White v Uber: limitation defence ends black cab group claims
When does the limitation clock start where a claim is based on alleged fraud or relevant facts have allegedly been deliberately concealed, and how does that question get tried across a large group claim?
Uber arrived in London in 2012, causing consternation among the taxi community. Protests, press campaigns and letters to the regulator followed, and in September 2017 Transport for London (TfL) refused to renew Uber’s licence. Uber appealed and, in June 2018, won its licence back. But when nearly 13,300 black cab drivers sued it in 2024, alleging that Uber had obtained its licence by misleading TfL, the court held in White v Uber [2026] EWHC 2651 (Comm) that they had left it too late.
Background
In 2024, nearly 13,300 London black cab drivers and the assignee of two private hire businesses sued Uber, alleging that it had conspired to cause them loss, principally by dishonestly misrepresenting to TfL how its booking system worked. Without the alleged conspiracy, they said, Uber would not have obtained a licence and competed with them. Uber denied the allegations.
It was common ground that the usual six-year limitation period had expired before the claims were issued. The claimants argued that Uber’s alleged fraud and concealment meant that time did not begin to run for limitation purposes until 2018. Uber challenged this, contending that the claims were all time-barred. A preliminary issue trial took place to decide the issue, on the assumption that the pleaded wrongdoing had occurred, with no finding as to the truth of the allegations.
The arguments turned on two key dates: 20 December 2017, when TfL’s September 2017 decision letter refusing to renew Uber’s licence became publicly available, and June 2018, when Uber’s appeal succeeded and the appeal evidence was released.
When does time start to run?
Section 32 of the Limitation Act 1980 postpones limitation where a claim is based on fraud or a relevant fact has been deliberately concealed. Time runs from when the claimant discovers the fraud or concealment (actual knowledge) or could have discovered it with reasonable diligence (constructive knowledge).
There are two formulations of the required knowledge test in the authorities: the ‘statement of claim’ test for fraud and the ‘worthwhile claim’ test for concealment. Uber was content for both to be applied, and the judge noted that the distinction was unlikely to make any practical difference. In substance, the claimants had to show that they had not discovered, and could not, with reasonable diligence, have discovered, enough to plead a case that would not be struck out.
The claimants’ case on limitation was the same for every claimant: none of them had discovered, or could reasonably have discovered, the fraud until a reasonable period after the appeal evidence was made public in June 2018. Twenty sample black cab claimants were selected for the trial, and three witnesses gave evidence. Birt J dealt first with what could reasonably have been discovered, observing that the preliminary issue was of greater utility on that question than on the actual knowledge of particular individuals, which is more difficult to ‘roll out’ across a wider claimant cohort. It also proved decisive: once the judge found that the fraud was reasonably discoverable, what anyone actually knew no longer mattered.
What could the claimants have discovered?
The central document was TfL’s September 2017 letter refusing to renew Uber London’s licence. In this letter TfL recorded its view that Uber’s descriptions of its booking process were “precise, but false” and that Uber had not been open with TfL about Greyball, a technology used by some Uber companies to evade regulatory enforcement elsewhere.
The claimants accepted that the letter put them on notice of a need to investigate, but maintained that it was not enough to plead dishonesty. A responsible pleader, they argued, needed to wait for the outcome of the appeal. Time therefore could not run until the appeal material became public in 2018.
The decision
The judge accepted that TfL had made no express finding of dishonesty, but held that the claimants could nonetheless have pleaded it, for three main reasons. First, there is nothing inherently objectionable in pleading fraud by reference to the reasoned conclusions of a third party such as a regulator, even without sight of the underlying material. Second, although TfL had left open whether Uber knew its statements were false, saying only that it either knew or did not understand its own systems, that supported a pleadable inference that Uber’s false statements were made knowingly. Third, the appeal evidence added reasoning, not new facts. Standing back, the judge concluded that a reasonably diligent claimant could have pleaded the claims by February 2018, which was more than six years before any of the claims were issued. As a result, all the claims were time-barred.
Comment
The judgment underlines the need to assess emerging information promptly against any claim under contemplation. Knowledge of an unspecified deception is unlikely to be enough to start time running, but once the essential facts of the alleged fraud are known or discoverable, time is likely to run. It does not wait for a regulator or a court to make a finding, or for further evidence that strengthens an inference already available.
This is consistent with the Court of Appeal’s approach in Gemalto Holding v Infineon Technologies [2022] EWCA Civ 782[DL1] [HF2] , a competition damages claim, where time was held to run from the European Commission’s statement of objections becoming public knowledge in April 2013, rather than its infringement decision sixteen months later. In both cases the message is the same: where a regulator’s investigation puts the essential facts into the public domain, a claimant cannot treat the eventual decision as the start date for limitation purposes.
For group claims, the case highlights that the entire group can be exposed to a preliminary issue. Where every claimant’s case on limitation rests on the same proposition, as it often will, constructive knowledge is an obvious candidate for early determination: it is an objective question, can be tried as a discrete issue and is decisive if lost.
Finally, it is worth noting an earlier decision in the same case on disclosure ([2026] EWHC 1610 (Comm)). The claimants’ solicitors had originally been instructed by a litigation funder in December 2017 to explore potential claims against Uber. As part of that investigation, the solicitors communicated with the black cab trade association and individual drivers, and some of that material had to be disclosed, although the solicitors’ advice to the funder itself remained protected by legal advice privilege. Third-party funding is a common feature of group claims, and funders are often involved before all claimants have been signed up. The case shows that material generated at early stages of group claims can become evidence of what was discoverable and when.