Vanessa Barnett, Consultant Solicitor at Keystone Law, unpicks some of the interesting nuances involved in two recent UK cases concerning the gig economy: firstly the Employment Appeal Tribunal decision involving Uber, and secondly the Central Arbitration Committee judgment in regards to Deliveroo.
If you were distracted by the more headline grabbing stories about bad behaviour and paying off hackers at Uber, you may have missed some interesting nuances afoot on the finer points of how the humans in the gig economy are being categorised and the implications of that for both the humans and those running a gig economy technology platform. In September Uber spent two days in an Employment Appeal Tribunal (‘EAT’), running an argument which is not new to any of us: in the spirit of Star Wars, let’s just call it the “This is not the taxi company you are lookingfor” argument. The original Employment Tribunal (‘ET’) case, from which the appeal was triggered, was brought by current or former Uber drivers in London who needed to be classified as ‘workers’ for various rights to benefit them.
Generally, being a worker is a better position for the human being involved, because that status gives some minimum rights (not as many as an employee, but more than being genuinely self-employed on a person’s own account). These rights include: the National Minimum Wage; protection against unlawful deductions from wages; the statutory minimum level of paid holiday; the statutory minimum length of rest breaks; to not work more than 48 hours on average per week or to opt out of this right if they choose; protection against unlawful discrimination; protection for whistleblowing; and to not be treated less favourably if working part-time. The relevant employment laws for the Uber case are:
- Section 230(3)(b) Employment Rights Act 1996 (‘ERA’) – contains a definition of a ‘worker’ (essentially someone who has entered into or works under a contract of employment or any other contract whereby the person has to do the work personally, subject to a few carve outs).
- Regulation 36(1) Working Time Regulations 1998 (‘WTR’) – contains the various rules which protect workers from working excessive hours, with this section describing how a worker is brought inside the WTRs’ benefits.
- Section 54(3) National Minimum Wage Act 1998 (‘NMWA’) – contains the rules around minimum wages for workers, again with this section describing the applicable worker.
Unsurprisingly, Uber appealed. It argued:
- The ET had erred in law in disregarding the written contractual documentation. (The argument was that there was no contract between the Uber drivers and ULL but there were written agreements between the drivers and Uber BV and passengers, which were inconsistent with the existence of any worker relationship. Uber said those agreements made clear that Uber drivers contractually provided transportation services directly to passengers; ULL it said provided its services to the drivers as their agent in finding the passengers. In finding otherwise, Uber said, the ET had disregarded the basic principles of agency law.)
- The ET erred in relying on regulatory requirements as evidence of worker status.
- It had also made a number of internally inconsistent and perverse findings of fact in concluding that the Uber drivers were required to work for Uber.
- It had further failed to take into account relevant matters relied on by Uber as inconsistent with worker status and as, on the contrary, strongly indicating that the drivers were carrying on a business undertaking on their own account.
Those arguments got short shrift from Judge Eady QC and the judgment is well worth a read. The EAT ruled that the ET was right to reject the characterisation of the relationship between drivers and Uber, specifically the role of ULL as agent, in the written contractual documentation. It’s a classic scenario of what I call ‘structure meets reality’: it doesn’t matter what the bits of paper say, or how the technology is set up, if what happens on the ground is obviously not that.
The ET had found that the reality of the situation was that the drivers were slipstreamed into the Uber business of providing transportation services, subject to arrangements and controls that pointed away from their working in business on their own account in a direct contractual relationship between the Uber driver and the passenger each timethey accepted a trip. The full judgment highlights the facts in more detail; here are some examples of facts which influenced the ET and were endorsed by the EAT: drivers are interviewed and accepted or rejected; the Uber driver’s use of the app is personal, they cannot share it or substitute another driver; Uber can accept or decline the bookings; Uber sets the price of the trip (not just the % of its commission); Uber controls (and to a degree hides from the driver) key information: passenger’s surname, contact details, and (until job accepted) intended destination; drivers can suffer financially if they deviate from the app’s navigation and a passenger complains; Uber monitors the driver’s feedback, cancellation rate and acceptance of jobs from passengers (with a benchmark of 80% acceptance to ‘maintain status’); Uber determines refunds and manages complaints from passengers; and drivers who do not accept jobs can be forcibly logged off the app.
Therefore, the EAT says, the ET was right to reject the contractual ‘fiction’ outlined by Uber – stating that the ET had not disregarded the principles of agency law but had been entitled to consider the true agreement between the parties was not one in which ULL acted as the driver’s agent. In carrying out its assessment like this, the EAT said that the ET was not obliged to disregard factors simply because they might be seen as arising from the relevant regulatory regime; that was part of the overall factual matrix the ET had to consider. In any event, in this case, the ET’s findings on control were not limited to matters arising merely as a result of regulation. In effect, according to the EAT, the control aspects of the life of an Uber driver had informed the ET’s conclusions not just on worker status but also on working time and as to the approach to be taken to their rights to minimum wage.
In contrast to this, Deliveroo has been in front of the Central Arbitration Committee (‘CAC’). CAC is an independent body who rule on whether or not to recognise a trade union. Here, a group of Deliveroo riders in a part of London, with the assistance of the Independent Workers Union of Great Britain, are seeking to have the union recognised for collective bargaining. Part of this process is a determination as to whether or not the Deliveroo riders are workers. The full judgment is worth a read. In the Uber decision a lot of weight was placed on the command and control aspects of the facts: the drivers pretty much have to do what Uber says, or they get kicked off the app. In particular, the relationship is personal: there is no right to substitution. At a first look, the ‘experience’ of a Deliveroo rider is much like that of an Uber driver: they make an application, they are interviewed, their bike has to meet certain standards, their work is controlled by the app, etc. Applying the principles and analysis of the ET/EAT in the Uber case, it would seem to point towards Deliveroo riders being workers.
But for Deliveroo we have a different structure meets reality scenario, with the CAC pulling out the following facts: “54. Deliveroo issued the New Contracts to existing Riders on 11 May 2017 with a covering letter which specifically drew attention to the substitution clause: ‘You will see that this agreement means you still have the ability to appoint another person to work on your behalf with Deliveroo at any time. A substitute working for you can log in using your phone or rider app details. But we request that you never “swap orders” with another app user as this can prevent the customer from receiving accurate GPS data to track where their order is.’
55. The covering letter also informed Riders that they could work for other companies including competitors: “That is fine with us: as an independent contractor you are free to work with whoever you choose and wear whatever kit you want to. There continues to be no requirement to wear Deliveroo branded kit while you work with us.”
Deliveroo was also able to show evidence of substitution in action. And here is the point: substitution is the death knell for being a worker. If you go right back to Section 230(3)(b) ERA, it’s about that human being having to do the work personally. If it’s not personal, then they are not a worker and all the various rights that flow from that do not arise. On top of that, the control aspects appear to be much more relaxed than under Uber: a Deliveroo rider need only accept one job every three months as a bare minimum to remain a viable rider with access to the app.
The CAC then went on to quote the Kalwak case, which was referenced by the ET in the Uber case: “95. It is important to spot the difference between form and substance – the oft quoted dicta of Elias LJ in Kalwak v Consistent Group Ltd  IRLR 560: ‘The concern to which tribunals must be alive is that armies of lawyers will simply place substitution clauses, or clauses denying any obligations to accept or provide work in employment contracts, as a matter of form, even where such terms do not begin to reflect the real relationship.’”
The CAC went on: “96. It follows that all the relevant evidence has to be examined as set out by Smith LJ, as approved and endorsed in Autoclenz: “To carry out [the exercise of discovering the actual legal obligations of the parties] the tribunal will have to examine all the relevant evidence. That will, of course, include the written term itself, read in the context of the whole agreement. It will also include evidence of how the parties conducted themselves in practice and what their expectations of each other were. Evidence of how the parties conducted themselves in practice may be so persuasive that the tribunal can draw an inference that that practice reflects the true obligations of the parties. But the mere fact that the parties conducted themselves in a particular way does not of itself mean that the conduct accurately reflects the legal rights and obligations. For example, there could well be a legal right to provide a substitute worker and the fact that that right was never exercised in practice does not mean that it was not a genuine right.”
That is the essence of it: on the structure meets reality scenario, the reality/evidence for Deliveroo riders is different. They really can just send a mate any time they like. The most interesting part of this debate now is not how to structure the platforms or what documentation to put in place. The real piece of lawyers’ work moving forward is the on the ground analysis of what’s actually happening. And when we are stress testing that for our clients, we also need to keep an eye on the politics of it all: the work of the Work and Pensions Committee (‘WPC’) and the Taylor Review of Modern Working Practices.
With an estimated 4% of British ‘employees’ in the ‘gig economy’ area, this will remain a hot topic. But with the Civil Service effectively monopolised by Brexit, we are unlikely to be able to rely on Government to come to a landing point here on a considered basis. It will likely be worker rights related cases that push this forward – and they may well be influenced by Frank Field’s comments as Chair of the WPC: “These companies parade the ‘flexibility’ their model offers to drivers but it seems the only real flexibility is enjoyed by the companies themselves.It does seem a marvellous business model if you can get away with it.” Only time will tell. But it’s certainly getting much harder to run the ‘I’m just a platform’ argument like a ‘traditional’ digital business.
This article was first published by Digital Business Lawyer
This article is for general information purposes only and does not constitute legal or professional advice. It should not be used as a substitute for legal advice relating to your particular circumstances. Please note that the law may have changed since the date of this article.