Generator Developments Ltd v LIDL UK GmbH

The issue on this appeal from the judgment of Mr Nicholas Lavender QC is whether Generator Developments LLP (“Generator”) is entitled to an equitable interest in land acquired for development by Lidl UK GmbH (“Lidl”). Mr Lavender decided that it was not. His judgment is at [2016] EWHC 814 (Ch). With the permission of Henderson LJ, Generator appeals. For the reasons that follow, I would dismiss the appeal.

The essence of Generator’s case is that it had reached an understanding with Lidl that Lidl would acquire the property in question for the joint benefit of both Generator and Lidl in furtherance of a joint venture development between them. Generator stood aside and allowed Lidl to buy the property on that basis. Those facts give rise to the creation of an equitable interest in the property under the principles known as the Pallant v Morgan equity. The judge rejected that case, largely on the ground that Generator had not established the necessary understanding to bring the principle into play. That was a conclusion of fact.

It is not suggested by Generator that the judge mis-stated the applicable principles of law. Rather, it is said that he did not apply them correctly to the facts that he found. The difficulty for an appellant in successfully appealing on that ground is formidable. This court is very reluctant to interfere with trial judges’ findings of fact. The recent cases on that subject are legion. In Henderson v Foxworth Investments Ltd[2014] UKSC 41, [2014] 1 WLR 2600, Lord Reed said [62] that “[w]hat matters is whether the decision under appeal is one that no reasonable judge could have reached”. At [67] Lord Reed said:

This approach applies equally in cases in which it is not said that the judge has applied the wrong legal test, but has applied the right legal test incorrectly to the facts: see Okotoks Ltd (formerly Spicerhaart Ltd) v Fine & Country Ltd[2013] EWCA Civ 672; [2014] F.S.R. 11 at [50] to [53].

I can take the primary facts largely from the judge’s careful judgment.

The property in question is Wates Way Industrial Estate, Ongar Road, Brentwood, Essex. At the material time it was owned by Sans Souci Enterprises Ltd, a company controlled by the Whight family. Mr Paul Whight was a friend of Mr Isaacs, the founder and managing director of Generator. Generator is a property development company with expertise in residential and mixed use development. The property was in use as an industrial estate, but the draft local plan allocated it for residential use.

Lidl was interested in establishing a presence on the property and instructed agents to help it acquire the site for use as a supermarket. Generator was also interested in the potential of the property for residential development. Both expressed interest to Mr Brelsford, Sans Souci’s agent. Mr Brelsford suggested to Generator that it should speak to Lidl. As a result there was a meeting between representatives of both on 29 October 2013, followed by a conference call on 4 November.

The judge also found that Generator's representatives understood that the proposed purchase required approval by both Generator's and Lidl's boards of directors and that it was not envisaged that that approval would be sought until after the joint offer had been accepted.

On 6 November 2013 Mr. Barnes on behalf of Generator wrote to Mr. Brelsford with an offer (subject to contract) to buy the property for £5,010,000. The letter, whose terms had been agreed with Lidl, began as follows:

Thereafter, it was Mr. Isaacs and Generator, rather than Lidl, who communicated with Sans Souci. This remained the case until some time after Generator's revised offer was accepted in December 2013. Generator’s offer was shortlisted by Sans Souci, but revised offers were requested. Generator made revised offers on 20 and 21 November (in terms agreed with Lidl) and a further offer on 28 November which Sans Souci accepted, subject to contract. The involvement of Mr. Isaacs and Generator made a material contribution to the success of the joint bid. The Whight family attached significance to the fact that they were dealing with Mr. Isaacs and Generator for a variety of reasons, including: Mr Whight's relationship with Mr. Isaacs; his confidence that Generator could be counted on to deliver; Generator's experience of mixed-use developments; the potential for future co-operation; and the Whight family's reputation in the local community in Brentwood (which Mr Whight considered would have been undermined by a sale to Lidl alone). These matters made Sans Souci more likely to accept the joint bid than it would have been if Lidl alone had made it, and all the more so if Generator alone had made a competing bid.

On 6 December 2013 there was a meeting at Lidl's offices between Mr Barnes and Mr Orr for Generator and Mr Beaumont and Mr Barber for Lidl, who proposed that Lidl would be the sole purchaser of the property. In telephone conversations on 11, 12 and 13 December 2013 there was discussion as to who should be named as purchaser in the Heads of Terms relating to the transaction with Sans Souci. The outcome of these discussions was that Heads of Terms which identified Lidl as the purchaser and Generator as the "Delivery Partner" were agreed by Generator and Lidl and sent to Sans Souci on 13 December 2013.

At Generator's request, on 18 December 2013 Lidl produced a first draft of Heads of Terms for a proposed agreement between them. This draft agreement went through a number of iterations. Further drafts were prepared on 5 and 10 January 2014, 5 and 17 February 2014 and 17 March and 26 March 2014. The basic structure, common to each draft, was the proposed transaction was to be conditional on the grant of planning permission for a development of the property including: (a) a retail store (with associated car parking spaces); and (b) a number of residential flats. The basic structure of the proposed transaction was that Lidl would buy the property and, if Generator succeeded in obtaining planning permission, Lidl would then sell the freehold of the property to Generator, Generator would build the store and the flats and Generator would then grant a 999-year lease of the store to Lidl. At [18] the judge noted a number of important features of the drafts.

At [62] the judge held, contrary to Generator’s case, that the requirement of board approval applied both to the proposed joint venture agreement and also to the proposed land transaction. The judge found that Mr Orr of Generator had been told of the need for Lidl’s board approval at the earlier meeting on 6 December; and he made no finding that any contrary information had been given to him or to any other representative of Generator.

On 23 December 2013 Sans Souci, Lidl and Generator entered into a lock-out agreement relating to the proposed land purchase. Lidl was described as “the Buyer” and Generator as “the Delivery Partner”. The lockout period was until 14 February 2014. The obligations normally imposed on the buyer in such an agreement were imposed on the Buyer and Delivery Partner. Generally, the Buyer and Delivery Partner were referred to together throughout the lockout agreement. Clause 6.1 of the Lockout Agreement provided as follows:

The date for exchange of contracts for the acquisition of the property was set for 14 February 2014, the date when the lockout period expired. It had been expected that there would be a simultaneous exchange of agreements between Generator and Lidl; but their negotiations were not advanced enough for that to happen. Nevertheless, on 14 February 2014 Lidl went ahead and exchanged contracts with Sans Souci for the purchase of the property for £6.81million. The purchase was completed on 17 March 2014.

The essence of the legal principle on which Generator relies takes as its starting point a pre-acquisition agreement or understanding. The critical time period is thus between 29 October 2013 and 14 February 2014. The judge examined this period in detail.

The first meeting between Lidl and Generator was on 29 October 2013. By the deadline of 6 November 2013, they had agreed that Generator would make a bid for the property which was expressed to be made "in conjunction with our joint venture partners Lidl". As Mr. Beaumont accepted, Generator was making the offer for both parties with a view to a development for both parties. Mr. Barber and Mr. Beaumont were likely to have described Lidl and Generator as "partners" throughout their discussions.

The possibility of Generator and Lidl developing the property as a joint venture was discussed in the meeting on 29 October 2013 and in the telephone conference on 4 November 2013. However, what had not been discussed at that stage was who the actual purchaser was to be. It might have been Generator; it might have been Lidl or it might have been a special purpose company. The judge concluded at [52]:

The judge found that first substantive discussions about the proposed joint venture took place at the meeting on 6 December 2013; and that the phrase “our joint venture” was frequently used. It was at that meeting that Mr Beaumont of Lidl proposed that Lidl should be the sole purchaser. The two reasons he gave were that that was the way Lidl usually did things; and that financing costs would be lower. Mr Barnes and Mr Orr of Generator accepted in the meeting that Lidl could be the sole purchaser. In practice this meant that Lidl would pay the whole of the purchase price out of its own funds; and that Generator would have no obligation to contribute to it. In addition, there was discussion at the meeting of the potential for future projects on which Lidl and Generator might work together. The judge commented that “Generator was no doubt encouraged by this, but it was merely potential for the future”.

At the time of the meeting on 6 December it was still envisaged that there would be simultaneous exchanges of contracts between Lidl and Sans Souci for the acquisition of the property, and between Generator and Lidl to regulate the terms of the joint venture. Thus the judge found at [95]:

By 10 December Generator had become aware that within Lidl there were different views on who the development partner should be. Although Mr Beaumont and Mr Barber of Lidl favoured Generator, others within Lidl favoured an alternative developer. Mr Orr of Generator reported to Mr Isaacs that this “does highlight that we cannot hang around or be complacent in getting our deal with Lidl agreed.”

The judge concluded at [98]:

During the period from 18 December 2013 to 14 February 2014, Lidl and its solicitors continued to deal directly with Sans Souci and its solicitors. But Generator remained involved and helped to prevent problems developing or becoming insuperable. Lidl continued to refer to Generator as its “Delivery Partner.” In the run up to exchange of contracts for the acquisition of the property Generator and Lidl spent much time on the scheme for the development of the property. Generator spent about £30,000 on architects, mechanical and engineering consultants and structural engineers.

As noted, during this time successive drafts of the Heads of Terms for the proposed joint venture agreement passed back and forth. There were a number of substantial points on which Generator and Lidl disagreed.

On 10 January 2014 Mr Barnes and Mr Orr of Generator met Mr Barber and Mr Beaumont of Lidl at the latter’s offices. It was common ground that no contract was made at that meeting. The judge also found that nothing was said by Mr Beaumont or Mr Barber which could properly be characterised as an assurance that Generator would acquire an interest in the property if Lidl bought it. The judge also found that Mr Barnes considered that it was a real possibility that Generator might fall out of the proposed deal with Lidl.

On 1 February 2014 Mr Barnes of Generator said in an email that the following week would be the week that Generator “agree the deal or not with Lidl”. On 7 February he asked Mr Barber to authorise Lidl’s lawyers (Clarke Wilmott) to speak directly with Generator’s lawyers (DAC Beachcroft). He attached a copy of what was to be the only draft of the Joint Venture Lockout Agreement, which had been prepared by DAC Beachcroft LLP. This draft concerned only Generator and Lidl; and was a different draft from the lock-out agreement relating to the land purchase.

It is an important document because it must be taken to reflect Generator’s understanding of where things stood. It began by reciting that the parties were in negotiations “with a view to agreeing” the proposed transaction (defined as the sale by Lidl to Generator). Clause 4.1 obliged Generator to “seek to agree” the terms for the proposed transaction with Lidl as soon as reasonably practicable. The draft agreement also provided as follows in clauses 5.2 and 5.3:

Clause 8.1 provided:

On 10 February Mr Barnes produced a document entitled “Stage 1 Report” for Generator’s board. It stated:

The judge found that Generator’s board understood that expenditure might be abortive because the deal with Lidl might not go through. He also found that the board was not asked to rely on any assurances that Lidl had given. The board informally approved the expenditure on that basis.

There were no other events of note before Lidl exchanged contracts for the acquisition of the property.

Even after exchange of contracts Generator and Lidl continued to negotiate the proposed heads of terms. They were still apart on questions such as the price that Generator would pay; how much of the price would be deferred; what would happen to the site if planning permission could not be obtained; and whether there would be any overage.

In an email of 28 March 2014 Mr Orr of Generator acknowledged to Lidl:

The Pallant v Morgan equity takes its name from that case: [1953] Ch 43. Two neighbours were interested in acquiring a piece of amenity land which was to be sold by auction. Each instructed an agent. Shortly before the auction the two agents agreed that one of the owners would refrain from bidding and that, if the other one succeeded in obtaining the land, he would divide it between them. Having secured the land, the new owner refused to divide it up. The two agents gave different accounts of what had taken place just before the auction. Harman J said that which of the two accounts was correct was “crucial”. He summarised the two accounts thus:

Harman J preferred Mr Mason’s account. Based on his finding of fact, Harman J went on to hold that:

In other words, he found that the defendant’s agent was, for the purpose of bidding, the agent for both parties. As an agent he would, of course, owe fiduciary duties to his principals. That is entirely in line with the previous authority which he considered, namely Chattock v Muller (1878) LR 8 Ch D 177 in which Malins V-C said:

In other words, liability was squarely based on the fiduciary obligations owed by an agent towards his principal. Although Harman J did not put it quite this way the agreement could have been analysed in contractual terms: not as a contract about the land, but as a contract of agency. In consideration of Mr Mason’s agreement not to bid, Mr James would bid as the agent for both principals. That is why the difference in the two accounts was, in Harman J’s words “crucial;” because on Mr James’ account there was simply the future prospect of an agreement.

Pallant v Morgan was considered by this court in Banner Homes Holdings Ltd v Luff Developments Ltd[2000] Ch 372. Chadwick LJ conducted a thorough review of the cases in which the Pallant v Morgan equity had been considered under the heading “constructive trusts”. He pointed out that in the then state of the law the principles applicable to this kind of constructive trust had much in common with those applicable to proprietary estoppel. He quoted the observation of Robert Walker LJ in what was then the very recently decided case of Yaxley v Gotts [2000] Ch 162 that:

However, Chadwick LJ went on to analyse both Chattock v Muller and Pallant v Morgan not by reference to concepts of agency and fiduciary obligations, but by reference to the principles applicable to proprietary estoppel. Chadwick LJ turned to consider Holiday Inns Inc v Broadhead which had first come before Megarry J on an interlocutory application. Megarry J held that:

He also went on to say that the Pallant v. Morgan equity had much in common with proprietary estoppel, as exemplified in Ramsden v Dyson(1866) LR 1 HL 129 and the other cases in that line of authority. When the case came on for trial, Goff J also adopted the analogy with proprietary estoppel: (1972) 232 EG 951.

Having considered further cases at first instance Chadwick LJ discussed the decision of Goulding J in Island Holdings Ltd v Birchington Engineering Co Ltd (unreported 7 July 1981). He summarised the facts of that case as follows:

I consider, with great respect, that Chadwick LJ did not correctly describe the reasons why Goulding J decided the case as he did. The first building block was Goulding J’s finding that the original agreement between the parties at the time when the council was proposing to sell a lease was itself a binding agreement. The second building block was that that binding agreement created a fiduciary relationship between the two parties. The third building block was that although that original contract was discharged when the parties entered into their “subject to contract” discussions about the freehold, the fiduciary relationship remained in being. It was that fiduciary relationship, continuing between the parties, that meant that when one of them acquired the freehold to the exclusion of the other the principle in Keech v Sandford (1720) Sel Cas T King 61 applied directly to the acquisition of the freehold. That was the principle that Goulding J said that he was applying. That was also the way in which Mummery LJ analysed the case in London & Regional Investments Ltd v TBI plc[2002] EWCA Civ 355 at [49]; in my judgment correctly.

At the end of his review of the authorities Chadwick LJ concluded that the Pallant v Morgan equity was a species of constructive trust. He then set out the conditions under which such an equity could be raised:

What was the foundation for this statement of principle? In my judgment it is to be found in Chadwick LJ’s assimilation of the principles applicable to cases of proprietary estoppel on the one hand and the imposition of a constructive trust on the other. The species of constructive trust is, nowadays, labelled a “common intention constructive trust”.

In addition, having regard to the way in which the case was actually decided, I do not consider that Island Holdings Ltd v Birchington Engineering Co Ltd supports the last part of Chadwick LJ’s second principle, namely that the equity can be invoked in a case in which the parties “plainly intend” the arrangement not to have “contractual effect”.

Banner Homes next fell for consideration by this court in London & Regional Investments Ltd v TBI plc. The essential question was whether a Pallant v Morgan equity had arisen where both parties were negotiating on an express “subject to contract” basis. This court held that it had not. The importance of the effect of the “subject to contract” formula was stressed by Mummery LJ at many places in his judgment. I quote some examples:

The next important case is the decision of the House of Lords in Cobbe v Yeoman’s Row Management Ltd[2008] UKHL 55; [2008] 1 WLR 1752. Mr Cobbe, an experienced property developer, orally agreed with Ms Lisle-Mainwaring, the sole director of Yeoman’s Row, to buy a property comprising a number of flats for redevelopment into six town houses. Yeoman’s Row already owned the property. The arrangement was that Ms Lisle-Mainwaring would obtain vacant possession of the property and Mr Cobbe would develop it and keep any profit, subject to overage under which each party would have 50% of the gross proceeds of the property over £24m. Acting in the belief, encouraged by Ms Lisle-Mainwaring on behalf of Yeoman’s Row, that the property would be sold to him, Mr Cobbe spent the next 18 months, engaging architects and other professionals, in applying for planning permission. Immediately after the grant of planning permission Ms Lisle-Mainwaring and Yeoman’s Row withdrew from the agreement. The House of Lords considered Mr Cobbe’s entitlement both to a proprietary remedy and to a personal remedy. So far as the proprietary remedy was concerned, they considered both proprietary estoppel and also constructive trust. Reversing the Court of Appeal ([2006] EWCA Civ 1139, [2006] 1 WLR 2964), the House decided that both claims failed; and it is important to understand why. Both Lord Scott and Lord Walker gave speeches. Lord Hoffmann and Lord Mance agreed with Lord Scott. Lord Brown agreed with both Lord Scott and Lord Walker.

In considering the claim in proprietary estoppel Lord Scott placed particular emphasis on the fact that Mr Cobbe knew that his agreement was not legally binding. He said at [25]:

Amongst the cases he referred to as supporting that proposition was London & Regional Investments which was, of course, a failed attempt to invoke the Pallant v Morgan equity as well as a failed claim in proprietary estoppel. Lord Scott specifically approved what Mummery LJ had said at [42]. It seems to me to be clear that in this respect Lord Scott did not distinguish between the two types of case.

At [27] he equiparated cases in which there was an express “subject to contract” reservation and cases in which it was known that the agreement was not legally binding. As he put it at [27]:

Thus the claim made in proprietary estoppel failed. Lord Scott went on to consider the claim based on constructive trust. At [30] he described the principle as follows:

He referred to Pallant v Morgan and Banner Homes as examples of when such a trust would be imposed.

Mr Gaunt QC, for Generator, relied strongly on this description of principle. However, particularly in the light of his earlier discussion of “subject to contract”, it is by no means clear that Lord Scott envisaged a case in which the putative joint venture was in the course of “subject to contract” negotiation, since he contrasted an agreement to embark on a joint venture on the one hand, and a lack of agreement about what is to be done with the land on the other.

Be that as it may, it is equally important to see why Mr Cobbe’s case under this head failed. At [33] Lord Scott said:

Why not? He explained at [37]:

Although it is true that one of the important factors was that Yeoman’s Row already owned the property, that was not the only reason why Mr Cobbe’s claim to have raised the Pallant v Morgan equity failed. It was equally important that Mr Cobbe’s expectation was based on an agreement that was, and was known to be, legally unenforceable; and that the expectation was that in due course there would be a legally binding contract. In the context of commercial negotiations about property development that is the expectation that any experienced developer would have.

One of the cases to which he referred in reaching his conclusion on this question was London & Regional Investments. The first point that Lord Scott made about that case was that it was a “subject to contract” case. That was not said to be an erroneous or irrelevant point.

At [53] Lord Walker quoted with approval a passage from Lord Cranworth’s speech in Ramsden v Dyson:

That was an entirely general observation about the reach of equity. Lord Walker commented that “These may be the first references to the notion that an arrangement which is expressly and deliberately acknowledged to be a “gentleman's agreement” may not be capable of giving rise to an estoppel.” The same must be true of a “subject to contract” agreement.

At [68] Lord Walker made an important distinction between the commercial context and the domestic context. He said (the emphasis is his):

At [75], like Lord Scott, Lord Walker saw no real difference between a case that was expressly “subject to contract” and one that was not. He said:

Having considered the Pallant v Morgan line of cases, Lord Walker said at [81]:

This is an important passage which underlines the general point that equity will not intervene in a case in which parties are consciously relying on honour alone. He repeated the point in his conclusion at [91]:

Earlier in his speech Lord Walker had referred at [60] to the decision of the Privy Council in Attorney General of Hong Kong v Humphreys Estate (Queen's Gardens) Ltd[1987] AC 114 which concerned “subject to contract” negotiations for an exchange of land between the Government of Hong Kong and a developer. Although the negotiations were well advanced, the developer withdrew. Lord Walker commented at [62]:

He returned to this point at [92] in which he concluded:

Finally, Lord Walker said at [93] that Mr Cobbe could gain no further assistance from the doctrine of constructive trust. By that I understand Lord Walker to have meant that the same principles which precluded a successful claim in proprietary estoppel would also preclude reliance on the doctrine of constructive trust, even of the kind discussed in Banner Homes to which he had referred earlier in his speech at [78].

The distinction that Mummery LJ drew in London & Regional between a “no contract” case and a “subject to contract” case (which he drew again in the Court of Appeal in Cobbe at [57]), and which I drew in Kilcarne Holdings Ltd v Targetfollow (Birmingham) Ltd[2004] EWHC 2547 (Ch), [2005] 2 P & CR 105, at [229] and [237] may not have survived the decision of the House of Lords, although my decision in that case was referred to both by Lord Scott at [35] and Lord Walker at [78] without any hint of criticism. But if that distinction has not survived, it does not detract from the potency of the “subject to contract” formula. Rather, it increases the potency of mutual knowledge that an “agreement” is not legally binding. In addition, in the light of the House of Lords’ rejection of Mr Cobbe’s case based on the Pallant v Morgan equity, and the reasons for that rejection, I find it hard to see how Chadwick LJ’s second proposition in Banner Homes can survive to its full extent, at least in the context of arms’ length “subject to contract” commercial negotiations.

In Herbert v Doyle[2010] EWCA Civ 1095; [2011] 1 EGLR 119 Arden LJ considered the decision of the House of Lords in Cobbe. At [57] she said:

I respectfully agree. It is therefore open to serious question whether the summary of principle contained in Snell’s Equity (33^rd ed) para 24-039 remains completely accurate in stating that the equity can arise even where the agreement “was not intended to have contractual effect.”

The final case that I need to consider is the decision of this court in Crossco No 4 Unlimited v Jolan Ltd [2011] EWCA Civ 1619, [2012] 1 P & CR 16. That case also considered a wide-ranging series of remedies to which the appellants claimed to be entitled. Etherton LJ examined the nature of the Pallant v Morgan equity in detail. He referred to some of the academic literature discussing the juridical basis on which the equity was founded. He said that in the light of developments in the law about common intention constructive trusts, the explanation of the Pallant v Morgan equity in Banner Homes was untenable. At [87] he pointed out that:

His preferred explanation was that the Pallant v Morgan equity was simply a case of a breach of an existing fiduciary duty. That, to my mind, is precisely the basis on which Pallant v Morgan itself was decided, applying ordinary principles of agency, as well as the basis on which Island Holdings Ltd v Birchington Engineering Co Ltd was decided, applying the rule in Keech v Sandford to a pre-existing and continuing fiduciary relationship. Etherton LJ went on to consider what the result should be if his legal analysis was wrong. He referred at [107] to the important fact that neither party in that case was prepared to enter into a legal commitment, which was important in a commercial context. At [108] he referred to each side’s expectation that their agreement should be embodied in a formal written contract.

Although Arden and McFarlane LJJ were attracted to Etherton LJ’s legal analysis, they both considered that it was not open to this court to treat Banner Homes as having any other ratio than the imposition of a constructive trust. Despite the temptation to build on Etherton LJ’s analysis, I consider that we must loyally follow the decision of the majority in Crossco in that respect.

At [133] Arden LJ said:

We were not addressed on the very recent decision of this court in Farrar v Miller [2018] EWCA Civ 172, but I do not consider that it changes the correct analysis.

The ultimate question that the judge posed for himself at [194] was this:

He answered that question in the negative, having particular regard to nine considerations that he set out between [195] and [204]. Mr Gaunt made sustained criticisms of the judge’s approach some of which, to my mind, were justified. Mr Gaunt submitted that the judge’s question was too narrow. The real question was not whether Generator would acquire an interest in the property if Lidl acquired it. The real question under this head was whether there was a sufficient understanding that the property would be acquired (whether by Lidl, Generator or a joint venture vehicle) for the joint benefit of both Lidl and Generator. I agree, subject to the important qualification that looking at the overall circumstances, it must be unconscionable for Lidl to keep the property for itself. Mr Gaunt also submitted, correctly in my opinion, that the judge concentrated too much on the question whether the parties had agreed the default position: that is to say, what would happen to the property in the event that the desired planning permission was not obtained.

That said, I do not propose to engage in a detailed discussion of the judge’s nine particular reasons for rejecting Generator’s claim; because, despite Mr Gaunt’s excellent oral submissions, standing back and looking at the facts overall, I regard this as a clear case; even if I would not have precisely followed the judge’s path to his conclusion. In my judgment Generator’s claim cannot succeed for a number of cumulative reasons.

First, this was a case of commercial parties, advised by lawyers, working at arms’ length towards the conclusion of an agreement for a purely commercial enterprise the terms of which were never agreed. Indeed, on many of the important terms the parties were far apart. The application of the principles underpinning the Pallant v Morgan equity, in so far as they rest on the doctrine of common intention constructive trust, operate quite differently in a commercial context from the way in which they operate in a domestic context. The principles as expounded by Chadwick LJ are firmly based on the supposed congruence between the principles of proprietary estoppel and the doctrine of common intention constructive trust. But we have seen from Cobbe that the House of Lords firmly denied the applicability of proprietary estoppel in a commercial case like this one where each party knows that they are not legally bound. In this case, as in Cobbe, there can have been no expectation on either side that the parties were legally bound to each other. If the principles underpinning the Pallant v Morgan equity are the same as those underpinning proprietary estoppel (as Chadwick LJ considered them to be) it follows logically that if a proprietary estoppel claim cannot succeed, nor can a claim based on the Pallant v Morgan equity. Moreover, in this case there was no common intention, because the parties were still in disagreement about the terms of the proposed enterprise.

Second, the proposed “joint venture” (if such it was) was expressly made “subject to contract”. That phrase appeared (three times) in all three versions of the heads of terms that passed back and forth between the parties well before contracts were exchanged for the land purchase. The meaning of that phrase is well-known. What it means is that (a) neither party intends to be bound either in law or in equity unless and until a formal contract is made; and (b) that each party reserves the right to withdraw until such time as a binding contract is made. It follows, therefore, that in negotiating on that basis both Generator and Lidl took the commercial risk that one or other of them might back out of the proposed transaction. As Lord Walker stressed in Cobbe, equity will not intervene in a case where the parties expressly agree that a putative agreement is binding in honour only. Likewise, in the Hong Kong case the Privy Council recognised that the use of the “subject to contract” formula means that the parties are not committed either in law or in equity. Like Mr Cobbe, Generator never expected to acquire any interest in the land otherwise than by way of a legally enforceable contract. To put the point another way, where negotiations in a commercial context are expressly made “subject to contract” there is no understanding or arrangement capable of bringing the first of Chadwick LJ’s requirements into play. The mere fact that parties have agreed to engage in good faith negotiations for the making of a joint venture agreement is insufficient to support a constructive trust: Kilcarne Holdings Ltd v Targetfollow (Birmingham) Ltd[2005] EWCA Civ 1355 at [15] and [23]. In short, a “subject to contract” agreement is no agreement at all. For the same reason, the consequence of an express non-agreement is that fiduciary duties will not arise. Mr Gaunt argued that there was a difference between an understanding in principle that there would be “a” joint venture, which was not “subject to contract”, and the detailed terms of such a venture which were. He also argued that although the “subject to contract” reservation applied to the terms of the proposed joint venture agreement, it did not apply to the acquisition of the land. I do not consider that that is a sound distinction. It would subvert the well-understood meaning and effect of negotiations “subject to contract” to differentiate the acquisition of the land, the principle of a joint venture and the terms of the putative joint venture agreement in that way. Moreover, it appears that Mr Barnes purported to draw that very distinction in paragraph 67 of his witness statement; but, as the judge recorded at [45], when asked about it in cross-examination he accepted that he struggled “to get to the finer points of what’s meant there”. That goes to show that the proposed distinction is an artificial one, which would not have been in the minds of the parties at the time. Moreover, it was always intended, right from the inception of discussions, that the parties’ respective rights and obligations would be regulated by written contracts.

Third, there was, as Generator knew, an ongoing risk that Lidl might do a deal with a different developer; and that provided the impetus for Generator to seek to agree terms with Lidl in a binding agreement.

Fourth, the draft lock-out agreement, drafted by Generator’s solicitors, was, according to Mr Barnes’ report to the board of Generator, the mechanism designed to be the protection for Generator. The board understood that any expenditure it incurred might prove abortive; and it did not rely on any assurances that Lidl might have given. That agreement stated in terms that neither party was committed to the sale and leaseback. To impose a constructive trust on Lidl based on the common intention of the parties would directly contradict Generator’s expressed intention in that draft agreement. In addition, the existence of the draft lock-out agreement shows once again that Generator was relying on the prospect of a legally binding contract to protect it, rather than on some ill-defined honourable conduct on the part of Lidl.

Fifth, there is some significance in the fact that (to the knowledge of Generator) Lidl’s board had not approved the joint venture. In order to be able to invoke the Pallant v Morgan equity it must in my judgment be possible to say that the agreement or understanding in question is one which has been assented to by a person capable of binding the party in question; or who at least has ostensible authority to do so. Generator knew that the employees of Lidl with whom they were dealing were not the decision-makers. Mr Gaunt argued that in approving the purchase of the land, Lidl’s board must have also implicitly approved the understanding on the basis of which the land was acquired. However, the judge made no such finding (indeed his finding at [62] suggests the contrary); and I do not consider that it is open to this court to make the necessary finding of fact which, on the materials we have been shown, would be pure speculation.

Sixth, it was not alleged in the Particulars of Claim that Lidl owed any pre-existing fiduciary duties to Generator, or that Lidl acted as Generator’s agent in buying the property. In an early part of his address Mr Gaunt confirmed that Generator’s case was not put on the basis of agency. Indeed, such an allegation is negatived by the judge’s finding that Lidl undertook the whole risk profile in buying the property without any commitment by Generator to contribute to the purchase price; and that Generator recognised that fact. So too is the statement in the draft lock-out agreement that neither party was committed to the transaction, and in particular by Generator’s express reservation of a right not to proceed. An agent is usually entitled to indemnity from his principal against costs incurred within the scope of his agency. It is clear that Generator was not prepared to commit itself to such a liability. The lack of a liability on the part of Generator to indemnify Lidl against any part of the purchase price is a clear signal that no agency was created.

Seventh, the judge made no finding that Generator relied on any assurance or expectation that it would acquire an interest in land. Indeed, his finding that the board was not asked to rely on any assurances that Lidl had given points in the opposite direction. Mr Gaunt argued that reliance was an irresistible inference from the judge’s findings of fact. However, when he was asked to describe the reliance, his initial answer was that it was reliance on the expectation that the “subject to contract” negotiations would ripen into a contract. But that species of reliance is no more than Mr Cobbe had. In the course of his reply he modified that explanation, submitting that Generator relied on the understanding that the property was being acquired by Lidl for the joint purposes of the two parties. Again, the judge made no such finding.

Eighth, it cannot be unconscionable to exercise a right which has been expressly reserved to both parties by means of the “subject to contract” formula; and which Generator had even more clearly reserved to itself in the draft lock-out agreement. As Lord Walker said in Cobbe (and as Arden LJ said in Crossco), it cannot be unconscionable for one party to follow a course which the other party has insisted was open to itself.

I would dismiss the appeal.

I agree.

I also agree.