London’s minicab regulator, TfL, has revoked Uber’s licence to operate in the British capital, one of its largest world markets. Getting that licence back may require the firm to finally confront a question it cannot afford to answer: Just what is Uber?
In October 2016, two men nervously waited to hear the decision on their employment tribunal case. Yaseem Aslam and James Farrar were both licensed London minicab drivers. They enjoyed the work, but they also felt that they were being treated unfairly by their employer.
What they were asking for was, they felt, simple and reasonable. They wanted a guarantee that they would earn the UK legal minimum wage and receive the minimum amount of paid annual leave required by law. That they had found themselves in court wasn’t just because their employer had refused to grant this, but because it refused to accept that they were its employees at all.
That employer was Uber…
…or was it?
Uber weren’t the men’s employer, Uber’s lawyers told the employment tribunal. Because Uber wasn’t… well… Uber. Not in the UK at least. It was, in fact, two entirely separate companies. Uber BV (UBV), a Dutch company that offered a rideshare app the men happened to use and Uber London Limited (ULL) , a TfL-licensed Private Hire Vehicle (PHV) operator based in the British capital that the men happened to have a relationship with.
The driver’s confusion, the lawyers insisted didn’t stop there. They were happy to accept that both men had a working relationship with ULL, but that didn’t mean they worked for it. At least not in any sense that would require Uber to grant them the legal rights guaranteed by the Employment Rights Act 1996, the Working Time Regulations Act 1998 or the National Minimum Wage Act 1998.
Caddies vs pole dancers
ULL, the lawyers argued, didn’t employ any drivers. It was simply a brand umbrella under which 30,000 independently licensed minicab firms chose to operate. It was true, they admitted, that the vast majority (if not all) of that 30,000 were single-operator firms like Aslam and Farrar, but they were still independent. The tribunal was shown the contracts the men and others had signed, and drew the judges’ attention to the careful wording within them that confirmed this.
Part of the benefit for signing this contract, the lawyers explained, was that operators were granted access to UBV’s app. Not ULL’s app. On this they were clear. UBV’s app. Through that app, passengers could contact those operators directly, negotiate a ride and agree a fare.
Neither ULL or UBV were involved in the ride itself, Uber’s lawyers were keen to stress. Yes, UBV took a small commission for facilitating the connection, but that was it. If there was a contractual arrangement, then it was solely made between passenger and operator, both of whom were free to set their own price (although yes, Uber’s lawyers admitted, the UBV app did suggest a price) and route (oh and yes, again, the UBV app did suggest an optimal route).
Uber’s lawyers argued that Aslam and Farrar (and the other 30,000 Uber drivers they indirectly represented) had misunderstood the relationship they had with Uber. They drew the judges’ attention to two previous similar examples in English case law. The two men, the lawyers argued, thought they were like golf caddies – technically independent, but so closely wedded to a golf club that they were, in effect, direct employees. Uber’s lawyers stressed that this was wrong. They had a different example in mind for the relationship between Uber and their drivers, one set by Quashie v Stringfellow.
The drivers weren’t like caddies at all, the lawyers explained. They were like pole dancers. Uber – or rather ULL and UBV separately – were simply providing them access to a space in which they could perform for money.
When it came, the tribunal’s ruling was unanimous: Whether Aslam and Farrar were direct employees of Uber or not was unclear. This was good news for Uber, but what came next was anything but. The tribunal ruled that what was clear was that the two men did, in a more general sense, work for Uber. Certainly to a level that entitled them, and indeed all Uber drivers within England and Wales, to the minimum wage and annual leave.
“This is,” the tribunal judges explained in their ruling, “we think, an excellent illustration of the phenomenon of which Elias J warned in the Kalwak case, of ‘armies of lawyers’ contriving documents in their clients’ interests which simply misrepresent the true rights and obligations on both sides.”
From Uber’s perspective, the obvious potential costs of this ruling were bad enough. The devil, however, is always in the detail and the tribunal wasn’t done. Over the subsequent pages of their ruling they gently, but firmly, tore a gaping hole in Uber’s legal definition of itself.
“[Uber’s] general case and the written terms on which they rely do not correspond with the practical reality.” The ruling explained.
It continued. “The notion that Uber in London is a mosaic of 30,000 small businesses linked by a common ‘platform’ is to our minds faintly ridiculous.”
The judges still weren’t finished.
“It is, in our opinion, unreal to deny that Uber is in business as a supplier of transportation services.”
This was a potential disaster for Uber. What they had stumbled into, perhaps unwittingly, was something that as an American firm ‘by birth’ they may not even have realised they needed to avoid: the Duck Test.
The Duck Test
If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck.
This maxim is known as the ‘Duck Test’. Unfortunately for Uber it is also a principle that, in part thanks to rulings by Baroness Hale (Bates van Winkelhof v Clyde & Co LLP) and Elias J (Consistent Group Ltd v Kalwak) is wired into the heart of English law.
The Duck Test means that what you do in the real world, and how you describe yourself there, trumps any clever words you put down on paper. What you do matters more than what you say. And having placed themselves in a position to be subject to the Duck Test, Uber were now failing it. Hard.
“We have been struck,” the tribunal said, “by the remarkable lengths to which Uber has gone in order to compel agreement with its (perhaps we should say its lawyers’) description of itself.”
“In her evidence,” they continued, “Ms Bertram [Uber’s lawyer] chose her words with the utmost care. But in publicity material and correspondence those speaking Uber’s name have frequently expressed themselves in language which appears to be incompatible with the case before us.”
The tribunal had looked at the fundamental contradiction lurking at the heart of Uber and decided that they were having none of it. They were presented with ample evidence that Uber consistently referred to ‘our drivers’. The firm constantly told passengers that it was Uber who was carrying them around. Uber also seemed happy to claim to TfL and the London Assembly Transport Committee that it employed people too. Indeed the tribunal were presented with testimony from the same team of lawyers that had presented to the Transport Committee. In it Ms Bertram herself referred to Uber creating jobs and having drivers.
“To our considerable surprise,” The ruling says, “Ms Bertram attempted before us to dismiss this as a typographical error.”
All of this led to one simple, potentially devastating legal conclusion for Uber. The tribunal was blunt:
“Any organisation (a) running an enterprise at the heart of which is the function of carrying people in motor cars from where they are to where they want to be and (b) operating in part through a company discharging the regulated responsibilities of a PHV operator, but (c) requiring drivers and passengers to agree, as a matter of contract that it does not provide transportation services (through UBV or ULL) and (d) resorting in its documentation to fictions, twisted language and even brand new terminology, merits we think a degree of scepticism.
“Reflecting on the Respondents’ general case, and on the grimly loyal evidence of Ms Bertram [Uber’s Counsel] in particular, we cannot help being reminded of Queen Gertrude’s most celebrated line:
‘The lady doth protest too much, methinks.’”
Uber, the tribunal was saying, was a big fat passenger-carrying duck.
Quack. Quack. Quack.
Uber’s foot-in-mouth (or perhaps wing-in-beak) encounter with English employment law may seem only tangentially related to the current situation. That situation is that TfL have – for the second time in under two years – refused to grant Uber a Private Hire License to operate in London. But the tribunal ruling is important, because it helps answer the question that should really be asked:
Just why is it so hard for Uber to meet the regulations in London?
The answer is that there is an existential dilemma lurking at the heart of Uber. It is Schrodinger’s taxi firm. To be profitable it needs to be an app. But to operate in London it needs to be a cab firm.
Until now, Uber has largely managed to walk a fine line between the two, in part by clinging firmly to the distinction between ULL and UBV. As the employment case brutally highlighted, however, there is no clever contractual language that, under English law, will allow them to maintain that distinction forever. It is their behaviour that matters and if they cannot force TfL to back down then their current predicament will only get worse.
It’s not about the app… again
The reasons why TfL have refused to grant Uber an operator license in London are relatively straightforward. Just as it was 18 months ago, when a similar decision was reached, it is about safety not the app.
The reasons for the license revocation are as follows:
- An identity fraud issue, which allowed unauthorised drivers to upload their photos to other drivers’ accounts. This allowed them to pick up passengers under false identities. At least 14,000 journeys are known to have occurred this way.
- System or process failures which allowed dismissed or suspended drivers to create an Uber account and carry passengers
- Serious hire or reward insurance issues related to the above and other failures (TfL successfully took Uber to court for breaches of insurance requirements in July 2019)
- A number of unspecified additional breaches
Based on these factors, TfL believe Uber’s current internal safeguards are not robust enough to prevent future issues on similar lines. They cannot, therefore, be classed as ‘fit and proper’ to hold a PHV licence.
As is clear, Uber’s licensing issue isn’t that TfL or the Mayor are blocking innovation. Nor are TfL anti-Uber. The problem is that Uber are struggling to demonstrate, to TfL’s satisfaction, that they have enough processes in place, under their current operational model, to ensure the safety of their passengers.
That ‘their’ is important. As the employment tribunal so eloquently put it, Uber are not ‘a mosaic of 30,000 small businesses’, they are a private hire firm. That means, as far as TfL are concerned, the responsibility for the safety of those passengers belongs to Uber.
Accepting this and making changes would be the easiest way for Uber to meet the regulatory requirements and regain their licence. Indeed this is, in essence, what they quietly did 18 months ago (despite claiming in the press that they had won a victory in court). Making more changes risks moving even further away from ‘app’ towards ‘cab firm’, however, which simply replaces one problem with another. Uber have already appealed against the employment tribunal decision once and lost and are now waiting on a Supreme Court appeal. Further concessions to TfL now would further increase the risk of losing that case. Uber’s existential crisis is firmly underway.
The existential crisis
Uber’s first response to their loss of London licence toed a familiar line. As was the situation 18 months ago, they questioned both the Mayor’s and the city’s commitment to innovation.
The logic behind this is understandable. Uber wishes to shift the debate away from Uber and make it about rideshare in general. Yet the truth is that the battle over rideshare apps has already been won or lost (depending on your point of view) in London. They are here to stay, and they are here to stay in numbers. Mytaxi (formerly Hailo), Gett, Kabbee, Kapten and now Ola (India’s homegrown Uber-beater) are all present on the streets of London in some way. They are all also regulated, where necessary, by TfL.
Uber’s second response has been to push a narrative to riders and drivers (via emails) and to the media that this decision represents a sudden reversal in TfL’s opinion. That, as Uber’s regional manager is quoted in a Huffington Post article as saying, TfL found Uber a: ‘fit and proper authority just two months ago’.
That this would be accepted by the recipients of an email from Uber is understandable. That so many media outlets have accepted it at face value is not. At best, it is an omission, at worst a lie. It is also one that is easy to disprove, simply by looking at TfL’s press releases for September. Uber weren’t so much granted a new license, as issued a stay of execution while all the above issues were investigated:
“Uber London Limited has been granted a two-month private hire operator licence to allow for scrutiny of additional information that we are requesting ahead of consideration of any potential further licensing application.”
Publicly massaging the truth over the current licensing situation and the reasons for its refusal may seem an odd tactic for Uber to take. Particularly given that their objective right now is to convince TfL that they can be trusted. But again, this situation highlights the corner that Uber have painted themselves into. Every change that requires Uber to behaves more like a cab firm moves them further away from ever achieving profitability.
The economics of Uber
We have written previously about the economics of Uber. Most notably the last time that the issue of Uber’s licence came up. There are two elements of their model that need to be understood:
- Uber’s path to profitability in any territory requires either a monopoly, allowing prices to be increased to cover costs, or a massive reduction in journey costs.
- Uber is reliant on venture capital and other forms of investment until profitability is achieved.
For Uber, securing a monopoly in London (or anywhere) is dependent on two factors – enough drivers to dominate the market and a low enough cost to force modal shift, or at least syphon off business from potential rivals (in London this is the existing minicab and black cab trades).
To achieve this, Uber uses investor capital to run at a significant loss in most territories. For example, in 2016 in New York (one of the few territories in which raw numbers were available), the average fare paid by a passenger covered only 40% of the journey cost.
Creating a sufficiently large pool of drivers is another reason why journeys rarely wash themselves in cost terms. London, like other territories, has a limited pool of licensed drivers and that means Uber have to make driving for them much more attractive than any alternative.
There are approximately 110,000 private hire drivers in London, a number that has largely held steady in recent years but is a significant increase on pre-Uber figures, which were closer to 60,000. It is fair to say that the explosion in private hire drivers caused significant disruption to London’s roads and considerable early friction between Uber and TfL.
Of these drivers, Uber claims that approximately 45,000 drive for them. Indeed stressing the job losses that would result from the permanent loss of their licence was a cornerstone of their public defence in 2017 and seems set to be so again. As with online games that include users who log in intermittently or have done historically to bolster their user numbers, so should Uber’s London driver figures be treated with some suspicion. In 2017 we cross-compared Uber’s various published numbers on journeys, drivers, profits and average income. Our conclusion was that Uber likely had somewhere in the region of 7,500 – 15,000 regularly active drivers annually. Either that or they were banking somewhere between £625m – £1.4bn of profits abroad, away from the eyes of Her Majesty’s Revenue and Customs (HMRC).
Rock meet hard place
Whatever the true figure, it still represents a considerable percentage of the available driver market. This is a market that the arrival of more rideshare apps, along with the measures TfL have put in place to limit PHV licensing, is making more competitive, not less. As a result, maintaining that level of driver coverage in London makes up a considerable part of Uber’s ongoing costs. Those costs will balloon further if the Aslam and Farrar ruling is confirmed (hence why Uber have appealed that decision all the way to the Supreme Court).
TfL’s licence revocation highlights the other problem that Uber face with maintaining their large driver base: it is the biggest risk to their licence. Drivers, or rather the potential safety issues bad actors among the driver pool can cause, are why Uber is now fighting for its ability to operate in London at all.
The problem for Uber is that the solutions to these two problems are mutually exclusive. To operate in London, TfL are ordering them to be more ‘hands on’, but to avoid greater financial burden due to employee obligations Uber needs to be as ‘hands off’ as possible with their drivers, to avoid them being classed as employees.
If you were wondering why Uber’s current PR blitz seems so committed to insisting that the steps they have taken already to meet regulations should be sufficient, this is the reason.
Enter the taxman
To make matters worse for Uber, there is another reason why they are so desperate for the current status quo to be preserved: a little thing called 20% Value Added Tax (VAT).
In October, ULL filed their accounts up to December 2018 with Companies House. The ‘liabilities’ section on accounts for internet startups (or in Uber’s case, unicorns) are almost always a fun read. There are a lot of things you can get away with including or omitting from an investor deck. Your accounts are different. Being creative with your accounts means breaking the law.
“The Uber Group” the accounts read, “is involved in an ongoing dialog with HMRC, which is seeking to classify the Uber Group as a transportation provider.”
There are few things that will terrify a business owner (or an accountant) more than the words ‘ongoing dialog with HMRC’. In Uber’s case, the second part of that statement makes things even worse. This is because being classified as a transport provider would make Uber, and not its drivers, liable for paying VAT in the UK.
This is another reason why Uber’s lawyers in the employment tribunal were so ‘grimly loyal’ to the idea that Uber is simply a ‘mosaic’ of small businesses. Because as long as this is true, each of Uber’s drivers are individually liable for VAT, rather than that liability sitting on Uber themselves. This allows Uber to commit tax-death by a thousand cuts.
The cumulative value of Uber’s bookings is high, but an individual Uber driver in London can expect to make £40,000 (you can find a good earnings vs costs breakdown here) at best. This is well below the VAT threshold of £85,000. Pushing the liability for VAT onto Uber would be a game changer. Not just because this would increase the tax burden on every journey from (effectively) nothing to 20%, something Uber would need to swallow to maintain its competitive price point in London, but also because HMRC would be able to collect VAT retrospectively on Uber. That’s potentially a £1.1bn liability (according to the Financial Times) currently hanging over Uber’s heads, which they can only avoid if they can persuade HMRC to accept that they aren’t a transport provider.
On that subject, just for variety, we will quote the judge from Uber’s first appeal of the Aslam and Farrar decision. The appeal was dismissed:
“I am satisfied the [tribunal] did not err either in its approach or in its conclusions when rejecting the contention that the contract was between driver and passenger and that ULL was simply the agent in this relationship, providing its services as such to the drivers. Having rejected that characterisation of the relevant relationships, on its findings as to the factual reality of the situation, the [tribunal] was entitled to conclude there was a contract between ULL and the drivers whereby the drivers personally undertook work for ULL as part of its business of providing transportation services to passengers in the London area.”
Quack. Quack. Quack.
The myth of scalability
The above provides some context for why Uber are struggling to meet their regulatory requirements. It also shows why the firm is so wedded to claiming that their current internal processes should be sufficient for TfL to grant them a licence. Uber’s problems, however, do not end there.
Whatever the outcome of the licensing process, it is likely to increase the cost to Uber of operating in London, as a greater level of driver oversight will be required. That cost will increase further should they lose the final appeal at the Supreme Court over Aslam and Farrar. Then there is the lurking VAT issue beyond. All of these increase the pressure on Uber to improve the financial return on every journey, but their ability to do so is much more constrained than they have sometimes tried to portray.
As with the majority of startups, particularly those claiming to be disruptors, at the core of Uber’s promise to investors is the principle of ‘economies of scale’. Just like Amazon (the argument normally goes) the larger you get as a technology firm, the more efficiencies you can find. This is another reason for the quest for monopoly – it’s not just about forcing your rivals out of the market, it’s about getting big enough to start seeing the efficiencies that size is meant to provide.
Taxis are not technology
As Uber have increasingly discovered, however, taxis are not technology. Every journey carries a number of fixed costs that simply don’t scale. Most particularly: driver and tax. Already for Uber the cost of the driver takes up the majority of the fare. Again, here the conflict between being an app and a cab firm counts against them.
The need to be ‘just an app’ means that for Uber the driver cost isn’t simply that of the person behind the wheel. Uber cannot maintain a fleet of cars themselves, nor can they buy fuel or most other services in bulk. The driver has to provide the car and the fuel, cover the cost of cleaning and a whole host of other services which have to be factored in to the total ‘commission’ that Uber pays out. A commission rate which now has to be more competitive after those costs, not less, because Uber has been unable to secure a monopoly on rideshare in London and the pool of available private hire drivers is topping out.
Should HMRC decide to make Uber liable for VAT, then the (effective) zero cost per-journey of tax is also about to get a whole lot bigger.
These costs do not scale, despite the traditional pitch to investors. It doesn’t matter whether Uber has four cabs on the streets of London or 40,000. They all require one driver, one car and have a fixed rate of tax. So the overall percentage of the revenue collected that is subsumed by operational costs will largely stay the same, no matter how large Uber gets
This is a problem not just in London, but across every territory. It is the reason that Uber are investing so heavily in driverless technology research. Because, for all of Uber’s talk of their contribution to the London (and UK) job market, the firm’s future profitability is dependent on making all of those jobs redundant.
The promise of driverless
Driverless technology is nowhere near ready for market though. If it can ever truly replace a cab driver at all. There is a reason why most driverless technology companies are based in America and put out videos of their cars (mostly) successfully dealing with traffic there. America is a car-first culture. Even in its cities roads are often designed to give priority to car use, sometimes to a point of active hostility to other forms of transport (such as cycling or buses).
The driverless technology problem is a cultural one as much as a technical one. It will be far easier for AI to navigate Los Angeles than the Lea Bridge Road. The Silicon Valley-centric nature of major startup investment – including Uber’s – has so far largely avoided acknowledging this is even a problem, let alone worked out how to tackle it. Indeed the betting money here at LR Towers is that the first workable driverless system that can deal with London is as likely to come out of Camden (where homegrown driverless firm Wayve are quietly beavering away) as it is from California.
Surviving the short term
All of this leaves Uber with a limited range of short-term options to squeeze out more revenue from their share of the fare. Indeed perhaps the only gap to exploit is the small one they have already identified: get more people paying money in the same vehicle.
If, as a Londoner, you have been wondering why your ears (podcasts) or eyes (Uber emails, social media and public advertising) are being bombarded with information and offers for UberPool then this is why. UberPool sees two or more people headed broadly in the same direction both get in the same Uber in return for a slightly lower fare.
It is Uber’s way of trying to beat the scalability problem, but it is not hard to see the limitations. The obvious one is that although it skews the fixed costs down a few percentage points, there are still limits to the effect it can have.
Pool driving is less appealing to drivers, as it adds complexity and uncertainty. It also further increases the risk that, in light of TfL’s current concerns, Uber can least afford to increase: passenger safety issues. There is a difference between getting into a car with a cab driver who (if the system is working) you can trust to have been vetted, and a complete stranger. Uber may argue that it is a conscious decision on the passenger’s part to place themselves in that situation, but that is hardly a viable defence in the court of public opinion, let alone the law, if something goes wrong.
That public opinion also remains another barrier to uptake. Again, culture matters. As seemingly every visitor from the north of England insists on pointing out, Londoners are not a particularly friendly bunch. This isn’t true, of course, but what is true is that socialising with strangers is not something Londoners have a strong tendency towards. It is no coincidence that Uber’s current round of podcast advertising for UberPool doesn’t claim it is an opportunity to talk to your fellow passenger. Instead it features internal monologues from two passengers, presumably sitting as far away from each other as possible on the back seat, quietly lost in their own thoughts.
It is difficult to know how heavily Uber are selling UberPool to investors as a potential solution (or at least one of them) to their current predicament in London or elsewhere. It is hard to see, however, how it can really do anything other than nudge the needle of revenue-per-journey slightly towards the black. To make drivers take it up, Uber are having to offer them better rates to do it. To make passengers use it, they are having to do the same.
Beyond all that, there is still an upper limit to how many ‘Poolers’ you can pack into a single journey anyway. Uber cars aren’t clown cars. There is a limit to how many passengers you can fit into one before you’re running a bus service, not a cab service. And if Uber start trying to edge into that territory then TfL will definitely want a word.
The quacks are starting to show
All of this serves to highlight the precarious situation that Uber now find themselves in, both in London and beyond.
Large American firms often seem to reach a stage where the legal department is the ultimate power in the company, often to the frustration of marketing. What the employment tribunal ruling clearly showed is that within Uber it is the opposite way round. It has to be, because the company’s continued existence depends on the image it presents to both passengers and investors.
That second group is arguably more critical right now than the first. It is investor money, not fares, that are keeping Uber alive. This is something that is worth its users remembering, next time Uber emails them and asks them to go on social media and tell TfL and the Mayor – or indeed the transport and political representatives of any city it operates in – just how much Uber means to them.
Right now they are not the customer. For now at least they are actually part of the product. They are a pressure group and a resource, a tool with which to lean on regulators and reassure investors that Uber are worth the long term gamble.
That’s important because the odds on that long term gamble paying off are getting longer. TfL’s decision to revoke their licence is part of the reason for that, but it should not be seen in isolation. It represents both an active threat to Uber’s continuing ability to function in London, and a contributing factor to the firm’s developing existential crisis.
Certainly, from a London perspective, Uber’s future looks uncertain to say the least. Their current best-case scenario is that they are able to restore the status quo without having to change a thing. This would require the Supreme Court to overturn the tribunal ruling, TfL to back down and grant them a London operator licence without any changes to the way they operate at all and HMRC to decide that they will drop their pursuit of Uber for VAT.
The chances of all three of those things happening are slim. Heading off the tribunal and HMRC, the two biggest long term threats is perhaps possible, but requires Uber to do something that they failed so spectacularly at last time round: keep quiet and convince a set of British judges that they are not the duck that they seem to be.
Uber’s problem is that TfL have now told them that if they want a licence to operate in London again, then they better start quacking very loudly indeed.