Last month, officials at the Illinois Department of Revenue said that there had been overpayments since 2014 totalling US$168 million to some 6,500 districts across Illinois. The Department says that it’s sensitive to the impact recovering the funds will have on the districts, and it will therefore allow them to repay the money slowly. But repay it they must, unless legislation forgives the debts. It is a stark example of the issues which I have been discussing on this blog in a series of posts on the principles underlying restitution claims made by the Executive. The series was prompted by media reports that the taxpayer has been hit with a €70m bill for overpaid farmers. In my first post in the series, I established the principle associated with Auckland Harbour Board v R [1924] AC 318; [1923] UKPC 92, [1923] NZPC 3 (18 December 1923) [Auckland] and Attorney General v Great Southern and Western Railway Company of Ireland [1925] AC 754 (HL) [GSWR]. This principle has two limbs: first, State payments must be authorised (this is the authorisation limb of the principle, which I discussed in my second and third posts); and second, unauthorised State payments can be recovered if they can be identified (this is the restitution limb of the principle, which I discussed in my fourth post). This is the fifth post; there will be one more post, for a total of six in the series on the Auckland/GSWR principle; and there will – eventually – be a footnote and an addendum on related issues). The restitution limb of the principle was established in the fourth post; it is a claim to restitution of unjust enrichment, because the recipient of the unauthorised payment has been unjustly enriched at the expense of the State; and I want to look in this post at three questions to which that claim gives rise: (i) what payments are caught by that limb; (ii) how do defences to that claim work; and (iiii) is there an imperative obligation on the State to seek to recover unauthorised payments.

The first of the three issues in this post is that there ought to be no limitation on the type of State payments which, if unauthorised, can be recovered. There is nothing in either limb of the principle, or in the authorities, to suggest that there should be such a limitation. Indeed, the constitutionalisation of the first – authorisation – limb of the principle strongly indicates there there should be no such limitation. So does analogical authority. In Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 (pdf) 177, Lord Goff derived the principle that overpaid tax can be recovered by the citizen as the converse of the second – restitution – limb of the Auckland/GSWR principle. In Child Poverty Action Group v Secretary of State for Work and Pensions [2011] 2 AC 15, [2010] UKSC 54 (08 December 2010) [16] Lord Brown acknowledged that the analogy also held the other way, so that considerations relating to the Woolwich principle could equally apply to the restitution limb of the Auckland/GSWR principle. It is clear that Woolwich principle applies to overpayments of all charges and levies by public bodies and is not restricted to overpayments of tax or of specific kinds of tax (see, eg, O’Rourke v Revenue Commissioners [1996] 2 IR 1 (Keane J); British Steel plc v Customs and Excise Commissioners [1997] 2 All ER 366; Waikato Regional Airport Ltd v Attorney General [2004] 3 NZLR 1, [2006] NZPC 7, [2003] UKPC 50 (30 June 2003) [80]; Harris v Quigley [2006] 1 IR 165, [2006] 1 ILRM 401, [2005] IESC 79 (01 December 2005); R (on the application of Hemming (t/a Simply Pleasure Ltd)) v Westminster City Council (Rev 1) [2013] EWCA Civ 591 (24 May 2013); Lindum Construction Co Ltd v The Office of Fair Trading [2014] EWHC 1613 (Ch) (19 May 2014) [84]). It means, for example, that there is no reason why the restitution limb of the Auckland/GSWR principle should not apply to the recovery of overpaid social welfare payments. As Refshauge J put it in Commonwealth of Australia v Davis Samuel Pty Ltd and Ors (No 7) [2013] ACTSC 146 (1 August 2013) [1731](1): “The absence of Parliamentary authority extends beyond the mere absence of a relevant appropriation and applies to any disbursement from the Consolidated Revenue Fund that is ultra vires for whatever reason …”.

Indeed, many unanswered questions relating to the restitution limb of the Auckland/GSWR principle might be settled by analogy with conclusions reached relating to the Woolwich principle. For example, in Deutsche Morgan Grenfell Group v Inland Revenue [2007] 1 AC 558, [2006] UKHL 49 (25 October 2006), the House of Lords that there is nothing in Woolwich to preclude the plaintiff from being able to assert other grounds for restitution, such as mistake, or vice versa. Nor should the State be confined to the restitution limb of the Auckland/GSWR principle if another ground for restitution, such as mistake, is also available, or vice versa. Newton J effectively held as much in Commonwealth v Burns [1971] VR 825, [1971] VicRp 100 (25 May 1971); Refshauge J in Commonwealth of Australia v Davis Samuel Pty Ltd and Ors (No 7) [2013] ACTSC 146 (1 August 2013) held that the Commonwealth could rely on either mistake or the restitution limb of the Auckland/GSWR principle, at its choice; and the analogy with Woolwich and DMG entirely justifies such a conclusion. Again, it is a matter of interpretation whether a statutory scheme for the recovery of overpaid tax excludes common law claims based on mistake or Woolwich (see, eg, Littlewoods v HM Revenue and Customs [2015] 3 WLR 1748, [2015] EWCA Civ 515 (21 May 2015)). In the Child Poverty Action Group case, the House of Lords held that it is similarly a matter of interpretation whether a statutory scheme for the recovery of overpaid State payments excludes common law claims based on mistake or the restitution limb of the Auckland/GSWR principle. It means, for example, that even if the restitution limb of the Auckland/GSWR principle can apply to the recovery of overpaid social welfare payments, legislation can provide that this claim is excluded and that the only basis for such recovery is a scheme provided by statute. This is the holding in the CPAG case (see also Minister for Social, Community and Family Affairs v Scanlon [2001] 1 IR 64, [2001] 2 ILRM 342, [2001] IESC 1 (16 January 2001); on the operation of the Irish scheme, see here and here).

The second of the three issues in this post is that the nature of the unjust enrichment claim in the restitution limb of the Auckland/GSWR principle has an impact on available defences. In Commonwealth v Burns [1971] VR 825, [1971] VicRp 100 (25 May 1971), (discussed in detail in my previous post), Newton J held that the Commonwealth could rely on Auckland to seek to recover overpaid pension, and continued:

… it was finally submitted that the Commonwealth was estopped from alleging that the payments in question had been made to her without lawful authority. … It was further submitted that Mrs Burns had spent the money now sought to be recovered in reliance upon the representation, so that a requirement that she repay the money would operate unjustly to her detriment … But a sufficient answer to this submission is, in my view, to be found in the well-established rule that a party cannot be assumed by the doctrine of estoppel to have lawfully done that which the law says that he shall not do: as earlier stated, nobody had, or could have had, any lawful authority to make the payments in question to Mrs Burns …

This has been followed in many subsequent cases (see, eg, AG v Gray [1977] 1 NSWLR 406 (NSW CA) 409-410; Formosa v Secretary to the Department of Social Security [1988] FCA 291 (17 August 1988) [45]; Roberts v Repatriation Commission [1992] FCA 642 (23 December 1992) [9]). In Commonwealth of Australia v Davis Samuel Pty Ltd and Ors (No 7) [2013] ACTSC 146 (1 August 2013) Refshauge J went further. He found that payments by the Commonwealth amounting to Aus$8.725million were recoverable, and he held that neither estoppel nor change of position could provide a defence:

… Unlike other species of restitution, if a payment is established to be unauthorised by Parliament, then a representation to the contrary cannot ground any defence based upon estoppel. This is because a representation of a state of affairs that is contrary to law cannot ground an estoppel …

Equally no change of position defence is available to a defendant subject to an action in restitution. This is because a change of position defence requires it to be shown that a person acted to his or her detriment on the faith of the receipt … Reliance cannot be placed upon the faith of a payment which was prohibited by law. ([1731](2)&(3); see also [1753]).

This is entirely consistent with well established principles that estoppel by representation has no application in public law (Arklow Holidays Ltd v An Bord Pleanala [2011] IESC 29 (21 July 2011) (Finnegan J); R v East Sussex County Council, ex parte Reprotech (Pebsham) Ltd [2003] 1 WLR 348, [2002] 4 All ER 58, [2002] UKHL 8 (28th February, 2002) [33] (Lord Hoffmann)), at least in the sense that a plea of estoppel cannot prevail as an answer to a well-founded claim that something done by a public body is ultra vires (Greendale Building Company v Dublin County Council [1977] IR 256 (Henchy J); Ashbourne Holdings v An Bord Pleanala [2003] 2 IR 114, [2003] 2 ILRM 446, [2003] IESC 18 (10 March 2003) [70] (Hardiman J); R (on the application of Anufrijeva) v Secretary of State for the Home Department [2004] AC 604, [2003] UKHL 36 (26 June 2003) [35] (Lord Steyn)).

We have seen that the authorisation limb of the Auckland/GSWR principle has been afforded constitutional status, especially in Australia. These considerations relating to estoppel suggest that the restitution limb of the principle probably also turns on the public law considerations behind the incapacity rather than on the incapacity itself. These public law considerations are clearest when they compel the denial of ordinary restitution defences, as illustrated in Burns and Davis Samuel. And, given the analogies between the restitution limb of the Auckland/GSWR principle and the Woolwich principle, it should come as no surprise to notice that the change of position defence is not available against Woolwich claims. Prof Elise Bant has argued that this can be justified by the “stultification” bar to the application of the defence:

This arises whenever the application of a particular rule would undermine an overriding policy of the law. In the context of change of position, the bar entails that the defence should be denied where its recognition would stultify the policy reason for ordering restitution.

Stultification ought to be the most significant hurdle to public authorities seeking to rely on the defence in circumstances where they have received a benefit pursuant to an unlawful demand. The Woolwich principle rests on judicial recognition of a longstanding and vitally important public interest in prohibiting unlawful demands by public authorities. The policy of prohibiting such demands will be undermined if a public authority is effectively permitted to keep the benefit of the unlawful demand by relying on the defence of change of position. (see “Restitution from the Revenue and Change of Position” [2009] LMCLQ 166, 172).

This reasoning was adopted by Henderson J in Test Claimants in the FII Group Litigation v HM Revenue and Customs [2014] EWHC 4302 (Ch) (18 December 2014) [310]-[315], who held that “to allow scope for the defence would unacceptably subvert, and be inconsistent with, the high principles of public policy which led to recognition of the Woolwich cause of action as a separate one in the English law of unjust enrichment”. This reasoning is equally applicable to explain the stultification of estoppel and change of position defences to claims on the restitution limb of the Auckland/GSWR principle.

The stultification policy would seem to apply only where the cause of action is itself policy-based – policy considerations relating to the security of the public purse underpin the Woolwich principle and both limbs of the Auckland/GSWR principle and thus justify the stultification of defences which undermine the policy basis of the cause of action. But the stultification policy would seem to apply to other causes of action which might be available as alternatives to Woolwich and Auckland/GSWR, which is one of the attractions of such causes of action to plaintiffs. Hence, in Murphy v AG [1982] IR 241 (rtf) 319 Henchy J held that

It is one of the first principles of the law of restitution on the ground of unjust enrichment that the defendant should not be compelled to make restitution, or at least full restitution when, after receiving the money in good faith, his circumstances have so changed that it would be inequitable to compel him to make full restitution.

The Supreme Court held that provisions of income tax legislation by which married couples paid more tax than unmarried couples were unconstitutional. O’Higgins CJ held that this striking down should be prospective only, but the other four members of the Supreme Court rejected this approach. Henchy J (Griffin and Parke JJ concurring) held that the plaintiffs and those in similar circumstances had restitution claims in duress and mistake to recover their overpaid taxes, but that the State could nevertheless rely on the defence of change of position against most of those claims. Kenny J also held that the plaintiffs had a claim in restitution, but did not consider whether the State could rely on the defence of change of position. Henchy J took an extremely generous understanding of when the State’s receipt is in good faith: in his view, it was only after each individual taxpayer objected that the State’s receipt of that taxpayer’s tax was not in good faith. And second, he took the view that the mere fact of the State’s expenditure was a sufficient change of position. These two factors combined to defeat the vast majority of possible claims. The rise of the policy-based Woolwich claim, and its concomitant stultification of the defence of change of position are developments subsequent to Murphy but not inconsistent with it. Murphy illustrates that where the cause of action is not policy-based (as it is not in the case of duress or mistake), then the state will not be precluded from asserting defences such as estoppel or change of position. But where the cause of action is policy-based (as it is in the case of Woolwich or Auckland/GSWR), then the defendant (whether the state in a Woolwich claim, or the citizen in an Auckland/GSWR claim) will be precluded from asserting defences such as estoppel or change of position. This is expressly stated by Henderson J in the Woolwich case of FII, and it explains the approach of Newtown J in the Auckland/GSWR case of Burns.

There are some dicta which do not consider the stultification of the defences. For example, in Social Services Appeal Board v Butler 1996 CanLII 6607 (NL SCTD) (29 March 1996) Barry J based the recovery of overpaid social assistance benefits on Auckland, but he also envisaged that the defence of change of position could apply against that claim. In Charles Terence Estates Ltd v Cornwall Council [2011] EWHC 2542 (QB) (07 October 2011), Cranston J held that the defendant Council had a restitutionary claim to recover rent paid to the plaintiffs on foot of ultra vires leases. He held that there was no question of any estoppel arising to give effect to the leases ([95] referring, inter alia, to Reprotech (above)); but, rather oddly, he also held that the payees could rely on the defence of change of position ([96]-[100]). The Court of Appeal ([2013] 1 WLR 466, [2012] EWCA Civ 1439 (13 November 2012)) held that the leases were valid, and did not need to reach the restitution issues. These dicta should not be followed; the weight of authority and principle is to the opposite effect: the policies underlying the restitution limb of the Auckland/GSWR principle stultify the operation of the estoppel and change of position defences.

The third of the three issues in this post is that the constitutionalisation of the authorisation limb of the principle raises the question whether the restitution limb is or ought to be similarly constitutionalised. For example, in Murphy v AG [1982] IR 241 (rtf) 313 Henchy J held that, in the case of an unconstitutional statute

Once it has been judicially established that a statutory provision enacted by the oireachtas is repugnant to the Constitution, and that it therefore incurred invalidity from the date of its enactment, the condemned provision will normally provide no legal justification for any acts done or left undone, or for transactions undertaken in pursuance of it; and the person damnified by the operation of the invalid provision will normally be accorded by the Courts all permitted and necessary redress.

There is no material distinction between an unconstitutional exercise of legislative power (in the purported enactment of an unconstitutional statute) and an unconstitutional exercise of executive power (in unauthorised public expenditure) – they are both unconstitutional, and the Courts will afford all necessary redress. In the case of unauthorised public expenditure, the redress consists in the recovery of such expenditure pursuant to the restitution limb of the Auckland/GSWR principle. The constitutionalisation of the first – authorisation – limb of the principle strengthens the policy underlying the restitution claim in the second limb. That in turn strengthens the policy-based stultification of defences such as estoppel and change of position. And that raises the question of whether the restitution limb is or ought to be constitutionalised its turn. And that, in turn, raises the question of whether unauthorised State payments not only can, but also must, be recovered.

For example, in Collins v Minister for Finance [2013] IEHC 530 (26 November 2013) a Divisional Court of the High Court (Kelly, Finlay Geoghegan and Hogan JJ, in a joint judgment delivered by Kelly J) dismissed the challenge to section 6 of the Credit Institutions (Financial Support) Act 2008. The Court concluded that Articles 11, 15, 17, 21, 22 and 28 of the Constitution, guard against unauthorised over-spending by the Government. It should not be possible to set these strict protections at naught simply by making the unauthorised over-spending and then deciding not to recover it. The obligation to recover such unauthorised over-spending is a necessary corollary of the very strict protections provided by this matrix of constitutional provisions. The facts of Collins almost, but not quite, presented this issue. The Court held that if “the Minister had acted wrongfully or in an unconstitutional fashion by issuing the promissory note in the first instance, then there would have been no lawful liability or obligation to the Central Bank”, and further dealings with the note would also have been unlawful ([31]). If the Minister had acted wrongfully or in an unconstitutional fashion by making any payments, then those payments would have been unauthorised for the purposes of the first limb of the Auckland/GSWR. The question of the recovery of such unauthorised payments would have arisen under the second – restitution – limb of the principle, and that would have provided the perfect context to conclude that such unauthorised State payments not only can, but also must, be recovered.

Another issue arising from the constitutionalisation of the restitution limb of the Auckland/GSWR principle is suggested by a consideration of the legislation proposed in Illinois to limit Auckland/GSWR claims and to forgive the debts of recipients of unauthorised state payments. In particular, the question arises whether such legislation is constitutional. For example, in Re Article 26 and the Health (Amendment) (No 2) Bill 2004 [2005] 1 IR 105, [2005] IESC 7 (16 February 2005) the Supreme Court held that provisions of a Bill which retrospectively abrogated the claims to restitution of unlawful nursing home charges were unconstitutional. Given the analogies between the Woolwich principle and the restitution limb of the Auckland/GSWR principle, if a statute retrospectively abrogating Woolwich claims is unconstitutional, it is likely that a statute retrospectively abrogating restitution claims under the second limb of the Auckland/GSWR principle would also be unconstitutional for the same reasons.

However, even though an arbitrary removal of the cause of action would be unconstitutional, and even though defences such as estoppel and change of position would be stultified, the courts might find some other limits to such claim. For example, in Murphy, Henchy J acknowledged that “considerations of economic necessity, practical convenience, public policy, the equity of the case, and suchlike matters, may require that force and effect be given in certain cases to transactions carried out under the void statute” (322; see also Griffin J at 328) and the Supreme Court in A v The Governor of Arbour Hill Prison [2006] 4 IR 88, [2006] 2 ILRM 481, [2006] IESC 45 (10 July 2006) has given full effect to this latter principle. Hence, in DPP v JC [2015] IESC 31 (15 April 2015) [68], O’Donnell J observed that in A

… the Court rejected an argument that once an act had been declared unconstitutional it necessarily followed that all steps taken on foot of it were void. As Murray CJ, as he then was, observed, “the law is too old and too wise to be applied according to a rigid abstract logic or a beguiling symmetry”. In this and in other areas the life of the law has not been logic but experience. That may involve recognising that a number of different values are involved. It is of course possible to present an argument in favour of an absolute rule as somehow more simple and therefore (and more dubiously) a more principled approach, but there are many problems which are complex and which require nuanced solutions. As Albert Einstein is reputed to have said, things should be made as simple as possible, but not more so. …

The potential constitutionalisation of the restitution limb of the Auckland/GSWR principle gives rise to a great many questions, and there are some indications of the principles and policies on which resolutions might be based, but there are as yet very few clear answers. My first post in this series was prompted by media reports that farmers had received unauthorised EU payments from the Department of Agriculture. My second post established the common law principle that, to be valid, state payments must be authorised; and my third post considered whether this principle has constitutional status by virtue of Article 11 of the Irish Constitution (by analolgy with a consideration of cases on sections 81 and 83 of the Australian Constitution). These posts established the first – authorisation – limb of the Auckland/GSWR principle. And, if the media reports are accurate, then the payments to the farmers were unauthorised and thus in breach of that limb of the principle. In my fourth post, I considered the ambit of a common law claim to recover such overpayments, suggesting that it seems that the farmers would be liable to such a claim. In this post, I discussed authorities that would deny the farmers the benefit of the defences of estoppel and change of position. These posts established the second – restitution – limb of the Auckland/GSWR principle. Again, if the media reports are accurate, then the farmers are liable to repay unauthorised EU payments received from the Department of Agriculture, and they are unlikely to be able to rely on defences of estoppel and change of position to belay or reduce that liability.

Furthermore, in this post, I also raised the question whether this claim should be constitutionalised, and whether this would then lead to the conclusion that the State must seek to recover the overpayments. If so, then it would be not be for the Minister to conclude that repayments should be funded through the exchequer rather than recovered from the overpaid farmers. Matters such as the passage of time and factual difficulties in assessing who was overpaid and by how much are not matters for Ministerial discretion; rather, they are quintessentially matters for the courts to decide on the basis of legal principles, and if such principles stultify some of the defences that might otherwise be available, so be it. Similar overpayments to Gardaí (Irish police officers) give rise to similar issues, which will be considered in an additional footnote post. Finally, overpayments to Irish farmers and Gardaí are not the only contexts in which these restitution issues have recently been in the news; an interesting example has recently arisen in South Africa; and I will discuss it in my sixth and final post in this series. Finally, in an addendum, I will consider related constitutional issues in the context of Collins v Minister for Finance [2013] IEHC 530 (26 November 2013). As the Illinois example with which I began this post makes clear, these issues are ubiquitous. It is therefore vital that the principles are clear. I hope that some light at least will have been shed by this series of posts.