Last week I watched Uber try to overturn an Employment Tribunal’s judgment that its tens of thousands of drivers were entitled to national minimum wage and basic employment rights. Alongside other interested observers of the ride-hailing app’s controversial legal battle, I live tweeted the morning’s arguments as they unfolded.


It was a difficult week in the Uber-land. The tech giant had just had its licence revoked by Transport for London and was also facing fresh claims based of sex discrimination in the UK, (adding to a string of sexual harassment allegations at the company’s Silicon Valley headquarters).

On its Employment Tribunal appeal, the company had some work to do. The judgment against it runs to some 40 pages and, to a large extent, is based on very specific factual findings about the relationship between the company and its drivers. The Tribunal was also highly critical of the“fictions” and “twisted language” found in the company’s small print.

Opening argument

Dinah Rose QC opened the case for Uber by casting aside TfL’s decision and the heated national debate about the gig economy as irrelevant sideshows. The case should not be about the development or expansion of the law to cover any existing gaps: that was Parliament’s role. She argued that Uber shouldn’t be lumped together with other potentially exploitative internet platforms that operate in very different ways.

The legal issues

The key issue, as with many recent ‘gig economy’ cases, was whether or not the drivers were ‘workers’ for the purposes of employment legislation (and thus entitled to a raft of basic rights).[1]

There are 2 sub-issues to consider:

  1. whether the claimed workers are personally obliged to do the work;
  2. whether, the claimed workers are operating as independent businesses.

No contract for services

Uber’s main point in the morning (under the first sub-issue) was that the drivers were under no personal obligation to work. Indeed, there was no contractual obligation for them to perform any work. Uber is simply a “powerful piece of technology putting riders in touch with drivers.” Drivers (and riders) are not obliged to use the app at all. Nobody is charged for signing up and not using the app.

This was essentially a re-run of an argument that failed in the Employment Tribunal.

The agency argument

The second, perhaps more interesting argument was that rather like a traditional mini-cab company, Uber acts as an agent for its drivers who are all self-employed businesses. The only, (quite radical) difference between Uber and a traditional mini-cab firm is scale, something that the Tribunal was wrong to baulk at.

Simply put, the “agency argument” is that Uber accepts the booking on the driver’s behalf and facilitates the introduction of rider and driver. There is a contract between Uber and the driver, but only for the services that Uber provides to the driver, for which the driver pays a fee or commission.  Uber works for the drivers, not the other way around.

The Tribunal misunderstood the effect of agency law when it disregarded the written contract between Uber and the drivers.

Uber’s appeal placed heavy reliance on decisions from the field of tax (specifically VAT) law, citing numerous decisions which in summary show that the Tax Tribunal has long accepted that mini-cab firms operate as agents on behalf of their drivers.

Uber also cited the case of Revenue and Customs Commissioners v Secret Hotels2 Ltd [2014] UKSC 16, in which an online company marketing hotel rooms was found to be an agent (who managed hotel bookings) rather than a principal (who bought hotel rooms and sold them on for profit) and was therefore exempt from paying VAT in the UK.

The Supreme Court in Secret Hotels2 went from examining the company’s website terms and conditions to assessing the commercial reality. It found that the website was in fact an agent despite various cancellation charges which HMRC argued were inconsistent with an agency relationship (at § 43 to 44). Such measures, the Court found, were designed to protect the company’s goodwill.

By analogy, Uber claimed that its driver rating system, rather than being a form of performance management or a disciplinary procedure, was intended to protect the goodwill and reputation of its brand in the same way.

Regulatory requirements

A further issue is whether or not a Tribunal assessing whether a person is a ‘worker’ should disregard factors that are due to regulatory requirements on the main business. For example, in the context of private hire vehicles in London, all drivers must work via a company and not on their own, that company must have a complaints procedure, a procedure for vetting driver, and so on.

According to Uber, such factors should be neutral rather than counted against the company as the Employment Tribunal did.

There is no authority on this point and it will be interesting to see which way the EAT goes. While there’s some guidance on the relevance of regulatory requirements to duties of care in negligence claims, the position in contracts of employment is less clear.

The afternoon’s arguments

Available tweets suggest that the rest of Day 1 (which I missed) saw Dinah Rose QC take aim at the Employment Tribunal’s factual findings in a determined effort to show that those findings were absurd.

First up was the Tribunal’s slightly opaque conclusion that in order for Uber’s defence to work, they’d need to show that drivers and riders formed ‘the business’.

Uber attacked the notion at the heart of the Tribunal’s reasoning that the drivers were only ‘workers’ under a personal obligation to drive for the company if they were willing to work.

Another spanner in the works was the idea that minimum wage payments and shift patterns would be impractical for Uber since they’d now have to check whether drivers felt willing to work while the app was switched on.

[Note that the Employment Tribunal in the recent Addison Lee case also found that their drivers were workers, but the Tribunal in that case was untroubled by the precise mechanics of how minimum wage, shift patterns and holiday pay would be worked out on the tarmac. It was happy to leave that issue to the parties to hash out (or, put another way, it was keen to duck the issue).]

You can also find a write-up of day 1 by Darren Newman on his blog, the Employment Solicitor blog and the FT.

Day 2

Day 2 saw Jason Galbraith-Marten QC for the drivers resisting Uber’s appeal and drawing attention to the absence of any written document in which the drivers gave their consent to Uber to act as their agent.

For those wanting to go more in-depth, the parties’ skeleton arguments were published online by Bates Wells Braithwaite here.

Next steps

The EAT’s judgment could take some months. It’s anticipated that Uber will ask for a leapfrog appeal to the Supreme Court in the event of defeat.

As some have pointed out, with the Supreme Court due to hear an appeal in a similar(ish) case concerning the rights of self-employed plumbers on 20 and 21 February 2018 (Pimlico Plumbers Ltd & Anor v Smith [2017] EWCA Civ 51), it looks like employment status in the gig economy will continue to glow brightly on the legal radar.

Whether the outcome will be welcomed by businesses, workers (or neither) remains to be seen, but so far the legal landscape has been broadly protective of basic workers rights, particularly where the drivers / plumbers / couriers are in a much weaker bargaining position than the businesses they work for.

[1] See section 230(3)(b) of the Employment Rights Act 1996 (‘ERA 1996’) plus the extended definition of a ‘worker’ in section 43K for the protection of whistleblowers; regulation 2(1) of the Working Time Regulations 1998 SI/1833 (‘WTR 1998’) and also the related definition of whether work amounts to “employment” for the purposes of section 83(2)(a) of the Equality Act 2010 (‘EA 2010’).

Posted by Ben Amunwa

Founder and editor of Ben is a commercial and public law barrister with The 36 Group. He gives expert legal advice on employment, public law and commercial disputes to a wide range of clients.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.