour Code-‘working... under the management, direction, and supervision.’ While subordination has been widely seen as a central feature of the employment relationship(Casale, 2011), it was introduced to Vietnamese labour legislation only recently and without explanation. As seen above, the absence of a cohesive, consistent understanding of this criterion is closely linked BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 61 with the existence of conflicting views about the subordinate nature of platform drivers. Significantly, it has allowed a strict view about subordination that contemplates the employer’s direct and constant control over the employee. This, in turn, has impeded the recognition of workers working under a subtler or indirect form of supervision and control and enjoying some degree of flexibility, like platform drivers, as employees. For a basic elaboration of the subordinate relationship between employee and employer, the Vietnamese authorities could consult a recent report of the ILO’s Committee of Experts on the Application of Conventions and Recommendations(CEACR)(ILO, 2020). Drawing from comparative law research, this report describes‘subordination’ as a situation in which ‘the employee is required to follow the instructions of the employer’(ILO, 2020). Citing Casale, it characterises this subordinate relationship by three powers that the employer exercises over the employee(Casale, 2011). The first is the ‘directional power’ which enables the former to assign tasks and give orders and directives to the latter(Casale, 2011). The second is the‘control power’ that subjects the performance of such tasks by the employee and his/her compliance with such orders and directives to the employer’s supervision(Casale, 2011). The last is the ‘disciplinary power,’ permitting an employer to sanction an employee for improper or negligent performance of the assigned tasks, given orders, and directives(Casale, 2011). Nonetheless, Casale also indicates that the application of such conventional concepts requires flexibility‘which goes against imperative and rigid labour law’ to adapt to the transforming global labour market 185 . Several facts indicate that the four investigated platforms have exercised the three powers described above over their drivers. These indicators are summarised in Table 9 below. The sole major fact that seems not to accord with such powers is the freedom of the platform driver to turn on and off his/her application. Nevertheless, as discussed earlier, such freedom is rather limited and, therefore, insufficient to reject the extensive subordination of the driver to the platform, especially where he/she relies on ride-hailing work for their main source of income. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 62 Table 9. Platforms’ hierarchical powers over drivers Dimension Indicators Directional Power - The platform allocates bookings to the driver and decides the route and price of each booking 186 - The platform provides directives to the driver as to how to perform his/her work, including but not limited to: Dress code; how to use the platform app; receipt and discharge of passengers; acceptance and delivery of orders; collection of charges and fees; and handling customers and other involved parties(e.g., restaurantpartners) 187 - The platform gives instructions or warnings to the driver via the platform app in particular circumstances, such as safety warnings in the case of excessively long working hours. 188 - The platform provides training, including compulsory training, to the driver, such as training in customer handling, order processing, traffic law and safe driving, and the company’s code of conduct. 189 - The driver must comply with rules enacted by the platform during his/her performance of work. These rules usually, but not solely, take the form of a code of conduct( Baemin COC; Be COC; Grab COC; Now COC). Control Power- The driver is not allowed to transfer his/her account to or have his/her work performed by anyone else( Baemin Driver Agreement art 4(2)(h); Be COC s 11; Grab COC s 13(3); Now Driver Agreement appx 1 s 4(2)(j)). - The platform oversees work performance by the driver and his/her compliance with its orders and directives by multiple means, including:(i) Technological instruments(like GPS, platform app, algorithms, and rating systems);(ii) customers’ feedback and ratings, and;(iii) supervision by company personnel(such as team leaders or‘undercover inspectors’) 190 Disciplinary Power - The driver will be sanctioned for failures to meet performance expectations set out by the platform, such as minimum acceptance, completion, and cancellation rates; minimum customer ratings; and revenue targets. Such sanctions include, but are not limited to, reducing assigned tasks, exclusion from bonuses, degradation or temporary/permanent deactivation. 191 - The driver will be sanctioned for breaches of the code of conduct or other company rules. Such sanctions include, but are not limited to, a reprimand, monetary fine, mandatory re-training, reduction of assigned tasks, and temporary/ permanent deactivation. 192 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 63 Accordingly, if the existing criteria to identify employment relationships is clarified in line with the above discussion, motorbike drivers working for platforms like Baemin, Be, Grab, and Nowespecially those depending principally on ridehailing work to make a living-would have a good chance to claim employment status, labour rights, and social protections. Administrative guidance as the main means Different means can be used to enhance the clarity of the existing criteria to differentiate between employment and self-employment, including legislative reforms, judicial decisions, and administrative guidance. In the context of Viet Nam, the most effective means appears to be administrative guidance, whether in the form of implementation regulations or specific guidance in particular cases. Compared with legislative change, the adoption of administrative guidance would generally take less time and, therefore, better serve the urgent need of platform drivers. Although judicial clarification has played a crucial role in the clarification of employment status criteria in numerous jurisdictions-including in relation to platform work(Valerio De Stefano et al , 2021)-it has a number of disadvantages in the Vietnamese context. Legal proceedings do not only take time, as exemplified in Vinasun vs Grab. The courts cannot initiate judicial processes by themselves, but have to wait until someone files a classification lawsuit. Considering financial and technical impediments for individual workers to take such action, this would require an effective, legally capable organisation to assist and represent workers before the courts. The problem is that platform drivers in Viet Nam lack such a representative. Further, since the doctrine of stare decisis is not recognised in Viet Nam, a court decision-such as that in Vinasun vs Grab would not have a binding force in similar cases, unless it was officially selected by the Supreme People’s Court as a legal precedent. Such selection would entail a lengthy, complicated process that requires a degree of support from several state actors. 193 Indeed, the Supreme People’s Court has been very cautious about introducing legal precedents, selecting only two labour cases since 2015. 194 For the reasons discussed above, administrative authorities, particularly MOLISA, should play a major role in clarifying the criteria used to identify an employment relationship. This does not rule out the possibility of legislative change in the future, especially for the sake of bold reforms that will be discussed below. Nor does it exclude the role of the courts. Judicial authorities can contribute to the enhancement of legal certainty through their consistent and coherent rulings on employment status over time. Reforming the legal framework for identifying employment relationships While clarifying existing criteria to determine employment status may increase the chance of platform drivers accessing labour and social protection, this cannot resolve all problems arising due to the growth of this new type of work. As illustrated above, the reluctance of Vietnamese regulators to decide the status of platform drivers has not merely stemmed from the lack of specificity of these criteria. It is also linked to their frequent adoption of rigid, formalistic approaches to distinguishing employment from self-employment. Additionally, they have struggled to classify ride-hailing work partly because this work involves greater flexibility and less direct, recognisable control over the worker than conventional employment. All of these factors suggest that Viet Nam’s legal framework to determine employment status needs to be thoroughly reformed and adapted to new emerging forms of work, such as platform work. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 64 The shift towards more flexible work arrangements and regulatory responses around the world Viet Nam is not the only country to have encountered the problems noted above. Nor are these problems related only to ride-hailing work. Rather, they represent a challenge at the global scale: The increasing blurring of lines between dependent and independent workers which has its roots in profound transformations in the world of work, caused by globalisation, technological innovations, organisational changes, and other factors, and attempts to disguise employment relationships(Casale, 2011; ILO, 2020). This has prompted national authorities around the world to review longstanding legal definitions, methods, criteria, and indicators used to determine the existence of employment relationships and adapt them to this new situation(Casale, 2011; ILO, 2020). Although countries have varied in their responses to the rapid transformation of the labour market, some general developments can be observed. Firstly, national regulators have increasingly based their decisions on employment status on a combination of multiple factors. While the ‘subordination’ element remains important, other factors are increasingly taken into consideration (Casale, 2011; ILO, 2020; De Stefano et al, 2021). At the same time, regulators in many countries have adapted the concept of subordination to the changing situation, departing from the traditional perception of the subordinate employee as a worker subject to the direct and constant control of the principal(De Stefano et al, 2021). Other major developments include designating specific groups of workers as employed or self-employed, easing the burden of proof for workers and introducing legal presumptions of the existence of an employment relationship where certain conditions or indicators exist(ILO, 2020). The growing mismatch between the legal scope of the employment relationship and the realities of working arrangements has also triggered regulatory attempts at the international level, including the adoption of Recommendation 198 by the ILO in 2006(ILO, 2020). The Recommendation is an outcome of about a decade of discussion within the ILO as to how to address the situation of workers excluded from the employment relationship and, therefore, the protection of labour law(ILO, 2020). By setting an international standard on determining the legal scope of the employment relationship, it aims to assist national states to develop policies that ensure the coverage of workers needing protection under the laws that regulate such relationships(ILO, 2020). To this end, the instrument seeks to balance the traditional role of labour law in‘address[ing] what can be an unequal bargaining position between parties to an employment relationship’ with its contemporary challenge in ensuring protection ‘accessible to all, particularly vulnerable workers.’ (De Stefano et al , 2019). Recommendation 198 contains three main parts, with the first devoted to the construction of a national policy on the employment relationship. The second part focuses on criteria to determine the existence of an employment relationship whilst the last addresses mechanisms for monitoring employment relationships and enforcing relevant rules. Recommendation 198 as a Generic Guide Recommendation 198 can be a helpful reference for Viet Nam to reform its legal framework for determining the existence of an employment relationship and strengthen protection for vulnerable workers in new forms of work and employment, such as app-based motorbike drivers. Firstly, the instrument offers relatively comprehensive guidelines in this respect whilst allowing state members to adapt them to national circumstances. Secondly, it was developed based on a wide-ranging comparative law study and extensive deliberations with tripartite constituents, national delegates, and high-profile experts(Casale, 2011; ILO, 2020). As a result, the instrument was drafted in light BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 65 of the laws, practices, needs, and proposals of a diverse range of jurisdictions that represent different regions, legal systems, and traditions (Casale, 2011). Lastly, it has been increasinglythough diversely-received in domestic law and jurisprudence across the world, including both advanced and emerging market economies(ILO, 2013). It is, arguably, a good starting point for a country looking for external ideas and models to modernise the legal scope of the employment relationship and ensure its compatibility with the transformation of an increasingly globalised labour market. Moreover, it is not completely new, but was previously introduced to the Vietnamese authorities(ILO, 2011). This report does not intend to have a lengthy discussion on Recommendation 198. It only highlights some major aspects of the instrument which are particularly relevant to how to address the situation of platform drivers in Viet Nam. One of these is the call for state members to set up a national policy to review and, when necessary, clarify and revise the scope of the employment relationship( Recommendation 198[1]). Although its primary objective is to ‘guarantee effective protection of workers who perform work in the context of an employment relationship,’ such a policy would also warrant‘fair competition’ between employers ( Recommendation 198 Preamble). For these purposes, the Recommendation suggests that the review of the employment relationship’s scope be conducted periodically in consultation with social partners. This suggestion is worth consideration by the Vietnamese authorities. As shown above, the country is not an exception to the global trend towards more flexible working arrangements. Although its national regulators have increasingly become aware of the challenge and recently attempted to clarify the scope of the employment relationship(Government, 2019), no specific policy has been institutionalised to warrant that such review and clarification be conducted regularly. Arguably, this has played a part in delays in modernising legal criteria to determine the existence of an employment relationship and in clarifying the employment status of platform drivers. These delays have not only contributed to the vulnerability of these workers. As indicated in the ruling of the HCMC People’s Court in Vinasun vs Grab, they have also enabled ride-hailing companies to benefit from unfair advantages over traditional transportation companies. 195 The review, clarification, and revision of the employment relationship’s scope does not necessarily take place through legislative processes. Though important, legislation is not the sole means to implement these tasks. Countries have employed many other means, like administrative sources, judicial decisions, collective agreements, codes of practice, or a combination thereof(ILO, 2020). In addition, they have conferred the responsibility on various bodies, including legislative organs, administrative agencies, judicial authorities, tripartite advisory bodies, and labour inspectorates. 196 Influenced by national law and practice, the functions of such bodies vary considerably between and within countries, ranging from reviewing legislation to clarifying existing laws, proposing new regulations, and providing guidance to the concerned parties. 197 Viet Nam may consider adopting a system which delegates these functions to multiple bodies. Legislative review and, if needed, revision would be periodically carried out by the National Assembly with the involvement of the existing tripartite institution-the National Committee of Labour Relations-or a more wideranging( ad hoc or permanent) consultative body coordinated by MOLISA. 198 Such a body may include, aside from delegates of MOLISA and social partners, those from other relevant authorities, such as judicial, social security, and tax authorities, and high-profile labour experts. Consistent with Vietnamese law and practice and to foster timely responses to labour market developments, MOLISA should be authorised to enact implementing guidelines regarding how to distinguish between employment and selfemployment and to provide information and advice to relevant parties, including(if requested) BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 66 in specific cases. Judicial authorities can also play a significant role. As discussed, consistent and coherent court rulings would enhance the clarity of legal criteria to determine employment status, in general, and the status of newlyemerging working arrangements in particular. Additionally, considering their experience in dealing with the question of employment status, judicial authorities should be more active in clarifying statutory provisions by introducing legal precedents, participating in legislative reviews, and proposing legislative change. Another major aspect of Recommendation 198 is the recognition of the principle of‘primacy of facts’. This suggests that the existence of an employment relationship be determined primarily based on‘the facts relating to the performance of work and the remuneration of the worker’ rather than how the relationship is contractually characterised( Recommendation 198[9]). That is to say, if there is any divergence between the established facts and contractual terms, whether textual or verbal, the former shall prevail (De Stefano et al, 2021). Well entrenched in statutory and case law of numerous jurisdictions, this principle is considered an important tool to tackle disguised employment(De Stefano et al, 2021). The‘primacy of facts’ principle has not been established in law or jurisprudence in Viet Nam. This has led regulators to depend heavily on formalistic contractual terms in deciding the employment status of app-based drivers, enabling platform companies to easily classify their workers as independent contractors and exclude them from labour law protection. Upholding the primacy of facts is, therefore, essential to attempts to claim employment and labour rights for platform drivers and other new types of workers. This will also remove incentives to disguise the employment relationship in general-a problem receiving growing attention from national regulators in recent labour law reforms(MOLISA, 2019). Recommendation 198 does not offer a definition of the employment relationship. However, it encourages state members to clearly define ‘the conditions(criteria) applied for determining the existence of[such a] relationship,’ listing ‘subordination’ and‘dependence’ as two examples( Recommendation 198[12]). These two factors-also known as‘legal subordination’ and‘economic dependency’-have been broadly used in national laws to describe the dependent relationship between an employee and an employer, i.e., the employment relationship (Casale, 2011; ILO, 2020). Although these terms are given the same meaning in some jurisdictions, they are usually understood differently(ILO, 2020).‘Economic dependency’ is deemed to exist where the remuneration of the worker constitutes his or her sole or principal source of income, or where such remuneration is paid by a person or enterprise in return for the worker’s activity, or where the worker is not economically autonomous and is economically linked to the area of activity in which the employer operates(ILO, 2020). While the definition of the employment relationship has traditionally focused on‘legal subordination,’ ‘economic dependency’ has become increasingly important and, in some countries, been used as an additional means of proof where there is doubt about an employment relationship(ILO, 2020). As suggested above, Viet Nam should better define legal criteria to determine the existence of an employment relationship, including two criteria: (i) The worker works under the management, direction, and supervision of another party, and; (ii) receives wages for such work. It should also consider introducing‘economic dependency’ as a factor in such determination. This is not simply due to the broad and increasing use of this factor in other jurisdictions. As the case of platform motorbike drivers has revealed, the economic dependence of a worker reduces his/her actual autonomy from the hiring party and increases his/her degree of submission to the latter’s power. Consequently, the more the worker is economically dependent on the party requesting work, the more he/she is likely to suffer from an unequal bargaining position and, therefore, the more he/she is in need of legal protection. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 67 Recommendation 198 also calls for state members to spell out in their laws, regulations, or any other means specific indicators of the existence of an employment relationship( Recommendation 198[13]). To foster this, it provides a list of such indicators, dividing them into two groups. The first contains indicators relating to the manner in which the work is performed by the worker, namely: i. whether the work is carried out according to the instructions and under the control of another party; ii. whether the work involves the worker’s integration in the organisation of the enterprise; iii. whether the work is performed solely or mainly for the benefit of another person; iv. whether the work must be carried out personally by the worker; v. whether the work is carried out within specific working hours or at a workplace specified or agreed by the party requesting the work; vi. whether the work is of a particular duration and has a certain continuity; vii. whether the work requires the worker’s availability, and; viii. whether the work involves the provision of tools, materials, and machinery by the party requesting the work. The second set of indicators is concerned with how the worker is remunerated, including: i. whether such remuneration is paid on a periodic basis; ii. whether it constitutes the worker’s sole or principal source of income; iii. whether the worker receives payment in kind, such as food, lodging, or transport; iv. whether he or she is provided with entitlements like weekly rest and annual holidays; v. whether the worker is paid by the party requesting the work for travel undertaken in order to carry out the work, and; vi. whether the worker bears any financial risk. These indicators are proposed to facilitate the assessment of subordination, dependency, or other conditions set out in national labour law. Nevertheless, they are not expected to be relevant to and have the same impact in all classification cases(ILO, 2020). None are supposed to be determinative or more important than others (ILO, 2020). They are neither compulsory nor exhaustive(ILO, 2020). Countries may adopt the list of indicators suggested by Recommendation 198, adapt it to local circumstances, or construct their own list in light of national law and practice. In any case, such a list should be reviewed periodically and, if needed, adapted to new and evolving working arrangements, including by way of introducing new indicators, or modifying or abolishing those that are no longer appropriate(ILO, 2020). National regulators are also free in how they assess indicators and give weight to each of them. In some jurisdictions, the employment relationship is deemed to exist where a certain number of indicators are met (ILO, 2020). Meanwhile, other countries have employed a‘multi-factor’ test in which each factor is assessed in connection with each other and all factors are considered in their entirety to determine employment status. 199 Despite variation between countries, it is usually not a single indicator, but a combination that confirms or rejects the existence of an employment relationship(ILO, 2020). It is worth briefly discussing some indicators set out in Recommendation 198, as they are regularly invoked in the context of platform work. One is the employer’s control over work performance and its power to supervise the employee. As illustrated above, the four platform companies investigated in this study rarely exercised direct, physical control over app-based drivers. More often, their control over workers was exerted through technological tools like algorithms, rating systems, and geo-localisation devices. For this reason, Vietnamese regulators have hesitated to confirm the subordinate relationship between app-based drivers and their platforms. Nonetheless, it should be noted that such indirect, technology-based control has been increasingly regarded as evidence of the control of the BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 68 employer over the employee by administrative and judicial authorities throughout the world. This has taken place-to different extents-in a good number of jurisdictions, including South Korea, Brazil, Uruguay, South Africa, Canada, the US, the UK, Belgium, France, Spain, and the Court of Justice of the European Union(De Stefano et al, 2021). Another indicator invoked in relation to platform work is the integration of the worker in the enterprise requesting work. Recognised in several jurisdictions, the‘integration’ test assesses whether the work carried out by the worker is an integral part of the organisation of the user enterprise(ILO, 2013). As changes in work organisation have rendered the control of the employer over the employee less apparent, the‘integration’ indicator has gained importance in national legislation and jurisprudence(Casale, 2011; ILO, 2013). More particularly, French courts have pointed to, among other things, the integration of app-based drivers to the services organised by platform operators to uphold the employee status of such workers(De Stefano et al, 2021). Proposals have also been made in other jurisdictions to increase the importance of this element in the employee test with a view to extending employee status to platform workers (De Stefano et al, 2021). In line with this, the CEACR has recently recommended that this indicator be carefully considered when assessing ‘new and emerging forms of work, such as platform work.’(ILO, 2020) As the services rendered by motorbike drivers play a vital part in Baemin, Be, Grab, and Now’s business, the introduction of the‘integration’ indicator to Vietnamese law would increase the chances of these workers claiming employment status. Although these companies have deliberately classified themselves as e-commerce platform operators and/or software application providers, the courts rejected such classification in Vinasun vs Grab, ruling that Grab is a transportation service operator. 200 Decree 10/2020/ND-CP, subsequently issued by the Government, also prescribed that a company that‘performs at least one of the main steps of transportation activities-namely, directly manages vehicles or drivers or determines transportation fares-in order to transport passengers or cargo on roads’ is a transportation service provider( Decree 10/2020/ND-CP art 3(2)). Although Decree 10/2020/ND-CP and the Vinasun vs Grab case only concern car services, there is no major legal impediment for the regulators to adopt a similar viewpoint about ride-hailing and delivery services rendered by motorbike platform drivers. Accordingly, if the ‘integration’ test is implemented, Baemin, Be, Grab, and Now motorbike drivers will likely be considered an integral part of the business of their companies. Other indicators that have triggered extensive debate in connection with platform work include the work being‘carried out within specific working hours’ and the request for the worker’s availability. Traditionally, these factors are considered indicative of employment whilst their absence is regarded as an indicator of independent contractor relations(ILO, 2013). Since platform workers are said to be free to schedule work and are not obliged to log on to the platforms at a fixed time, these indicators have been repetitively cited to argue for the independent contractor status of such workers. As discussed, this has discouraged Vietnamese regulators from recognising platform drivers as employees. Elsewhere, administrative and judicial authorities in Australia, Brazil, Italy, and the USA have also relied on the free choice of working hours to classify platform workers as independent contractors(De Stefano et al, 2021). Notwithstanding this, the argument about flexible work schedules has not prevented the courts in several other countries-including Belgium, France, Germany, the Netherlands, and Spain-from re-classifying platform workers, including drivers, as employees(De Stefano et al, 2021). Frequently, these courts reasoned that such flexibility does not, in itself, exclude an employment relationship if other facts, such as the worker’s subordination to the platform or his/her integration into the platform’s services, sufficiently indicate an employment relationship BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 69 (De Stefano et al, 2021). Equally important, it has been pointed out that the freedom of platform workers is largely theoretical since their work schedules are substantially controlled by platform operators through variable, incentivebased remuneration and various methods of rating and evaluation(De Stefano et al, 2021). As national regulators differ, there is an increasing awareness that workers’ schedules are no longer helpful in determining the employment status of new, highly casualised types of work, like platform work(De Stefano et al, 2021). The availability of a large pool of workers, the disaggregation of work into short-term tasks, and the use of technological tools and incentive systems have enabled platform operators to reduce control over the work schedules of their workers, especially in direct ways. Consequently, the excessive focus on and rigid application of indicators concerning working time would exclude one of the most vulnerable groups of workers from legal protection. More broadly, these indicators have also diminished its general significance in the context of the modern workplace owing to the rise of new working models and advanced technologies. The last indicator relates to who supplies work tools and equipment. Under the laws of several countries, a substantial capital investment by the worker is usually considered indicative of independent contractor status(ILO, 2013). Though not recognised in Vietnamese labour legislation, local judges have occasionally applied this indicator to decide employment status (Pham, 2010). While rarely mentioned explicitly, the ownership of tools and equipment has had effects on how Vietnamese regulators perceive the employment status of platform drivers. 201 On the one hand, platform drivers’ reliance on their own vehicles and other equipment to perform work has led some state officials to lean towards an independent contractor status. 202 On the other, many others believe that the use of platform uniforms suggests that app-based motorbike drivers are employees(Trần, 2021). Beyond Viet Nam, authorities in some jurisdictions-including Australia, Chile, and the USA-have relied on platform workers’ non-use of company names, logos, uniforms, and the like and/or their investment in capital equipment (such as vehicles, smartphones, and data plans) to reject their employment status(De Stefano et al, 2021). And yet, several other national regulators-including those in Australia, Canada, Spain, the UK, and Uruguay-take a different view. For them, the tools supplied by platform workers, such as mobile phones and motorcycles, are not a substantial investment in capital equipment, especially when such investment is compared with the technology investment made by platform companies(De Stefano et al, 2021). In the same vein, the CEACR rightly commented that‘the material provided by[platform] workers does not necessarily constitute the essence of the business.’(ILO, 2020). Consistent with the law of several countries, including most jurisdictions in Europe(ILO, 2013), which only allows a fairly limited role for this indicator, the Committee recalls that“this element alone could not be taken as the only determinant of an employment relationship.”(ILO, 2020). The Vietnamese Labour Code sets out three criteria(conditions) to differentiate employment from self-employment without providing any explanation. This lack of specificity has resulted in uncertainties, inconsistencies, and rigidities in determining the employment status of platform drivers. Introducing specific indicators, such as those proposed by Recommendation 198, would help to resolve these problems. Enhancing specificity and clarity, such indicators would concurrently foster consistency and predictability in the determination of employment status. Significantly, the deployment of a list of multiple suggestive indicators would reduce the rigidity of traditional concepts and parameters that have become increasingly inappropriate for new and emerging forms of work, such as ride-hailing/ platform work. Nonetheless, as the experience of other jurisdictions suggests, such indicators require careful consideration, flexible adaptation, and periodic review to ensure compatibility with labour market developments. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 70 To sum up, the Vietnamese authorities should go further than clarifying the current criteria to determine the existence of employment relationships and reform the legal framework for such determination more comprehensively. Such a bold reform is recommended because the problem with platform drivers stems not only from a lack of specificity of these criteria, but also from the increasing mismatch between the traditional concepts and approaches used to classify workers and rapid changes in work relationships. To address this mismatch, major reforms should be undertaken to:(i) Institutionalise a national policy for the periodic review of the scope of the employment relationship;(ii) recognise the principle of‘primacy of facts’;(iii) revise criteria to differentiate employment from self-employment, including introducing the‘economic dependency’ criterion and better defining the‘receiptof-wages’ and legal subordination criterion, and;(iv) adopting a list of specific suggestive indicators to facilitate the application of these criteria. These proposed reforms are, however, only some of the lessons that Viet Nam can learn and adapt from ILO Recommendation 198 and legal developments in other jurisdictions. Introducing a Rebuttable Presumption of Employment Status Although the revision of the general framework and criteria to identify employment relationships will increase the accessibility of platform workers, including app-based drivers, to labour protection and social security, a major hurdle may still exist for the realisation of their rights. That is, platform companies may keep classifying their workers as independent contractors and, thus, platform drivers will have to undergo costly, complex, and time-consuming legal processes to claim employment status(Miriam A. Cherry and Antonio Aloisi, 2017). For a rough illustration, it took almost three years for Vinasun to pursue its lawsuit against Grab. 203 Particularly, since Grab relies heavily on digital technology to operate its business, the traditional taxi company had to extensively use bailiff and expert services to establish evidence of how the platform was operating like a transportation service provider. 204 These services cost Vinasun billions of Viet Nam dong, only a small part of which was compensable. 205 Undoubtedly, platform workers like app-based motorbike drivers cannot afford such expenses, let alone others. In a more comparable example, two GrabCar drivers brought the platform to a district court in HCMC in 2019 after their apps were deactivated(Phan Thương, 2022). Both accepted amicable solutions, as their cases had taken a long time. In one case, the process lasted about three years with the first instance hearing remaining unresolved after many delays and adjournments. 206 On another (but related) note, following the issuance of Decree 10/2020/ND-CP, which sets out a broad definition of‘car transportation service provider’ to include ride-hailing platforms, Grab redrafted its contractual arrangements to continue disguising its transportation business and evading transportation business regulations(Minh Khánh, 2023; Trần Duy, 2022). Suffice to say that a mere revision of the legal framework and/or criteria to identify employment relationships will be inadequate to make meaningful changes to the vulnerability of app-based motorbike drivers. To address the misclassification of platform workers, several jurisdictions have introduced a rebuttable legal presumption in favour of employment status for platform workers. The Californian legislature pioneered this innovation. Its well-known‘AB5 law,’ published in 2019, extended the employee classification status to numerous platform workers by adopting the‘ABC test’ under which a worker is legally presumed to be an employee, unless the hiring entity could prove that the worker: ▪ is free from control or direction in the performance of the work; ▪ performs work which falls outside the usual course of the hiring entity’s business, and; BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 71 ▪ is customarily engaged in an independently established trade, occupation, or business of the same nature as that involved in the work performed(Legislative Counsel Digest, 2019). The impact of the AB5 law was subsequently limited due to the passage of‘Proposition 22’ in September 2020. While granting some protections to platform drivers, this Proposition explicitly designates these drivers as independent contractors and, thus, excludes a large proportion of the platform workforce from the ABC test(De Stefano et al, 2021). Though being challenged at home, the AB5 initiative appears to have inspired legislative proposals in other jurisdictions, such as the Netherlands 207 and Ontario(Canada)(Andrew Graham, 2021). Meanwhile, Proposition 22 has triggered a constitutional battle in Californian courts, the outcome of which remains to be seen (Kellen Browning, 2023). At present, European jurisdictions are leading reforms in this direction. In 2021, the Spanish parliament enacted the so-called‘Riders’ law,’ establishing a rebuttable legal presumption of an employment relationship for app-based delivery workers upon the satisfaction of three conditions: (i) The work involves the delivery of goods to final consumers;(ii) the platform directly, indirectly, or implicitly exercises the powers of organisation, management, and control through a digital platform or tool, and;(iii) the platform uses algorithms to manage the service or determine the working conditions(Willem Waeyaert, Karolien Lenaerts, and Dirk Gillis, 2022). A year later, Belgium and Portugal introduced similar legal presumptions. However, unlike their Spanish counterpart, these presumptions apply to all platform workers and require the fulfilment of different sets of conditions(Chris Van Olmen, 2022; Macedo Vitorino, 2023). In December 2021, the European Commission published a draft directive on improving working conditions in platform work(“ Platform Work Directive”). This proposed directive contains a rebuttable presumption under which a platform is legally deemed to be an employer if it meets at least two of the following criteria: ▪ The platform limits the level of the worker’s remuneration; ▪ The platform monitors the performance of work through electronic means; ▪ The platform limits the worker’s freedom to schedule working time, to accept or refuse tasks, or to use subcontractors or substitutes; ▪ The platform requires the worker to comply with mandatory rules on appearance, conduct towards end-users, or work performance, and; ▪ The platform limits the worker’s ability to work for other parties(European Commission, 2021). More recently, the European Parliament and its Employment Committee endorsed a stronger version for such a presumption which does not require any criteria to be satisfied(European Parliament, 2022; Foo Yun Chee, 2023). A legal presumption of employment relationship for platform workers is also among reform options being considered in other(non-European) jurisdictions, such as Argentina and Australia (Jaan Murphy, 2023; Francisca Pereyra and Lorena Poblete, 2022). The Vietnamese authorities could follow other jurisdictions to introduce a rebuttable legal presumption of employment status for platform workers. By shifting the burden of proof from platform workers to platform operators, it will enable the former to more easily access labour rights and social protections(Cherry and Aloisi, 2017; Miriam Kullman, 2022). At the same time, it will reduce incentives to disguise employment among the latter and, therefore, save platform workers from protracted, expensive, and uncertain legal battles in which they are usually financially and technically disadvantaged(Cherry and Aloisi, 2017; Miriam Kullman, 2022). A presumption of an employment relationship will, therefore, not only increase legal certainty regarding the status of platform workers, but also help them‘counterbalance the unequal bargaining power’ between the two parties(ILO, 2020). These benefits are particularly meaningful for platform workers in Viet Nam, as they lack an organisation that can represent and assist them BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 72 in misclassification disputes. Additionally, since the doctrine of stare decisis is not recognised in Viet Nam, the success of some platform workers in a misclassification case does not necessarily lead to the re-classification of similar workers. As evidence, the confirmation of the employee status of GrabCar drivers in Vinasun vs Grab has not resulted in the re-classification of app-based drivers by this platform. Moreover, the adoption of a legal presumption in favour of platform workers will facilitate law enforcement and dispute management. As mentioned earlier, the vague status of appbased drivers has discouraged labour authorities from intervening in labour conflicts in ridehailing and delivery platforms. Further to this, it has caused difficulties for these authorities in obtaining information about the labour force and‘managing labour relations’ in ride-hailing and delivery platforms(Trương et al, 2021). 208 The establishment of the presumption of an employment relationship between platform companies and their workers will create a legal basis for labour and other(e.g., social security) authorities to discharge their functions more effectively. The legal presumption of employment status has, however, faced two major criticisms. One is that the presumption would re-classify truly selfemployed workers and, thus, deprive them of the flexibility and autonomy that self-employed status provides(Reshaping Work, 2023; People; 2022). Though it contains some truth, this argument is largely exaggerated. Proponents of the argument, usually platform companies, reason that most people working for digital platforms are genuine independent contractors and prefer to be classified as such(Benoit Le Bret, 2023). However, mounting evidence on platform work does not support this claim, especially in sectors like ride-hailing and deliveries(Aude Cefaliello, 2023; De Stefano et al, 2021). As this study has demonstrated, despite being classified as independent contractors, Baemin, BE, Grab, and Now workers-particularly those relying heavily on platform work for their earningsoperate substantially like subordinate employees. Of note, our survey included a question asking worker-respondents if they wanted to have an employment contract with their platforms. Only 29.7 percent of the 192 workers answering the question said‘no.’ A majority(62.5 percent) indicated‘yes’ whilst the rest(7.8 percent) chose ‘unsure.’ Related to this point, it should also be emphasised that regulatory authorities around the world have, in recent years, increasingly and consistently re-classified platform workers as subordinate employees. Furthermore, the employment relationship presumption will not convert all platform workers into employees. Since the presumption is rebuttable, platform operators can initiate administrative or judicial processes to demand re-classification if they have sufficient evidence to support this. Additionally, the risk of misclassifying truly self-employed workers can be substantially alleviated in multiple ways. One is setting out conditions for the activation of the presumption, as exemplified by the draft Platform Work Directive and the laws of Belgium (Van Olmen, 2022), Portugal(Macedo Vitorino, 2023), and Spain(Waeyaert, Lenaerts and Gillis, 2022). These may include, among other things, a minimum threshold of working hours(say, e.g., the worker has worked for the platform for at least a month, and has worked, on average, for 24 hours a week). Such a threshold would help to exclude workers who usually depend less on the platform financially and, therefore, enjoy more autonomy and flexibility in actuality from a presumed employment relationship. Another possible solution is piloting the presumption first with sectors with substantial evidence of misclassification and vulnerable working conditions, such as ride-hailing, deliveries, and domestic work(De Stefano et al, 2021; Paula Rodríguez-Modro ñ o, Astrid AgenjoCalderón, and Purificación López-Igual, 2022), before extending it to other sectors. This gradual approach seems to be suitable for Viet Nam, where digital labour platforms remain concentrated mainly in a few sectors. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 73 Finally, the benefits of an employment relationship presumption considerably outweigh the problems it may cause. This is not merely because the likelihood of platform workers being misclassified as independent contractors by their platforms is greater than the likelihood of platform workers being misclassified as employees due to the operation of such a presumption. As discussed, misclassified platform workers usually refrain from taking legal action to be reclassified due to various practical impediments. The presumption would shift the burden to prove the nature of the working relationship from the platform worker to the platform entity who is in a much better position to bear this burden, considering its relative advantages in time, expertise, financial capacity, and access to evidence. Significantly, the onus should lie upon the platform- the party that mainly determines the terms and conditions of the contractual relationship. In short, the presumption would be an effective means to address the massive asymmetry between platforms and workers, without which the latter would effectively be deprived of the rights and protections to which they are entitled. The other argument that has been made against an employment relationship presumption is that granting employee status to platform workers would increase labour costs, damaging the development of new, innovative business models and pressing platform companies to reduce employment. We will discuss this argument below. Platform workers as thirdcategory workers? Introducing a legal presumption of employment status is not the only way to resolve the classification of platform workers. Another proposed option is constructing a‘third’, ‘intermediate’, or‘hybrid’ category between subordinate employees and independent contractors(De Stefano et al, 2021; Zhenxing Ke, 2022). This is exemplified by a bill submitted to the Columbian Congress in 2019(De Stefano et al, 2022). The bill proposes a new category for economically dependent platform workers and a number of protections for these workers, some of which are adapted to the unique nature of platform work. These protections include, inter alia, a guarantee of collective labour rights; social security protections, including insurance against work-related accidents and compulsory contribution to the health and pension system; and prohibition against compulsory allocation of work. This bill, however, seems to have been sidelined since the Columbian national election in 2022- as the new government has advanced a more radical bill that seeks to broaden existing labour protections to platform workers(Fairwork, 2022). Until recently, China appears to be the only jurisdiction to have created a third category specifically for platform workers. Following the explosion of strikes organised by app-based drivers in 2020, Chinese authorities enacted several legal instruments to regulate this newlyemergent form of work. The most important instrument is the Guiding Opinion on Protecting Workers in the New Form of Employment (“ Guiding Opinion”), issued in 2021(Ke, 2022). It introduces a new employment category for those‘who do not meet the employee status standard but are subject to some degree of control from the company,’ and gives these workers some protections, including: A written agreement requirement; prohibition of taking deposits from workers and restricting them from working for other platforms; anti-discrimination; minimum wage guarantee; and consultation of worker representatives on work rules and platform algorithms directly involving workers. Further, the Guiding Opinion suggests that local governments and platform companies work to improve the working conditions of platform workers, including in relation to remuneration, holiday leave, and occupational injury insurance. It also calls for labour unions and inspectorates to actively extend their organisations and activities into the emergent platform sector. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 74 France represents a somewhat different example in which platform workers are de facto treated as third-category workers. The European state introduced many laws and ordinances to provide platform workers with some labour rights and social protections. Of these instruments, the most notable is the so-called‘ El Khomri Law ,’ adopted by the French parliament in 2016(Dirk Gillis, Karolien Lenaerts, and Willem Waeyaert, 2022). It grants self-employed platform workers the right to form or join a trade union and engage in strike action. Further, the El Khomri Law requires that the platform assume responsibility regarding occupational accidents and vocational training for workers who have incomes exceeding a particular threshold. Another important piece of legislation is the‘ Mobility Orientation Law ,’ introduced in 2019 to regulate platform work in the transport sector(Dirk Gillis, Karolien Lenaerts, and Willem Waeyaert, 2022). It gives platform workers a right to refuse assignments and a right to disconnect. Notably, the Mobility Orientation Law introduces the possibility for platform operators to enact a charter which defines rights and obligations between themselves and their workers. This enabled the former to establish a legal presumption of non-employment through such charters, but such a presumption was subsequently declared unconstitutional by the French Constitutional Court(Dirk Gillis, Karolien Lenaerts, and Willem Waeyaert, 2022). By giving platform workers some-but not all-labour protections without resolving their employment status, France is considered to have‘create[d] a third status without actually naming it.’(Isabelle Daugareilh , 2019). The notion of a third legal category, neither employed nor self-employed, is not entirely new. Such categories have long existed in many jurisdictions, i.e., before the growth of platform work(Miriam A. Cherry and Antonio Aloisi, 2018). Examples include‘dependent self-employed workers’ in Canada,‘employeelike workers’ in Germany,‘quasi-subordinate workers’ in Italy, and‘workers’ in the UK. Among these jurisdictions, the UK has recently utilised the third category to classify platform drivers. In a landmark case, Uber BV vs Aslam, the UK Supreme Court ruled that Uber drivers are‘workers’- an intermediate category between‘employees’ and‘self-employed’ contractors, recognised under Section 230(3)(b) of the Employment Rights Act 1996( Uber BV v Aslam[2021] UKSC 5). This has enabled these drivers to access some protections, such as the national minimum wage, paid leave, and antidiscrimination and collective labour rights, but not the full range of labour rights afforded to employees( Employment Rights Act 1996(UK) s 230(3)(b)). The existence of a third employment category in some countries, and the ruling of the UK Supreme Court in the Uber case, have, in part, inspired proposals for a new employment category for platform workers(Ke, 2022). A number of reasons have been cited to support the classification of platform workers as a third category of employment. One is that these workers do not fit neatly into the traditional binary classification of employment, as they have similarities and differences with both employees and independent contractors(Cherry and Aloisi, 2017). Accordingly, it is contended that developing a hybrid category would resolve legal confusion and allow those failing to meet employment status criteria to enjoy certain protections(Cherry and Aloisi, 2017). Another major argument is that some employment rights, like minimum wage protection and occupational accident liability, do not suit the realities of the platform employment model(Cherry and Aloisi, 2017). Finally, classifying platform workers as third-category rather than subordinate workers will minimise increases in labour costs and, in doing so, foster newly-emerged, innovative platform businesses to further develop and create more jobs in the future(Cherry and Aloisi, 2017). Notwithstanding this, the proposed treatment of platform work as a third category of employment should be considered cautiously. First of all, the creation of a new, intermediate category of employment does not guarantee greater legal clarity. Rather, it may produce more legal uncertainty if clear criteria are not established to distinguish third-category workers from BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 75 employed and self-employed workers(Cherry and Aloisi, 2017; Kullman, 2022). Adding more complication, platform work arrangements vary considerably, including with regard to the degree of control afforded to workers. Confusion may emerge as to whether a platform worker has employee or third status. Take China, for example. Since the Guiding Opinion differentiates third-category workers and employees based on the degree of control(Ke, 2022), it is not easy to distinguish these two categories from each other, particularly considering that conventional employees may also enjoy some autonomy and flexibility. The same problem can also be seen in the UK, even though the third employment category is more established there. As pointed out by some commentators, the UK Supreme Court did not rule out the employee status of the drivers in the Uber case, but indicated that Uber exercised the functions of a typical employer (Ewan McGaughey, 2019). What’s more, the introduction of a hybrid category would prompt a complex policy question: Which protections should be afforded to and excluded from this category? If the third category grants modest rights and benefits to workers, it will likely incentivise employers to disguise genuine employment(Cherry and Aloisi, 2017; Kullman, 2022). This happened in Italy following the adoption of the‘quasisubordinate’ category. The category was widely misused by businesses to evade employment law, leading to subsequent legal interventions to minimise and discourage its use(Cherry and Aloisi, 2019). As comparative studies have revealed, the creation of a third category seems to work well only when it is used to enlarge the coverage of labour protection and social security beyond standard employment rather than to provide a cheaper alternative to it(Cherry and Aloisi, 2017; Cherry and Aloisi, 2019). In addition, the contention about the incompatibility of some labour protections with the platform employment model is exaggerated. Protections like minimum wage guarantee and occupational injury liability can be, and have been, extended to platform workers without major obstacles (Cherry and Aloisi, 2017; Rodrigo Simonet and Jorge Garcia, 2022). Furthermore, the distinct features of platform work can be addressed by adapting existing regulations without the need to develop a new legal category(as exemplified in the recent labour law reform in Chile)(Simonet and Garcia, 2022). Lastly, but importantly, the argument for lower standards for platform work, whether to protect employment or foster(business/technological) innovation, is not convincing. It creates unfair privileges for platform enterprises over other companies(Cherry and Aloisi, 2017), sacrifices the interests of platform workers in favour of business interests, and discriminates against platform workers compared to those working for traditional(non-platform) businesses. Further, if platform enterprises are granted such privileges on the grounds of technological or business innovation or employment protection, several other enterprises would also invoke the same or similar reasons to demand exemption from labour law. Also, the likely negative impact of employment status on the platform business model is-to a large extent-speculative and overstated. In fact, several platforms have operated despite their(re-)classification of workers as employees(Cherry and Aloisi, 2017). In short, treating platform workers as a third category of workers is likely to cause more problems than it solves, especially if it is to create a discounted employment status with fewer rights. Although the adoption of a legal presumption of an employment relationship is not a perfect solution, it is arguably a better and more effective solution to the prevalent misclassification of platform workers, including app-based motorbike drivers in Viet Nam. Enacting platform-specific protection A further solution to the vulnerability of app-based motorbike drivers is to introduce protections specifically applicable to these workers or to platform workers more broadly. This is required because the current legal framework for labour protection has failed to fully address the needs BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 76 of platform workers. One reason for this is that it has focused chiefly on employment relationships whilst self-employed platform work also demands some form of protection. In addition, the growth of new types of work has generated issues that are specific to platform work and, therefore, require special treatment. Of these issues, three need immediate attention and intervention beyond the traditional classification of workers. The first issue that has to be addressed is algorithmic management. This study has demonstrated that, while algorithmic management played a crucial role in the management, coordination, and supervision of platform drivers and had important effects on their remuneration, working time, flexibility, job security, and other working conditions, ridehailing and delivery platforms rarely explained how it works to their workers. In addition, the algorithmic management systems of digital platforms often failed to ensure an equitable allocation of work and provide workers with informed decision-making and sufficient opportunities to appeal and voice concerns about decisions affecting their interests. The findings of this research regarding algorithmic management in ride-hailing and delivery platforms in Viet Nam are unsurprising. They accord with other research on digital labour platforms elsewhere(Lenaerts et al, 2021). In several jurisdictions, regulatory innovations were introduced to improve the transparency and fairness of algorithmic management in digital platform work. In Spain, the so-called‘Riders’ Law’ compels all digital platforms to inform workers’ legal representatives of the inner logic of the algorithms that lead to(semi-)automated decisions‘influencing working conditions and work allocation.’(Waeyaert, Lenaerts, and Gillis, 2022). A new piece of legislation in Portugal also requires that job applicants are informed of the use of algorithms and artificial intelligence and that decision-making based on such artificial intelligence systems conforms to general rules on equality and anti-discrimination(Macedo Vitorino, 2023). In the same vein, China’s Guiding Opinion obliges platform companies to consult labour unions or worker representatives when formulating or revising work rules and platform algorithms directly involving workers (Ke, 2022). Meanwhile, Chile’s recently revised Labour Code grants platform workers the right to access the personal data collected by their platforms and the right to data portability while prohibiting discrimination by automated decision-making systems(Simonet and Garcia, 2022). Arguably, the EU’s proposed directive on improving working conditions in platform work offers the most radical innovation, demanding that platform operators implement measures to ensure algorithmic transparency, human oversight over working conditions, and workers’ right to appeal automated decisions. Together, these innovations can provide good examples for the construction of regulations on algorithmic management in Viet Nam. Another issue that demands specific regulations to protect platform workers is occupational safety and health(“ OSH”). As shown above, despite the exposure of platform drivers to high risks to their safety, health, and dignity, the investigated platform companies have done little to reduce and share such risks with their workers. These high risks do not only stem from the fact that delivery and passenger transport services usually involve significant OSH risks(Lenaerts, 2021). The emergent literature on platform work reveals that platform workers often face higher OSH risks than those performing similar tasks in the traditional labour market due to multiple factors, like inadequate professional and OSH training; inappropriate work tools and protective equipment; lack of support from colleagues and supervisors; irregular and unpredictable work schedules; and strong pressures to work faster or accept extra work- to name a few(Lenaerts, 2021). These factors, in turn, have their roots in the distinctive features of platform work, such as the on-demand nature of the rendered service, the disaggregation of work into time-bound tasks, algorithmic management and digital surveillance, gig- and incentive-based remuneration, dispersed workforce and workplace, and the supply of work tools and equipment by workers. This suggests the necessity of specific OSH regulations BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 77 for platform workers(or, at least, those working in dangerous occupations like delivery and ridehailing), particularly considering that Viet Nam’s Law on Labour Safety and Hygiene offers very limited protection to non-employee workers. Many jurisdictions have begun to enact OSH regulations that apply to platform workers, irrespective of their employment status. France is a pioneer on this front, as the El Khomri Law obliges platform companies to bear the costs of insurance against occupational accidents and diseases for workers with annual incomes at or above a prescribed threshold(Gillis, Lenaerts, and Waeyaert, 2022). The Taipei City Governance and Self-Regulatory Ordinance on Delivery Platform Business, issued by the Taipei City Government(Taiwan) in 2019, offers another illustration. Under this Ordinance, delivery platforms are mandated to undertake three OSH measures: Purchasing private accidental insurance for delivery workers; discontinuing services in dangerous conditions; and providing pre-duty OSH training for workers and reporting accidents to labour authorities(Bo-Shone Fu, 2021). An experimental occupational injury insurance was also initiated under the Guiding Opinion issued by the Chinese government in 2021(Ke, 2022). More recently, the 2022 labour law reform in Chile imposed an obligation on platform companies to provide OSH training and protective gear and to purchase insurance for workers’ personal items(Simonet and Garcia, 2022). These legal developments do not only illustrate the necessity of platform-specific OSH regulations. They can also serve as food for thought for legislators and policymakers in Viet Nam. The third issue requiring immediate attention is the exclusion of platform workers from collective labour rights. As discussed, since app-based motorbike drivers in Viet Nam are not officially recognised as employees, they are not allowed to form labour unions and WROs. Rather, they can only join‘syndicates’ which do not have the right to represent members in collective bargaining, bipartite dialogue, or dispute resolution processes. For the same reason, platform motorbike drivers do not have the right to strike. These legal loopholes have not only inhibited the capability of these workers to address precarious and vulnerable working conditions, but also caused dilemmas and difficulties for local regulators in managing escalating labour conflicts in the platform sector. Platform workers in other countries have also faced legal obstacles to the exercise of collective labour rights, especially in relation to the right to collective bargaining(De Stefano and Aloisi, 2020). Several platform workers are excluded from the protective framework of collective agreements because of their self-employed status or the existence of strict antitrust regulations (De Stefano and Aloisi, 2020). Complicating matters further, the mobilisation of workers in the platform sector encounters practical hurdles, including dispersed workplaces, regular turnover, fragmented workforces(owing to the diverse economic and social background of workers and their different degrees of commitment and dependency on platform work), and the frequent focus of trade unions on standard workers- to name a few(De Stefano and Aloisi, 2020). Despite this, recent years have seen the growing efforts of traditional unions to extend their reach to platform workers and the emergence of new-style workers’ associations in this emergent sector(De Stefano and Aloisi, 2020). These organisations have actively represented the interests of platform workers by negotiating for collective contracts, assisting workers in‘court battles,’ lobbying for legal reforms, and organising collective action and public campaigns(De Stefano and Aloisi, 2020). In addition, significant legal changes have taken place to support platform workers’ collective labour rights(De Stefano and Aloisi, 2020). Examples are diverse. Thanks to the adoption of the El Khomri Law, platform workers in Franceregardless of their employment status-now enjoy the right to form and join a trade union and engage in strike action(Gillis, Lenaerts, and Waeyaert, 2022). Chile has also reformed its Labour Code to recognise platform workers’ right to organise and bargain collectively with platform companies(regardless of their BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 78 employment status)(Simonet and Garcia, 2022). In a similar vein, regulations newly introduced in China suggest that trade unions extend their organisations and activities to the platform sector and negotiate with platform companies for collective agreements(Ke, 2022). In Japan, the Tokyo Labour Commission upheld the right of Uber Eats delivery workers to form a trade union and ordered the platform company to respond to their request for collective bargaining(The Asahi Shimbun, 2022). Beyond the national level, ILO Committees and supervisory bodies of human rights treaties, like the European Convention on Human Rights and the Inter-American Court of Human Rights, have consistently confirmed that the right to collective bargaining is an essential element of freedom of association, a universal right applicable to all workers, including selfemployed workers(De Stefano and Aloisi, 2020). Considering the urgent need of local workers, Vietnamese regulators should follow those in several other jurisdictions to extend collective labour rights to platform workers. Bold reforms should be taken to guarantee platform workers’ right to organise, right to collective bargaining, and right to strike. It should be noted that over half(52.7 percent) of the 203 workers participating in our survey stated that they want to join a trade union or WRO. Moreover, while divided on the question of the employee status of platform drivers, state and union officials with whom we talked indicated strong support for platform drivers’ collective labour rights. Summary and Recommendations In light of the problems identified, this section proposes four solutions to strengthening legal protection for app-based motorbike drivers in Viet Nam. The first and simplest action to be taken is clarifying the criteria to differentiate employment and self-employment currently set out in Article 3(1) of the 2019 Labour Code. It should be made clear that, to be classified as an employee, a worker is neither necessarily paid on a periodic basis nor prevented from receiving pay in the form of revenue or profit sharing. Guidance should also be published to elaborate the‘subordination’ criterion. Characterised by the employer’s directional, control, and disciplinary powers, the criterion should, nonetheless, be interpreted flexibly in light of the fast-transforming labour market. The second suggestion is thoroughly reforming the legal framework to identify employment relationships. As pointed out above, the problem with app-based motorbike drivers not only stems from uncertainties in the existing criteria used for such identification. It also has its roots in the increasing mismatch between traditional concepts and approaches used to classify workers and rapid changes in work relationships. A comprehensive review of the existing legal framework, rather than a mere clarification of Article 3(1) of the 2019 Labour Code, would help tackle this root cause. Viet Nam is not alone in facing this problem and it can draw upon considerable international experience. ILO Recommendation 198 can be a helpful starting point for Vietnamese legislators to reform the legal framework to determine the existence of employment relationships and, in doing so, strengthen protections for vulnerable workers in new forms of work. In light of this Recommendation, possible reforms should include introducing mechanisms to periodically review the scope of the employment relationship, recognising the principle of‘primacy of facts,’ incorporating‘economic dependency’ into the criteria to distinguish employment from selfemployment, and adopting a list of suggestive indicators to facilitate the application of these criteria. The third proposed solution is introducing a rebuttable legal presumption of an employment relationship for platform workers. Shifting the burden of proof to show that app-based drivers are independent contractors to the platforms would be a good option to implement. Although this may not work for all app-based drivers, the benefits outweigh the costs. Our recommendation is to pilot the presumption first in sectors with substantial evidence of misclassification and vulnerable working conditions, such as ride-hailing, deliveries, and BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 79 domestic work, before extending it to other sectors. This gradual approach seems to be suitable for Viet Nam, where digital labour platforms remain concentrated mainly in a few sectors. An often-mentioned option is to classify platform workers as a‘third,’‘intermediate,’ or‘hybrid’ category between subordinate employees and independent contractors, as has been proposed and/or occurred in the United Kingdom, France, and China. Despite this, we urge caution for its adoption in Viet Nam. The complications it could further create may outweigh the potential benefits. Particularly, if the third category offers a cheaper alternative to employment, it will motivate platform operators to continue disguising genuine employment. Thus, a presumption of employment status proves to be a better solution to the misclassification of platform work. Finally, the Vietnamese government should introduce protections specifically applicable to platform workers. As discussed earlier, these protections are needed because the traditional labour law framework has centred on employment relationships and largely disregarded the particular needs of platform workers, including app-based drivers. In this respect, Viet Nam can look at legislation and regulations proposed or already passed in jurisdictions such as Chile, China, the EU, Portugal, Spain, and Taiwan and introduce regulations that ensure fair and transparent algorithmic management, provide OSH protections for app-based drivers, and grant platform workers collective labour rights. These regulations should apply to all platform drivers or workers, whether or not they are classified as employees. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 80 CONCLUSION The frequent portrayal of app-based motorbike drivers as equal‘partners’ of digital platforms in Viet Nam is fundamentally untrue. In reality, these drivers have little control over their pay and conditions whilst having to work under the close control of platform operators. In stark contrast to the narrative about a desirable job with high earnings and flexible schedules, they often work in precarious conditions. As our survey revealed, the average income of app-based motorbike drivers in HCMC was 9,290,340 VND (380 USD) per month, after costs. Although this is considerably higher than local minimum and liveable wage rates, these drivers had to work, on average, 11.1 hours a day and 28.03 days a month to achieve it. If they had worked a standard 48-hour working week, they would have only earned an average of 6,210,672 VND(253 USD) per month, i.e., 16.6 percent below the living wage of 7,446,294 VND(304 USD)-though still above the minimum wage of 4,420,000 VND(180 USD) applicable to most areas in HCMC. However, this is a very rough comparison. It does not take into account the fact that app-based drivers are not entitled to several benefits received by ordinary employees. If, for example, we gross up the income of employees with social security contributions paid by their employers and deduct 23 days of public holidays and annual leave from the income of app-based drivers, the situation of the latter would be much less desirable. The average income of platform drivers who worked 48 hours each week would be just 5,752,834 VND(235 USD) per month, almost the same as the minimum income of a trained employee(5,746,221 VND per month) and 22.7 percent lower than the living wage, without several other employment entitlements. Even worse, their take-home earnings are highly vulnerable to policy changes unilaterally made by platform companies, customer ratings, and market conditions. Moreover, platform motorbike drivers usually work under extreme conditions that feature long working hours; insufficient rest; and high risks to their safety, health, and dignity. Frequently suffering from job insecurity and unfair management, they are, however, largely excluded from social protections. Even though platform companies have introduced their own accident and/or health insurance programmes, these are not sufficient to offset what app-based drivers would be entitled to under state-run social security schemes. Despite this and other problems, platform drivers have faced significant hurdles in collectively organising and fighting for their rights. This has caused them to resort to informal activism and wildcat protests which have, so far, produced little success. The vulnerability of app-based motorbike drivers is closely linked to the lack of a decisive answer on the employment status of these workers. This has not only excluded them from basic labour rights and social security in practice. It has also enabled platform operators to dominate bargaining processes and impose unfair contractual arrangements on their drivers. Furthermore, the ambiguous status of platform drivers has prevented and discouraged them from seeking to redress perceived injustices, whether by way of collective bargaining, bipartite dialogue, dispute resolution, or collective action. As analysed above, the unclear status of app-based drivers, in turn, has its roots in the absence of established, well-defined criteria to determine the existence of employment relationships. In addition, although work relationships have undergone enormous changes, regulators continue to rely heavily on rigid, formalistic concepts and approaches in such determination. The impact of uncertainties regarding the legal status of platform drivers and its direct consequence on their vulnerability and exclusion from coverage of labour law and social security underlines the acute need to provide legal protection for these workers. To tackle the problem, Viet Nam should enhance the clarity of legal criteria to distinguish between employment and self-employment, especially BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 81 the‘receipt-of-wages’ and‘subordination’ criteria. In the long run, it should also consider reforming the legal framework to determine employment status more thoroughly. Utilising Recommendation 198 of the ILO as a starting point, Vietnamese policymakers could also look to other jurisdictions for potential reforms. Major changes to be considered include introducing a national policy for a periodic review of the scope of the employment relationship; legalising the principle of‘primacy of facts;’ giving more weight to‘economic dependency’ in determining the existence of employment relationships; and adopting a list of suggestive indicators for such determination. Another possible reform is introducing a rebuttable legal presumption in favour of the employment status of platform workers. In shifting the burden of proof to platform companies, such a presumption would deincentivise digital platforms from disguising employment whilst increasing the actual access of app-based drivers to labour rights and social protections and easing the work of the authorities in charge of law enforcement and dispute resolution. Although this solution has been criticised, much of this criticism is exaggerated and ill-supported by evidence. While it does have some merits, the concern about misclassifying truly self-employed workers can be alleviated by setting conditions for the presumption to be activated, including, for example, a threshold of minimum working hours. Additionally, Viet Nam could adopt a gradual approach in which the presumption of an employment relationship is applied to some sectors, such as ride-hailing and deliveries, first before being extended to others. We note that platform drivers have been classified as a third category of worker in some jurisdictions. However, this option would create more problems than it would solve, and we are unconvinced of its merit for Viet Nam as it already grapples with addressing ambiguities in existing laws. In short, although the adoption of a legal presumption of an employment relationship is not perfect, it is a better solution to the misclassification of platform workers, including app-based motorbike drivers, in Viet Nam. Apart from addressing the legal status of appbased motorbike drivers, Viet Nam should also consider enacting protections specifically applicable to such drivers or, more broadly, platform workers. This could be a temporary solution, pending the resolution of the debate over the employment status of platform workers. Nonetheless, the more important reason for the introduction of platform-specific legislation is that the emergence of platform work has generated issues that the traditional labour law framework has failed to sufficiently address. As we have suggested above, prompt action should be made to ensure the fairness and transparency of algorithmic management, afford minimum OSH protections, and guarantee collective labour rights for platform drivers. These measures should be taken regardless of their legal status. Vietnamese policymakers are increasingly aware of the situation of platform drivers and have begun discussions on how to regulate platform work. The country is not alone in tackling this issue: Regulatory innovations have been proposed and undertaken at the international level and several other jurisdictions have started to address the mismatch between traditional labour law and newly-emergent forms of work. In this spirit, our recommendations draw not only from regulatory experiences but, eventually, should give food for thought to make platform work fairer and more just. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 82 Endnotes 1 The exchange rate used in this report is 1 USD= 24,500 VND. 2 Literally meaning“technology-based motorbike taxi drivers.” The word xe ôm originally refers to motorbike taxi services only. Yet, the term xe ôm công nghệ is widely used to include both ride-hailing and delivery services. 3 Many other terms have been used to describe the same phenomenon, including‘creative economy,’‘collaborative economy,’‘sharing economy,’ and‘gig economy.’ The term‘platform economy’ is used in this report due to its emphasis on the central role of digital platforms in these new business models(ILO, 2022). 4 We originally received responses from 211 workers, including eight women. The Covid-19 outbreak in HCMC in 2021 prevented us from approaching more female drivers in later stages. To avoid data errors, we excluded the eight female respondents from survey analysis. Our interview data, nonetheless, did include in-depth interviews with female workers. 5 This includes the days-off that drivers normally take every month, but excludes occasional leave, such as leave for sickness, holidays, and broken vehicles. 6 Decree 90/2019/ND-CP of the Government dated 15 November 2019 on Regional Minimum Wage Rates for Employees. The only exception is the remote suburban district Cần Giờ. 7 Decision 595/QD-BHXH of Viet Nam Social Security dated 14 April 2017 on Procedures to Collect Social Insurance, Health Insurance, Unemployment Insurance, Work-related and Occupational Disease Insurance Contributions. 8 Labour Code 2019 arts 112, 113, 115; Law on Labour Safety and Hygiene 2015 art 38. 9 Varying from one platform to another and between different services, additional fees may include night fees; bad weather fees; door-to-door delivery fees; service fees; and parking fees-to name a few. 10 As published on platform websites, the deduction rates of Baemin, Be, Grab, and Now in 2021 were 20%, 25%, 27%, and 10%, respectively. 11 This percentage has been reduced to 25.926% since 2023, thanks to the Government’s adjustment of the VAT rate. 12 Interview with driver D6(HCMC, 5 June 2019). Also Interviews with drivers D2(HCMC, 15 May 2019), D7-D8(HCMC, 6 June 2019), and D26(HCMC, 5 June 2019). 13 Interviews with D9(HCMC, 6 June 2019), D10-D11(HCMC, 5 June 2019), and D35(HCMC, 19 January 2021). 14 Interviews with D10-D12(HCMC, 5 June 2019), D18(HCMC, 6 June 2019), and M1(HCMC, 11 September 2020). M1 is a former manager of two surveyed platforms. 15 Interview with D26(HCMC, 5 June 2019); surveyors’ fieldnote(HCMC, February-March 2021). 16 Interview with D11(HCMC, 5 June 2019). 17 Interviews with D10-D12(HCMC, 5 June 2019), D27(HCMC, 6 June 2019), D37-D39(HCMC, 9 June 2020), and D47 (HCMC, 3 September 2020). See also Võ and Lê(2018); Duy(2020). 18 Interview with D11(HCMC, 5 June 2019). 19 Interview with D11(HCMC, 5 June 2019). 20 Labour Code 2019, art 98. 21 Interview with D44(HCMC, 9 April 2021). 22 Interviews with D6 and D28(HCMC, 5 May 2019), D6(HCMC, 6 May 2019), D44(HCMC, 9 April 2021), D45(HCMC, 18 March 2021), and D47(HCMC, 3 September 2020). 23 Interviews with D6(HCMC, 5 May 2019), D47(HCMC, 3 September 2020) and M1(11 September 2020). 24 Interviews with D7, D8, and D27(HCMC, 6 June 2019); D12(HCMC, 5 June 2019); D29(HCMC, 20 July 2020); D34 and D35(HCMC, 19 January 2021); D39(HCMC, 9 June 2020); D43(HCMC, 31 March 2021); and M1(11 September 2020). BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 83 25 Interview with D43(HCMC, 31 March 2021). 26 Interviews with D43(HCMC, 31 March 2021), D45(HCMC, 18 March 2021), and D46(HCMC, 21 March 2021). 27 Interview with D43(HCMC, 31 March 2021). 28 Interview with D6(HCMC, 5 June 2019). See also Interview with D45(HCMC, 18 March 2021) for a similar example. 29 Interview with D35(HCMC, 19 January 2021). 30 See, e.g., ZingNews(2024) reporting ten cases in which GrabBike drivers died due to traffic accidents(eight cases) and robberies(two cases). 31 Interview with D46(HCMC, 20 March 2021). 32 Interview with D43(HCMC, 31 March 2021). 33 Interview with D45(HCMC, 18 March 2021). 34 Interviews with D43(HCMC, 31 March 2021) and M1(HCMC, 11 September 2020). 35 Interviews with D16, D19, and D25(HCMC, 5 June 2019); D27(HCMC, 6 June 2019); D35(HCMC, 19 January 2021); D43(HCMC, 31 March 2021); and M1(HCMC, 11 September 2020). 36 Interview with D8(HCMC, 6 June 2019). 37 Interviews with D11 and D24(HCMC, 5 June 2019), D43(HCMC, 31 March 2021), D44(HCMC, 9 April 2021), and D47 (HCMC, 3 September 2020). 38 See Baemin COC ss 12-14; Be COC s 37; Grab COC s 15.1; Now COC ss 4.1-4.2. 39 Ibid. See also Interviews with D29(HCMC, 20 July 2020), D42(HCMC, 13 June 2020), and D46(HCMC, 20 March 2021) indicating that the execution of such sanctions is immediate and automated. 40 Be COC s 37. 41 Interviews with D7(HCMC, 6 June 2019), D25(HCMC, 5 June 2019), D44(HCMC, 9 April 2021), and D45(HCMC, 18 March 2021). 42 Interview with D7(HCMC, 6 June 2019). 43 Interview with D45(HCMC, 18 March 2021). 44 Interview with D13(HCMC, 6 June 2019). 45 Interviews with D6(HCMC, 5 June 2019), D17(HCMC, 6 June 2019), and D44(HCMC, 9 April 2021). See also Baemin Bonus Policy; Be Bonus Policy; Grab Bonus Policy; and Now Bonus Policy. 46 Interviews with D44(HCMC, 9 April 2021) and M1(HCMC, 11 September 2020). See also Baemin Bonus Policy; Be Bonus Policy; Grab Bonus Policy; and Now Bonus Policy. 47 Interviews with D6(HCMC, 5 June 2019), D17 and D18(HCMC, 6 June 2019), D25(HCMC, 5 June 2019), and D49 (HCMC, 28 March 2021). See also Baemin, Hợp đồng Hợp tác Kinh doanh[Business Cooperation Contract](“Baemin Driver Agreement”) art 5(1)(c) allowing the platform to supervise the driver through the mobile app. The contract was provided by a driver-interviewee. 48 Interviews with D6(HCMC, 5 June 2019) and D49(HCMC, 28 March 2021). 49 See Baemin COC ss 12-14; Be COC s 37; Grab COC s 15.1; Now COC ss 4.1-4.2. 50 See Baemin Bonus Policy; Be Bonus Policy; Grab Bonus Policy; and Now Bonus Policy. 51 Interviews with D8(HCMC, 6 June 2019), D27(HCMC, 6 June 2019), D32(HCMC, 18 January 2021), and D44(HCMC, 9 April 2021). 52 Interview with D32(HCMC, 18 January 2021). 53 Interview with D35(HCMC, 19 January 2021). 54 Interviews with D4(HCMC, 15 May 2019), D12(HCMC, 5 June 2019), D31(HCMC, 2 February 2021), D36(HCMC, 18 January 2021), and D44(HCMC, 9 April 2021). BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 84 55 Interviews with D9, D13, and D17(HCMC, 6 June 2019); D11, D12, D14, and D28(HCMC, 5 June 2019); D29(HCMC, 20 July 2020); and D49(HCMC, 28 March 2021). 56 Interviews with D7, D8, and D27(HCMC, 6 June 2019); D12(HCMC, 5 June 2019); D29(HCMC, 20 July 2020); D34 and 35(HCMC, 19 January 2021); D39(HCMC, 9 June 2020); and M1(HCMC, 11 September 2020). 57 Interviews with D2(HCMC, 15 May 2019), D4(HCMC, 15 May 2019), and D39(HCMC, 9 June 2020). 58 Interview with D39(HCMC, 9 June 2020). 59 Interviews with D2 and D4(HCMC, 15 May 2019). 60 Interviews with D6(HCMC, 5 May 2019) and D47(HCMC, 3 September 2020). 61 Interview with M1(11 September 2020). 62 Interviews with D4(HCMC, 15 May 2019); D6, D14, D22, and D28(HCMC, 5 June 2019); and D9 and D17(HCMC, 6 June 2019). See also Baemin Driver Agreement art 4(2)(k). 63 See Baemin COC; Be COC; Grab COC; Now COC. 64 Interview with D26(HCMC, 5 June 2019). 65 Interviews with D26(HCMC, 5 June 2019), D32(HCMC, 18 January 2021), and D44(HCMC, 9 April 2021). 66 Interview with D32(HCMC, 18 January 2021). 67 Ibid. 68 Ibid. 69 Interviews with D29(HCMC, 20 July 2020), D42(HCMC, 13 June 2020), and D46(HCMC, 20 March 2021). 70 Interviews with D6(HCMC, 5 June 2019) and D44(HCMC, 9 April 2021). 71 Interviews with D5(HCMC, 2 June 2019), D6(HCMC, 5 June 2019), D7(HCMC, 6 June 2019), and D35(HCMC, 19 January 2021). 72 Interview with M1(11 September 2020). 73 Interview with D5(HCMC, 2 June 2019). 74 Interview with D7(HCMC, 6 June 2019). 75 Interviews with D29(HCMC, 20 July 2020), D42(HCMC, 13 June 2020), and D46(HCMC, 20 March 2021). 76 Interviews with D6 and D10(HCMC, 5 June 2019), D8(HCMC, 6 June 2019), D32(HCMC, 18 January 2021), D33 (HCMC, 26 January 2021), D35(HCMC, 19 January 2021), and D44(HCMC, 9 April 2021). 77 Interview with M1(HCMC, 11 September 2020). 78 Interviews with D9(HCMC, 6 June 2019), D10 and D11(HCMC, 5 June 2019), and D35(HCMC, 19 January 2021). 79 Interviews with D5(HCMC, 2 June 2019); D8 and D27(HCMC, 6 June 2019); D10, D12, D19, D23, and D26(HCMC, 5 June 2019); D32(HCMC, 18 January 2021); D35(HCMC, 19 January 2021); and D44(HCMC, 9 April 2021). 80 Interview with D44(HCMC, 9 April 2021). 81 Baemin, Be, Gofast, and Grab use different names for these contracts, including, respectively:‘Business Cooperation Contract’(Hợp đồng Hợp tác Kinh doanh,“Baemin Driver Agreement”);‘Cooperation Contract’(Hợp đồng Hợp tác, “Be Driver Agreement”);‘Independent Contractor Agreement’(Hợp đồng Nhà thầu Độc lập,“Now Driver Agreement”); and‘Cooperation Agreement’(Thoả thuận Hợp tác,“Grab Driver Agreement”). 82 Baemin Driver Agreement art 3(1); Be Driver Agreement arts 6(1), 7(4); Now Driver Agreement s 3. 83 Grab Driver Agreement art 1.2. 84 Now Driver Agreement s 4. 85 See Baemin Driver Agreement art 8(1); Be Driver Agreement art 6(2)-(3); Now Driver Agreement s 5; Grab COC. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 85 86 Baemin Driver Agreement; Be Driver Agreement; Grab Driver Agreement; Now Driver Agreement. 87 Interviews with D6 and D26(HCMC, 5 June 2019), D17(HCMC, 6 June 2019), D32(HCMC, 18 January 2021), D33 (HCMC, 26 January 2021), and D44(HCMC, 9 April 2021). 88 Interview with D44(HCMC, 9 April 2021). 89 Interviews with D5(HCMC, 2 June 2019); D10, D12, and D19(HCMC, 5 June 2019); D32(HCMC, 18 January 2021); D34(HCMC, 19 January 2021); D43(HCMC, 31 March 2021); and D44(HCMC, 9 April 2021). 90 Interview with D34(HCMC, 19 January 2021). 91 Pay rate depends on whether the accident falls within category(i),(ii), or(iii). 92 Pay rate depends on whether the accident falls within category(i),(ii), or(iii). 93 The eligibility of drivers and their maximum eligible days and pay rates depend on whether their accidents fall within category(i),(ii), or(iii). 94 Pay rate depends on whether the accident falls within category(i),(ii), or(iii). 95 Pay rate depends on whether the accident falls within category(i),(ii), or(iii). 96 The eligibility of drivers and their maximum eligible days and pay rates depend on whether their accidents fall within category(i),(ii), or(iii). 97 Interviews with union cadres U2(Zoom, 4 February 2021), U3(HCMC, 6 April 2021), U4(HCMC, 16 February 2021), and U5(HCMC, 30 March 2021). 98 Interviews with union cadres U2(Zoom, 4 February 2021), U3(HCMC, 6 April 2021), U4(HCMC, 16 February 2021), and U5(HCMC, 30 March 2021). 99 Interviews with U1(Hanoi, 20 May 2019), U2(Zoom, 4 February 2021), U3(HCMC, 6 April 2021), and U4(HCMC, 16 February 2021). 100 Interviews with O4(HCMC, 20 March 2021), U4(HCMC, 16 February 2021), and U5(HCMC, 30 March 2021). 101 Interviews with U4(HCMC, 16 February 2021) and U5(HCMC, 30 March 2021). 102 Ibid. 103 Ibid. 104 Interviews with D29-D49(HCMC, 14 June 2020-9 April 2021). 105 Interviews with U4(HCMC, 16 February 2021), U5(HCMC, 30 March 2021) and M2(Hanoi, 31 March 2021). See also ‘TAGs nghiệp đoàn xe ôm’[TAGs Motorbike Driver Syndicate], Người Lao động[The Labourers]. 106 Interviews with U4(HCMC, 16 February 2021), U5(HCMC, 30 March 2021), and M2(Hanoi, 31 March 2021). 107 Interviews with D6 and D19(HCMC, 5 June 2019), D7-D9(HCMC, 6 June 2019), D35(HCMC, 19 January 2021), D42 (HCMC, 13 June 2020), M1(HCMC, 11 September 2020), M2(Hanoi, 31 March 2021), and M3(HCMC, 7 July 2021). See also Baemin(2022b); Be(2022e); Grab(2022b); ShopeeFood(2022e). 108 Interviews with D6, D10, D16, and D19(HCMC, 5 June 2019); D7 and D8(HCMC, 6 June 2019); D35(HCMC, 19 January 2021); and D42(HCMC, 21 June 2020). 109 Interview with D6(HCMC, 5 June 2019). 110 Interviews with D8, D9, and D27(HCMC, 6 June 2019); D10, D11, D14, and D19(HCMC, 5 June 2019); and M2(Hanoi, 31 March 2021). 111 Interview with D19(HCMC, 5 June 2019). 112 Interview with D27(HCMC, 6 June 2019). 113 Interviews with D8, D19, and D27(HCMC, 6 June 2019); D12, D16, and D22(HCMC, 5 June 2019); D33(HCMC, 26 January 2021); D35(HCMC, 19 January 2021); D38(HCMC, 9 June 2020); and D47(HCMC, 3 September 2020). BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 86 114 Interview with D35(HCMC, 19 January 2021). 115 Interviews with D1(HCMC, 12 May 2019); D4(HCMC, 15 May 2019); D6; D10-D12 and D22(HCMC, 5 June 2019); D8, D9, D13, and D18(HCMC, 6 June 2019); D30(HCMC, 14 June 2020); D31(HCMC, 2 February 2021); D33(HCMC, 26 January 2021); D34 and D35(HCMC, 19 January 2021); D36(HCMC, 18 January 2021); D38(HCMC, 9 June 2020); D44(HCMC, 9 April 2021); and D45(HCMC, 18 March 2021). 116 Interviews with D6, D10-D12, D16 and D22(HCMC, 5 June 2019); D9, D13, and D18(HCMC, 6 June 2019); D30 (HCMC, 14 June 2020); D31(HCMC, 2 February 2021); D33(HCMC, 26 January 2021); D34 and D35(HCMC, 19 January 2021); and D45(HCMC, 18 March 2021). 117 Interviews with D12 and D25(HCMC, 5 June 2019), D30(HCMC, 14 June 2020), D31(HCMC, 2 February 2021), D33(HCMC, 26 January 2021), D34 and D35(HCMC, 19 January 2021), and D44(HCMC, 9 April 2021). See also, e.g., Hội Grabbike Sài Gòn[GrabBike Saigon Group][accessed 22 December 2022]; Hội AE Grabbike Sài Gòn[Group of Saigon GrabBike Brothers][accessed 22 December 2022]; Cộng đồng Grab Việt Nam(4 bánh và 2 bánh)[Grab Viet Nam Community(Car and Motorbike)][accessed 22 December 2022]. 118 Interviews with D2(HCMC, 15 May 2019), D31(HCMC, 2 February 2021), D32(HCMC, 18 January 2021), and D44 (HCMC, 9 April 2021). 119 Interviews with D2(HCMC, 15 May 2019); D8, D13, D18, and D27(HCMC, 6 June 2019); D12, D24, and D25(HCMC, 5 June 2019); D31(HCMC, 2 February 2021); D32(HCMC, 18 January 2021); D33(HCMC, 26 January 2021); D34 (HCMC, 19 January 2021); D42(HCMC, 13 June 2020); and D44(HCMC, 9 April 2021). See also Buckley, above no. 115. 120 Interview with D34 and D35(HCMC, 19 January 2021); fieldnotes(HCMC, 5-20 December 2020). 121 Interviews with D8(HCMC, 6 June 2019), D31(HCMC, 2 February 2021), D33(HCMC, 26 January 2021), and D44 (HCMC, 9 April 2021). 122 Interviews with D36(HCMC, 18 January 2021) and D44(HCMC, 9 April 2021). 123 Interviews with D11, D12, D22, and D26(HCMC, 5 June 2019); D18(HCMC, 6 June 2019); D30(HCMC, 14 June 2020); and M2(Hanoi, 31 March 2021). 124 Interviews with D11 and D22(HCMC, 5 June 2019), D36(HCMC, 18 January 2021), D44(HCMC, 9 April 2021), and M2(Hanoi, 31 March 2021). 125 Interviews with D29(HCMC, 20 July 2020), D31(HCMC, 2 February 2021), D32(HCMC, 18 January 2021), D34(HCMC, 19 January 2021), D44(HCMC, 9 April 2021), and M1. See also Buckley, above no. 115. 126 Interviews with D2(HCMC, 15 May 2019); D12, D24, and D25(HCMC, 5 June 2019); and D13(HCMC, 6 June 2019). 127 Interviews with D29(HCMC, 20 July 2020), D31(HCMC, 2 February 2021), D32(HCMC, 18 January 2021), D34 (HCMC, 19 January 2021), and D44(HCMC, 9 April 2021). See also Do Hai Ha, Toan Le, Dinh Thi Chien, and Luong Minh Son,‘Platformisation as Informalisation: Decomposition, Recomposition and the Struggle of Platform Drivers for Labour Rights in Viet Nam’(Paper presented at Asian Law and Society Association Annual Meeting 2022, Hanoi, 9-10 December 2022) for a detailed account of these protests. 128 Interviews with O1(HCMC, 23 March 2021), O3(HCMC, 5 March 2021), O4(HCMC, 20 March 2021), U1(Zoom, 4 February 2021), U2(6 April 2021), and U3(HCMC, 16 February 2021). 129 Ibid. 130 Interviews with officials O1(HCMC, 23 March 2021) and U3(HCMC, 6 April 2021). 131 Interviews with O1(HCMC, 23 March 2021), O3(HCMC, 5 March 2021), and O4(HCMC, 20 March 2021). 132 Interviews with U2(Zoom, 2 April 2021), U3(HCMC, 6 April 2021), and U4(HCMC, 5 June 2019). 133 Interview with lawyer L1(Hanoi, 20 May 2019) and U2(Zoom, 2 April 2021). See also, e.g., Đoàn and Hồ(2019); Đoàn (2019). 134 Interviews with O1(HCMC, 23 March 2021), O3(HCMC, 5 March 2021), O4(HCMC, 20 March 2021), U2(Zoom, 2 April 2021), and U3(HCMC, 6 April 2021). See also, e.g., Nguyễn and Uyên(2021); Trần(2021); Văn(2020). BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 87 135 Interviews with D44(HCMC, 9 April 2021) and U2(Zoom, 2 April 2021). See also Ngô and Minh(2020); Trần(2020). 136 Interview with O1(HCMC, 23 March 2021). See also MOJ, above no. 281, 2-3 noting that the report was drafted in consultation with MOLISA. 137 See HCMC People’s Court, Judgment 1910/2018/KDTM-ST(28 December 2018); Superior People’s Court in HCMC, Judgment 05/2020/KDTM-PT(10 March 2020). 138 Ibid. 139 Ibid. 140 See Superior People’s Court in HCMC, above no. 283, 15. 141 Ibid 16. 142 See the Law on Road Traffic 2008 ch VI; Decree 10/2020/ND-CP of the Government dated 17 January 2020. 143 Stare decisis is a legal principle that requires courts to follow rules or principles laid down in earlier judicial decisions when resolving a similar case(Solanki, 2023). 144 Interviews with U2(Zoom, 2 April 2021) and U3(HCMC, 6 April 2021) . See also Bảo and Trần(2020). 145 Ibid. 146 Ibid. 147 Interviews with U2(Zoom, 2 April 2021) and U3(HCMC, 6 April 2021). 148 Interviews with U1(Hanoi, 20 May 2019), U2(Zoom, 2 April 2021), U3(HCMC, 6 April 2021), and U4(HCMC, 5 June 2019). 149 See especially interviews with O1(Hanoi, 20 May 2019; HCMC, 23 March 2021); O2(HCMC, 3 May 2019, 11 Feb 2021); and U4(HCMC, 5 June 2019, 16 Feb 2021). 150 See Baemin Driver Agreement art 1(3); Now Driver Agreement s 18. 151 Interviews with O1(HCMC, 23 March 2021), O3(HCMC, 5 March 2021), L1(Hanoi, 20 May 2019), and U2(Zoom, 4 February 2021). 152 Interviews with O1(HCMC, 23 March 2021), O3(HCMC, 5 March 2021), L1(Hanoi, 20 May 2019), and U2(Zoom, 4 February 2021). 153 For further information, see the third section of Part Three. 154 See Part Two above. 155 Baemin Driver Agreement art 2(2); Be Driver Agreement art 3(2); Grab Driver Agreement art 2(1); Now Driver Agreement appx 1 s 2(2); interview with M1(HCMC, 11 September 2020). 156 Interviews with O1(HCMC, 23 March 2021) and U3(HCMC, 6 April 2021). 157 See Labour Code 2019 arts 3(1), 13(1). 158 See Labour Code 2012 art 3(1). 159 Interviews with O1(HCMC, 23 March 2021), O3(HCMC, 5 March 2021), U3(HCMC, 6 April 2021). See also MOJ, above no. 281, 16. 160 Grab COC s 15(2) 161 Baemin COC ss 12, 13; Be COC s 37; Now COC arts 4(1), 4(2). 162 Interviews with D22 and D23(HCMC, 5 Jun 2019). 163 Interviews with D43(HCMC, 31 March 2021) and D44(HCMC, 19 April 2021). 164 Interview with D44(HCMC, 19 April 2021). 165 Labour Code 2019 arts 148(2), 166, 167. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 88 166 Interviews with O1(HCMC, 23 March 2021), O2(HCMC, 11 February 2021), U2(Zoom, 4 February 2021), and U3 (HCMC, 6 April 2021). See also, e.g., Superior People’s Court in HCMC, Judgment 05/2020/KDTM-PT(10 March 2020) 15-16; Trần(2020); Văn(2020). 167 Interviews with O1(HCMC, 23 March 2021) and U3(HCMC, 6 April 2021). 168 Interviews with O3(HCMC, 5 March 2021) and L1(Hanoi, 20 May 2019). 169 Interviews with D2(HCMC, 15 May 2019); D6, D21, D22, and D24(HCMC, 5 June 2019); D17(HCMC, 6 June 2019); D29(HCMC, 20 July 2020); M1(HCMC, 11 September 2020); and M4(7 July 2021). 170 Baemin Driver Agreement art 4(2); Be Driver Agreement arts 3(3), 3(6), 3(9); Now Driver Agreement appx 01[4(2)]. 171 Baemin Driver Agreement arts 1(3), 9; Be Driver Agreement arts 4(7), 5(7), 5(8), 5(15); Now Driver Agreement ss 18, 21 and appx 01[4(2)(g)]. 172 Baemin Driver Agreement art 2; Be Driver Agreement art 2; Now Driver Agreement appx 01[2]; Grab Driver Agreement art 2. 173 Ibid. 174 Baemin Driver Agreement arts 4(2), 5(1); Be Driver Agreement art 5; Now Driver Agreement arts(2),(5(1); Grab Driver Agreement arts 3-4. 175 Interview with D44(HCMC, 9 April 2021). 176 VGCL Charter art 16. 177 Interviews with D3(HCMC, 17 May 2019), D6(HCMC, 5 June 2019), and D8(HCMC, 6 June 2019). 178 Interview with D6(HCMC, 5 June 2019). 179 Interview with D33(HCMC, 26 Jan 2021). 180 Interview with D17(HCMC, 6 June 2019). 181 Interview with U2(Zoom, 2 April 2021), U4(HCMC, 5 June 2019), and O4(HCMC, 20 March 2021). 182 Interviews with O2(HCMC, 11 February 2021), O3(HCMC, 5 March 2021), O4(HCMC, 20 March 2021), and U4 (HCMC, 16 February 2021). 183 The Labour Code is the sole piece of legislation that recognises the right to strike and only applies to employees. 184 Interviews with O1(HCMC, 23 March 2021), O2(HCMC, 11 February 2021), O3(HCMC, 5 March 2021), O4(HCMC, 20 March 2021), U1(HCMC, 6 April 2021), U2(Zoom, 4 February 2021), and U4(HCMC, 16 February 2021). 185 T Treu,‘Labour Law and Social Change,’ public lecture(Geneva, International Labour Office, November 2002) cited in Casale(2011) 186 Baemin Driver Agreement arts 1(1), 2, 5(1)(b); Be Driver Agreement arts 4(2), 5(2,5); Grab Driver Agreement art 3(6); Now Driver Agreement appx 1 s 1(1);... 187 Baemin Driver Agreement art 4(2); Be Driver Agreement art 5(2); Grab Driver Agreement art 3; Now Driver Agreement appx 1 s 4(2). 188 Interviews with D9(HCMC, 6 June 2019) and D43(HCMC, 31 March 2021). 189 Baemin, above no. 371; Be, above no. 371; Grab,‘Đào tạo Đối tác Mới(GrabBike)’[Training New Partners(GrabBike)], Grab[accessed 20 March 2023]; ShopeeFood, above no. 371. See also Baemin Driver Agreement art 4(2)(d); Be COC s 5; Now Driver Agreement appx 1 s 4(2)(e) on the driver’ obligation to undertake training provided by the platform. 190 See Part Two above, especially Algorithmic Control and Unfair Management. 191 Baemin COC; Be COC; Grab COC; Now COC. 192 Baemin COC; Be COC; Grab COC; Now COC. BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 89 193 See Resolution 04/2019/HDTP of the Judges’ Council of the Supreme People’s Court dated 18 June 2019 on Selection, Announcement, and Application of Precedents. 194 See Precedent 20/2018/AL; Precedent 70/2023/AL. 195 HCMC People’s Court, Judgment 1910/2018/KDTM-ST(28 December 2018). 196 Ibid. 197 Ibid. 198 See Decision No. 1413/QD-TTg of the Prime Minister dated 18 August 2021 on the Functions, Duties, Organisation, and Operation of the Labour Relations Committee. 199 Ibid. 200 HCMC People’s Court, Judgment 1910/2018/KDTM-ST(28 December 2018); Superior People’s Court in HCMC, Judgment 05/2020/KDTM-PT(10 March 2020). 201 Interviews with O1(HCMC, 23 March 2021), O3(HCMC, 5 March 2021), and U3(HCMC, 6 April 2021). 202 Interviews with O1(HCMC, 23 March 2021) and O3(HCMC, 5 March 2021). 203 Interview with lawyer L2(HCMC, 11 February 2023). 204 Ibid. 205 Ibid. 206 Interview with Judge J(HCMC, 25 November 2022). 207 Ibid 25-26. 208 Also see interview with O4(HCMC, 20 March 2021). BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 90 REFERENCES Articles, Books, Papers and Reports 1. Anh Trọng(2020a),‘Số phận hàng vạn lái xe công nghệ đi về đâu?’[What Are the Future of Dozens of App-based Drivers?], Tiền phong, 10 March 2020 2. Anh Trọng(2020b),‘Tài xế diễu hành phản đối Grab khấu trừ doanh thu’[Drivers Marching to Protest Grab’s Deductions of Their Income], Tiền phong, 7 December 2020 3. Anwar, Mohammad Amir and Graham, Mark(2021),‘Between a Rock and a Hard Place: Freedom, Flexibility, Precarity and Vulnerability in the Gig Economy in Africa’ 25(2) Competition& Change 237 4. Bảo Hân and Trần Kiều(2020), Tổng LĐLĐVN sẽ nghiên cứu kỹ về kiến nghị đối với NLĐ là xe ôm công nghệ[The VGCL Will Carefully Study for Suggestions regarding Platform Motorbike Workers], Lao động, 12 December 2020 5. Berg, Janine, Furrer, Marianne, Harmon, Ellie, Rani, Uma and Silberman, M. Six(2018), Digital Labour Platforms and the Future of Work: Towards Decent Work in the Online World(International Labour Office) 6. Bình Minh(2020),‘Mùa ế khách của tài xế công nghệ’[Low Season of App-based Drivers], Zing News, 15 February 2020 7. Browning, Kellen(2023),‘California Court Mostly Upholds Prop. 22 in Win for Uber and Other Gig Companies,’ New York Times(13 March 2023) 8. Buckley, Joe(2020),‘Mapping Ride-hailing App Driver Strikes in Vietnam,’ New Mandala, 16 December 2020 9. Buckley, Joe(2022),‘Time for Vietnam to Apply Minimum Wage to Gig Economy Workers,’ FULCRUM, 10 August 2022 10. Casale, Giuseppe(2011),‘The Employment Relationship: A General Introduction,’ in Giuseppe Casale(ed), The Employment Relationship: A Comparative Overview(International Labour Office) 1 11. Chee, Foo Yun(2023),‘EU Countries Amend Draft Proposal on Gig Workers’ Rights, Companies Unhappy,’ Reuters(13 June 2023)< https://www.reuters.com/markets/europe/eu-countriesamend-draft-proposal-gig-workers-rights-companies-unhappy-2023-06-12/> BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 91 12. Cherry, Miriam A., and Aloisi, Antonio(2017),‘“Dependent Contractors” in the Gig Economy: A Comparative Approach’ 66(3) American University Law Review 635 13. Cherry, Miriam A. and Aloisi, Antonio(2018),‘A Critical Examination of a Third Employment Category for On-Demand Work’ in Nestor M. Davidson, Michèle Finck and John J. Infranca(eds), The Cambridge Handbook of the Law of the Sharing Economy(Cambridge University Press) 316 14. Công Trung(2023),‘Công ty giao đồ ăn Baemin rút khỏi thị trường Việt Nam, vì sao?’[Why Will Baemin Exit Vietnam?], Tuổi trẻ, 24 November 2023 15. Daugareilh, Isabelle(2019),‘France’ in Isabelle Daugareilh, Christophe Degryse and Philippe Pochet (eds), The Platform Economy and Social Law: Key Issues in Comparative Perspective(European Trade Union Institute) 51 16. De Groen, Willem Pieter, Kilhoffer, Zachary, Lenaerts, Karolien and Mandl, Irene(2018), Employment and Working Conditions of Selected Types of Platform Work(Eurofound) 17. De Stefano, Valerio(2016), The Rise of‘Just-in-time’ Workforce: On-demand Work, Crowdwork and Labour Protection in the‘Gig Economy’(International Labour Office) 18. De Stefano, Valerio and Aloisi, Antonio(2019),‘Fundamental Labour Rights, Platform Work and Human Rights Protection of Non-standard Workers’ in Janice R. Bellace and Beryl ter Haar(eds), Research Handbook on Labour, Business and Human Rights 359 19. De Stefano, Valerio, Durri, Ilda, Stylogiannis, Charalampos and Wouters, Mathias(2021), Platform Work and the Employment Relationship(ILO) 20. DFS(2019), The Future of Work in the Digital Era: The Rise of Labour Platforms(DFS, December 2019) 21. Do Hai Ha(2016a), Employment Agreements under the 2012 Labour Code, unpublished ILO report 22. Do Hai Ha(2016b), The Dynamics of Legal Transplantation: Regulating Industrial Conflicts in PostĐổi mới Vietnam(PhD thesis, University of Melbourne) 23. Do Hai Ha, Toan Le, Dinh Thi Chien and Luong Minh Son(2022),‘Platformisation as Informalisation: Decomposition, Recomposition and the Struggle of Platform Drivers for Labour Rights in Vietnam’ (Paper presented at Asian Law and Society Association Annual Meeting 2022, Hanoi, 9 — 10 December 2022) 24. Đoàn Thị Phương Diệp(2019),‘Nhận diện quan hệ pháp luật lao động trong bối cảnh cách mạng công nghiệp lần thứ tư(4.0)’[Identifying Employment Relationships in the Context of the Fourth Industrial Revolution](2) Tạp chí Khoa học Pháp lý 37 25. Đoàn Thị Phương Diệp and Hồ Đức Hiệp(2019),‘Nhận diện chuyển đổi quan hệ lao động công nghiệp: Những thách thức pháp lý đặt ra cho Việt Nam’[Identifying the Transformation of Labour Relations: Legal Challenges for Vietnam](709) Tạp chí Tài chính 31 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 92 26. Đức Hùng(2017),‘Tăng chiết khấu vô lý, GrabBike không còn là mảnh đất kiếm sống màu mỡ’ [Unreasonable Increase of Deduction Rates: GrabBike Is No Longer a Fertile Land for Living Purpose], Chất lượng& Cuộc sống, 21 August 2017 27. Duell, Nicola(2020), Platform Work(Background Paper, European Network of Public Employment Services) 28. Duggan, James, Sherman, Ultan, Carbery, Ronan, McDonnell, Anthony(2017),‘Algorithmic Management and App-work in the Gig Economy: A Research Agenda for Employment Relations and HRM’ 30(1) Human Resource Management Journal 114 29. Duy Anh(2020),‘Đời khổ xế công nghệ: Nghịch lý cuốc xe giá đắt, từng ngày kiệt sức’[The Hard Life of Platform Drivers: Exhausted Drivers Despite High Fares], Vietnamnet, 6 December 2020 30. Evans, Peter C. and Gawer, Annabelle(2016), The Rise of the Platform Enterprise: A Global Survey (The Center for Global Enterprises, January 2016) 31. Fairwork(2022), Fairwork Columbia Ratings 2022 32. Fiebauln, Andrés and Tirachini, Alejandro(2020),‘The Sharing Economy and the Job Market: The Case of Ride-Hailing Drivers in Chile’ 48(5) Transportation 2235 33. Fu, Bo-Shone(2021),‘The Contemporary Challenge and Government Responses on Delivery Platform Worker Rights and Benefits in Taiwan: The“MOL Guidance 2019” and the“Taipei City Ordinance 2019”’ 5(32) Japan Labor Issues 125 34. Gandini, Alessandro(2018),‘Labour Process Theory and the Gig Economy’ 72 Human Relations 1 35. Gawer, Annabelle(2021), Online Platforms: Societal Implications of the New Dominant Business Models of the Digital Economy(Hertie School) 36. Gillis, Dirk, Lenaerts, Karolien and Waeyaert, Willem(2022), France: Lessons from the Legislative Framework on Digital Platform Work(European Agency for Safety and Health at Work) 37. Global Living Wage Coalition(2021),‘Living Wage for Ho Chi Minh City, Vietnam,’ Global Living Wage Coalition 38. Google, Temasek and Bain& Company(2020), E-conomy SEA 2020. At Full Velocity: Resilient and Racing Ahead 39. Governance and Tripartism and European Labour Law Network(2013), Regulating the Employment Relationship in Europe: A Guide to Recommendation No. 198(International Labour Office) 40. Government(2019), Submission Paper regarding the Revised Labour Bill(Hanoi, 17 May 2019) BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 93 41. Graham, Andrew(2021),‘Ontario NDP Bill Looks to Protect Gig Workers,’ Global News(25 October 2021) 42. Graham, Mark, Woodcock, Jamie, Heeks, Richard, Fredman, Sandra, Du Toit, Darcy, Belle, Jean-Paul van, Mungai, Paul and Osiki, Abigal(2019),‘The Fairwork Foundation: Strategies for Improving Platform Work’ in Proceedings of the Weizenbaum Conference 2019“Challenges of Digital Inequality- Digital Education, Digital Work, Digital Life”(Berlin) 43. Hà Kiều(2019),‘Thời đại 4.0: Lương xe ôm cao gấp 3 — 4 lần viên chức nhà nước’[Industry 4.0 Era: Motorbike Taxi Drivers’ Salaries Are from Three to Four Times Higher than Public Servants’ Salaries], 24h.com, 03 August 2019 44. Hiếu Đam(2020),‘Tài xế công nghệ: Cầm về được nửa tiền mỗi cuốc xe’[Platform Drivers: Taking Home Half of the Money Earned from Each Ride], Dân Việt, 13 December 2020 45. Hoàng Mạnh(2018),‘Nghề lái xe Grab: Đang nằm ngoài sự điều chỉnh của Luật Lao động?’ [Grab Ride-hailing Work: Beyond the Scope of the Labour Code?], Dân trí, 22 June 2018 46. Hoàng Nam(2020),‘Kết quả kinh doanh của Grab như thế nào sau 6 năm vào Việt Nam?’ [Grab’s Business Results after Six Years of Operation in Vietnam], Dân Việt, 9 December 2020 47. Hoàng Thuỳ(2023),‘Ứng dụng Be lại tăng giá cước và chiết khấu với tài xế’[Be Has Increased Fares and Commission Rate], Đầu tư Doanh nghiệp, 31 March 2023 48. Hottenhuis, Stephanie, Blom, Frans, Boekhout, Roland, Knibbe, David, Lemstra, Derk, Miesen, Rob and Zijlema, Peter(2018), Unlocking the Value of the Platform Economy: Mastering the Good, the Bad and the Ugly(Dutch Transformation Forum, November 2018) 49. ILO(2007), The Employment Relationship: An Annotated Guide to ILO Recommendation No. 198 (10 October 2007) 50. ILO(2011), Quan hệ Việc làm: Tài liệu hướng dẫn Khuyến nghị số 198 của Tổ chức Lao động Quốc tế[The Employment Relationship: An Annotated Guide to ILO Recommendation No. 198](ILO in Vietnam) 51. ILO(2020), Promoting Employment and Decent Work in a Changing Landscape: Report III(Part B) (International Labour Office) 52. ILO(2021), World Employment and Social Outlook 2021: The Role of Digital Labour Platforms in Transforming the World of Work(International Labour Office) BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 94 53. ILO(2022), Decent Work in the Platform Economy(Geneva, 10 — 14 October 2022) 54. Ke, Zhenxing,(2022)‘A Third Employment Category 55. for Platform Workers in China: A Tough Start’ 10(2) The Chinese Journal of Comparative Law 297 56. Khánh Vy(2024),‘Ra đời 1 năm đã lên như diều gặp gió’[Fast Growth after One Year], CafeF, 15 March 2024 57. Kim Dung(2022),‘Tài xế công nghệ chưa mặn mà với bảo hiểm xã hội’[Platform Drivers Are Not Interested in Social Insurance], VOV, 26 May 2022 58. Kullman, Miriam(2022),‘“Platformisation” of Work: An EU Perspective on Introducing a Legal Presumption’ 13(1) European Law Journal 66 59. Lạc Duy(2020),‘Grab và tài xế: Mối quan hệ“nửa vời” giữa hợp đồng hợp tác kinh doanh và lao động?’[Grab and Drivers: An In-Between of Partnership Contracts and Labour Contracts?], Kinh tế Sài Gòn Online, 24 December 2020 60. Le, Lam and Nguyen, Nhung(2024),‘ShopeeFood Miscalculates Delivery Distances. Gig Workers in Vietnam Are Paying the Price,’ Rest of World, 7 March 2024. 61. Le Bret, Benoit(2023),‘A Bridge Too Far: The Unintended Consequences of the European Platform Work Directive,’ Euractiv(25 January 2023) 62. Lê Tuyết(2023),‘Tài xế công nghệ“tranh cuốc gay gắt vì lực lượng tăng nhanh”’[App-based Drivers Compete for Rides Due to a Quickly Expanding Workforce], VnExpress, 27 October 2023 63. Lee, Min Kyung, Kusbit, Daniel, Metsky, Evan and Dabbish, Laura(2015),‘Working with Machines: The Impact of Algorithmic and Data-Driven Management on Human Workers’ in Association for Computing Machinery: Conference on Human Factors in Computing Systems, 18 April 2015, 1603 64. Lenaerts, Karolien, Waeyaert, Willem, Smits, Ine and Hauben, Harald(2021), Digital Platform Work and Occupational Safety and Health: A Review(European Agency for Safety and Health at Work) 65. Macedo Vitorino(2023),‘New Legal Framework for Portuguese Labour Code – Decent Work Agenda,’ Macedo Vitorino(18 April 2023) BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 95 66. Mai Hà(2020),‘6 năm Grab tại Việt Nam: Tài xế từ“trái ngọt” tới“trái đắng”[Six years of Grab in Vietnam: Drivers Turning from“Sweetness” to“Bitterness”], Thanh Niên, 10 December 2020 67. McGaughey, Ewan(2019),‘Uber, the Taylor Review, Mutuality and the Duty Not to Misrepresent Employment Status’(2019) 48(2) Industrial Law Journal 180 68. Meijerink, Jeroen, Jansen, Giedo and Daskalova, Victoria(2021),‘Platform Economy Puzzles: The Need for a Multidisciplinary Perspective on Gig Work’ in Jeroen Meijerink, Giedo Jansen and Victoria Daskalova(eds), Platform Economy Puzzles: A Multidisciplinary Perspective on Gig Work (Edward Elgar Publishing) 2 69. Minh Hạnh(2024),‘Không còn là công việc hấp dẫn, xe ôm công nghệ tính bỏ nghề’[App-based Motorbike Drivers Intend to Quit Job, As It Is No Longer a Good Job], Lao động, 11 January 2024 70. Minh Khánh(2022),‘Nhiều tài xế xe công nghệ quấy phá ứng dụng’[Several Platform Drivers Disrupting Applications], Zing News, 5 July 2022 71. Minh Khánh(2023),‘Đà Nẵng đề nghị làm rõ hợp đồng hợp tác kinh doanh của Grab’[Da Nang authorities Suggest Clarifying Grab’s Business Cooperation Contracts], Zing News, 8 May 2023 72. Mitchell, Richard and Murray, Jill(2002),‘Book Review: Legal Regulation of the Employment Relation’ 31(2) Industrial Law Journal 199 73. MOLISA(2019), Report No. 50/BC-LDTBXH on Appraisal of the Implementation of the Labour Code(Hanoi, 7 May 2019) 74. MOJ(2021), Report No. 45/BC-BTP on review of the civil law framework for transactions, contracts, ownership, property rights and dispute resolution in sharing economic models(17 March 2021) 75. MPI(2020), Dự thảo báo cáo đánh giá tác động của một số loại hình kinh tế chia sẻ chính tới nền kinh tế[Draft Report on Assessment of the Impact of Major Sharing Economic Models on the Economy],(Hanoi, November 2020) available at 76. Ng Hải(2021),‘Grab, Gojek, Baemin tăng phụ phí dịp Tết đến 15.000 đồng/chuyến’[Grab, Gojek and Baemin Charge Additional Fees Up to 15,000 VND/Ride during Tết Holidays], Người Lao động, 8 February 2021 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 96 77. Ngô Cường and Minh Ánh(2020),‘Tài xế Grab đề nghị công nhận là người lao động, chứ không phải đối tác’[Grab Drivers Request to Be Recognized as Employees, Not Partners], Lao động, 10 December 2020 78. Ngọc Bình(2022),‘Hơn 5 năm khoác áo tài xế công nghệ: Khi niềm vui“tiếp lửa” cho nghề’ [More Than Five Years of Being an App-based Driver: Happiness Making the Job], Thanh Niên, 15 August 2022 79. Nguyễn Nhật Tiến(2017),‘Thói quen gọi taxi ở Việt Nam đã thay đổi’[Habits of Taxi Customers Have Changed], Vietnamnet, 24 June 2017 80. Nguyễn Sinh and Uyên Hoà(2021),‘Nhập nhằng quan hệ giữa tài xế và hãng taxi công nghệ’ [Uncertainty regarding the Relationship between Drivers and E-taxi Companies], Pháp luật Việt Nam, 8 April 2021 81. Nguyễn Thành Nhân(2018),‘Tân binh GoViet“đại chiến” gã khổng lồ Grab: Khách hàng nên cẩn trọng’[Rookie GoViet Fights Against Giant Grab: Customers to Be Careful], Người Đưa tin, 20 October 2018 82. Nguyễn Trang(2018),‘Tài xế Grab, Uber có được Luật Lao động bảo vệ không?’[Are Grab, Uber Drivers Be Protected by Labour Law?], VOV, 13 March 2018 83. Nguyệt Anh(2022),‘Những“đặc quyền” khi làm tài xế công nghệ’[“Privileges” of App-based Drivers], Thanh Niên, 21 September 2022 84. Nhat, Minh(2020),‘Pandemic Pushes Up Food-delivery Apps’ Business,’ Hanoi Times, 20 December 2020 citing a survey conducted by Q&Me in December 2020 85. Nhật Minh(2024),‘Tài xế xe ôm công nghệ chạy hơn 10 tiếng vẫn không đủ lo cho gia đình’[Despite Working More Than Ten Hours, App-based Motorbike Drivers Do Not Earning Sufficient Living Pay], Lao động, 3 March 2024 86. OECD(2019), An Introduction to Online Platforms and Their Role in Digital Transformation(OECD Publishing) 87. Pereyra, Francisca and Poblete, Lorena(2022), Regulating Platform Delivery Work in Argentina. Tensions between Regulations and the Priorities of Workers, 234 AFD Research Papers BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 97 88. Pham, Cong Bay(2010), Determining the Existence of an Employment Relationship at the Court: Practices, Difficulties and Solutions, unpublished ILO report(Hanoi, November 2010) 89. Phạm Hoài Huấn(2021),‘Định danh cho các tài xế Grab: Những yếu tố cần cân nhắc’[Identifying Grab Drivers: Factors to Consider], Kinh tế Sài Gòn Online, 4 April 2021 90. Phan Thương(2019),‘Hai tài xế kiện Grab ra Toà vì bị khoá tài khoản vĩnh viễn,’ Thanh niên (3 August 2019) 91. Phương Dung(2020),‘Bước chuyển mình của“Kỳ lân Hàn Quốc” sau 1 năm ra mắt tại Việt Nam’ [The Development of the“Korean Unicorn” a Year after Its Commencement in Vietnam], Công nghệ số& Truyền thông, 9 June 2020 92. PV(2018),‘Thu nhập lên đến chục triệu nhờ những cuốc xe công nghệ’[Incomes Up to Ten Million[Dong] Thanks to Ride-hailing Trips], Lao động, 20 November 2018 93. Reshaping Work(2023), Back to What Truly Matters: Platforms, AI, and Youth in the Workplace 94. Rodríguez-Modro ñ o, Paula, Agenjo-Calderón, Astrid and López-Igual, Purificación(2022),‘Platform Work in the Domestic and Home Care Sector: New Mechanisms of Invisibility and Exploitation of Women Migrant Workers’ 30(3) Gender& Development 619 95. Sheposh, Richard(2020),‘Platform Economy’ in Salem Press Encyclopedia(Salem Press) 96. Schwellnus, Cyrille, Geva, Assaf, Pak, Mathilde and Veiel, Rafael(2019), Gig Economy Platforms: Boon or Bane?(OECD, Economics Department Working Papers No. 1550) 97. Shaheen, Susan, Cohen, Adam and Zohdy, Ismail(2016), Shared Mobility: Current Practices and Guiding Principles(US Department of Transportation) 98. Simonet, Rodrigo and Garcia, Jorge(2022),‘Chile’s Legislative Solution for Platform Work: Does It Fit the Bill?’ Comparative Labor Law and Policy Journal 99. Srnicek, Nick(2017), Platform Capitalism(Cambridge University Press) 100. Stewart, Andrew and Stanford, Jim(2017),‘Regulating Work in the Gig Economy: What Are the Options?’ 28(3) The Economic and Labour Relations Review 420 101. The Asahi Shimbun(2022),‘Labor Body Recognizes Union Rights for Uber Eats Deliverers’ The Asahi Shimbun(25 November 2022) BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 98 102. Thế Lâm(2019),‘Hàng trăm ngàn xe ôm công nghệ sẽ“vướng” khấu trừ trong thuế thu nhập’ [Hundreds of Thousand of App-based Motorbike Drivers Will Be Subject to Income Tax Deductions], Lao động, 3 September 2019 103. Thế Lâm(2022),‘Tài xế xe ôm công nghệ quay cuồng trong“cơn lốc” giá xăng’[App-based Motorbike Drivers Are Dizzy Due to Petrol Price Surge], Lao động, 17 June 2022< https://laodong. vn/ban-doc/tai-xe-xe-om-cong-nghe-quay-cuong-trong-con-loc-gia-xang-1057331.ldo> 104. Thiên Trang(2020),‘Now dành 100 triệu đồng mỗi tháng cho Quỹ chăm sóc tài xế’[Now Donates VN$ 100 Million Each Month for the Driver Care Fund], Hà Nội mới, 26 December 2020 105. Tiểu Phượng(2020),‘Chạy GrabBike để mua ô tô Mercedes: Giấc mơ viển vông hay mục tiêu khả thi?’[Driving GrabBike to Buy a Mercedes Car: Pipe Dream or Feasible Goal?], VietnamBiz, 4 May 2020 106. Trần Duy(2022),‘Hoạt động chui nhiều nơi, vì sao Grab chưa bị xử lý?,’ Giao thông, 28 September 2022 107. Trần Phương(2021),‘Gọi tài xế công nghệ là đối tác: cách gọi luồn lách’[Calling Platform Drivers Partners: A Way to Evade[the Law]], Tuổi trẻ, 23 March 2021 108. Trần Thị Hồng Minh(2021),‘Tác động của mô hình kinh tế chia sẻ tới nền kinh tế Việt Nam’[The Impact of the Sharing Economy on the Vietnamese Economy],(2021)(4) Tạp chí Kinh tế và Dự báo 36 109. Trần Thị Thanh Huyền(2021),‘Bàn về vấn đề việc làm và người lao động trong nền kinh tế chia sẻ’[Employment and Workers in the Sharing Economy],(2021)(1) Tạp chí Kinh tế và Dự báo 29 110. Trương Quốc Lâm, Nguyễn Đức Lộc, Nguyễn Trần Phượng Trân and Nguyễn Xuân Trường(2021), Đánh giá thực trạng công tác quản lý và các phương thức tập hợp người lao động tham gia cung ứng dịch vụ xe sử dụng nền tảng ứng dụng công nghệ trên địa bàn thành phố Hồ Chí Minh (Research Report, HCMC, February 2021) 111. Văn Hưng(2020),‘Tài xế là lao động của Grab’[Drivers Are Grab’s Workers], Zing News, 18 December 2020 112. Van Olmen, Chris(2022),‘Belgium: Labour Deal 2022: Better Protection for Digital Platform Workers,’ L&E Global(9 December 2022) BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 99 113. Văn Hưng and Thanh Hương(2020),‘Hiện thực cay đắng của giấc mơ“chạy Grab kiếm 30 triệu đồng/tháng”’[The Bitter Reality of the Dream of“Driving Grab to Earn 30 Million Dong Per Month], Zing News, 13 December 2020 114. Viễn Thông(2020),‘Thị trường gọi xe: Khe cửa hẹp cho các tân binh’[Ride-hailing Market: A Narrow Door for Newcomers], VnExpress, 9 October 2020 115. Viễn Thông(2021),‘Gojek có 200.000 tài xế ở Việt Nam’[Gojek Has 200,000 Drivers in Vietnam], VnExpress, 20 April 2021 116. Việt Hưng(2023),‘Chia lại thị phần gọi xe công nghệ’[Reallocating the Ride-hailing Market Share], The Leader, 13 November 2023 117. Võ Thủy and Lê Phan(2018),‘Rủi ro nghề tài xế công nghệ’[Risks of App-based Motorbike Drivers], Tuổi trẻ cuối tuần, 11 December 2018 118. Waeyaert, Willem, Lenaerts, Karolien and Gillis, Dirk(2022), Spain: The‘Riders’ Law,’ New Regulation on Digital Platform Work(European Agency for Safety and Health at Work) 119. Wood, Alex J., Graham, Mark, Lehdonvirta, Vili and Hjorth, Isis(2019),‘Networked but Commodified: The(Dis)Embeddedness of Digital Labour in the Gig Economy’ 53(5) Sociology 931 120. World Bank(2016), World Development Report 2016: Digital Dividends Cases 1. HCMC People’s Court, Judgment 1910/2018/KDTM-ST(28 December 2018) 2. Superior People’s Court in HCMC, Judgment 05/2020/KDTM-PT(10 March 2020) 3. Supreme People’s Court, Precedent 20/2018/AL regarding Conclusion of an 4. Employment Contract upon the Expiration of Probation 5. Supreme People’s Court, Precedent 70/2023/AL regarding Termination of the Employment Contract of a Part-time Union Cadre 6. Uber BV v Aslam[2021] UKSC 5 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 100 Code of Interviews Driver Date D1 12/5/2019 D2 15/5/2019 D3 17/5/2019 D4 15/5/2019 D5 15/5/2019 D6 5/6/2019 D7 6/6/2019 D8 6/6/2019 D9 6/6/2019 D10 5/6/2019 D11 5/6/2019 D12 5/6/2019 D13 6/6/2019 D14 5/6/2019 D15 5/6/2019 D16 5/6/2019 D17 6/6/2019 D18 6/6/2019 D19 5/6/2019 D20 5/6/2019 D21 5/6/2019 D22 5/6/2019 Place HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC Driver D23 D24 D25 D26 D27 D28 D29 D30 D31 D32 D33 D34 D35 D36 D37 D38 D39 D40 D41 D42 D43 D44 Date 5/6/2019 5/6/2019 5/6/2019 5/6/2019 6/6/2019 5/6/2019 20/7/2020 14/6/2020 2/2/2021 18/1/2021 26/1/2021 19/1/2021 19/1/2021 18/1/2021 9/6/2020 9/6/2020 9/6/2020 21/6/2020 21/6/2020 13/6/2020 31/3/2021 9/4/2021 25/7/2023 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ Place HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC HCMC Driver D45 D46 D47 D48 D49 D50 Date 18/3/2021 20/3/2021 3/9/2020 17/1/2021 28/3/2021 25/7/2023 Lawyer L1 L2 Date 20/5/2019 11/2/2023 Platform Manager M1 M2 M3 M4 Date 11/9/2020 31/3/2021 7/7/2021 7/7/2021 Place HCMC HCMC HCMC HCMC HCMC HCMC Place Hanoi HCMC Place HCMC Hanoi HCMC HCMC State Official O1 O2 O3 O4 Union Cadres U1 U2 U3 U4 U5 Date 20/5/2019 23/3/2021 3/5/2019 11/2/2021 5/3/2021 20/3/2021 Date 6/4/2021 4/2/2021 6/4/2021 16/2/2021 30/3/2021 101 Place Hanoi HCMC HCMC HCMC HCMC HCMC Place HCMC Zoom HCMC HCMC HCMC BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 102 Legal Instruments 1. Civil Code 2015 2. Employment Rights Act 1996(UK) 3. ILO Protection of Wages Convention, 1949 (No. 95) 4. ILO Employment Relationship Recommendation, 2006(No. 198) 5. Labour Code 1994 6. Labour Code 2012 7. Labour Code 2019 8. Law on Employment 2013 9. Law on Investment 2020 10. Law on Labour Safety and Hygiene 2015 11. Law on Health Insurance(Revised) 2014 12. Law on Road Traffic 2008 13. Law on Social Insurance 2014 14. Law on Trade Union 2012 15. Decree 10/2020/ND-CP of the Government dated 17 January 2020 on Business Conditions for Automobile Transport Services 16. Decree 90/2019/ND-CP of the Government dated 15 November 2019 on Regional Minimum Wage Rates for Employees 17. Decision 595/QD-BHXH of Vietnam Social Security dated 14 April 2017 on Procedures to Collect Social Insurance, Health Insurance, Unemployment Insurance, Work-related and Occupational Disease Insurance Contributions 18. Decision 1413/QD-TTg of the Prime Minister dated 18 August 2021 on the Functions, Duties, Organization and Operation of the Labour Relations Committee 19. Resolution 04/2019/HDTP of the Judges’ Council of the Supreme People’s Court dated 18 June 2019 on Selection, Announcement and Application of Precedents 20. VGCL Charter 2020 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 103 Others 1. Baemin, Hợp đồng Hợp tác Kinh doanh[Business Cooperation Contract](“ Baemin Driver Agreement”) 2. Baemin(2019),‘Bộ Quy tắc Ứng xử Tài xế Baemin’[Code of Conduct for Baemin Drivers], Blog Tài xế Baemin, 17 December 2019(“ Baemin COC”) 3. Baemin(2020),‘Tổng hợp chương trình thưởng Baemin’[Consolidation of Baemin Bonus Programs], Blog Tài xế Baemin, 15 January 2020(“ Baemin Bonus Policy”) 4. Baemin(2021),‘Baemin dành tặng gói bảo hiểm“Trợ cấp thu nhập khi nằm viện” cho đối tác tài xế Baemin’[“Income Protection for Hospitalization” Insurance for Baemin Drivers], Blog Tài xế Baemin, 2 December 2021 5. Baemin(2022a),‘Bảo hiểm tai nạn UIC- Người bạn đồng hành vô hình của xế Min’[Accident Insurance UIC – Invisible Fellow Traveller of Baemin Drivers], Blog Tài xế Baemin, 3 August 2022 6. Baemin(2022b),‘Cẩm nang dành cho đối tác’[Manual for Driver-Partners], Blog Tài xế Baemin [accessed 22 December 2022] 7. Be, Hợp đồng Hợp tác[Cooperation Contract](“ Be Driver Agreement”) 8. Be(2020a),‘Bảo hiểm chăm sóc sức khoẻ beHealthCare dành cho tài xế’[Healthcare Insurance for Drivers], Be, 14 October 2020 9. Be(2020b),‘Chính sách bảo hiểm tai nạn Tài xế và Hành khách’[Accident Insurance Policy for Drivers and Passengers], Be, 29 June 2020 10. Be(2020c),‘Quy định& Chế tài về Chất lượng, An toàn& Gian lận’[Rules and Sanctions regarding [Service] Quality, Safety and Fraudulence], Be, 24 February 2020 11. Be(2022a),‘Bộ Quy tắc Ứng xử Dành cho Quý Tài xế Be’[Code of Conduct for Be Drivers], Be (accessed 30 November 2022)(“ Be COC”) 12. Be(2022b),‘Chương trình đào tạo – beAcademy’[Training Program- beAcademy], Be[accessed 22 December 2022] BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 104 13. Be(2022c),‘Chương trình“tài xế Be thân thiết” 2022’[Loyal“Be Drivers” Program 2022], Be, 18 September 2022 14. Be(2022d),‘Chương trình thưởng’[Bonus Programs], Be accessed 30 November 2022(“ Be Bonus Policy”) 15. Be(2022e),‘Cộng đồng Tài xế be’[Be Driver Community], Be [accessed 22 December 2022] 16. Be(2022f),‘Cước phí dịch vụ vận chuyển’[Charges for Ride Services], Be accessed 30 November 2022 17. Bùi Xuân Hải and Hà Thị Thanh Bình(2020), Statements at Conference on“Some Legal and Governance Issues in the Sharing Economic Model”(HCMC Law University, 9 October 2020) 18. Cefaliello, Aude(2023),‘Platforms and their Presumptions of Power’ Social Europe(22 February 2023) 19. CLBBeBikeSàiGòn[BeBikeSaigonClub] [accessed 22 December 2022] 20. Cộng đồng Grab Việt Nam(4 bánh và 2 bánh)[Grab Vietnam Community(Car and Motorbike)] [accessed 22 December 2022] 21. European Commission(2021), Proposal for a Directive of the European Parliament and of the Council on Improving Working Conditions in Platform Work(9 December 2021) 22. European Parliament(2022),‘Digital Workers: Better Working Conditions and Protection of Rights’(Press Release, 12 December 2022) 23. Gofast, Hợp đồng Nhà thầu Độc lập[Independent Contractor Agreement](“ Now Driver Agreement”) 24. Gofast(2020), Bộ Quy tắc Ứng xử[Code of Conduct](1 August 2020)(“ Now COC”) 25. Grab, Thoả thuận Hợp tác[Cooperation Agreement](“ Grab Driver Agreement”) 26. Grab(2014),‘Các chế độ đãi ngộ dành cho đối tác’[Benefits for Partners], Grab, 3 November 2014 27. Grab(2017),‘Grab áp dụng dịch vụ GrabBike tại 6 thành phố mới!’[Grab Introduce GrabBike Services in Six New Cities], Grab, 12 December 2017 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 105 28. Grab(2020a),‘Grab 2 bánh TPHCM: Cập nhật chương trình thưởng ngọc, áp dụng từ 18/3/2020’ [GrabBike HCMC: Updated Gem Bonus Program, Effective from 18 March 2020], Grab, 16 March 2020(“ Grab Bonus Policy”) 29. Grab(2020b),‘Siêu ưu đãi: Gói bảo hiểm chăm sóc sức khoẻ dành cho Đối tác Grab’[Discount: Healthcare Insurance Package for Grab Partners], Grab, 16 October 2020 30. Grab(2021),‘Bộ Quy tắc Ứng xử Dành cho Đối tác Grab’[Code of Conduct for Grab Partners], Grab, 2 November 2021 (“ Grab COC”) 31. Grab(2022a),‘Cập nhật chính sách bảo hiểm tai nạn cá nhân Chubb năm 2022’[Update on Accident Insurance Policy for Individuals Chubb 2022], Grab, 12 March 2022 32. Grab(2022b),‘Chúng tôi có thể giúp gì cho bạn?’[What Can We Do for You?], Grab[accessed 22 December 2022] 33. Grab(2022c),‘Grab Blogs,’ Grab[accessed 22 December 2022] 34. Grab(2022d),‘Thông báo giới thiệu sản phẩm mới:“Bảo hiểm Đối tác”[Introduction of a New Product:“Partner Insurance”], Grab, 31 May 2022 35. Grab(2023a),‘Đào tạo Đối tác Mới(GrabBike)’[Training New Partners(GrabBike)], Grab[accessed 20 March 2023] 36. Grab(2023b),‘Grab 2 bánh: Thay đổi thuế suất thuế giá trị gia tăng theo Nghị định 44/2023/ NĐ-CP’[VAT Adjustment under Decree 44/2023/ND-CP], Grab, 1 July 2023 37. Hội AE Grabbike Sài Gòn[Group of Saigon Grabbike Brothers][accessed 22 December 2022] 38. Hội Grabbike Sài Gòn[Grabbike Saigon Group] [accessed 22 December 2022] 39. Hội Tài xế Hub – ShopeeFood[Hub – ShopeeFood Driver Group][accessed 22 December 2022] 40. Legislative Counsel Digest(2019),‘Assembly Bill No. 5’, http//Bill Text- AB-5 Worker status: employees and independent contractors.(ca.gov), 19 September 2019 BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ 106 41. Người Lao động(2024),‘TAGs nghiệp đoàn xe ôm’[TAGs Motorbike Driver Syndicate], Người Lao động[The Labourers][accessed 30 June 2024] 42. Murphy, Jaan(2023), Regulating the‘Gig’ Economy As a Form of Employment(Parliament of Australia, Law and Bill Digest)< https://www.aph.gov.au/About_Parliament/Parliamentary_ departments/Parliamentary_Library/pubs/BriefingBook47p/GigEconomy>(accessed 4 October 2023) 43. People(2022),‘We Are Not Platforms! Industries Deeply Concerned by EU Proposal,’ RU(3 November 2022) 44. ShopeeFood(2020),‘Chính sách thưởng ngày’[Daily Bonus Policy], ShopeeFood Help Center, 31 December 2020(“ Now Bonus Policy”) 45. ShopeeFood(2022a),‘Bảo hiểm tai nạn dành riêng cho Tài xế ShopeeFood’[Accident Insurance for ShopeeFood Drivers], ShopeeFood Help Center, 12 January 2022 46. ShopeeFood(2022b),‘Cẩm nang đào tạo’[Training Manual], ShopeeFood[accessed 22 December 2022] 47. ShopeeFood(2022c),‘HUB là gì?’[What Is HUB?], ShopeeFood Help Center, 19 December 2022 48. ShopeeFood(2022d),‘Thu nhập của tài xế được tính như thế nào?’[How to Calculate Drivers’ Earnings?], ShopeeFood Help Center, 11 November 2022 49. ShopeeFood(2022e),‘Xin chào! Bạn cần được hỗ trợ?’[Hello! Do You Need Help?], ShopeeFood Help Center[accessed 22 December 2022] 50. Trần Đức Ân(2020),‘Thưa ACE tài xế thân mến’[Dear Driver Brothers and Sisters][status update], Facebook, 10 December 2020 51. Zing News(2024),‘Tài xế GrabBike tử vong’[GrabBike Drivers Died], Zing News [accessed 30 June 2024] BEHIND THE MYTH OF‘BUSINESS PARTNERSHIPS’ BEHIND THE MYTH OF BUSINESS PARTNERSHIPS VULNERABLE APP-BASED DRIVERS IN URGENT NEED OF LEGAL PROTECTION IN VIET NAM Responsible for publishing, contents Director- Editor- in- Chief: DR. TRAN CHI DAT Editors: Graphic editor: Designer: Nguyen Long Bien- Bui Thi Nga Luck House Co. LTD. Luck House Co. LTD. INFORMATION AND COMMUNICATIONS PUBLISHING HOUSE Website: www.nxbthongtintruyenthong.vn, book365.vn, ebook365.vn Address: 6 th floor, 115 Tran Duy Hung street, Cau Giay District, Hanoi City Editor’s telephone: 024.35772143/35772141 Issuer’s telephone: 024.35772138/40 E-mail: nxb.tttt@mic.gov.vn Branch Ho Chi Minh City: No. 211 Nguyen Gia Tri Road, Ward 25, Binh Thanh District, Ho Chi Minh City Telephone: 028.35127750 E-mail: cnsg.nxbtttt@mic.gov.vn Branch in region Trung- Tay Nguyen: Adress: No. 42 Tran Quoc Toan, Hai Chau District, Da Nang City No. 46 Y Jut, Buon Ma Thuot City, Dak Lak Province Telephone: 0236.3897467/0262.3808088 E-mail: cndn.nxbtttt@mic.gov.vn Printed: 70 copies, dimension 21 x 29.7 cm at LUCK HOUSE CO. LTD. Address: 4/6/518 Doi Can, Cong Vi, Ba Dinh, Ha Noi No.: 5294-2024/CXBIPH/3-182/TTTT Publishing License No: 736/QĐ-NXBTTTT dated 31/12/2024 Finished printing and depositing in the 4 th quarter of 2024 ISBN: 978-604 4997-78-0 About the authors Do Hai Ha is a research associate at Asian Law Centre, Law School, University of Melbourne, and a lecturer at Ho Chi Minh City University of Law. His research intersects between labour law, law and development, law and society and the legal dimension of Chinese globalism in Vietnam. Toan Le is a senior lecturer at Swinburne Law School in the School of Business, Law and Entrepreneurship, Swinburne University of Technology. His research focuses on regulatory issues concerning labour, property and taxation, and he present the issues from socio-legal and law and society perspectives. Huynh Thi Ngoc Tuyet is a retired researcher of the Southern Institute of Social Science, belonging to Vietnam Academy of Social Sciences, located in Ho Chi Minh City. She researches labour issues in the platform companies and sharing economy, and rural-urban migration. Nguyen Thi Minh Chau is a researcher at the Southern Institute of Social Science, belonging to Vietnam Academy of Social Sciences, located in Ho Chi Minh City. Her interest lies in internal rural- urban migrations, labour issues and access to social protection. Dinh Thi Chien is a lecturer at Ho Chi Minh City University of Law. She teaches and researches in labour law issues with a focus on comparative labour law and the employment contract. Luong Minh Son is a lecturer at Ho Chi Minh City University of Law. He teaches and researches in labour law. Nguyen Quoc Thang is a lecturer at the University of Labour and Social Affairs(Ho Chi Minh City Campus). He teaches and researches on labour issues. Imprint 7 Ba Huyen Thanh Quan, Ba Dinh Hanoi, Vietnam IPO Box 44 Responsible: Timo Rinke| Resident Representative vietnam.fes.de FriedrichEbertStiftung.Vietnam The views expressed in this publication are not necessarily those of the Friedrich-Ebert-Stiftung. Commercial use of all media published by Friedrich-Ebert-Stiftung(FES) is not permitted without the written consent of the FES. Publications by the Friedrich-Ebert-Stiftung may not be used for the electioneering purposes. NOT FOR SALES