Two decades of litigation about how UK dividend taxation treated overseas subsidiaries produced, almost as a side effect, one of the most consequential limitation decisions in modern tax law. The Supreme Court overturned the House of Lords' own earlier approach and made it significantly harder to reach back decades for a mistake-of-law tax refund.
On this page
Full name: Test Claimants in the Franked Investment Income Group Litigation and others v Commissioners for Her Majesty’s Revenue and Customs
Citation: [2020] UKSC 47
Court: Supreme Court (Lord Reed, Lord Hodge, Lord Briggs, Lord Sales, Lord Hamblen)
Judgment: 20 November 2020
Subject: When the extended limitation period for restitution claims based on a mistake of law begins to run under s32(1)(c) Limitation Act 1980
Result: Appeal allowed on the limitation point. The Court overruled Deutsche Morgan Grenfell v IRC [2006] UKHL 49 on when time starts to run, adopting a discoverability test.
The Facts
The Franked Investment Income (FII) Group Litigation was one of the largest and longest-running pieces of UK tax litigation, spanning proceedings from the 1990s onward and multiple references to the Court of Justice of the European Union. It concerned the way UK corporation tax and advance corporation tax (ACT) were charged on dividends received by UK-resident companies from non-resident subsidiaries, compared with the more favourable treatment of dividends from UK-resident subsidiaries. The claimant groups argued this differential treatment breached EU law on freedom of establishment and free movement of capital.
Once the underlying EU law incompatibility was established in earlier stages of the litigation, a separate and critical question arose: for how many years back could the claimants recover tax paid under the unlawful regime? Ordinary limitation would confine claims to six years before the claim form, but the claimants relied on section 32(1)(c) of the Limitation Act 1980, which extends time where a claim is for relief from the consequences of a mistake, providing that the period does not begin to run until the claimant discovered the mistake, or could with reasonable diligence have discovered it. On the earlier House of Lords authority in Deutsche Morgan Grenfell v IRC [2006] UKHL 49, this had been understood to mean that time did not start running until an authoritative court ruling established that the tax had indeed been unlawfully charged, potentially allowing recovery of tax paid decades earlier.
Procedural History
- High Court and Court of Appeal: the FII litigation proceeded through multiple stages over more than a decade, including references to the CJEU, addressing both the substantive EU law incompatibility and various remedial and procedural questions.
- Supreme Court [2020] UKSC 47: heard a discrete but critical limitation appeal, considering whether the House of Lords' own decision in Deutsche Morgan Grenfell on the meaning of "discovered" under s32(1)(c) was correctly decided, applying the Supreme Court's power under Practice Statement (Judicial Precedent) [1966] 1 WLR 1234 to depart from an earlier decision of the House of Lords where it is right to do so.
The Issues
- Was Deutsche Morgan Grenfell v IRC [2006] UKHL 49 correctly decided on when a claimant "discovers" a mistake of law for the purposes of s32(1)(c) Limitation Act 1980?
- If not, what is the correct test, and how does it apply to a claimant whose case rests on a developing and, for a long period, genuinely uncertain area of EU law?
- What is the standard of "reasonable diligence" against which discoverability is to be judged?
The Statutory Framework
Section 32(1) of the Limitation Act 1980 postpones the start of the limitation period in three situations: where the action is based on the defendant's fraud (limb (a)), where a relevant fact has been deliberately concealed from the claimant (limb (b)), and where the action is for relief from the consequences of a mistake (limb (c)). In each case, time does not begin to run until the claimant "has discovered the fraud, concealment or mistake ... or could with reasonable diligence have discovered it."
The Ratio: The Discoverability Test
The Court held that the previous approach conflated two distinct things: the moment a claimant could recognise a worthwhile claim existed, and the moment the law was authoritatively settled in the claimant's favour. Those are not the same. A sophisticated claimant, properly advised, could often identify a strongly arguable claim well before the point of law was finally and authoritatively resolved by an appellate court, particularly where the direction of travel in EU law was already becoming clear through earlier stages of the same or comparable litigation. The correct question is when a reasonably diligent claimant in the taxpayer's position could have recognised the claim was worth pursuing, not when doubt was entirely removed.
On the standard of reasonable diligence, the Court held it is objective: how a person carrying on a business of the relevant kind, with adequate but not unlimited resources, and a reasonable but not excessive sense of urgency, would have acted in the circumstances. For large, sophisticated corporate claimants pursuing multi-jurisdictional group litigation, advised throughout by specialist tax counsel, that standard could be met at a materially earlier point than for a smaller, less resourced taxpayer.
Authorities Considered
Deutsche Morgan Grenfell Group plc v IRC [2006] UKHL 49
The House of Lords decision the Supreme Court overruled on this point. It had held that a mistake of law is not "discovered" for s32(1)(c) purposes until it is judicially established that the payment was not due, because until an authoritative ruling exists, a taxpayer cannot be said to have made a legally relevant "mistake" at all, only to have held a view later shown to be wrong. The Supreme Court in FII held this reasoning wrongly collapsed the concept of a "mistake" into the concept of "certainty," and was inconsistent with the ordinary, purposive reading of s32(1)(c).
Practice Statement (Judicial Precedent) [1966] 1 WLR 1234
The mechanism by which the Supreme Court can depart from an earlier decision of the House of Lords or itself where adherence to precedent would cause injustice or unduly restrict the proper development of the law. The Court considered the criteria for departure carefully before concluding it was appropriate here, given the practical unworkability of the previous test in long-running, multi-stage litigation.
Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349
The House of Lords authority establishing that money paid under a mistake of law, not just a mistake of fact, is recoverable in restitution, and the doctrinal foundation on which the entire FII mistake-based restitution claim rested.
Canada Square and Section 32(1)(b)
Three years after FII, the Supreme Court returned to section 32 in Canada Square Operations Ltd v Potter [2023] UKSC 41, this time addressing the deliberate concealment limb, s32(1)(b), in the context of an unfair credit relationship claim. The Court held that "concealment" for these purposes does not require deliberate wrongdoing in the sense of dishonesty; it is enough that a fact relevant to the claimant's cause of action was withheld in circumstances where it ought to have been disclosed, and the claimant did not know and could not reasonably have discovered it. Time then runs from when the claimant did discover, or could reasonably have discovered, the concealed fact.
Relevance to Tax Practice
FII is the leading modern authority on limitation for restitutionary claims to recover overpaid tax based on a mistake of law, whether the mistake arises from an EU law incompatibility, a subsequently corrected HMRC misapplication of statute, or any other basis where the taxpayer paid tax it later transpires was not properly due. It applies squarely to advising a client who believes they may have overpaid tax years or decades ago because of a legal position that has since been shown to be wrong, whether by a later tribunal decision, a change in HMRC guidance, or a wider group litigation outcome.
Practitioner Application
Assessing a client's limitation position
- Identify the earliest point a "worthwhile claim" existed, not the date the point was finally, authoritatively settled. This often requires working out when the direction of travel in the relevant law first became reasonably clear to a properly advised person in the client's position.
- Assess the client's own resources and sophistication against the objective reasonable diligence standard. A well-advised corporate group is held to a materially higher standard of what it could have discovered than an unrepresented individual.
- Distinguish limitation limbs carefully. A concealment argument under s32(1)(b) following Canada Square is analytically different from a mistake argument under s32(1)(c) following FII; conflating them weakens both.
- Do not assume a long-dormant claim is safe. Clients (and sometimes advisers) who assume time only starts running once a case is definitively won elsewhere are relying on exactly the reasoning the Supreme Court rejected.
Common mistakes
- Advising a client that limitation only starts once an appellate court authoritatively rules in a comparable case. This is the discredited Deutsche Morgan Grenfell position.
- Ignoring how sophisticated the particular claimant was, and what advice was actually available to it, when assessing reasonable diligence.
- Overlooking that ordinary six-year limitation under s5 Limitation Act 1980, or the relevant statutory tax appeal time limits, may separately and additionally apply and bar a claim regardless of the s32 analysis.
Frequently Asked Questions
What did Test Claimants in FII v HMRC decide about limitation?
The Supreme Court held that the extended limitation period under s32(1)(c) Limitation Act 1980, for mistake-based claims, runs from when a claimant discovered, or with reasonable diligence could have discovered, a worthwhile claim, not from when a later court decision authoritatively established the mistake. This overruled the House of Lords’ earlier approach in Deutsche Morgan Grenfell.
What is the standard for "reasonable diligence" under section 32(1)(c)?
An objective standard: how a person carrying on the relevant kind of business, with adequate but not unlimited resources and a reasonable sense of urgency, would have acted. The test asks when such a person could have recognised they had a worthwhile claim, not when they actually realised it.
Why does this matter for tax restitution claims against HMRC?
It curtails how far back a mistake-of-law restitution claim can reach. A reasonably diligent taxpayer or adviser who could have identified a worthwhile challenge earlier, even before the point was authoritatively settled, will find their claim time-barred correspondingly earlier than under the previous, more generous approach.
How does this compare to the approach in Canada Square v Potter?
Canada Square Operations Ltd v Potter [2023] UKSC 41 concerned deliberate concealment under s32(1)(b), a different limb, and confirmed a similarly fact-sensitive approach to when time starts running. The two decisions are analytically distinct but both reject rigid, formalistic tests in favour of asking what the claimant could reasonably have known, and when.