Hostname: page-component-7b7567c8c-fvjsw Total loading time: 0 Render date: 2026-11-14T09:16:54.836Z Has data issue: false hasContentIssue false
  • English
  • Français

Regulatory Risks Faced by the Transportation Sharing Economy: Workforces at Stake

Published online by Cambridge University Press:  08 February 2019

Ana Maria CORRÊA
Rights & Permissions [Opens in a new window]

Extract

The rise of sharing economy platforms is ubiquitous throughout the world.1 Their success is attributed to the fact that they directly connect supply and demand by enabling individuals to share goods and personal services, with minimal initial barriers to entry. In addition, they outsource their workforce – making it less expensive and easier to expand internationally. Similar to other global supply chains, workers are mediated by a third party and technology is fundamental to that.2

Information

Type
Articles
Information
European Journal of Risk Regulation , Volume 9 , Issue 4 , December 2018, pp. 641 - 654
Copyright
© Cambridge University Press 

I Introduction

The rise of sharing economy platforms is ubiquitous throughout the world.Footnote 1 Their success is attributed to the fact that they directly connect supply and demand by enabling individuals to share goods and personal services, with minimal initial barriers to entry. In addition, they outsource their workforce – making it less expensive and easier to expand internationally. Similar to other global supply chains, workers are mediated by a third party and technology is fundamental to that.Footnote 2

Studies have demonstrated that the internet has been related to the decentralisation of the workforce in modern economies even before the emergence of online platforms.Footnote 3 Fifteen years ago, the World Summit of the Information Society recognised that Information and Communication Technology (ICT) was changing working practices.Footnote 4 Online platforms are the apex of this trend. In 2016, the European Commission reported that 32% of Europeans provided services through online platforms.Footnote 5 In Europe and the United States alone, it is estimated that sharing economy businesses have fast been increasing their number of independent workers.Footnote 6 The inevitability of online platforms in our modern economy led the UK government to release a “Good Work Plan” in February 2018, addressing, among other things, the challenges of the changing workforce conditions in our technological era.Footnote 7

Currently, Transportation Network Companies (TNC) are one of the most established examples of the sharing economy. Uber is an arguably notorious case available in the market, but it has direct and specialised competitors around the world, such as Lyft, Sidecar, Didi Chuxing, Kabbee, Hailo, Flywheel and Ola Cabs among many others. As with any new industry, TNCs are facing regulatory challenges that could jeopardise their current structure and their further development.Footnote 8 Although some sharing economy platforms have had global ambitions, their success and activities depend on national regulations and institutions. Apart from the EU, the fact that one platform can work under the label of an intermediary information society service in one country does not imply that the same platform, structured in the same way, will be able to work as an intermediary information society service in other jurisdictions. Transplanting sharing economy platforms to different markets has regulatory risks. One of the main barriers to transplanting them across countries is the way different traditions address the legal perception and conditions of an employment relationship. This issue has raised particular attention in the shared transportation businesses. In that regard, the labour class actions faced by transportation sharing economies may shape how they develop their activities and manage their workforce across countries.Footnote 9 Notwithstanding the fact that labour law is traditionally a national matter, global sharing companies have created similar labour problems in several countries almost simultaneously. In this context, comparative law is an appropriate tool with which to consider the regulatory risks concerning the workforces of online platforms.

This article aims to explore the regulatory risks that transportation sharing economies face when they outsource their main service – transportation. TNCs have defined themselves as a technology intermediary to connect drivers and passengers. They accomplished this by way of an ingenious legal framework of having all their main suppliers not as employees but as independent contractors. TNCs are not alone; the “on-demand” economy depends, in general, on independent contractors to provide services and goods. This article will primarily address the core question of whether the drivers are employees of TNCs and how they deploy legal strategies to avoid that classification for their contractors.

Despite other sharing economy companies experiencing legal problems regarding their drivers,Footnote 10 the article will focus on just one company, Uber, considering its global reach and its representativeness in the sector. By reviewing recent judicial cases in the United States and the United Kingdom, we will explore how the company recruits its drivers, develops its contracts and terms of service (part II); the challenges to define employment (part III); and some local legal strategies to address misclassification disputes (part IV).Footnote 11

It will be argued that the employment status of the drivers is threatened by the technological identity forged by the company. The aim of analysing these two jurisdictions is to demonstrate how different labour law traditions have addressed the misclassification of workers imposed by a single global company. Surprisingly, the answer to the issue of misclassification was noticeably similar.

II Drivers at stake: labour litigation across jurisdictions

A giant legal battle started in the United States between 2013 and 2014. In eastern and western states of the country, in California and Massachusetts, two class actions representing an average of 300,000 drivers challenged the business configuration of Uber.Footnote 12 In the cases of O’Connor v Uber Technology and Yucesoy v Uber Technology,Footnote 13 the core question was crucial for guarantees and benefits for Uber’s main workforce: “Are the drivers employees or independent contractors?”Footnote 14 The consequences of requalifying the drivers can massively change the company’s structure and, most importantly, the drivers’ rights. In these two class actions, drivers claimed that they were misclassified as independent contractors; they were denied reimbursement of their necessary business expenses, which would include gas, smartphones, phone data and insurance.Footnote 15 The company retroactively would have to pay approximately $800,000,000 to its drivers.Footnote 16

In 2015, two years after the class action was filed in California, Yaseen Aslam, James Farrar and 19 other drivers filed a case in the Employment Tribunal in London. The dimensions of the case were limited when compared with the California and Massachusetts class actions, which together represented thousands of drivers. However, the core litigation was identical. The claimants accused the platform of misclassifying their contracts and failing to pay minimum wage, paying working time and paid leave. In the case Aslam, Farrar v Uber BV, Uber London Ltd, and Uber Britannia Ltd, the Employment Tribunal had the chance to analyse in detail the conditions of work, the contracts, and the constraints of the drivers in London.Footnote 17 The Tribunal also ruled, for the first time in Europe, that drivers are employees based on the legal design that linked the drivers to the platform and the control the company had over the drivers.

In a domino effect, in February 2017, a Labour Tribunal Court decided, for the first time, in the state of Minas Gerais, Brazil, that Uber driver Rodrigo Leonardo Silva Ferreira was an employee of the company.Footnote 18 In France, the Tribunal des Prud’Hommes in Paris decided in February 2018, that an Uber driver was not an employee. Even if the driver had provided more than 4,000 rides through the application and worked most of the time for the platform, the Court considered the driver was free to choose his working time. In addition, the Court reinforced that Uber is not a transportation company but only a platform to put supply and demand in contact with each other.Footnote 19

Similar to other sharing economies, the platform identifies itself worldwide as a technology company whose only purpose is to put passengers and drivers in contact with each other. Despite being a technological intermediary, the company decides the fare rates,Footnote 20 receives payment for all rides and retains a percentage of it,Footnote 21 sets the cancellation policy, and selects the driver when a ride is requested.Footnote 22 In relation to its main suppliers, it requires, in general, drivers to have an appropriate driver’s licence (in some jurisdiction it requires years of driving experience), an eligible vehicle with four doors, and insurance certificates.Footnote 23 In addition, a criminal record check and a personal interview with the driver is also conducted.Footnote 24 Each driver has an independent contract, making Uber a platform connected to thousands of independent drivers who provide the main and final service available to the passenger via the platform.

The way that the platform is legally structured made courts in several countries face an existential dilemma: does it provide a transportation or information society service? The company has faced massive litigation thus far due to its identity.Footnote 25

1 To be or not to be a technology company

It seems that the company is more than a digital platform. At the end of 2017, the European Court of Justice decided that Uber provides transportation services and its activity is not limited to intermediation or information society services. This ruling was set in the context of a competition law dispute, but it could serve as inspiration for other matters. The Court justified the transportation service classification based on the fact that the platform created a supply system of transport services, encompassing the app, the selection of the drivers, and the regulation of the price of the service.Footnote 26

Long before the CJEU decision, the entire legal argument for reclassifying the drivers’ status was rooted in the technological identity of the company in the US and the UK. Even if control over the drivers should be the main criterion to classify them as employees, according to national legislation and precedents, the courts explored the identity of the company in several pages of their orders.Footnote 27 In its terms of service, the platform has always been presented as a software technology company headquartered in San Francisco, California. Uber has emphatically denied that it is a transportation business, “and consequently, it does not have any vehicles or employ drivers. Uber provides technologic services that allow its users to locate a participating driver ‘on demand’ through a mobile application”.Footnote 28

The company provides two types of contract in the United States and the United Kingdom: a software licence and online services agreement and/or a transportation provider service agreement, specific to collective partners or transportations companies, such as taxi-cab drivers.Footnote 29 Drivers declare that they are not in an employment relationship under these contracts. Moreover, the terms of service provided to riders clearly state that Uber has no obligation to provide its users with rides at all and that the “services constitute a technology platform that enables the user to pre-book and schedule transportation, logistic, delivery, and/or vendors services with independent party providers”.Footnote 30 In summary, the company defines its role in the terms of services as that of an intermediary.

Drivers do not perceive Uber in the same way that the company does. They claim that despite the fact that Uber legally denies that it is a transportation business, it was implicated deeply in marketing transportation services and it has referred to itself before as an “On-Demand Car Service”, it had the slogan “Everyone’s Private Drive”, and it had stated “Uber provides the best transportation system”.Footnote 31 The drivers also claim that Uber does not sell its software as typical technologic companies do, and that the company revenue mainly is due to its transportation services. The software itself does not cost anything to the users. For those reasons it would be completely illogical to deny its transportation nature.Footnote 32

The Employment Tribunal and the United States District Court in California denied Uber the status of simply being a digital platform because recognising it as such would mean reaffirming Uber as an intermediary for transportation purposes. The matter of whether Uber is a technology or transportation company, in the US, is not res judicata, but the District Court of Northern California already stated and claimed in the early stages of the litigation that Uber “is ultimately a transportation company, albeit a technologically sophisticated one”.Footnote 33 The argument was based on three elements: (1) Uber’s existence is due to transportation services; (2) it does not sell any software; (3) it sells and provides transportation. The definition as a technology company centres only on the way the platform operates – the use of the internet and applications – instead of focusing on what the company truly does, which is to enable customers to book and receive rides. The intermediary frame is too narrow to define the larger context of the business.

In the UK, the Employment Tribunal recognised Uber as a transportation company based on marketing to promote its transportation services.Footnote 34 On Uber’s website, the platform chooses its words carefully to identify the drivers as partners and emphasises that Uber just makes recommendations for drivers.Footnote 35 The same does not happen when they directly address the drivers or send internal emails. In the case of Aslam v Uber, the company clearly stated that it offered a business opportunity and not jobs in the organisation. However, in a private contact with the drivers, Uber boasted of “providing job opportunities to people who had not considered driving work and potentially generating tens of thousands of jobs in the UK”.Footnote 36

2 Driver recruitment and monitoring

The manner in which Uber recruits its drivers was used by the plaintiffs to strengthen the employment argument. The drivers became partners with Uber after completing an application process. The applicants had to upload a copy of their drivers’ licence, their vehicle registration and their insurance. They had to pass a background check. They were also required to pass a city knowledge test and attend an interview with an Uber employee. They were advised “to bring their car, dress professionally and be prepared to stay for an hour”.Footnote 37 The background check, city knowledge exam, vehicle inspection and personal interview were considered essential measures by the company because “Uber provided the best transportation service and to keep it this way, they would be taking some major steps to improve both driver and vehicle quality in the Uber system”.Footnote 38

When applicants were finally successful in their application requirements and interview, they signed a contract stating that the relationship between the transportation provider and Uber was solely that of independent contracting parties.Footnote 39 Both parties specifically agreed that there was no employment agreement or employment relationship.

Uber drivers claim that the company controls them in two different manners: (1) prescribing how they should act with clients; and (2) monitoring their activities. In all employment litigations, as in all countries where Uber operates, the drivers’ performance is a condition to keep them connected and using the platform. The company regards the quality of the service provided as indispensable to keep its clients loyal to the application. Drivers consider the performance monitoring Uber imposes as typical of an employment relationship.

Several documents attached by David O’Connor’s lawyers, for instance, showed how Uber tried to constrain its drivers, with commands or orders related to the way they should dress, the manner in which they should treat the passengers, and what they expected the atmosphere in the car to be. Uber instructs: “(1) make sure you are dressed professionally; (2) send to the client a text message when 1–2 minutes away from the pickup location (this is very important); (3) make sure the radio is off or on soft jazz or on National Public Radio; (4) make sure to open the door for your client”.Footnote 40

In the UK, the claimants Aslam and Farrar alleged that Uber exerted excessive control over the activity of its independent drivers. They claimed that Uber instructed, managed, and controlled them. Conversely, Uber stated that the company merely gives instructions to guarantee the common interest of ensuring a good ride experience for passengers. When drivers start working for Uber UK, they receive a welcome pack containing instructions of what is expected from them. As part of these instructions, Uber states that it expects high-quality service, demonstrated by the rating given by the clients, and low cancellation rates or high acceptance rates, in which the driver should accept at least 80% of trip requests to retain their account status. Drivers are forcibly logged out of the Uber app for 10 minutes when they reject three trips in a row. Uber also lists its safety and quality preferences: “polite and professional drivers; zero tolerance to any form of discrimination; avoid inappropriate topics of conversation; do not contact the rider after the trip has ended”.Footnote 41 All of these recommendations are followed by “please remember that there are some recommendations that if not followed, may constitute a breach of your partner terms or license conditions”.Footnote 42

In any jurisdiction, the company can terminate its contract with the drivers based on their low performance, which is solely evaluated by passengers, who are encouraged to write comments about how the drivers behaved, the cleanliness of the car, any unpleasant smells, the music, and the trip in general.Footnote 43 The fact that passengers constantly evaluate drivers allowed Uber to have an enormous amount of surveillance and control over the manner of its workers’ performance. Several documents reported by the judge in the O’Connor case revealed that the company regularly terminated the accounts of drivers who did not comply with Uber performance standards. An email from an Uber San Francisco manager to an employee stressed this point: “We will be deactivating Uber accounts regularly of drivers who are in the bottom 5% of all Uber drivers and not performing up to the highest standards. We believe that the removal of underperforming drivers will lead to more opportunities for our best drivers”.Footnote 44

Monitoring performance is treated as evidence of a relationship between employers and employees in the two cases analysed.Footnote 45 In the US, for instance, in a case involving FedEx, a precedent was established as a matter of law stating that monitoring job performance was essential in determining that FedEx’s drivers were employees. In Uber’s case, the drivers could be monitored many times a day.Footnote 46

Regarding working hours, Uber claims that it gives complete freedom to its drivers to choose when and how much they want to work, with the condition that they do at least one trip in every 180 days to keep the contract valid. In addition, Uber alleges that the company exerted no pressure on or control over its drivers’ working hours. There is no question of permitting or authorising breaks because drivers are free agents, according to Uber. Thus, they can take breaks whenever they feel it is necessary.Footnote 47 Despite this Uber claim, the break-time aspect was a subject of litigation because Uber established a limitation to it by contract. If the drivers were logged into the application and they refused more than three trips, they would be compulsorily logged out of the platform for a certain amount of time, regardless of whether the drivers refused the trips for a fair reason such as a problem with the car or even a time break. For this reason, drivers believed that Uber had control over their break time.

III Defining employment in the us and the uk

Defining who is an employee is not an easy task. The problem of the misclassification of workers remains a reality in different sectors. In the US, courts and government agencies at the state and federal levels use different criteria and standards to classify employees; the tests to distinguish independent contractors from employees are not homogeneous; and statutory law, in general, does not provide any clear definition of employment relations or employee.Footnote 48

Two important national labour regulations vaguely define “employee”. The National Labour Relations Act (NLRA), which regulates collective bargaining, has a definition section (§153(3)), which includes the terms person, employer and employee. The definitions are not self-explicatory; “the term ‘employer’ includes any person acting as an agent of an employer” and “the term ‘employee’ shall include any employee, and should not be limited to the employees of a particular employer”.Footnote 49 The Fair Labour Standards Act (FLSA), which regulates minimum wages and overtime, among other rights, defines employees as “any individual employed by an employer”.Footnote 50

The National Labour Relations Board (NLRB)Footnote 51 and courts seem to have case-based definitions of employment. The Supreme Court recognised that there is no uniform and easily applicable test to define who employees are. In total, courts and the NLRB created and applied four different tests to distinguish employees: the common-law agency test (focused on the employer’s right to control the tasks accomplished by the worker), which is the most common test;Footnote 52 the economic realities test (more inclusive approach and focus on the employer’s power over the economic conditions of work and not the details of how the work is performed);Footnote 53 the primary purpose test (focused on the purpose of the work); and the hybrid test, which combines the common law test with the economic realities test. The tests evolved because the common law test resulted in inconsistent rulings at several times and was considered arbitrary by some scholars.Footnote 54

The Supreme Court has set a precedent limiting the definition of the term employment and adopting the common law test for all statutes that do not explicitly provide a definition for the term and excludes the purpose test for its width.Footnote 55 In that sense, the implementation of the NLRA is conditioned to the common law test when the status of the worker is in doubt. Conversely, the FLSA specifically relates the definition of employee to economic realities and focuses on: (1) the opportunity for profit or loss; (2) investment in equipment or materials; (3) the extent to which the service rendered is an integral part of the employer’s business.Footnote 56

Factors that generally are considered among the criteria in the tests are as follows: (1) the control that the business has over the worker’s performance; and (2) the entrepreneurial potential the worker has, which means the possibility to handle different activities and clients and not perform functions that are an essential part of the company’s normal operations.Footnote 57

Contrary to several atypical work contracts such as temporary work, independent contractors do not have the right to be protected by any employment law and are therefore outside the reach of minimum wage, worker’s and unemployment compensation, occupational safety and health laws, collective bargaining laws, social security, disability and antidiscrimination laws.

To define the status of Uber’s drivers, Judge Edward Chen analysed the facts of the case in light of some conditions imposed in the Supreme Court of California precedent SG Borello & Sons, Inc v Dep’t of Industries Relations. The Supreme Court enumerated several indicia of an employment relationship, which is known as the Borello test.Footnote 58 These indicia are helpful in determining an employment status. In O’Connor v Uber Technology, it was decided that the status of the drivers was ambiguous according to the Borello test. The test did not prove to be conclusive. There was evidence that showed the drivers as independent contractors, such as the complete freedom to choose their worktime (the only condition is to be offline), their breaks, the fact that they owned their own vehicles, and the possibility of employing other drivers to drive on their behalf.Footnote 59 However, there also was evidence that showed they were employees, such as the frequent monitoring (even if the control over the drivers was controversial) and performing a regular and integral part of Uber’s business. As long as there were disputes over the material question of fact, related to whether Uber had the right to control the manner and means of its drivers, the case could not be decided by a judge, but otherwise, it should be submitted to a trial and decided by a jury. There were factors under the Borello analysis that supported both status qualifications of the drivers.

In the UK, the criteria to distinguish employees from independent contractors in British labour law comprise subordination, allocation of risk, and degree of independence.Footnote 60 However, often the definition of the term “employee” ends up with the courts considering that the legal definition is not very clear from the statutes. According to the Employment Rights Act 1996, an employee is “an individual who has entered into or works under a contract of employment”.Footnote 61

The UK courts, similarly to those in the US, apply the common law test to determine employment status, which includes aspects such as control, economic dependency, and subordination or mutuality of obligation.Footnote 62 The mutuality of obligation criteria has been the main guide used to identify employment status since the late 1970s. It considers the existence of the mutual obligation to provide work and to accept any work that is offered; the duration of the employment; the regularity of employment; the right to refuse work; and custom in the trade.Footnote 63 In the Uber case, the Employment Tribunal ruled that to have skilled drivers, Uber recruits, interviews and instructs drivers. It ruled that because drivers cannot establish the fee for the service they want to provide to the client, this makes them dependent.Footnote 64 It is true that in the terms of service, it is stated that the driver can ask for a lower fee, but not a higher one.Footnote 65 If the driver asks a lower fee, Uber would have the same amount as if the entire fee was charged. In reality, the drivers do not ask for lower fees. In addition, the Employment Tribunals considered that Uber managed and controlled the drivers by the rating system. The rating system has direct consequences for the drivers, varying from a warning email to removal from the platform. Uber additionally administered passengers’ complaints about the drivers in this monitoring role. The Tribunal decided that the contractual terms that linked Uber and its drivers did not correspond with the reality. “The notion that Uber in London is a mosaic of 30,000 small business linked by a common platform is to our minds faintly ridiculous”.Footnote 66 For Uber, each car with a driver constituted an independent business, and in this business, the driver and passenger would have a binding contract without knowing details retrospectively, including full names or phone numbers. Uber also retains all the information about the passengers to be picked up, including their credit cards. The driver has none of this information, only the first name and a photo when it is provided by the client. In addition, this same contract provides that the driver will start a ride without knowing the destination until the ride begins with a route prescribed by a stranger to the contract (UBV) and for a fee also set by a stranger to the contract. Because of these facts, the Tribunal considered that the contract between driver and passenger was a fiction with no relation to the real legal reality.Footnote 67

For all these reasons, it was ruled that the drivers provided work for Uber and were workers of ULL, the Uber branch in London. The Tribunal also held that Uber could have independent contractors, but not within the current contractual framework. Uber appealed to the Employment Appeal Tribunal against the decision in 2016, which was upheld in 2017.Footnote 68

IV Local legal strategies to address misclassification challenges

In the US, considering the financial risks of going on trial and after three years of extremely active and highly contested litigation, a settlement between the drivers of both cases and Uber was submitted to the District Court of California in March 2016. The agreement consisted of two primary parts, a monetary payment and non-monetary relief.Footnote 69

Uber proposed a nonrevisionary payment of $100,000,000 that would not be distributed equally to all class members but would notably be attributed in function of the number of miles drivers had transported Uber passengers or the “on trip” mileage. The non-monetary agreement consisted of several changes to Uber’s business practices, including more transparency in company policies with the creation of a “Comprehensive Deactivation Policy”, a manner to seek redress from deactivation (Driver Appeal Panel), and better bargaining power with Uber in case of future disputes (Driver Association)Footnote 70.

According to the proposed “Comprehensive Deactivation Policy”, Uber will no longer be able to deactivate drivers at will. The agreement states that the contract with drivers only can be terminated for sufficient cause, including low ratings. Before the termination of the contract and therefore deactivation, drivers would receive two prior written warnings with an explanation justifying the deactivation and a chance to fix any problematic issues.Footnote 71 This clause limits the possibility of Uber deactivating accounts and provides protection to the drivers to remain in the platform. Uber partially loses its power to dispose of its main workforce, operating more as a platform between passengers and drivers.

To facilitate fair contract termination, Uber proposed creating a Driver Appeal Panel, composed of highly-rated drivers. Drivers who think that their contract has been terminated unjustly can bring their reasons to the panel. If the aggrieved drivers are not satisfied with the result of the panel, Uber will pay a neutral arbitrator to decide whether there was sufficient cause for termination of the contract and deactivation of the account. Uber also will fund and facilitate the formation of a Driver Association, which is not a union, but which will have elected driver leaders. The Driver Association will take up issues regarding drivers’ concerns with Uber management. The management, in turn, has committed to meet, discuss and address the issues facing Uber drivers.Footnote 72

The practical consequence of this settlement was that the case that was scheduled for 20 June 2016 did not go on trial. Nevertheless, the settlement was rejected in June 2016 because it was considered “not fair, adequate, and reasonable”,Footnote 73 mainly because the amount of money was seen as insufficientFootnote 74 and deactivation continued to be a bone of contention, even if the company proposed clearer standards in a comprehensive written deactivation policy.Footnote 75

Uber agreed to lose its discretionary power when deactivating accounts or drivers. According to their new policy, a driver can only be deactivated for sufficient cause. The fact that a driver accepts few ride requests was not included among these sufficient causes. It would be considered excessive control under an independent contractor activity. Drivers can accept as many rides as they want. However, Uber imposed a condition considered by the judge to be inadequate to the liberty of the drivers. In cases where a driver declined a ride request, Uber could temporarily log the driver out of the application for a limited period of time. Uber suggested that this helped to maintain the quality of the service and that the drivers spontaneously had the option to turn off the application, instead of declining rides. “Where drivers do have low acceptance rates …, we will alert them to the issue. If things do not pick up, we may log them out of the app for a limited period of time”.Footnote 76

The Uber strategy was to create a similar situation to an employment relation, but without the same benefits. The company wanted to facilitate the creation of an Association of Drivers, to bargain for driver entitlements, but it does not recognise this as a union. The drivers do not have the right to minimum wage and hour claims.

V Conclusion

Globally, since Uber was launched in 2009, it has changed a few characteristics to adapt its original identity – of a technology platform – to different jurisdictions where it operates. It is true that the company has not given up its technology identity until now, and it insists on differentiating itself from taxis and from normal private transportation companies. It is also true that the company has been trying to mitigate the way it manages its drivers and its marketing. Considering the way Uber has subtly been changing, it is unlikely that it will keep its way of operating if its workers have their employment recognised. It is more likely that Uber will adapt its contracts and operations in a way that drivers will be recognised entirely as independent contractors. I make such a normative statement based on the following facts:

Uber has changed its marketing strategies drastically over time, and they now differ vastly from when they started. According to evidence in the litigation, one could find communication messages such as “On-Demand Car Service; Everyone’s Private Drive; Uber provides the best transportation system”Footnote 77 on its website in the past. Since then, Uber has changed its communication strategy and it now promotes itself through messaging “Make money when you want: Uber is an app that connects riders to drivers; Set your own schedule; Make money on your own terms”Footnote 78, thus emphasising its intermediary/platform role.

Another fact brought up repeatedly in the different cases was the potential control the company has over the drivers’ performance and working hours. As noted above, control is a test of employment in the two analysed jurisdictions. Uber is struggling to change this aspect, and according to witnesses in the legal cases, the company no longer requires that the drivers play a certain kind of music during the trip, open the door for the passengers, or offer water or treats.Footnote 79 Another noticeable change regarding control is the deactivation policies. Uber cannot terminate the contract with the driver at mere will. The fact that the company could do so before characterised its role as a manager and not as a mere intermediary between drivers and passengers. The new deactivation policies are much clearer, and drivers can only be deactivated for sufficient cause, including low ratings. The evaluation of the rating depends on the average in the city where the driver works, but ratings under four stars often are considered low. This aspect is still a sensitive issue under the control aspect, because the fact that Uber is concerned with the drivers’ performance implies some form of control by the company.

Even if Uber adapts its legal contracts to the best possible scenario with its drivers, to maintain their status of independent contractors, the fact is that the company will have thousands of suppliers with no status of employment, even though many of them work exclusively for the platform and depend on the platform for their livelihood. These issues raise concerns as to the future of employment on sharing economy platforms and the role of labour laws in new technological arrangements.

The problem of misclassification of workers is not new, especially in the transportation sector.Footnote 80 The innovation resides in the use of technology to connect consumers to workers and how it has helped transportation network companies to decentralise their workforce. In this case, technology enables these businesses to completely outsource their services to independent contractors around the world. This outsourcing pattern has scaled in an unprecedented manner the number of workers outside the protection of labour laws, because independent contractors are commonly under special contractual regimes and are not entitled to minimum wage, compensatory time payment or unemployment benefits, for example. The expansion of sharing economy platforms raises the question of whether traditional labour law is adapted to protect workers in the new technological landscape and the evolving relationship between businesses and their workforce.

Footnotes

*

PhD Researcher at the Université Libre de Bruxelles, Perelman Centre. Her research is funded by CAPES Foundation. Email: amacedoc@ulb.ac.be. The author would like to thank Arnaud Van Waeyenberge (HEC Paris), David Restrepo (HEC Paris), Benoit Frydman (ULB), Isabelle Rorive (ULB) and Joana Vasconcelos (UCP Lisboa) for their insightful comments and suggestions.

1 See for different regulatory problems faced by online platforms/sharing economies, Lobel, O, “The Law of the Platform” (2016) 101 Minnesota Law Review 88.

2 Wexler, M, “Reconfiguring the Sociology of the Crowd: Exploring Crowsourcing” (2010) 31 International Journal of Sociology and Social Policy 6; Felstiner, A, “Working the Crowd: Employment and Labor Law in the Crowdsourcing Industry” (2011) 32 Berkeley Journal of Employment & Labor Law 143; A Kittur, “The Future of Crowd Work” Proceedings of the 2013 Conference on Computer Supported Copperative Work, available at <www.lri.fr/~mbl/ENS/CSCW/2012/papers/Kittur-CSCW13.pdf >.

3 Agrawal, A and others, “Digitization and the Contract Labor Market: A Research Agendatle” in A Goldfarb, SM Greenstein and CE Tucker (eds), Economic Analysis of the Digital Economy (University of Chicago Press 2015); Abramovsky, L and Griffith, R, “Outsourcing and Offshoring of Business Services: How Important Is ICT?” (2006) 4 Journal of the European Economic Association 594.

4 WSIS Declaration of Principles – Building the Information Society: a Global challenge in the new millennium – Document WSIS-03/Geneva/Doc/4-E/2003, para 47, available at <www.itu.int/net/wsis/docs/geneva/official/dop.html >.

5 European Commission (2016) The use of collaborative platforms – Flash Eurobarometer 438, in <ec.europa.eu/COMMFrontOffice/PublicOpinion/index.cfm/ResultDoc/download/DocumentKy/72885 >.

6 De Stephano, V, “The Rise of the Just-in-Time Workforce: On-Demand Work, Crowdwork, and Labor Protection in the Gig-Economy” (2016) 37 Comparative Labor Law & Policy Journal 471; Aloisi, A, “Commoditized Workers: Case Study Research on Labor Law Issues Arising from a Set of on-Demand/Gig Economy Platforms” (2016) 37 Comparative Labor Law & Policy Journal 653.

7 UK Government, Good Work: A Response to the Taylor Review of Modern Working Practices, February 2018, available at <assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/679767/180206_BEIS_Good_Work_Report__Accessible_A4_pdf >.

8 Rogers, B, “The Social Costs of Uber” (2015) 82 University of Chicago Law Review Dialogue 85; Milles, S, “First Principles for Regulating the Sharing Economy” (2016) 53 Harvard Journal on Legislation 147, 165; Ranchordas, S, “Does Sharing Mean Caring? Regulating Innovation in the Sharing Economy” (2015) 16 Minnesota Journal of Law 413, 433; Cohen, M and Sundararajan, A, “Self-Regulation and Innovation in the Peer-to-Peer Sharing Economy” (2015) 82 University of Chicago Law Review Dialogue 116, 116; Katz, V, “Regulating the Sharing Economy” (2015) 30 Berkeley Technology Law Journal 1067, 1070.

9 O’Connor v Uber Technology Inc, 82 F Supp 3d 1133, 1135-38 (ND Cal 2015) (hereinafter, O’Connor v Uber); Yucesoy et al v Uber Technology Inc et al, Civil Action No 1:14-cv-13938-IT, US District Court for the District of Massachusetts (hereinafter Yucesoy v Uber); Cotter v Lyft Inc, 60 F Supp 3d 1067, 1070 (ND Cal 2015).

10 Cotter v Lyft Inc, supra, note 9.

11 The TNC Lyft was also sued in a class action for misclassification of the drivers as independent contractors: see Cotter v Lyft, supra, note 9. I chose Uber because of its global dimensions.

12 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, p 8.

13 In 2015, the District Court in Massachusetts transferred the Yucesoy case to California, to be decided with the O’Connor case, because the plaintiffs had agreed by contract that the state and federal courts in San Francisco, California, enjoy exclusive jurisdiction over all claims arising out of or in connection with the licensing agreement or plaintiff’s use of Uber’s technology: Yucesoy v Uber, supra, note 9, p 14.

14 Yucesoy v Uber, supra, note 9, Memorandum of Law in Support of Defendants’ Motion to Dismiss or in the Alternative to Transfer Venue, p. 4.

15 California Labor Code § 2802 states that “an employer shall indemnify his or her employee for all necessary expenditures or losses incurred by the employee in direct consequence of the discharge of his or her duties, or of his or her obedience to the directions of the employer, even though unlawful, unless the employee, at the time of obeying the directions, believed them to be unlawful”.

16 Amount estimated by the plaintiffs: O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, p 23.

17 Aslam v Uber London Case 2202551/2015, Employment Tribunal, London, pp 3–7 (hereinafter, Aslam v Uber London).

18 33a Vara do Trabalho de Belo Horizonte, Rodrigo Leonardo Silva Ferreira v Uber do Brasil Tecnologia LTDA, Processo n 0011369-34.2016.5.03.0112.

19 Conseil de Prud’Hommes de Paris, Florian Menard v SAS Uber France, Uber BV, n F 16/11460.

20 See <www.uber.com/legal/terms/gb/ >, “Terms and Conditions”, under “Payment”.

21 ibid.

22 ibid, “Terms of Use”, under “Acceptance of Booking as Agent of the Transportation Provider”.

23 See <www.uber.com/en-GB/drive/resources/get-started/ >. These requirements can vary in different jurisdictions, although they are very similar.

24 These requirements were found in the judicial decisions: O’Connor v Uber Technology Inc, supra, note 9, p 3 and Aslam v Uber London, supra, note 17, pp 3–16.

25 Cases around the world about the company’s technology identity were contested for different regulatory purposes: unfair competition and lack of licences: Spain (Asociacion Madrileña del Taxi v Uber Technologies INC, Juzgado de lo Mercantil No 2, Madrid, Recurso 707/2014, 9/12/2014 and Case C-434/15, Asociación Profesional Elite Taxi v Uber System Spain Case); Belgium (AS Radio Taxi Bruxellois v SPRL Uber Belgium, Tribunal de Commerce de Bruxelles, Chambre des Actions en Cessation, 31/03/14); France (Uber France SAS, Conseil Constitutionnel, Décision n° 2015-484 QPC, 22/09/2015 & Uber France, Cour de Cassation, Arrêt n 699, 23/06/2015); Italy (Soc Coop. Taxiblu e altri v Uber international holding BV a altri, Tribunale di Milano, ordinanza, 9/07/2015); Germany (Taxi Deutschland Servicegesellschaft v Uber BV, Landgericht Frankfurt am Main, Zivilkammer, Aktenzeichen 2-03 O 329/14, 25/08/2014); Canada (City of Toronto v Uber Canada Inc, Uber BV and Rasier Operations BV, Ontario Superior Court of Justice, 2014); Brazil (Simtetaxi, sp Sindicato dos Motoristas e Trabalhadores nas Empresas de Taxi no Estado de São Paulo v Uber do Brasil Tecnologia Ltda, Tribunal de Justiça do Estado de São Paulo, 12a Vara Cível, Processo digital 1040391-49.2015.8.26.0100); Taxation: Australia (Uber BV v Commissioner of Taxation, Federal Court of Australia, NSD 904 of 2015, 17/02/2017); discriminatory treatment: USA (National Federation of the Blind of California v Uber Technologies, Inc, United States District Court, California, 2015; Thomas Liu v Uber, Equal Employment Opportunity Commission, 2014); misclassification of workers: Brazil (Rodrigo Leonardo Silva Ferreira v Uber do Brasil Tecnologia LTDA, 33a Vara do Trabalho de Belo Horizonte, 13 de Fevereiro de 2017. Processo n. 0011369-34.2016.5.03.0112); UK (Aslam v Uber London), USA (O’Connor v Uber and Yucesoy et al v Uber Technology Inc et al, Civil Action No 1:14-cv-13938-IT, US District Court for the District of Massachusetts).

26 C-434/15 Asociación Profesional Elite Taxi ECLI:EU:C:2017:981.

27 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, pp 11–15; Aslam v Uber London, supra, note 17, p 27.

28 Yucesoy v Uber, supra, note 9, Memorandum of Law in Support of Defendants’ Motion to Dismiss or in the Alternative to Transfer Venue, p 1.

29 See <www.uber.com/fr/legal/terms/us/;https://www.uber.com/fr/legal/terms/gb/ >; <www.uber.com/fr/legal/terms/br/ >.

30 See <www.uber.com/fr/legal/terms/gb/ >.

31 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, pp 4–5.

32 ibid, p 9.

33 ibid, p 11.

34 Aslam v Uber London, supra, note 17, p 27.

35 See <www.uber.com/en-GB/ >.

36 Aslam v Uber London, supra, note 17, p 17.

37 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, p 3.

38 In an internal document called “SF Hiring Freeze & Quality Push”, Uber stresses that these screening measures are important because Uber provides the best transportation service “… and to keep it this way, we will be taking some major steps to improve both driver and vehicle quality on the Uber system”: O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, p 12.

39 Service Agreement at 7, Other Provisions, see <www.uber.com/legal/terms/us/ >.

40 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, p 21.

41 Aslam v Uber London, supra, note 17, p 14.

42 ibid, p 15.

43 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment; Aslam v Uber London, supra, note 9.

44 Email from UBER SF Community Manager instructing a fellow Uber employer to “get rid of this guy. We need to make serious cuts of guys below”: O’Connor v Uber, Order Denying Defendants Motion for Summary Judgment, p 11.

45 ibid; Aslam v Uber London, supra, note 17.

46 FedEx controlled the appearance of its drivers and their vehicles, their worktime, and how and when they should deliver the packages: Alexander v FedEx Ground Package Sys Inc, Case No 05-cv-00038-EMC (ND Cal 2016), pp 16–20.

47 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, pp 18–19.

48 Rubistein, M, “Our Nation’s Forgotten Workers: The Unprotected Volunteers” (2006) 9 Journal of Labour and Employment Law 147. See the tensions about the classification of workforce in the “on-demand” economy: Means, B and Seiner, JA, “Navigating the Uber Economy” (2016) 49 University of California Davis Law Review 1511; Cunningham-Parmeter, K, “From Amazon to Uber: Defining Employment in the Modern Economy” (2016) 96 Boston University Law Review 1673.; Cherry, MA, “Beyond Misclassification: The Digital Transformation of Work” (2016) 37 Comparative Labor Law & Policy Journal 577.; Rogers, B, “Employment Rights in the Platform Economy: Getting Back to Basics” (2016) 10 Harvard Law & Policy Review 479.

49 NLRA, s 2 §152: “The term ‘employee’ shall include any employee, and shall not be limited to the employees of a particular employer, unless this subchapter explicitly states otherwise”.

50 FLSA, s 203 § 203. Other Federal regulations define employees in the same way: the Employment Retirement Income Security Act (ERISA) of 1974; the Family and Medical Leave Act (FMLA) of 1993; the Civil Rights Act of 1964, title VII, the Age Discrimination in Employment Act (ADEA) of 1967; American with Disabilities Act, 1990.

51 The NLRB is a US governmental agency responsible for investigating and prosecuting unfair labour practices mainly in the cases that have interstate commerce. For more information see <www.nlrb.gov >.

52 Rubistein, supra, note 48, 161.

53 Zatz, N, “Working Beyond the Reach or Grasp of Employment Law” in A Bernhardt and others (eds), The Gloves-Off Economy: Workplace Standards at the Bottom of America’s Labor Market (Cornell University Press 2008) p 35.

54 Rubistein, supra, note 48.

55 Nationwide Mutual Insurance Co v Darden 503 US 318 (1992).

56 Davidov, G, “The Three Axes of Employment Relationships: A Characterization of Workers in Need of Protection” (2002) 52 The University of Toronto Law Journal 357, 368–369.

57 Jost, M, “Independent Contractors, Employees, and Entrepreneurialism Under National Labor Relations Act: A Worker-by-Worker Approach” (2011) 68 Washington & Lee Law Review 311, 334–335.

58 SG Borello & Sons, Inc v Department of Industries Relations, Case No S003956 (SC Cal 1989).

59 O’Connor v Uber, supra, note 9, Order Denying Plaintiffs Mot for Preliminary, p 17.

60 Byrne Brothers (Formwork) Ltd v Baird [2002] IRLR 96; Bates van Winkelhof v Clyde & Co LLP [2014] UKSC 32; Cotswold Developments Construction Ltd v Williams [2006] IRLR 181.

61 Employment Rights Act 1996, s 230(1).

62 Deakin, S, “Interpreting Employment Contracts: Judges, Employers, Workers” (2004) 20 International Journal of Comparative Labour Law and Industrial Relations 201.

63 ibid, p 202.

64 Aslam v Uber London, supra, note 17, p 29.

65 See <www.uber.com/fr/legal/terms/gb/ >.

66 Aslam v Uber London, supra, note 17, p 28.

67 ibid, p 40.

68 Appeal No UKEAT/0056/17/DA.

69 The entire terms of the agreement can be found at O’Connor v Uber, supra, note 9, Notice of Motion and Motion For Preliminary Approval and Memorandum of Points and Authorities In Support Thereof.

70 Other factors, less pertinent to this paper’s analysis, were also considered, such as: (a) whether they drove in California or Massachusetts; (b) if they drove in California, whether they were a member of the certified class in O’Connor; (c) whether they opted out of Uber’s arbitration clause: O’Connor v Uber, supra, note 9, Notice of Motion and Motion for Preliminary Approval and Memorandum of Points and Authorities in Support Thereof, p 6.

71 Except in cases of fraud, safety, discrimination or illegal conduct: ibid, p 8.

72 ibid, p 9.

73 ibid, p 2.

74 Based on evidence, it was estimated that in mileage expenses alone, drivers were due a reimbursement at an average of $700 million. A sum of $30 million was due for reimbursement of data usage charges on their phones, $122 million on tips and $2.4 million for overtime work. In total, Uber should pay its drivers in the United States, if they were reclassified as employees, $854.5 million, instead of the $100 million offered in the settlement. In addition, in 2015, the California Labor Commission decided that a driver who had filed a complaint should earn five times more than what was offered in the settlement: O’Connor v Uber, supra, note 9, Order Denying Plaintiffs’ Motion For Preliminary Approval, p 8.

75 ibid, p 10.

76 The judge considered that the platform preserved it management power in cases of a driver’s declination of a ride. In cases of low acceptances rates, Uber still has the power to log out the driver: O’Connor v Uber, supra, note 9, pp 24–25.

77 O’Connor v Uber, supra, note 9, Order Denying Defendants Motion for Summary Judgment, pp 4–5.

78 See <www.uber.com/en-GB/drive/ >.

79 Tribunal Regional do Trabalho da 3a Região. 09a turma, Rodrigo Leonardo Silva Ferreira v Uber do Brasil Tecnologia LTDA 23 May 2017.

80 In the field of transportation particularly, many judicial disputes related to the misclassification of the drivers have been raised over the decades. In the United States, see Rutherford, W, “Fitting a Square Ped Into a Round Hole: Alexander v. Fedex Ground Packages Systems & the Sharing Economy” (2016) 11 Liberty University Law Review 33, 42.

* PhD Researcher at the Université Libre de Bruxelles, Perelman Centre. Her research is funded by CAPES Foundation. Email: . The author would like to thank Arnaud Van Waeyenberge (HEC Paris), David Restrepo (HEC Paris), Benoit Frydman (ULB), Isabelle Rorive (ULB) and Joana Vasconcelos (UCP Lisboa) for their insightful comments and suggestions.