Home / Insights / Blog / Building Schemes Law Clarified – Birdlip v Hunter [2016]

Building Schemes Law Clarified – Birdlip v Hunter [2016]

Building Schemes Law Clarified – Birdlip v Hunter [2016]

Author: Martin Hutchings QC

The enforceability of restrictive covenants by and against successors in title of the original parties (‘transmissibility’) remains one of the more complicated areas of property law. The law has developed for well over 200 years in a not altogether logical fashion. As readers will know, there are 3 basic methods by which successors in title can seek to enforce restrictive covenants – by assignment; on the principle of annexation, and thirdly, where a building scheme can be shown (i.e. a scheme of mutual covenants or ‘local law’ existing within a defined estate area). Of the three methods, proving the existence of a building scheme (or ‘scheme of development’) is the one in which the law has developed in the least clear way.

The first significant case on building schemes to reach the Court of Appeal for over 20 years: Birdlip Ltd v Hunter and another [2016] EWCA Civ 603 therefore provides welcome clarification, particularly of the evidential requirements necessary for the establishment of a scheme. For those wishing to understand building schemes and advise clients on enforceability, the judgment of Lewison LJ will now form compulsory reading.

Birdlip is not a decision which turns simply on its own facts. Potentially it sets a new course in terms of the advice which practitioners give clients as regards proof of the existence of building schemes.

Birdlip is of more general interest, in showing the sort of circumstances in which an appeal court may be prepared to admit evidence which was not before the lower court (see the ‘postscript’ below).

The facts

The question of enforceability arose because Birdlip wished to build two extra houses on a plot on which a house had been built in about 1910 – shortly after a series of covenants had been imposed on the land. The relevant covenant: ‘not to build more than one dwelling’ was one of a number of familiar restrictions imposed in a 1910 conveyance when the land (situated in Gerrards Cross) was originally sold off by the common vendor. Land Registry searches showed that a considerable area of surrounding land had been sold off by the common vendor with similar (though not always identical) covenants.

Birdlip had secured planning permission for its intended development, but Mr and Mrs Hunter, whose land adjoined Birdlip’s, objected to the development. The difficulty for the Hunters in seeking to enforce the restrictive covenant was that the key 1910 root conveyance had also conveyed the Hunters’ plot to their predecessor – in other words the 1910 conveyance had been of a single plot, title to which had subsequently been split. So for this and other reasons, the Hunters could not rely on annexation; and assignment was a non-starter too. This left them with the possibility of establishing a building scheme – as to which on a very initial view, the evidence might have seemed promising for the Hunters.

The Hunters could point to:

  • At least 20 conveyances whereby the common vendor had sold off numbered plots each containing similar restrictions over a defined period (1906 – 1914)
  • Contemporaneous plans showing the estate. Although the 1910 conveyance did not show the defined area within which the scheme was intended to operate, nor did its words allow its identification from extrinsic evidence, nevertheless, two original agreements for sale (which preceded the conveyances of 2 of the other plots within the ‘estate’) were put in evidence. They were dated 1908 and 1914. The reverse of each of these agreements contained a plan and a prominent note on the face of the plan, which read ‘The Estate Boundaries are edged in red, and the sites as at present staked out for sale are colored (sic) pink’. The 1910 conveyance land fell squarely within the estate boundaries as shown on these plans
  • Clear evidence from these agreement plans, of the prior lotting of an apparently defined area (prior lotting being one of the badges of a building scheme and often considered to be cogent evidence of an intention for the covenant to be for the common benefit of purchasers)
  • The fact that the covenants had been in existence for over 100 years and a local ‘Covenants Association’ had been successful in upholding the covenants
  • Two cases in the 1970s and 1990s before the Lands Tribunal, involving plots within the estate, where the applicants had sought modification of the ‘one house per plot’ covenant but had failed, in the face of a large number of objectors. (Clearly the fact that modification had been sought suggested that the applicants, in each case, must have accepted the covenants’ validity and enforceability).

However, an apparent obstacle for the Hunters was that the 1910 conveyance referred only in the most general terms to an ‘agreement’ preceding the conveyance and, the conveyance plan identified only the plot being sold and gave no clue as to the geographical extent of any defined estate area. More significantly, the plan of the estate attached to the 1908 agreement differed significantly from the 1914 agreement plan; yet, as indicated above, both were described unambiguously on their face as showing ‘the Estate Boundaries’.

The history of the proceedings

Birdlip took the view that its claim was suitable for summary disposal – because the question of the existence or otherwise of a building scheme was a matter of inference from agreed facts and, the interpretation of the 1910 conveyance and other original conveyances. It sought summary judgment before Master Bowles, but lost. Master Bowles said that not only would he not grant summary judgment in Birdlip’s favour, but, he felt that the evidence of the existence of a mutual scheme of covenants was strong.

Birdlip appealed to the Judge but also asked for the trial to be heard at the same time – as the trial would in effect, be a re-run of the summary judgment appeal. The appeal and trial were heard before Judge Behrens in the Chancery Division in March 2015. He agreed with the Master. He dismissed the appeal and granted a declaration in the Hunters’ favour as to the existence of a building scheme. This gave the Hunters the clear right to prevent Birdlip’s intended development.

One of the difficulties that the Hunters had had to overcome in persuading the Master and the Judge to grant declarations in favour of a scheme, was the fact that the two agreements for sale which were before the court were not agreements relating to the key 1910 conveyance. Furthermore, as indicated above, and much more significantly, the ‘Estate Boundaries’ on each of the plans was significantly different in at least two respects. First, the 1914 plan showed only about half of the land shown on the 1908 plan. Secondly, what had been shown as plots 20 to 50 on the 1908 plan had been excluded from the boundaries shown on the 1914 plan.

One of Birdlip’s key points therefore was that it was now impossible for the Hunters to prove a defined geographical area in which the scheme was to operate. How could it be said that the 1910 conveyance would have been preceded by an agreement containing a version of the 1908 plan, as opposed to a version of the plan as attached to the 1914 agreement? And if it could not now be proved on the balance of probabilities which (if any) of the two available versions of the estate plan was attached to the agreement which preceded the 1910 conveyance, how could the Hunters satisfy the primary requirement of showing a known, defined area in which the scheme was to operate?

But both the Master and the Judge had dismissed Birdlip’s objection regarding the differences between the two plans. The Master and the Judge both felt able to draw the inference that a version of the 1908 plan would have been attached to the first conveyance off in 1906 when the scheme ‘crystallised’. Furthermore, both felt able also to infer that the ‘agreement’ referred to in the 1910 agreement for sale that must have preceded the 1910 conveyance, would have contained a version of the 1908 plan attached to it – i.e. the plan which showed the estate to its full extent and which (unlike the 1914 plan) did not ‘miss out’ at least two important geographical areas.

The Court of Appeal Judgment

The CA unanimously allowed the appeal and granted a declaration that Birdlip’s land was not burdened by a building scheme.

How did it reach this conclusion in stark disagreement with the judgment of two experienced judges (Master Bowles and Judge Behrens) whose judgments displayed little doubt, in their minds, as to the existence of a building scheme?

One clear answer is that the CA looked more carefully at the authorities which identified the relevant criteria for the existence of a building scheme and, the limits of permissible inference in this area of law.

First, Lewison LJ set out the 6 characteristics of a building scheme: (i) it applies to a defined area (ii) it relies on a common vendor (iii) each property must be burdened by restrictions intended to be mutually enforceable (iv) the limits of the defined area must be known to each of the purchasers (v) the common vendor is himself bound by the scheme from the point it crystallises – meaning he cannot dispose of plots within the defined area otherwise than on the terms of the scheme (vi) the effect of the scheme is that it will bind future purchasers within the area, potentially for ever.

Next, Lewison LJ set out the two familiar prerequisites for a scheme derived from Jamaica Mutual Life Assurance Society v Hillsborough Ltd [1989] 1 WLR 1101: (i) the identification of the land to which the scheme relates (ii) an acceptance by each purchaser of part of the lands from the common vendor that the benefit of the covenants will enure to the vendor and to others and that he correspondingly will enjoy the covenants entered into by other purchasers.

Lewison LJ went on to point out that whether the two components above exist has been said to be a question of fact or, a question of intention or, a combination of both. But in a key paragraph, he then went on to say:

’25. One would have thought, a priori, that in the case of a scheme of mutual covenants designed to last potentially for ever, that that intention would be readily ascertainable without having to undertake laborious research in dusty archives searching for ephemera more than a century old. In almost all cases to which we were referred where a scheme of mutual covenants was found to exist, the area of land to which the scheme applied was ascertainable from the terms of the conveyance or other transactional documents in question. Conversely where the conveyance or other transactional documents gave no indication of the land to which the scheme applied, no scheme was found.’

He then reviewed a series of cases starting in 1878, up to the Jamaica Mutual case in 1989, and said:

Thus far, the cases in which schemes of mutual covenants have been found to exist have been cases where there is something in the conveyance or other transactional documents to alert a purchaser to the existence of a scheme. However, there are undoubtedly statements in the cases that the existence of a scheme may be inferred purely from the circumstances surrounding the initial sales…’.

But as Lewison LJ pointed out, the cases in which the wider statements regarding inference had been made, tended to be cases such as auction sales, where the surrounding circumstances at the time of the first sale off, showed that there was ‘no point in …taking restrictive covenants (which he [i.e. the common vendor] will not be able to enforce) unless they were intended to be mutually enforceable by the purchasers [e.g.]: Nottingham Patent Brick and Tile Co v Butler (1885) 15 QBD 261…..’.

The importance of Lewison LJ’s explanation is that it shows that the correct approach to determining the existence of a scheme is to focus primarily on the terms of the conveyance itself – particularly if no other transactional documents are in fact available in relation to the particular plot conveyed.

Thus whilst the textbooks[1] may tell us that whether a scheme exists or not is a ‘question of fact to be determined from the terms of the titles and [all] the relevant circumstances surrounding the sales by the common vendor…’ this does not mean that a scheme can be proved by speculative or indirect evidence nor by reliance on inference from surrounding circumstances or such vague matters as the fact that a series of conveyances appear to contain similar covenants and that the sales off appear to have occurred at the same time. The sort of wide evidential enquiries that were previously encouraged by those seeking to prove the existence, or otherwise, of a scheme may, in future, therefore play a far more limited role. The inference of a scheme which relies heavily on surrounding circumstances is not likely to be justified. And that is good news for those seeking certainty in this area of law.

The 1910 Conveyance

So the CA’s starting point was to focus carefully on the precise terms of the 1910 conveyance. Lewison LJ made the following key points about it: (i) there was no reference in the verbal descriptions of the parcels clause to any estate of which the land was said to form part (ii) the conveyance plan showed no lots but simply the property conveyed (iii) there was no reference in the conveyance to any other plan (iv) the covenants were expressed to be given for the benefit of a far wider area than the land now said to be the subject of the scheme (i.e. the 1908 agreement plan land) (v) there was no express provision that the covenants were to be mutually enforceable (vi) the restrictions themselves include a number of different geographical descriptions (vii) some of the covenants were positive and therefore the benefit was unlikely to have been other than personal to the common vendor (vii) the covenants requiring ‘vendor’s surveyor’s consent’ to building work, pointed against a scheme .

From this review, he stated that ‘…I would provisionally conclude that no scheme has been established. What, then, of the extrinsic evidence?….‘. Whilst then going on to doubt whether a scheme could ever be proved by extrinsic evidence alone, Lewison LJ nevertheless considered what that evidence showed. He pointed out that even if it were permissible to prove a scheme by extrinsic evidence alone, it would at very least ‘require cogent evidence to do so’.

And in his ensuing review of the extrinsic evidence, he found that the major stumbling block to the existence of a scheme was the different definitions of the geographical areas of the ‘estate’ when comparing the 1908 agreement plan and the 1914 agreement plan. And Lewison LJ concluded that the judge’s speculation as to the fact that a version of the 1908 plan would have been attached to the 1910 conveyance was anyway ‘…no more than speculation which went far beyond permissible inference…‘ and further, that, even if the Judge (and, implicitly, the Master) had been right to speculate as to the fact that ‘mistakes’ had been made when the 1914 plan was prepared – this did not help the Hunters because the requirement that a purchaser must know of the existence of the defined area within which the mutual scheme was to apply – could not be shown. The reason why mutuality could not be shown (i.e. whether the plan differences were the result of a mistake or not) was not relevant.

Practical Conclusions

The following key conclusions may be drawn from Birdlip

1.The CA was anxious to point out the overriding need nowadays for purchasers of land potentially burdened by a scheme to be able easily to ascertain its existence and geographical extent from the conveyance itself or, other transactional documents. This in a sense chimes with the modern approach to annexation of restrictive covenants, exemplified in Crest Nicholson v McAllister [2003] 1 All ER 46, where Chadwick LJ set out the clear policy reasons of certainty which dictated that s78(1) of the LPA 1925 annexes the benefit of covenants only to such land as the conveyance has identified by express words or necessary implication.

2.Statements in some earlier scheme cases suggesting that the intention of the common vendor to establish a scheme can involve a consideration of a wide range of evidence, must now be viewed with caution. The starting point is always the conveyance itself and the transactional documents – if available. Extrinsic evidence may have a role to play – but it is a secondary one and, it is certainly very doubtful that a scheme can ever be established from extrinsic evidence alone.

3.In considering extrinsic evidence, there is a clear distinction between permissible inferences and impermissible speculation – which those of us who advise clients on enforceability questions should always keep in clear focus.


In Birdlip the appellant also sought to introduce new evidence in the CA which had not been before the lower courts. In the end, because the CA concluded that the Judge and Master were wrong in their approach to the evidence that was before them, it was not necessary for the CA to pay heed to this new evidence (in particular, another conveyance from 1911 which tended to point against the existence of a scheme). What was significant, though, was the CA’s ‘flexible’ approach to the Ladd v Marshall test for the introduction of new evidence in an appellate court.

CPR 52.11(2)(b) allows an appellate court to receive new evidence but it is generally thought that the Ladd v Marshall criteria must be satisfied before the evidence is received. These dictate that the party seeking to introduce new evidence must show (i) that the evidence could not, with reasonable diligence, have been obtained before the trial (ii) the evidence might have had an important, though not necessarily decisive, influence on the outcome of the trial (iii) that the evidence is plainly credible.

The CA in Birdlip allowed the evidence in – but what proved a key influence on their decision was that the trial had been conducted without live evidence and most importantly perhaps ‘the question for decision would, in practice, have affected a wide area so that it would have been unsatisfactory to have decided the appeal on the basis of part only of the available material.’ Again therefore the CA showed that it is prepared to be flexible in applying the apparently ‘strict’ Ladd v Marshall test where the particular circumstances of the case merit flexibility.

Martin Hutchings QC acted for the successful appellant in Birdlip v Hunter

Andrew Olins acted for the successful appellant in Birdlip v Hunter

Real Estate Dispute Resolution

To find out more about our property dispute resolution service for both commercial and private clients, call our solicitors today on 01895 207988 or email propertydisputes@ibblaw.co.uk.

[1] E.g. Francis: Restrictive Covenants and Freehold Land (2013) para 8.105.