2017/10 Uber drivers found to be workers (UK)

Summary

In a much publicised case, Uber drivers have won a first instance employment tribunal finding that they are ‘workers’ and not self-employed contractors. This decision means that they are entitled to basic protections, such as the national minimum wage, paid holiday (under the Working Time Directive) and protection against detriment for ‘blowing the whistle’ on wrong doing. The decision could have substantial financial consequences for Uber, which has around 40,000 drivers in the UK but Uber has already confirmed that it will appeal the decision, so we are unlikely to have a final determination on this question for some time.

Background

Under UK law, people contracted to do work for someone else can be an employee, a worker or a self-employed contractor. ‘Worker’ is a category which includes employees but does not include the genuinely self-employed who are in business on their own account. Some legal protections (such as protection from ‘unfair dismissal’) only apply to employees. Certain other legal protections (such as the national minimum wage and rights to paid holiday) are granted to ‘workers’, even if they do not fulfil the strict criteria to be ‘employees’ but do not apply to the genuinely self-employed.

A ‘worker’ is someone who works under:

• A contract of employment, or

• Any other contract under which the individual undertakes to perform work personally for someone who is not the client or customer of any profession or business undertaking of the individual.

According to the earlier case of Byrne Brothers (Formwork) Ltd – v – Baird & Others [2002] ICR 667, the intention behind the distinction is to create an intermediate class of protected workers: those who are not employees but are also not really operating their own business. The purpose is to grant protection to those who are in a subordinate and dependent position, unlike the genuinely self-employed who are in a more ‘arm’s length’ relationship. Drawing the distinction between workers and those who are operating a profession or business undertaking on their own account involves the same considerations as when distinguishing between an employee and someone self-employed but with the “boundary pushed further in the putative worker’s favour”. So, it is relevant to consider factors like the degree of control exercised by the person claimed to be the ‘employer’, whether the engagement is exclusive, the duration of the engagement, the method of payment, what equipment is supplied by the individual, and who takes the risk of failure (the possible worker, or the possible employer).

In the case of Cotswold Developments Construction Ltd – v – Williams [2006] IRLR 181 the judge said that “a focus on whether the purported worker actively markets his services as an independent person to the world in general (a person who will thus have a client or customer) on the one hand, or whether he is recruited by the principal to work for that principal as an integral part of the principal’s operations, will in most cases demonstrate on which side of the line a given person falls.”

In Autoclenz Ltd – v – Belcher [2011] ICR 1157, the Supreme Court held that individuals could be workers, even if their contractual terms said expressly that they were not. The question to be determined was: what was the true agreement between the parties? Written terms that did not reflect the true agreement would be substituted with what the court determined the true agreement actually was.

Facts

The claimants were drivers bringing claims against Uber for not paying the national minimum wage and for compensation for failure to provide paid holiday. They were test claimants, selected from a group of others for the purposes of this preliminary hearing to determine if they were ‘workers’ and therefore entitled to bring the claims at all.

The tribunal made a number of findings about Uber’s business model and the terms and conditions operating between Uber and the drivers and between Uber and the passengers.

Uber’s business model is dependent on recent technological advances, as it relies upon a smartphone telephone application (‘App’). Passengers download the App onto their phones and register, giving their credit card details. They can then use the App to hail a cab. Uber drivers in the vicinity get notified by the App that there is a customer (though they are not told the intended destination) and can choose to accept or decline the fare. If they accept and take the passenger, the route is planned by the phone’s satellite navigation technology. The driver indicates using the App when they have arrived at the destination. The fare is determined by Uber from the phone’s satellite tracking data (based upon distance travelled and time taken). In ‘surge’ areas when there is higher demand than supply of drivers, a multiplier is applied to fares. Fares are taken by Uber directly from the credit card. Uber pays the drivers weekly, keeping a percentage of the fee for itself as ‘a service fee’ for the use of the App.

During the tribunal hearing, Uber tried to argue that its business was a technology platform which put self-employed drivers in touch with customers rather than a business providing driving services. It claimed that it was helping the drivers grow their own business and that the passenger was the driver’s customer rather than Uber’s customer. The tribunal looked at a range of contractual documentation which Uber said supported its position and made various findings about the reality of the arrangements.

Relevant findings of fact made by the tribunal included the following:

• That Uber marketed a variety of ‘products’, including UberX, a lower cost product; UberXL for larger vehicles which had a capacity of at least six; and UberExec which is a more premium product with a higher specification and price.

• The Uber fare is a recommended fare only and drivers can agree a lower (though not a higher) fare with passengers but would still be liable to pay the service fee to Uber based on its recommended fare.

• Uber generates paperwork which has the appearance of an invoice addressed to the passenger from the driver. It only has the passenger’s first name on it and does not include any contact details. It does not fulfil the ordinary function of an invoice as it is never sent to the passenger (who receives a receipt from Uber after the fare has been taken from the credit card).

• Although Uber claimed drivers could accept tips from passengers, the tribunal saw documents which showed that Uber disapproved of drivers soliciting tips.

• If a passenger complained that they had been overcharged (for example, because the route was inefficient) the matter was considered by Uber. Uber could decide to compensate the passenger without referring the matter to the driver and might reduce the payment to the driver accordingly. However, when Mr Farrar (a former Uber driver that was selected as a test claimant for the preliminary hearing) queried this on one or two occasions, Uber compensated him for the deduction. This led the tribunal to conclude that Uber determined the question of refunds at its discretion but Uber accepted that, where there was no proper ground for holding the driver at fault, it must bear the loss.

• If Uber rides are procured by fraud (e.g. by identity theft) Uber generally accepts the loss and does not try to pass it on to the driver (unless the driver was somehow at fault).

• Uber sometimes paid drivers’ cleaning costs incurred by them when passengers soiled the car.

• The terms between Uber and the passenger (which the passenger has to accept when logging on to the App) state that Uber is not a transportation provider, that it is acting as agent for the driver and is not a party to the contract between driver and passenger.

• The terms between Uber and the driver also state that Uber is not a transportation provider and that it offers a tool to connect customers seeking driving services to drivers who can provide them. The terms provide that passengers can rate drivers through the App and that if a driver’s average rating falls below a minimum acceptable level, the driver might be warned and, if the rating does not improve, might be barred from using the App.

• The driver’s right to use the App is personal and non-transferrable.

• Drivers interested in working with Uber must sign up online, provide certain documents and undergo a form of induction. A 2015 email from an Uber email address invited applicants to ‘book an interview slot now’. The tribunal found that there was an interview, albeit not a searching one and that drivers unable to speak English would be excluded.

• Uber restricts what make, model and age of car it will accept. It prefers them to be black or silver. The driver is responsible for the car and the costs of running it and must fund their own private hire licence.

• Uber reviews the driver rating given by passengers in their feedback and discourages drivers from cancelling rides (after acceptance) or rejecting ride requests whilst on duty. An Uber document stated that drivers must accept at least 80% of trip requests to retain their account status. Drivers who decline three trips in a row are liable to be forcibly logged off the App by Uber for ten minutes, after having received a warning. A similar system of warnings culminating in a ten minute log-off penalty applied for cancellations by drivers.

• Drivers were encouraged to be polite, professional and avoid inappropriate topics of conversation. Uber had a ‘zero tolerance’ approach to discrimination and harassment and prohibited drivers from contacting passengers after the end of the trip.

• The Uber App determined the route. If the driver took a different route and an issue arises as to whether it was the most efficient route, Uber starts from the position that it is for the driver to justify any departure from the route determined by the App.

• Drivers whose passenger ratings fall below a certain level are given a graduated series of ‘interventions’ to help them improve. If they do not improve sufficiently, the drivers are removed from the platform and their accounts deactivated. For more serious driver conduct, Uber operates a ‘driver offence process’ with ‘warnings’ being sent before deactivation.

• Uber sent numerous messages to drivers (collectively and individually) which were presented as ‘recommendations’, ‘advice’, ‘tips’ and ‘feedback’ seeking to modify their behaviour to improve the ‘rider experience’.

• In various communications, Uber referred to ‘our drivers’, ‘Uber drivers’, that Uber had provided ‘job opportunities’ and that Uber paid the drivers ‘commission’.

• Drivers can work for other organisations, including direct competitors.

• Drivers treat themselves as self-employed for tax purposes.

The claimants argued that Uber instructed, managed and controlled the drivers. Uber denied this and claimed that, to the extent that they seemed to be directing drivers’ behaviour, it was to ensure a ‘satisfactory rider experience’, which was in the common interest of Uber and the driver or it was “to preserve the integrity of the platform.” Uber argued that the drivers were never under any obligation to switch on the App and that this was incompatible with any contract to provide a service to Uber.

Judgment

The tribunal accepted that there was no obligation to switch on the App and that whilst the App was switched off, there was no contractual obligation to provide driving services. However, the tribunal held that when a driver had the App switched on, was within the territory within which they were authorised to work and were able and willing to accept assignments then they were Uber ‘workers’ for as long as those conditions were satisfied.

In reaching their decision, the tribunal was influenced by what it saw as ‘twisted language’ and ‘fictions’ in Uber’s contractual documentation (e.g. the passenger ‘invoice’ which was not an invoice and never sent to the passenger). The tribunal essentially thought that Uber’s characterisation of the contractual relationship could not be trusted. The tribunal also believed that, despite what Uber claimed, the reality of the situation was that Uber ran a transportation service and used drivers to do it rather than simply being a technology platform enabling the drivers to run their own businesses. If Uber was not in the business of providing transportation services, how could it be marketing a ‘product range’ of driving services? No individual driver had a ‘product range’. It also didn’t make sense to talk about Uber helping drivers ‘grow’ their businesses or providing them with ‘leads’. No driver could ‘grow’ their business through Uber other than by driving more hours. Passengers are not ‘leads’ because drivers cannot negotiate terms with them. Drivers are offered and accept the trips purely on Uber’s terms.

Uber’s written terms stated that the driver enters into a contract with the passenger. However, the tribunal did not find that credible, pointing out that this rested upon an assumption that the driver enters into a binding agreement with a person whose identity he or she will never know, to undertake a journey to a destination the driver won’t know until the journey begins, by a route prescribed by a stranger to the contract, from which it is difficult to depart, for a fee which is set by a stranger to the contract and is not known by the passenger (who only knows the total sum they pay).

The tribunal found that the supposed driver/passenger contract was a pure fiction and that the only sensible interpretation was that Uber provided a transportation service and that the drivers worked for Uber. Finally, the tribunal found that the drivers were Uber’s ‘workers’. It was not in dispute that they undertook to provide their work personally. If there was no contract between driver and passenger there must be a contractual link between Uber and driver. The ‘essential bargain’ between driver and Uber is that the driver makes him or herself available to carry (and does carry) Uber passengers to their destinations. It is not a contract under which Uber is a client or customer of a business carried on by the driver. The agreement between Uber and driver is one of ‘dependent work relationship’, not a contract at arm’s length between two business undertakings. And, as in the case of Autoclenz, Uber cannot rely upon its ‘carefully crafted’ documentation because ‘it bears no relation to reality’.

The tribunal was careful to point out that it would have been possible for Uber to devise a business model that did not result in a worker relationship between itself and the drivers, it was just that the model they chose failed to achieve this.

Commentary

It is hard to disagree with the tribunal’s assessment that the Uber documentation did not seem to accurately reflect the relationship between the various parties; having said that, the decision does throw up some difficulties. Firstly, there is the question of ‘mutual obligation’. Someone is only a worker if there is a contractual relationship between principal and worker in which there is mutuality of obligation. In an employment relationship this means an obligation on the worker to be available for work (and to do it, if required) and on the employer to pay them. In this case, the driver was never under an obligation to turn on the App at all. The tribunal’s finding that the driver was a ‘worker’ when the App was switched on does not seem wholly satisfactory and the tribunal does not seem to have given much thought to mutuality of obligation. The tribunal’s decision was focussed very much on the degree of control being exercised by Uber over the drivers.

A second issue arises because of the status of London black cabs (hackney carriages). Black cab drivers are self-employed. They must have done ‘the knowledge’ (i.e. have passed a test on London roads) before they can be licensed. Black cabs are allowed to cruise looking for passengers who can flag them down, unlike other cabs which must be booked. There is an App which people can use to catch black cabs which does not seem hugely different from the Uber App but there is no suggestion that black cab drivers are ‘workers’.

New technology and the growing ‘gig’ economy have thrown up all sorts of new working patterns which are difficult to analyse in terms of traditional working relationships (as this case highlights). These new types of employment relationship are controversial, with some commentators viewing them as eroding employment protections. For that reason, this decision was being watched very closely. It was decided very much on its own facts, however, and it cannot be taken as a definitive analysis of working relationships in ‘the gig economy’.

The decision is only a first instance decision and not binding on other tribunals (although it is persuasive). It is being appealed by Uber.

In a similar decision, Dewhurst – v – CitySprint UK Ltd, a different employment tribunal has also recently ruled that a bicycle courier was a worker of the courier firm, entitling her to the same protections.

Comments from other jurisdictions

Comment from an EU/academic perspective (Bartłomiej Bednarowicz, PhD Researcher at the University of Antwerp): From the perspective of EU law, the judgment of the tribunal should be particularly welcome, as it bridges another gap in recognising workers’ rights in the gig economy. The English tribunal applied a functional definition of an employer and focused its assessment on factors such as the degree of management and control exercised by Uber over drivers – criteria previously affirmed in ECJ case law.

Yet up to now, under under EU law, it has been unclear whether individuals working in the gig economy, such as Uber drivers, could be regarded as workers. As EU law is silent on this, the general concept of a worker and its usual application has prevailed. Well-established ECJ case law has it that a worker is someone who, for a certain period of time, performs services for and under the direction of another, in return for pay. Therefore, if an employment relationship lacks subordination, an individual is more likely to be regarded as self-employed, which fundamental legal implications in terms of which EU legal provisions apply. Workers enjoy free movement, whereas the self-employed enjoy the freedom to provide services.

In the landmark case of Kunsten Informatie en Media (C-413/13 FNV), the ECJ looked at the issue of false self-employment with regard to the application of EU competition law to self-employed individuals. The Court ruled that even if an individual is regarded by national law as self-employed, if the way the person works becomes dependent on someone else, that status could be lost. This is particularly likely if the service provider does not bear any financial or commercial risk from the activity and operates as an auxiliary within the other person’s business. Moreover, the status of ‘worker’ is unaffected by the fact that service providers have more independence and flexibility to choose the time, place and content of their work than employees – just as was the case with the English Uber drivers.

What both judgments have in common is that they look at the employer/employee relationship in terms of function. The tribunal focused on the degree of control and overall management of Uber drivers and the ECJ looked at the commercial and financial risks that are either shared or not by the service provider and the other person. And as the question of employment versus self-employment is to the fore at the moment, more case law can be expected.

Denmark (Christian K. Clasen, Norrbom Vinding): Since its entry into the Danish market, Uber has been at the centre of great controversy in the media as well as politically. Uber faced immediate opposition from the taxi industry when it launched in 2014 and also from public authorities. The taxi industry used the media and arranged public gatherings, whilst the Danish Transport Authority reported Uber to the police. The police soon began to fine Uber drivers for performing activities in conflict with the Danish Taxi Act. The drivers did not accept the fines and so test cases were brought before the Copenhagen City Court. The main question before the Court was whether or not the Uber drivers were performing cab-driving activities. If so, they would be required to have taxi drivers’ licences – which they had not applied for. The Court ruled against the Uber drivers, arguing that they were performing cab-driving activities within the meaning of the Danish Taxi Act and consequently, the fines imposed on them were upheld. The decision was appealed to the Danish Eastern High Court, which upheld the decision by the Copenhagen City Court.

Whether Uber was an accessory to the Uber drivers’ illegal actions was a separate question. However, the Danish Prosecution Service has not yet commenced proceedings against Uber because of the political situation. It is expected that the Prosecution Service will not pursue the matter until it is clear whether certain new legislation will be passed by the Danish Parliament. The proposed legislation concerns the fact that in February this year, the Danish Minister of Transport announced that new legislation would relax the requirements for obtaining a taxi driver’s licence, so that new parties could enter the market. Under existing law, there are many strict requirements for obtaining a licence, involving both the vehicles and insurance.

It is expected that Uber drivers operating within Uber’s current business model will still not be able to perform services under the new legislation. This has caused great frustration among Uber’s supporters. It will be interesting to see how the situation develops, but in any event, the Uber case in Denmark is an interesting example of how technology can challenge existing law.

With regard to whether Uber drivers are employees or self-employed, there has been no similar question in the Danish courts as yet, but as the criminal case described above resulted in the conclusion that Uber driving was to be considered as cab driving subject to the Danish Taxi Act, it seems likely that Uber drivers will be treated as cab owners.

There is an important distinction in the applicable collective agreements between taxi cab owners and taxi cab drivers. Taxi cab owners are employers or self-employed contractors. By contrast, taxi cab drivers are employees within the meaning of the collective agreements. However, the position is complicated by the fact that taxi cab owners will normally also be drivers of the taxi and share it with other taxi drivers. The recent Parliamentary compromise that is leading to new legislation refers to independent cab owners as being subject to the law. Even though this compromise does not explicitly refer to Uber, there seems little doubt that this is a reference to Uber drivers, since there is a new statutory requirement that independent cab owners must provide their services through a system facilitator. Therefore, within the meaning of the existing collective agreements, Uber drivers are likely to be considered as independent cab drivers, i.e. self-employed contractors. Time will tell if the new legislation will cause major changes to the collective agreements.

It should be noted that during the current legal vacuum, Uber is still in operation and, to our knowledge, Uber’s services are widely used.

Bulgaria (Rusalena P. Angelova, Djingov, Gouginski, Kyutchukov & Velichkov): In contrast to UK law, Bulgarian law does not distinquish between workers and employees. In Bulgaria, employees/workers work under employment (labour law) contracts and contractors (including self-employed contractors) provide services under civil (consultancy/service) contracts. Further, Bulgarian law does not specify a category of individuals or a range of services, the performance of which requires an employment contract to be formed. However, the main principle set out in the Bulgarian Labour Code is that relationships connected to the provision of a labour force must be formalised under employment contracts.

In order to work out whether a given relationship should be regulated under an employment contract or a civil one, the whole of the factual background needs to be considered, including the nature of the services to be performed. The major differences between an employment and a service agreement are as follows:

• Subject matter: the subject matter of a civil law contract is the achievement of a particular result, whilst that of an employment contract is the labour process itself, requiring continuous and repeated performance of the duties.

• Relationship between the parties: a service provider under a civil contract is independent of the person assigning the work and the performance of the work is not related to working time. The individual organises and conducts the work independently. By contrast, an employee under an employment contract is dependent on the employer with regard to the organisation of the working process and the way he or she performs the work.

As a result of a decision of the Bulgarian Supreme Administrative Court to the effect that it had breached competition law, Uber has stopped operating in Bulgaria. However, with respect to taxi drivers, Bulgarian law expressly stipulates that taxi services may be provided by individuals on behalf of a taxi company, but for the account of the individuals as independent contractors.

In my view, if a similar case were brought before the Bulgarian courts, they would assess how much control the company exerts over the individuals and whether the relationship between them was aimed at the continuous provision of labour or the completion of particular tasks. Given the factual background, it is more likely that the Bulgarian courts would not find Uber drivers to be workers.

I adhere to the view that if Uber holds itself out as nothing more than a neutral technological platform, designed simply to enable drivers to transport passengers, Uber drivers are not workers, but rather self-employed contractors.

Belgium (Karel Devloo, Van Olmen & Wynant): Contrary to the approach opted for in the UK, Belgian labour law only identifies two types of individuals contracted to do work for pay: ‘employees’ and ‘self-employed contractors’. However, the main criteria for assessing a person’s status are comparable to those applied under UK law, and all relate to the existence or absence of a relationship of subordination between the contracting parties.

The intention of the parties, as evidenced by the contract, constitutes the essential factor in determining ‘employment status’. But it can be rebutted if the facts do not match the way the parties have chosen to classify it, for example if the individual is not actually free to organise his or her own work or working time or if one party can exercise hierarchical control over the other.

However, there has been much debate about whether to alter the law, as the gig economy begins to move the goal posts. At present, the focus of the law is on the creation of economic and social security for people performing limited services for themselves (i.e. up to € 5,100 of earnings per year).

Although the current criteria may not be entirely up to the job of working out which category a person should be assigned to, no one has yet litigated on employment status before a Belgian labour court. This may be because the legal and factual context in Belgium is different from countries such as the UK. First of all, Uber only operates in a few places in Belgium and there are not many drivers involved. In addition, many services offered by the gig economy are provided by people who are not wholly dependent on that work (e.g. students) and who may value the flexibility more than a fixed and guaranteed income. As their earnings were, until recently, not taxed or subject to social security contributions, there was less reason for them to complain about their employment status. In addition, social services are easily accessible in Belgium, including for those who do not have a steady job. Lastly, Belgian labour law does not have class action procedures or preliminary hearings with test claimants, which reduces the potential impact of a ‘reclassification judgment’. That being said, foreign case law reminds us that even in Belgium, the gradual move towards labour flexibility is likely to challenge the classic divide between employees and self-employed contractors.

Subject: Miscellaneous; employment status

Parties: Aslam and Farrar and others – v –Uber B.V. and others

Court: Employment Tribunal

Date: 28 October 2016

Case number: 2202550/2015

Internet publication: https://www.judiciary.gov.uk/wp-content/uploads/2016/10/aslam-and-farrar-v-uber-reasons-20161028.pdf