By means of the doctrine of subrogation, one person is substituted for another in the exercise of that other’s rights against a third person. In an oft-repeated (if not always apt) metaphor, the person receiving the benefit of subrogation is said to stand in the shoes of the other as against the third party (eg, Patten v Bond (1889) 60 LT 583 (Ch) 585 (Kay J); In re Bell Lines Ltd [2006] IEHC 188 (28 April 2006) (Dunne J); Lowick Rose LLP v Swynson Ltd [2018] AC 313, [2017] UKSC 32 (11 April 2017) [62] (Lord Mance)). However, the shoes are sometimes an imperfect, even an uncomfortable, fit. So, in Banque Financière de la Cité v Parc (Battersea) Ltd [1999] AC 221, [1998] UKHL 7 (26 February 1998) Lord Hoffmann commented that “the subject of subrogation is bedevilled by problems of terminology and classification which are calculated to cause confusion”, and a great many academic and judicial computer screens have been filled in seeking to allay that confusion. Now comes news of another labourer in the vineyard: Rory Gregson Subrogation and Marshalling (Hart Publishing, 2024); and a very welcome addition to the literature it is too. I am grateful to Hart Publishing for providing me with a pdf download to review here.

As a matter of description (and making no normative claims whatsoever), there are three main occasions on which subrogation can be seen to arise: under statute (marine insurance is the leading example; see also In re Customs House Capital Ltd [2020] IEHC 510 (30 April 2020)), pursuant to contract (especially in the context of insurance; see, eg, FM Conway Ltd v Rugby Football Union [2023] EWCA Civ 418 (19 April 2023)), and by operation of law. Much of the confusion to which Lord Hoffmann refers happens with attempts to explain this third occasion on which subrogation can arise. Worse, this confusion is often then compounded by seeking to apply an explanation from the third occasion to the first two. In this third occasion, subrogation arises by operation of law where a creditor has rights against a debtor, and the plaintiff is subrogated to the rights of the creditor against the debtor. Hence, for example, where a surety has paid off the debtor’s debt to the creditor, the surety can be subrogated to the creditor’s rights against the debtor; where a lender has advanced a loan for the purchase of real property, the lender can be subrogated to the vendor’s lien against the purchaser; and where a plaintiff has paid off a mortgage, the plaintiff can be subrogated to the mortgage. Similar principles apply where lenders of invalid loans subrogate to the rights discharged by the application of the loans. In such categories, subrogation has been said to arise by operation of law (i) on settled principles, in defined circumstances, in well-recognised categories of case (without further explanation of the categories or of connections between them); (ii) in these categories, on the basis of the actual or presumed intentions of the parties; (iii) to reverse the unjust enrichment of the defendant; (iv) when reason and justice demand that it should be; and (v) at the discretion of the court. At first blush, these five explanations of when subrogation arises as a matter of law may seem very different, and disputes over their application and reach are the kinds of classification issues that are calculated to cause the confusion to which Lord Hoffmann refers. Here, there are many possible non-theories, theories and meta-theories – and, whilst they are all offered in the spirit of rationalisation to reduce confusion, they do not all succeed in the endeavour. There are many categories of subrogation: non-theories are content simply to describe them; theories try to explain them; and meta-theories try to derive or impose a single overarching explanatory hypothesis.

I have argued several times on this blog that these various formulæ are not competing principles at all, and that confusion can best be allayed by reconciling them (much as, for example, Neuberger LJ sought to do in Cheltenham & Gloucester plc v Appleyard [2004] EWCA Civ 291 (15 March 2004)). The great majority of cases where subrogation arises by operation of law will fall into one of the established categories. And many, if not most, of the categories can be explained as turning on the actual or presumed intentions of the parties – the former covers the cases where the plaintiff deals the debtor; the latter explains the cases where the plaintiff pays the creditor directly (for more on this, see the second small refinement I suggest below, and chapter 4 here (pdf)). And the intention explanation may be relied upon to extend the doctrine to new or developing categories of case. Some, at least, of the cases may be explained as reversing the unjust enrichment of the defendant; this is a further category of case not explained by, and by way of an addition to, the intention explanation; and that principle against unjust enrichment may also be relied upon the extend the doctrine of subrogation to new or developing categories of case. Where subrogation arises by operation of law, an approach that starts with the categories, and works out first to the actual or presumed intentions of the parties, and then – in the alternative – to restitution for unjust enrichment, defines both when reason and justice demand that subrogation should arise and when the discretion of the court should be exercised in its favour. In this way, the various formulæ may be reconciled to operate, not in competition, but in collaboration with each other.

Most workable theories of subrogation are built upon such reconciliations, though they are not straightforward. Moreover, when such a reconciliation is eschewed in favour of assertions either of a non-theory that no reconciliation is possible or of a meta-theory that only one of the formulæ correctly reconciles the full sweep of the doctrine of subrogation, that analysis becomes even more difficult and confusing. This is not to say that there are not able elucidations of such positions. For example, on the one hand, Dyson Heydon, Mark Leeming & Peter Turner (eds) Meagher, Gummow & Lehane’s Equity: Doctrines and Remedies (5th ed, LexisNexis Austraia, 2015) [MGL] (chapter 4) embrace an atomistic approach that focusses on the existing categories without seeking to generalise in any way from them, and the High Court of Australia followed the lead of that non-theory in Bofinger v Kingsway (2009) 239 CLR 269, [2009] HCA 44 (13 October 2009). On the other hand, Charles Mitchell has long argued that subrogation is restitutionary, responding to unjust enrichment (see, eg, Charles Mitchell The Law of Subrogation (Clarendon Press, 1995); Charles Mitchell & Stephen Watterson Subrogation. Law and Practice (OUP, 2007)), and Lord Hoffmann followed the lead of that meta-theory in BFC v Parc (above); see also Bank of Cyprus UK Ltd v Menelaou [2016] AC 176, [2015] UKSC 66 (4 November 2015) (Lord Clarke; noted by Cleaver and Watterson on this issue); Revenue & Customs v Investment Trust Companies [2018] AC 275, [2017] UKSC 29 (11 April 2017); Lowick Rose v Swynson (above)). Nor is this to say that one or the other of the MGL/Bofinger atomistic non-theory or the Mitchell/BFC unjust enrichment meta-theory is necessarily wrong. This simply to say the competition of well-articulated non-theories, theories, and meta-theories, of subrogation, is a recipe for confusion, and that any approach to resolve the confusion faces a very high bar to do so.

Striding boldly into this debate is Rory Gregson [pictured right] Subrogation and Marshalling (Hart Publishing, 2024)) [building upon his Oxford DPhil thesis (2021) (pdf)]. First, following Mitchell, he acknowledges the distinction between (simple) subrogation to subsisting rights and (reviving) subrogation to extinguished rights, and confines his analysis to the latter. This, quite properly, puts insurance subrogation and other examples of contractual subrogation to one side; it ensures that there is no cross-pollination, or even cross-contamination, between that context and other cases of subrogation; and it directs the analysis to the key issue of when subrogation can arise by operation of law. I, therefore, entirely agree with this first stage in Gregson’s reasoning.

Second, in part 2 of the book, chapters 3 to 7, he argues that subrogation is not a remedy for unjust enrichment, since it is not governed by the same restitutionary rules as other remedies for unjust enrichment. A similar argument has been made in Ireland by Daniel Donnelly The law of subrogation (TCD PhD thesis, 2000) (see also Nathan Tamblyn “Separating Unjust Enrichment and Subrogation” (2017) 44 Exeter Law Review 1 (pdf)). In Bank of Cyprus v Menelaou (above) [108], Lord Cranwarth was less convinced than Lord Clarke

with respect of the case for “rationalising” the older cases “through the prism of unjust enrichment”, as Lord Clarke suggests was done in BFC v Parc (above), thus in effect conflating the two doctrines. As Lord Millett explained in Foskett v McKeown [2001] 1 AC 102, 129, there is a clear distinction of principle between a claim to enforce property rights and a claim for unjust enrichment. Earlier in the same judgment (at p 127F) he had emphasised that property rights are to be determined “by fixed rules and settled principles”, not by discretion or policy. Subrogation to a vendor’s lien is a claim to a property right, but it is, as Lord Clarke acknowledges, a less than straightforward concept. It should not be extended, nor should the established rules be distorted, without good reason.

Similarly, although Lord Neuberger expressly agreed with Lord Clarke, he was also attracted to the view that that the outcome “could be justified on the alternative basis of an orthodox proprietary claim rather than on unjust enrichment” ([59]). And in the subsequent Lowick Rose v Swynson (above) [30] Lord Sumption considered that it would be “unwise … to try to fit the subrogation cases into any broader category of unjust enrichment”.

Of course, the matter will not be resolved by duelling dicta. However, Gregson’s careful analysis supports these dicta. In particular, in chapter 3, he argues that, the existing categories of subrogation by operation of law do not reverse a defendant’s factual enrichment (see also Lowick Rose v Swynson (above) [30] (Lord Sumption)); and, insofar as subrogation might be said to prevent unjust enrichment, it responds to quite different notions of enrichment than are to be found in direct restitutionary claims. Similarly, in chapter 4, he demonstrates that subrogation does not respond to the requirement that the defendant be enriched “at the expense of the plaintiff” that emerges from ITC (above; though Irish law might differ in this regard). In chapter 5, he demonstrates that, whilst many of the cases in the existing categories can be explained as turning on restitutionary causes of action, such as mistake or failure of consideration, many others cannot. He returns to the inadequacy of such causes of action in chapter 8. In chapter 6, he demonstrates that there is no clear authority that restitutionary defences – such as, in particular, change of position – apply to the existing categories of subrogation. Hence, in chapter 7, he demonstrates that the effects of subrogation are quite different to those of restitution for unjust enrichment. Consequently, Gregson, quite properly, puts the restitutionary meta-theory of subrogation aside. Subject to the first small refinement I suggest below, I entirely agree with this second stage in Gregson’s reasoning.

Third, he addresses the principles underlying the well-recognised categories of subrogation arising by operation of law. In chapter 1, “What is Subrogation”, Gregson fairly and unjudgmentally describes the contexts in which subrogation arises. Furthermore, in chapter 8, “Previous Attempts to Justify Subrogation”, Gregson carefully analyses leading non-theories, theories, and meta-theories, that have been advanced in this context, including unjust enrichment, unconscionably, and general principles of equity. To the extent that he puts these aside too, I agree with this third stage in Gregson’s reasoning. However, to the extent that he also sets aside the intention explanation and certain dicta of Lord Sumption, I reserve my position for the discussion of the second small refinement I suggest below.

Fourth, having put aside what he regards as all of the previous attempts to justify subrogation, in chapter 10, “Properly Distributing the Burden of a Debt”, Gregson suggests a theory of his own, though only for the categories of subrogation by operation of law. He argues that, in those categories, subrogation ensures the proper distribution of the burden of debts. It is a fair and workable description of the law built upon a careful and good faith reading of the authorities. It is certainly a better explanation for subrogation than reason and justice, or judicial discretion, or (subject a small point I make below) restitution for unjust enrichment. And it avoids the MGL/Bofinger objection that the Mitchell/BFC/unjust enrichment thesis is imposed upon the cases from the top town, rather than arising organically. So, subject to the second small refinement I suggest below, I also agree with this fourth stage in Gregson’s reasoning.

It is important to appreciate the extent of this theory. It is not a meta-theory, insofar as it does not attempt to include all of occasions on which, as a matter of description, subrogation can be observed as arising: under statute, pursuant to contract, and by operation of law. Instead, it is confined to cases where subrogation arises by operation of law. Hence, explaining these categories as turning on the proper distribution of the burden of debts avoids the confusion to which Lord Hoffmann refers in BFC v Parc (above). Moreover, it can be accommodated in the blend of approaches with which this post began. It is an excellent advance on the explanations that have gone before, and it has deepened our understanding of the categories of subrogation arising by operation of law. Nevertheless, I think Gregson’s approach can be refined in two small ways.

My first small suggested refinement focusses on principles of restitution for unjust enrichment. Whilst I agree with Gregson – and Donnelly – that such principles are inapt to provide a meta-theory that rationalises the law in the existing categories of subrogation, I think that this does not mean that such principles cannot underpin a new category. So, rather than attempting to explain the full sweep of subrogation on foot of a meta-theory of restitution for unjust enrichment, BFC should be seen, not as a reinterpretation of the existing categories, but far more narrowly simply as the basis for an additional category of subrogation. There are hints of a distinction between “a cause of action based on unjust enrichment” and “subrogation following the discharge of a debt, which is arguably based on a different principle” in Prudential Assurance Company Ltd v Revenue & Customs [2018] UKSC 39 (25 July 2018) [68]; see also Menelaou [108] (Lord Cranwarth) (above)); and this distinction could, with profit, be built upon. Moreover, in Mitchell’s hands, some of the older intention cases may indeed be better explained as responding to unjust enrichment; and, even if this re-explanation does not compel all of the traditional categories to be understood in unjust enrichment terms, some at least of those cases could be accommodated within the new unjust enrichment category of subrogation without destabilising the existing categories (an example may be the blending of unjust enrichment principles with the the surety’s right to subrogation to the principal creditor’s rights against the principal debtor in Lehman Brothers Holdings Scottish LP 3 v Lehman Brothers Holdings plc [2021] WLR(D) 537, [2021] EWCA Civ 1523 (20 October 2021) (Lewison LJ)).

Of course, for the reasons that Gregson gives that unjust enrichment does not easily map onto subrogation, there will be very few cases in this category, but it must not be excluded. In the Irish Court of Appeal in Comcast International Holdings Inc v The Minister for Public Enterprise [2021] IECA 325 (01 December 2021) Haughton J described this category variously as “subrogated restitution” ([28]), “restitutionary subrogation” ([45], [74], [77]), “subrogated restitutionary relief” ([47}(2)), “restitutional subrogation” ([66]), or “restitution subrogation” ([105]); and there are some examples in the caselaw. In Niru Battery Manufacturing Company v Milestone Trading Ltd (No 2) [2004] 2 All ER (Comm) 289, [2004] EWCA Civ 487 (28 April 2004) [52]-[62], Clarke LJ used the BFC unjust enrichment analysis to justify subrogation in circumstances beyond the recognised categories of case in which subrogation has been recognised. Again, in Croxen v Gas And Electricity Markets Authority [2022] EWHC 2826 (Ch) (11 November 2022), without reference to the existing categories, Zacaroli J considered that the plaintiff had a claim in unjust enrichment against the defendant, satisfied by an equitable right of subrogation to the defendants’ customers’ claims against the defendant.

So, I disagree with Gregson and Donnelly that principles of restitution for unjust enrichment must be entirely displaced here. Even if they must be abandoned as a meta-theory explaining the full breadth of the doctrine of subrogation, my first small refinement to Gregson’s analysis is that principles of restitution for unjust enrichment can quite happily supply the analytical basis for a new, post-BFC, category of subrogation, separate and distinct from, and additional and complementary to, the existing categories which he suggests turn on the proper distribution of the burden of a debt.

My second small suggested refinement focuses on the intentions of the parties. The intention explanation is not without its difficulties (see, eg Jennifer Palmer “Unjust Enrichment, Proprietary Subrogation and Unsatisfactory Explanations” (2016) 28 Singapore Academy of Law Journal 955-983 (SSRN); Bolwell v NWC Finance Pty Ltd [2024] VSC 30 (9 February 2024) (Osborne J)). But it is repeatedly advanced in the cases. Before the Mitchell/BFC unjust enrichment analysis muddied the waters, the existing well-recognised categories of subrogation arising by operation of law were traditionally said turn on the actual or presumed intentions of the parties (see, eg, Highland Finance (Ireland) v Sacred Heart College of Agriculture [1998] 2 IR 180 (SC) (Blayney J)). In chapter 8, Gregson deals with the intention explanation in very brief compass, and finds it wanting. He concludes that, in BFC, “the House of Lords rightly rejected intention as the justification for subrogation” (94; see BFC: there is a “danger that the contractual requirement of mutual consent will be imported into the conditions for the grant of the restitutionary remedy” (Lord Hoffmann), so that “the concept of mutual intention, whether actual or presumed, can be artificial” (Lord Hutton)).

However, as with the unjust enrichment explanation (above), this matter will not be resolved simply by duelling dicta, and I think that there is more to be said about the intention explanation. First, let us distinguish between the two species of intention. On the one hand, the presumed intentions of the parties can be understood to explain the cases where the plaintiff pays the creditor directly. In such cases, the limiting factor which justifies the presumption is the plaintiff’s status as surety or owner of an interest in property, where it is more than appropriate to presume that the parties intended the plaintiff would be subrogated to the creditor’s security against the debtor; and, when the courts embark upon an analysis of the particular facts of the case in these direct payment cases, they are seeking to determine whether that presumed intention has been rebutted. On the other hand, the actual intentions of the parties can be understood to explain the cases where the plaintiff deals not with the creditor but with the debtor, and the plaintiff’s right to subrogation turns on the actual intentions of the plaintiff and the debtor. In such cases, rather than the status of the claimant, the limiting factor is instead seen to be the actual intention of the parties; and, when the courts embark upon an analysis of the particular facts of the case in these indirect payment cases, they are seeking to determine the parties’ actual intentions. Hence, in both the direct-payment cases where the claimant’s status generates the presumption of intention, and indirect-payment cases which turn on the actual intentions of the parties, subrogation arises as a matter of law (for more on this distinction, see chapter 4 here (pdf)).

Gregson dismisses the latter cases, where the courts seek the actual intentions of the parties, as fictions, on the basis that law now recognises that subrogation arises by operation of law (15). However, this gets the issue the wrong way round: rather, subrogation arises by operation of law in these cases because it represents the actual intentions of the parties. In BFC, Lord Hoffmann considered that it was important to avoid the danger that the concept of “mutual consent” might be “imported into the conditions for the grant of the restitutionary remedy”. I agree. But that is because I think that intention explains the traditional categories of subrogation, that restitution for unjust enrichment underpins a new category, and that these are separate doctrines subject to different principles. In Lowick Rose v Swynson ([30]) Lord Sumption suggested that “the real basis” of (equitable) subrogation “is the defeat of an expectation of benefit which was the basis of the payer’s consent to the payment of the money for the relevant purpose”. This principle of the defeat of an expectation of benefit serves, not as a restitutionary cause of action, but as an explanation of some at least of the actual intention cases. In chapter 8, Gregson quite properly rejects this as an explanation for all cases of subrogation, but I think it can be used in the much narrower context to explain some at least of the actual intention cases. So, my second refinement of Gregson’s analysis would be either to carve the actual intention indirect payment cases out of Gregson’s theory that subrogation ensures the proper distribution of the burden of debts, or to accommodate it within that theory by glossing Lord Sumption to say that the actual intention that has to be found on the facts is that the parties actually intended to achieve the proper distribution of the burden of the debt. The former would hollow out Gregson’s theory, so the latter may be the better route.

As to the cases where intention is presumed, where the plaintiff deals with the directly the creditor, the intention that is presumed can be said to be that the parties intended to achieve the proper distribution of the burden of the debt. In other words, Gregson’s theory that subrogation ensures the proper distribution of the burden of debts is an excellent explanation of the cases in which subrogation is said to arise on the basis of the presumed intentions of the parties. So, whilst I entirely agree with Gregson’s analysis of the flaws in many of the previous attempts to justify subrogation, I think that his analysis is incomplete insofar as it fails to engage fully with the intention explanation, and in particular with the distinction between the actual and presumed intentions of the parties. But, as the previous paragraph suggests, this distinction can be accommodated within his analysis.

Incorporating these two suggested refinements – distinguishing between restitution and intention and explaining the non-restitutionary categories of subrogation arising by operation of law as turning on the (actual and presumed intentions of the parties to achieve the) proper distribution of the burden of the debt – still avoids the confusion to which Lord Hoffmann refers in BFC v Parc (above).

One of the difficulties in the way of any coherent explanation of those categories is a distinction between the law and equity. For example, where the rules underpinning these categories are said to arise both at law and in equity, there is still resistance to aligning the rules between law and equity, and insistence that ne’er the twain shall meet – on this view, gaps in one category on one side of the line cannot be filled by analogy with rules in another category on the other side of the line. Again, there are many statements that subrogation arises only on equitable principles, but it is impossible to accept these claims when they arise exclusively from equitable exceptionalism and are not otherwise justified by the nature of the categories of subrogation at issue. In Lowick Rose v Swynson (above), Lord Sumption effectively treated all of the categories of subrogation by operation of law as equitable, and this may be the best solution.

Sometimes, the equitable principle engaged here is described as unconscionability (Boscawen v Bajwa [1996] 1 WLR 328, [1995] EWCA Civ 15 (10 April 1995) (Millett LJ)); but if it is a synonym for reason and justice, or discretion, then, for the reasons Gregson gives in chapter 8, it should be eschewed. As Lord Sumption put it in Lowick Rose v Swynson ([34]), “the law of equitable subrogation” should not be transformed “into a general escape route” from relevant principles of the law. Rather, the general equitable principles should be treated either, in the existing categories, as responding to the (actual and presumed intentions of the parties to achieve the) proper distribution of the burden of the debt, or, in some new cases, as providing restitution for unjust enrichment.

Subject to the two refinements I suggest above, Gregson’s proposal – in the existing categories of subrogation that turn on the actual or presumed intentions of the parties, subrogation enables the proper distribution of the burden of a debt – is an elegant rationalisation of a wide range of difficult cases. And it fits comfortably with the blend of approaches with which this post began. His book makes for an excellent belated Christmas present for the subrogation lawyer in your life. Highly recommended.