Discoverability under s.32(1)(c) can be triggered by a CJEU tipping point — and even an Advocate General’s Opinion: BAT Industries v Commissioners of Inland Revenue [2025] EWCA Civ 1271

Introduction

This Court of Appeal decision is the latest landmark in the Franked Investment Income (FII) Group Litigation, a two-decade saga addressing the United Kingdom’s tax treatment of dividends paid from overseas subsidiaries to UK parent companies. It engages the intersection of domestic limitation rules and European Union freedoms, with colossal sums at stake and significant consequences beyond the FII litigation.

The appeal focuses on a single but pivotal limitation question: when did time start running, under section 32(1)(c) of the Limitation Act 1980, for claims to recover corporation tax paid under a mistake of law? The Supreme Court in FII SC 2 ([2020] UKSC 47; [2022] AC 1) rejected the earlier “wait for final appellate confirmation” approach in Deutsche Morgan Grenfell (DMG) and replaced it with a practical test: the limitation period starts when, using reasonable diligence, a claimant knew or could have known that there was a real possibility of a mistake, such that a “worthwhile claim” could be pursued or the claimant had “sufficient confidence to embark on the preliminaries” to litigation.

Applying that test, Richards J had held that the “date of discoverability” was 6 June 2000, the date of the CJEU’s judgment in Verkooijen (Case C‑35/98). HMRC (the Revenue) appealed, arguing for a much earlier date (11 July 1996), while FCE Bank cross-appealed for a later date, at least to the CJEU’s judgment in Hoechst/Metallgesellschaft (8 March 2001). The Court of Appeal dismissed both the appeal and the cross-appeal, endorsing the High Court’s evidential methodology and legal analysis, while adding an important clarification: an Advocate General’s (AG’s) opinion can, in principle, be enough to trigger discoverability under s.32(1)(c).

Summary of the Judgment

The Court of Appeal upheld the High Court’s conclusion that a reasonably diligent, well-advised UK-based multinational would have discovered its mistake of law (in the sense that a “worthwhile claim” arose, or that there was sufficient confidence to begin the preliminaries to litigation) by 6 June 2000, when the CJEU decided Verkooijen. The Court:

Rejected HMRC’s contention that discoverability arose as early as 1996;
Rejected FCE’s contention that, so far as the ACT Challenge is concerned, discoverability arose no earlier than Hoechst (2001) or later;
Affirmed the proper approach to the s.32(1)(c) inquiry: identify what a “well-advised multi-national group” would have discovered by taking advice from an “Appropriate Adviser” (with EU law and UK direct tax expertise) working to a standard of reasonable diligence and adequate but not unlimited resources;
Clarified the meaning of a “worthwhile claim”: a legally viable claim, not merely a hypothesis “worth investigating”;
Accepted that the CJEU’s reasoning in Verkooijen dismantled the then-prevailing professional consensus that had treated such challenges as hopeless, by undermining the prior understanding of “comparability,” “cohesion,” and the necessity of EU-level harmonisation;
Observed (obiter) that the second Advocate General’s Opinion in Verkooijen (December 1999), which corrected a factual misunderstanding and engaged the decisive legal reasoning later adopted by the CJEU, would also have sufficed to trigger discoverability — although this did not affect the outcome.

The Court also emphasised orthodox appellate restraint (Volpi v Volpi) in relation to evaluative findings and expert evidence: the High Court’s reasoned acceptance of Mr Gammie’s evidence on professional consensus and legal thinking at the time could not be displaced.

Analysis

Precedents Cited and Their Influence

Kleinwort Benson v Lincoln City Council [1999] 2 AC 349: Confirmed that s.32(1)(c) applies to mistakes of law. This remains intact; HMRC’s broader attack on Kleinwort Benson had already failed in FII SC 2.
Deutsche Morgan Grenfell (DMG) [2007] 1 AC 558: The “final appellate decision” test for discoverability was overruled by FII SC 2. The present Court of Appeal applies the FII SC 2 test, focusing on a claimant’s ability to recognise a legally viable claim earlier than final appellate resolution.
FII SC 2 [2020] UKSC 47: The governing test. Discoverability occurs when a claimant, using reasonable diligence, knows (or could know) that there is a real possibility of a mistake of law such that a “worthwhile claim” arises or there is “sufficient confidence” to begin preliminaries. The Court of Appeal faithfully applied this, resisting HMRC’s attempt to dilute it and FCE’s attempt to raise the bar.
OT Computers v Infineon [2021] EWCA Civ 501 and Gemalto v Infineon [2022] EWCA Civ 782: Cited for the approach to “reasonable diligence,” the objective standard, and the notion that the exercise does not require fine-grained probability assessments of success at the outset.
Volpi v Volpi [2022] EWCA Civ 464, Henderson v Foxworth [2014] UKSC 41, Perry v Raleys [2019] UKSC 5, Byers v Saudi National Bank [2022] EWCA Civ 43, Griffiths v TUI [2021] EWCA Civ 1442: Appellate discipline on interfering with primary and evaluative fact-finding and on handling expert evidence. These underpinned the Court’s refusal to revisit the High Court’s assessment of the expert tax/EU-law consensus of the period.
CJEU case law shaping “comparability” and “cohesion”:
- Bachmann (C‑204/90): Established a narrow “cohesion of the tax system” justification requiring a direct link between advantage and levy, often pleaded to defend domestic fiscal coherence.
- Schumacker (C‑279/93): Residents versus non-residents can be comparable in certain tax contexts; it prompted further probing of direct tax restrictions but did not, by itself, unsettle the professional consensus regarding dividend taxation.
- Verkooijen (C‑35/98): The critical inflection point. It rejected arguments that harmonisation was a prerequisite, declined a non-comparability plea, and decisively narrowed cohesion-based justifications. The Court of Appeal agreed this dismantled the UK professional consensus and triggered discoverability.
- Hoechst/Metallgesellschaft (C‑397/98, C‑410/98): Confirmed the comparability analysis and rejected cohesion arguments in the ACT context. The Court held that the AG’s Opinion (Sept 2000) would also have been sufficient to trigger discoverability for ACT claims, which undermined FCE’s cross-appeal even on its own case.
- Other EU authorities cited (Avoir Fiscal, Daily Mail, Commerzbank) were analysed but found not to have, as at 1996, opened a viable route to the specific DV/ACT challenges at issue.

The Court’s Legal Reasoning

The Court took the Supreme Court’s twin formulations (“worthwhile claim” and “sufficient confidence to embark on preliminaries”) as equally authoritative and mutually illuminating. It rejected attempts to recast “worthwhile” as “worth investigating”; the standard is legal viability robust enough to get past a strike-out, not initial curiosity.

The inquiry is objective. It asks what a claimant could have discovered using reasonable diligence, with adequate but not unlimited resources, and a reasonable sense of urgency. Crucially:

The claimant is to be envisaged as a “well-advised UK-based multinational group” taking legal advice;
That advice is to be sought from an “Appropriate Adviser”: a team (or adviser) with both EU law and UK corporate tax expertise, cognisant of the detailed workings of the relevant domestic regime;
The initial advice is necessarily “high level” in the sense that it distils and applies existing principles; the exhaustive merits assessment and evidence collection happen at the “preliminaries” stage once a viable claim is identified.

Applying this framework, the Court accepted Richards J’s factual findings—supported by Mr Gammie’s expert evidence and contemporaneous sources (notably Farmer & Lyal and the Ruding Committee)—that up to July 1996 there was a settled professional consensus that DV/ACT challenges were not viable because:

It was thought that only EU-level harmonisation or treaties could address double taxation frictions;
Non-UK EU subsidiaries were not viewed as “comparable” to UK subsidiaries for the purposes of the asserted discrimination;
The “cohesion of the tax system” defence from Bachmann was widely thought to justify the differential treatment.

Verkooijen shifted the ground. The CJEU’s reasoning rejected, or seriously weakened, each of the consensus pillars: it found cross-border dividend rules could be incompatible with Treaty freedoms without harmonisation; refused to treat foreign and domestic dividends as non-comparable; and applied the cohesion doctrine narrowly, requiring a direct link that was absent. At that point, a well-advised multinational would, as a matter of reasonable diligence, have recognised a viable claim.

Two amplifications are significant:

AG Opinion as a trigger: The Court indicated that the second AG Opinion in Verkooijen (Dec 1999), once the factual premise was corrected, deployed the same line of reasoning later adopted by the CJEU and could itself have sufficed to trigger discoverability. This obiter clarification will be important in future s.32(1)(c) disputes: the clock can start with an AG Opinion where it credibly signals a decisive legal shift making a claim legally viable.
Same date for DV and ACT challenges: Although ACT (an advance payment of corporation tax) and DV (taxation of foreign dividends) are not identical, the Verkooijen reasoning undercut the consensus obstacles across both categories. The Court agreed with the Judge that a single discoverability date applies for both DV and ACT challenges here. Moreover, even had ACT required Hoechst to crystallise, the AG Opinion in Hoechst (Sept 2000) would have sufficed—still too early for FCE’s cross-appeal theory.

The Court also addressed and rejected several technical criticisms:

“Constructive discovery” as a shorthand did not misstate the test; the Judge correctly focused on what the claimant could have discovered using reasonable diligence;
The burden of proof was correctly applied: claimants had to show they could not have discovered earlier than the candidate date without unreasonable steps;
HMRC’s attempt to re-try the evaluative exercise on the paper record (substituting its preferred reading of EU cases and literature) fell foul of Volpi principles; the High Court’s acceptance of Mr Gammie’s evidence and methodology was rational and well-explained.

Impact and Implications

The decision has wide ramifications for limitation in mistake-of-law restitution claims, especially in complex public law/tax settings where legal landscapes evolve through EU (or comparable apex) jurisprudence:

AG Opinions matter: Discoverability can be triggered by an AG Opinion that clearly signals the legal tipping point—claimants cannot safely wait for a CJEU judgment, still less a domestic final appellate decision. The limitation clock may start earlier than many have assumed.
“Worthwhile claim” calibrated to legal viability: The threshold is not low. It is more than “worth looking into,” but less than “probable success.” It maps onto “good grounds for supposing” a valid claim that can survive a strike-out. This refined calibration should influence internal decision-making and advice records.
Evidence-led reconstruction of legal thinking: Courts will ask what a competent, dual-skilled (EU and domestic tax) adviser would have said, informed by contemporaneous professional consensus and literature. Parties should marshal expert and documentary evidence of the state of legal thought at the relevant time, including how professional consensus evolved.
Unitary dates across related legal theories: Where the same EU-law “tipping point” dismantles shared obstacles (comparability, cohesion), courts may align discoverability dates across different but related causes (DV and ACT here). Attempts to slice the dates more finely must be grounded in strong contemporaneous evidence.
Appellate restraint: The Court underscores that discoverability under s.32(1)(c) is an evidentially grounded evaluative finding. Appeals will rarely succeed absent clear legal misdirection or irrationality.

For practitioners, the operational message is clear: in areas of unsettled law, especially where supranational jurisprudence is in play, firms should:

Track not only judgments but also AG Opinions and authoritative scholarly/official commentary;
Record contemporaneous advice identifying when a claim crossed from “possible” to “legally viable” (with reasons), mindful that later courts may reconstruct that timeline;
Appreciate that postponing action until final appellate clarity risks being out of time under s.32.

Complex Concepts Simplified

Section 32(1)(c) Limitation Act 1980: Normally, a six-year limitation applies to claims in restitution for money paid by mistake. Section 32(1)(c) postpones the start of that six-year period until the claimant discovers (or could with reasonable diligence discover) the mistake. The “discoverability” test is objective and evidence-driven.
“Worthwhile claim” and “sufficient confidence”: The Supreme Court equates these. A claim is “worthwhile” when there are good legal grounds to suppose it is viable (capable of surviving strike out), prompting preliminaries like notifying HMRC, instructing counsel, and gathering evidence. It is not enough merely to spot an arguable idea worth investigating.
“Appropriate Adviser”: A hypothetical competent adviser (or team) with both UK direct tax and EU law expertise, applying a high-level doctrinal analysis initially, and able to assess whether the basis of the payment is legally questionable.
“Comparability” (EU law): In free movement analysis, a state cannot treat cross-border situations less favourably than domestic ones if they are objectively comparable in light of the measure’s purpose and context. The consensus previously assumed non-comparability for foreign dividends; Verkooijen unsettled that assumption.
“Cohesion of the tax system” (Bachmann): A narrow justification permitting a restriction on a fundamental freedom if a direct link exists between a tax advantage and an offsetting levy within the same tax system and taxpayer. Verkooijen and Hoechst emphasised how rarely that justification applies.
ACT and FII (UK tax): Advance Corporation Tax (ACT) was a prepayment of corporation tax due when UK companies distributed dividends (1973–1999). Franked Investment Income (FII) was the system crediting that ACT to reduce double taxation of corporate distributions within the UK. Foreign dividends did not generate FII, creating asymmetry for cross-border structures.
DV vs ACT Challenges: The DV Challenge attacked the treatment of foreign-source dividend income (taxed under Case V) compared to domestic dividends (exempt). The ACT Challenge attacked the exclusion of foreign-source dividends from FII. Both ultimately turned on EU-law constraints (free movement of capital and establishment).

Detailed Discussion of Key Authorities in Context

Verkooijen (C‑35/98)

At the time, practitioners assumed that domestic systems could maintain differential treatment absent EU harmonisation; that foreign and domestic dividend situations were not comparable; and that the cohesion justification from Bachmann would defend coherent dividend tax architectures. Verkooijen undermined each strand:

Harmonisation was not a prerequisite: national measures could breach free movement rules in their own right;
Comparability was accepted in substance: the resident shareholder receiving foreign dividends was comparable to receiving domestic dividends for purposes of the exemption;
Cohesion failed for lack of the requisite direct link.

The Court of Appeal agreed that this “dismantled” the extant consensus and would have led, in 2000 (indeed as early as the corrected AG Opinion in Dec 1999), a competent adviser to say a viable claim now existed.

Hoechst/Metallgesellschaft

In ACT terms, the UK argued both non-comparability (because foreign parents were not within the ACT net) and cohesion (group income election prevented ACT collection if extended cross-border). The AG Opinion, and then the CJEU, rejected both defences, tightly applying the Verkooijen logic and treating ACT as an advance payment of corporation tax. The Court of Appeal reasoned that, if Verkooijen did not already take ACT claims over the “worthwhile” line, the AG Opinion in Hoechst certainly did—again showing that waiting for a CJEU judgment is not necessary to start the s.32 clock.

Other EU Authorities

Avoir Fiscal, Daily Mail, and Commerzbank revealed that EU freedoms could, in principle, intersect with direct tax, but they did not, by 1996, provide a basis for concluding that DV/ACT-style claims were viable. Schumacker prompted interest but was then seen as an exception rather than a measure that unseated the prevailing consensus on dividend taxation. It was Verkooijen that altered the legal landscape decisively.

Why the Appeals Failed

HMRC’s appeal failed because the High Court had correctly applied the FII SC 2 test, adopted a permissible “Appropriate Adviser” methodology, and made rational evaluative findings (including endorsing Mr Gammie’s evidence on professional consensus, supported by Farmer & Lyal and the Ruding Committee). The Court rejected attempts to lower the “worthwhile claim” threshold to “worth investigating,” and confirmed that what a claimant could have discovered tracks what competent advice would have said at the time.
FCE’s cross-appeal failed because the Court accepted that Verkooijen’s reasoning was sufficient to make an ACT claim legally viable at the preliminary stage. Even if not, the AG Opinion in Hoechst (Sept 2000) would have done so, which would still render FCE’s timing position untenable.

Conclusion

BAT Industries v Commissioners of Inland Revenue consolidates and operationalises the Supreme Court’s reorientation of s.32(1)(c) in FII SC 2. It teaches three durable lessons.

Discoverability is an evidence-based tipping point, not a judicial finish line. It occurs when a well-advised claimant, using reasonable diligence, could see a legally viable route to a claim—not when a final court confirms it. That tipping point can be a CJEU judgment like Verkooijen or, in appropriate cases, an Advocate General’s Opinion that cogently sets out the decisive reasoning.
“Worthwhile claim” means legal viability—enough to survive strike-out and justify preliminaries—rather than a mere investigatory hunch. The yardstick is the advice a competent, EU-and-tax-informed “Appropriate Adviser” would have given at the time, assessed against contemporaneous professional thinking and literature.
Appellate courts will not re-try the evaluative fact-finding inherent in s.32 inquiries. Where the judge has set the right test, weighed expert evidence, and avoided hindsight bias, challenges will fail.

On the facts of the FII litigation, Verkooijen represented the moment when the professional consensus against DV and ACT challenges was dismantled and a “worthwhile claim” emerged. The Court’s recognition that an AG Opinion can be enough to start the clock is a notable and pragmatic development. For future mistake-of-law limitation disputes—within and beyond tax—this judgment underscores the need for vigilant monitoring of legal developments short of final judgments, contemporaneous recording of legal assessments, and timely action once a claim becomes legally viable.