Restitution
front page

What's new?

Archive front page

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2007

2006

2008

2009

Dear Charles:

The difficulties you have with the mistake analysis show why absence of basis is the simplest most coherent view of unjust enrichment. Although the province of unjust enrichment would be smaller under this view, what was covered would hold together better.

Cheers,

Charles Mitchell wrote:

Stimulated by Rob’s comments I have a few further thoughts about the case:

(1) The "cause of action" issue: does English law recognise a restitutionary claim for tax paid under a mistake of law?

I share Rob’s view that the CA's treatment of this issue was weak and unconvincing. Jonathan Parker LJ’s methodology was very strange. However, it doesn’t follow from the fact that the CA did a poor job that no respectable arguments can be marshalled to support the proposition that recovery of ultra vires tax payments should be governed by an 'exclusive regime'. The public character of tax authorities, and the special nature of the relationship between citizen and state, might lead one to conclude that such claims are sufficiently different from claims between private individuals that they should be fenced off and treated differently. Rebecca Williams has developed this line of thought more eloquently and expertly than I can manage, and I'm not sure that I find it completely convincing: it troubles me, for example, that the public/private borderline is so fuzzy that cases would often arise where it was unclear whether or not the claim fell within the scope of an exclusive regime. Without pursuing this any further, though, I would just say that it may not be entirely fair to counsel for the IRC to suggest that they relied solely on the CA's narrow textual approach: at [131]ff Lord Walker indicates that arguments from principle were also made, albeit that he and the others seem to have found these unconvincing.

(2) The "mistake" issue: did DMG make the payments of ACT under a mistake of law?

(a) Do judicial decisions changing the law retrospectively render payments made in line with the old law mistaken?

The HL were obliged to say yes in light of KB v Lincoln. In my view, however, the majority’s reasoning in KB was deeply flawed because the reach of legal fictions cannot be determined by taking them at face value and discovering in them an 'inherent logic' which they do not possess. In KB the majority started with the fiction that the effect of a case reversing a previous decision is retrospectively to make the law governing the relationship between the parties different from the law which actually governed their relationship at the time when they entered it. It can be argued that this fiction is defensible because without it, the court could not adjust the parties’ relationship in a way that ex hypothesi the court now considers appropriate; moreover, litigants would otherwise have no incentive to press for changes to the law, as they would be unable to take personal advantage of such changes. But these reasons do not necessarily hold good when it asked whether the fiction of retrospective legal change also affects the relationship between other people who were not parties to the case. If the pursuer in Donoghue had had an auntie who paid her money because she felt sorry for her and correctly believed at the time when she made the gift that the pursuer had no claim in tort against the defender, the question whether the auntie should be allowed to recover the money because she should retrospectively be deemed to have made a mistake can be sensibly answered only by starting from first principles: it won't do simply to say 'one fiction fits all' and that the auntie must have been mistaken because retrospective changes to the law must affect everybody if they are going to affect anybody. But if the majority in KB had a better reason than this for its findings on retrospective mistakes then I can’t find them; nor can I find anything in DMG which makes me feel better, notwithstanding Lord Hoffmann's attempt to cheer me up at [23] by saying that the mistake in a retrospective mistake case is a 'deemed' mistake. I guessed that! The question is why should the claimant be deemed to have made a mistake? Answer: 'practical considerations of fairness'. Hey ho.

(b) Was DMG mistaken in its belief that the money was due?

I agree with Rob that this was the IRC's best point and that counsel seems to have blown it. As Lord Scott holds (and as Monica Chowdry and I observed in [2005] RLR 1, at 16-18) DMG was not mistaken in its belief that a taxpayer which failed to make a group income election was liable to pay ACT: its mistake was to think that it was not entitled to make such an election. This mistake led DMG not to make an election, with the inexorable result that the money was due. This suggests that DMG's mistake was neither here nor there and that it shouldn’t have been allowed to recover. Otherwise a taxpayer who pays money pursuant to a completely valid tax regime might just as well say 'Oh I didn’t realise that if I put my assets offshore I wouldn’t have had to pay this tax, so now that my friendly accountant has explained this to me I should be entitled to a rebate because I paid under a mistake'.

At first instance Park J recognised the problem at [25] but sought to get round it by holding that DMG 'would have made' an election if it had known that it was entitled to do so, and then asserting that it followed from this that the money wasn't due. This was a non sequitur. In the CA Jonathan Parker LJ mistook the facts when he held at [231] that 'the regime gave rise to no obligation to pay' because it was unlawful the only bit of the statute which was unlawful was the bit which disabled DMG from making an election; DMG's liability to pay MCT in the event that it failed to make an election arose under a statutory section that was not unlawful at all. In the HL Lord Hoffmann agrees with Jonathan Parker LJ at [32] and is wrong for the same reason. He also dismisses Park J's recognition of the problem as 'rather sophisticated', a criticism echoed by Lord Walker at [143] ('over-analytical'). Given the general expertise in tax law which they both undoubtedly possess (especially Lord Walker) I find this language surprising; they would have done better to heed the philosopher’s advice endorsed by Lord Hope at [41]: 'Seek simplicity and distrust it.' However, at [62] while Lord Hope correctly rejects Jonathan Parker LJ's analysis, he also agrees with Park J, and is wrong for the same reason: the fact that there was 'an unbroken causative link between the mistake and the payment' is true, but it still doesn’t mean that the payment wasn’t due (as Lord Scott correctly observes at [89]).

So I agree with Rob that much more should have been made of this - but still, I wonder whether there isn’t an answer which the majority of the HL might have given in response to Lord Scott - namely that the English courts were required by the ECJ in Metallgesellschaft at [96] to award restitution to parties in DMG's position in order to give full effect to its Article 52 rights, and that this direction mandated a departure from the rule that normally governs restitutionary claims in unjust enrichment, that they are debarred by the existence of valid statutory rules requiring the claimant to pay the defendant. I’m afraid that my EC law isn't good enough to answer this question, but perhaps there is someone out there who knows the answer: where a statutory tax regime is made up of several interconnected sections, some of which are expressly declared void for incompatibility with the Treaty, but some of which are not, can an English court override the tax authority's right to keep money paid under one of the surviving valid sections on the basis that one of the void sections would otherwise be indirectly effective, contrary to the declaration of incompatibility? (At [82] Lord Scott indicates that the answer is 'no'!)