Funder Files, Fiduciary Duties, and Disclosure: What Uber v White Decided About Who Controls a Litigation File

The Commercial Court has ruled that a firm’s pre-retainer work for a funder can become disclosable to the clients it later signs up. The reasoning reshapes how firms and funders should approach privilege, control, and confidentiality.

The facts

Uber London Ltd v White [2026] EWHC 1610 (Comm) is an extempore ruling of Birt J, given two weeks before trial of a preliminary issue. Black cab drivers, and the assignee of two dissolved private hire companies, allege Uber’s corporate group ran an unlawful means conspiracy between mid-2012 and March 2018, including fraudulent misrepresentations said to have secured Uber’s London private hire licence. Claimed losses exceed £340 million.

The claims were issued in 2024, after the six-year limitation period had expired. The claimants rely on section 32 of the Limitation Act 1980, contending they could not reasonably have discovered the fraud until after Uber’s licensing appeal judgment in June 2018 — the question tried as a preliminary issue.

The dispute concerned the period before the retainer. In December 2017, before any driver was a client, the funder Harbour engaged Mishcon de Reya to investigate a potential claim against Uber, corresponding with the drivers’ trade association, the LTDA, and opening a file in Harbour’s name. Claimants were not signed up until October 2018; Harbour funded the claim’s early stages before withdrawing in November 2019.

Uber sought disclosure of this pre-retainer material, the “Harbour Communications,” as relevant to what claimants knew, or could reasonably have discovered, before the limitation cut-off. The claimants resisted on relevance, privilege, and control, and lost on all three, subject to limits on scope.

The English position

Relevance. Communications shedding light on what individual drivers actually knew were held plainly relevant to the section 32 issue, as was material reflecting Mishcon and Harbour’s own assessment of the claim’s prospects, though more marginally.

Privilege. Legal advice privilege was accepted to apply to some documents. The live dispute was litigation privilege, which under Three Rivers (No.6) requires communications made for the dominant purpose of conducting litigation. Al Sadeq v Dechert confirms a non-party, including a funder, can in principle claim litigation privilege, subject to that test. Birt J held it was not met: Harbour’s purpose was to decide whether to fund a claim it was not itself going to conduct, which differs from deciding whether to litigate. The communications therefore did not attract litigation privilege, consistent with Excalibur Ventures v Texas Keystone and Winterthur v AG (Manchester) on funding documents.

Control. This was the most significant finding. Under the ordinary disclosure test, a party must disclose documents it holds, has held, or has a right to inspect. The claimants argued this fell outside that test, since Mishcon created the documents while acting for Harbour, no driver was yet a client when the file was opened, and the material sat within a retainer to which the claimants were strangers.

Birt J took a different approach, asking not about possession but whether Mishcon owed the claimants a duty to disclose the material once their retainer began. Drawing on Moody v Cox and Hilton v Barker Booth & Eastwood, he held that a solicitor owes a client a duty to disclose everything material to that client’s matter, even where the solicitor learned it while acting for another party under a separate duty of confidence. A solicitor who owes conflicting duties may not prefer one over another, and must perform both as best he can.

Mishcon investigated the claim for Harbour before later taking on individual drivers as clients to bring that same claim. From that point it owed those clients a duty to disclose everything material to their claim, including matters learned while acting for Harbour. That duty gave the claimants a right to the material from the moment their retainer began, and such a right itself amounts to control, regardless of possession or timing. Harbour itself raised no obstacle, asserting privilege but claiming no wider confidentiality.

The claimants also relied on a confidentiality clause in a later retainer, negotiated through an intermediary, RGL, arguing it gave informed consent for Mishcon to withhold the Harbour material. The court disagreed: RGL’s authority to bind individual claimants was thinly evidenced; accepting portal posted terms is not informed consent, which requires understanding what is being surrendered; the claimants’ right had already accrued when the clause was agreed, so it could not remove that entitlement without clear consent to its loss; and the wording, limited to persons in a similar sector, probably did not extend to Harbour, a funder, or to the claimants’ own information.

Proportionality. Having found relevance, no litigation privilege, and control established, the court ordered disclosure, but limited its scope to documents bearing on individual drivers’ actual knowledge, excluding material reflecting only Mishcon and Harbour’s own assessment of the claim’s merits, which was more marginally relevant and likely, in any event, largely covered by legal advice privilege.

The position in the BVI

The BVI’s Civil Procedure Rules are closely modelled on the English CPR, and the BVI Commercial Court regularly draws on English authority, including Three Rivers and Al Sadeq v Dechert. Hilton v Barker and Moody v Cox are both part of the common law applied in the territory, so a BVI court would likely adopt the same approach.

This scenario is familiar in BVI practice: funders, office holders, and claims aggregators routinely instruct a firm to investigate a claim before any formal retainer exists with the eventual claimant, creditor, or beneficiary, particularly in insolvency and asset tracing work. Under this reasoning, a firm’s pre-engagement file could be treated as within a later client’s control, regardless of confidentiality undertakings to the original instructing party, unless clear and timely informed consent has been obtained.

Points of note for practitioners

• The dominant purpose test remains narrow. A funder’s own due diligence is unlikely to attract litigation privilege, even under a genuine solicitor and client relationship.
• Control can arise independently of possession. A fiduciary duty of disclosure, arising after documents were created, can itself be a source of control.
• Timing governs confidentiality clauses. Such a clause will be judged against when it was agreed relative to when the client’s right to the information arose.
• The reasoning is readily transposable. Given the alignment between English and BVI procedure, firms and funders should expect the same principles to apply.