Gladman Developments Ltd v Canterbury City Council

- - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - -

Did an inspector who allowed an appeal under section 78 of the Town and Country Planning Act 1990 against the refusal of planning permission for a development of housing misinterpret and misapply relevant policies of the development plan? That is the basic question in this appeal. It raises no new issue of law.

With permission granted by Sales L.J. on 14 September 2018, the appellant, Gladman Developments Ltd., appeals against the order of Dove J., dated 26 June 2018, quashing the decision of the inspector appointed by the interested party, the Secretary of State for Housing, Communities and Local Government, to allow Gladman’s appeal against the refusal of planning permission by the respondent, Canterbury City Council, for a development of up to 85 dwellings on land at Blean Common, close to the settlement of Blean in Kent. The inspector held an inquiry into Gladman’s appeal in February and March 2017. His decision letter is dated 11 July 2017. The city council’s challenge was made by an application under section 288 of the 1990 Act. Dove J. upheld it on all three grounds.

The Secretary of State defended the inspector’s decision in the court below, but has taken no part in this appeal.

The appeal concerns two saved policies of the adopted local plan for the city council’s area – Policy H1 and Policy H9 of the Canterbury District Local Plan First Review, adopted in July

At the time of the inquiry into Gladman’s appeal the development plan comprised the saved policies of the adopted local plan. The saving direction made by the Secretary of State for Communities and Local Government on 30 June 2009 stated that the “extension of saved policies listed …” was “intended to ensure continuity in the plan-led system and a stable planning framework locally, and in particular, a continual supply of land for development”.

In Chapter 1 of the plan, “Key Vision and Strategic Development Objectives”, paragraph 1.1, under the heading “OUR DEVELOPMENT OBJECTIVES”, stated that the plan had “a long term vision for the District to concentrate development within the urban areas of Canterbury, Whitstable and Herne Bay thus enabling urban (and suburban) renaissance …”. Under the heading “HOUSING”, paragraph 1.7 said that the city council’s “Urban Housing Capacity Study (2003) … indicates a significant proportion of the housing land requirement for the Plan period can be met by the release of previously-developed land in the urban areas”, but that “some “greenfield” sites will also need to be released to ensure the full housing requirement is met”. Paragraph 1.14 said that “[the] rural areas of the district are defined essentially as all those areas outside the built-up areas of the towns and villages”, that in those areas “the general countryside policies set out in this Plan, the South East Plan and Kent & Medway Structure

In Chapter 2, paragraph 2.1 under the heading “OUR OBJECTIVES” stated six “objectives for providing decent housing”, one of them “[to] maximise housing development on land that has previously been developed, is derelict or underused (brownfield land) within the urban areas”.

In the section headed “HOUSING DEVELOPMENT WITHIN THE URBAN AREAS”, under the sub-heading “Urban Housing Capacity Study (UHCS)”, paragraph 2.11 stated:

“2.11

The UHCS demonstrates that there is the potential and capacity within our existing urban areas to achieve the strategic housing requirements set out by the Structure Plan, until 2011. On the basis of this Study, the Council does not need to allocate, or grant planning permission for large new housing development outside the urban areas before

and paragraph 2.13:

“2.13

There are some sites outside urban areas but within villages that are previously developed, used, underused or derelict. Such sites could come forward as large windfall sites if they do not have an adverse impact on the social and physical infrastructure of the villages and surrounding areas and are acceptable in all other respects. These will be assessed against policy H9. Housing development on previously developed land outside the villages will not be acceptable unless there are exceptional circumstances, and where it is sustainable.”

In the section headed “HOUSING OUTSIDE URBAN AREAS” paragraph 2.53 said that “[based] upon [the UHCS], existing planning permissions and allocations and the trend of windfall development that is likely to come forward, it will not be necessary to allocate sites outside the existing urban areas for any housing redevelopment in the Plan period …”. Paragraph 2.54 said that “[in] rural areas, outside the urban areas, housing provision is restrained by national and Structure Plan policies”, and that “[some] minor development is permissible within existing villages and exceptionally in the open countryside, outside these villages …”. Under the heading “New Housing in Villages”, paragraph 2.55 said that “[some] villages may have the potential for some limited minor housing development or infill development, consistent with the scale of the village …”. Paragraph 2.56 explained:

“2.56

… Given the sequential approach to the location of new development as set out in PPG3, the City Council has sought to concentrate new residential development on previously developed land within the three main urban areas of Canterbury, Herne Bay and Whitstable. Therefore, the City Council considers that new residential development in all those villages listed below should be limited to minor development only.

…”.

The listed villages included Blean. Paragraph 2.57 said this:

“2.57

There will be some instances where brownfield land within villages becomes available for development, where the nature of the proposals constitutes more than minor residential development. In these circumstances, the impact of a housing scheme would need to be fully assessed prior to the proposal being acceptable in principle.”

Saved Policy H9 stated:

“Planning permission for new residential development, in excess of minor development, on previously developed sites within villages, will only be granted where:

(a)

An appraisal has been carried out to ascertain that the development will not have an adverse impact upon the existing social and physical infrastructure of the village and surrounding area;

(b)

The development has regard to the character and appearance and historic environment of the village;

(c)

The development does not conflict with other Local Plan design or environmental objectives;

(d)

A Development Brief has been prepared in advance of any determination of a planning application to ensure the proper planning of the area.”

Under the heading “Housing for Local Needs in the Countryside”, paragraph 2.58 said that the city council “recognises that in certain circumstances housing should be provided in the countryside to meet an identified housing need”, and that “[this] need should be based on an up-to-date housing needs survey carried out in conjunction with the Parish Council or local residents in places where no Parish Council exists”.

In chapter 5 of the local plan, “Promoting Our Countryside”, under the heading “A PROTECTED COUNTRYSIDE”, paragraph 5.27 confirmed that “[one] of the City Council’s objectives is to protect and enhance the countryside …”. Paragraph 5.29 acknowledged that the structure plan “provides protection for the countryside: policy EN1 protects the countryside for its own sake; policy EN3 conserves and enhances Kent’s landscape and wildlife habitats …”; and that “[all] these policies will be applied in the District”.

The examiner’s final report on the draft replacement local plan was published in June 2017. He recommended that it should not be adopted as submitted, but indicated that it could be made sound with modifications. In its modified form, current at the time of the inspector’s decision, Policy SP4, under the heading “Strategic approach to location of development”, stated:

“The urban areas of Canterbury, Herne Bay and Whitstable will continue to be the principal focus for development, with a particular focus at Canterbury, together with development at the rural service centres and at local centres. …

In addition to the development allocations set out in this plan:

1.

In the urban areas of Canterbury, Herne Bay and Whitstable, new housing development will be supported on suitable sites where this would be acceptable in terms of environmental, transport and other planning factors, and would not result in the loss of sites identified for business and other specific uses;

2.

Provision of new housing that is of a size, design, scale, character and location appropriate to the character and built form of the rural service centres of Sturry and the local centres of … Blean … will be supported provided that such proposals are not in conflict with other local plan policies related to transport, environmental and flood zone protection and design, and the Kent Downs AONB Management Plan, where applicable;

…

5.

In the open countryside, development will be permitted if required for agriculture and forestry purposes (see Policy EMP13).”

Paragraph 2 of the policy in its previous draft form had referred to “Small-scale provision of new housing that is of a design, scale, character and location appropriate to the character and built form of the service centres of … Blean …”. In the later draft the words “Small-scale” were deleted, and the word “size” was added before “design”.

The inspector identified three “main issues”, the first of which was “whether the proposed development would accord or conflict with the existing and emerging planning policies for the area” (paragraph 15 of the decision letter).

His understanding of Policy H1 was that it was a permissive policy, not restrictive. In a passage headed “Accordance with policy” he said (in paragraphs 16 to 22):

“16.

In the Adopted LP, Policy H1 states that residential development will be permitted on allocated sites and on previously developed land within urban areas. As such, the policy is permissive of development within these locations, and silent on development elsewhere.

17.

The appeal site is not allocated for development, nor is it previously developed land. And although the village of Blean does not have any defined boundary, there is no dispute that the site is outside the existing built up area. The proposed development therefore does not fall within any of the categories where development is expressly permitted by Policy H1. However, given the policy’s purely permissive nature, this does not amount to a conflict.

18.

The Council argues that there is an implicit ‘negative corollary’: that because some locations are identified for development, it must follow that all others are to be precluded. But nothing within the policy itself, or its explanatory text, supports that interpretation. It is well established case law that planning policies are to be read objectively, having regard to their language and context, and it is difficult to see how the concept of an implicit policy could sit comfortably with this doctrine.

19.

Furthermore, as the Council’s planning witness acknowledged, there is nothing in the National Planning Policy Framework (the NPPF), to support the proposition for a negative corollary. Indeed in the circumstances, it seems to me that this would effectively amount to a negative presumption, against any development other than that expressly proposed in a local plan. Such a presumption would run counter to the NPPF’s presumption in favour of sustainable development.

20.

It is also argued that broadening the scope of Policy H1 in the way the Council suggest, is necessary so as to give effect to the LP’s strategic objective and ‘long term vision’ of protecting the countryside. But the strategic objectives and vision are not identified amongst the plan’s saved policies. Nor is Policy H1 specifically linked to these by anything in the plan. It may well be unfortunate that, as a result of the revocation of the former Kent Structure Plan and South East Regional Strategy, the Council now finds itself with no policies to protect the countryside. But that does not justify mis-applying Policy H1.

21.

I appreciate that the appellants themselves failed to argue the point about Policy H1 in their Planning Statement at the application stage, but that does not preclude them from doing so now. I am aware of the Daventry judgement [Gladman Developments Ltd. v Daventry District Council [2016] EWCA Civ 1146], but in that case one of the policies in question expressly precluded development in the countryside, whereas Policy H1 does not. I also note the comments of the Inspector in the written representations appeal for 8 dwellings at Thanington. But in deciding the present appeal, I must have regard for all the evidence before me, and the balance of the evidence leads me to the view that I have expressed above.

22.

For these reasons therefore, although the appeal proposal does not specifically accord with Policy H1, neither do I find any conflict with that policy.”

On Policy H9 he said (in paragraphs 23 to 25):

“23.

Policy H9 states that permission for new residential development, in excess of minor development, on previously developed sites within villages, will only be granted where various requirements are complied with. Read in conjunction with the accompanying text at paragraph 2.56, it seems likely that what this was intended to mean is that the requirements specified in the policy are to be applied to any residential proposal which is either outside a village, or exceeds minor development, or is not on brownfield land. Consequently, despite some ambiguity, I agree with the Council that Policy H9 is relevant to the appeal proposal.

24.

However, the Council is wrong, in my opinion, in suggesting that the policy seeks to prevent such developments. The policy is permissively worded, and does no more than to set out a list of relevant considerations. These include the effects on social and physical infrastructure, character and appearance, the historic environment, and the LP’s design and environmental objectives. Provided a scheme is assessed with regard to these matters, and they are taken into account in any decision, it seems to me that Policy H9 will be satisfied. In the present case the matters specified, where relevant, have all been considered, through the present appeal.

25.

Consequently, subject to my findings on the relevant matters, on which I will comment later in my decision, I find no in-principle conflict with Policy H9.”

Later, when considering the proposal’s “[compliance] with Policy H9 criteria”, he said (in paragraph 91):

“91.

Earlier in this decision, I found that the only adopted or emerging local plan policy relevant to the principle of development on the appeal site is Policy H9 of the adopted LP. That policy specifies four matters that are to be considered.”

Having concluded (in paragraphs 92 to 95) that all four criteria were satisfied, he found “no conflict” with Policy H9 (paragraph 96).

As for Policy SP4 of the draft replacement local plan, the inspector said (in paragraph 34):

“34.

Draft Policy SP4 provides that the principal focus of development will be at the main urban areas, together with some development at the rural service centres. Blean is identified as one of the latter. At the service centres, the policy gives general support to small-scale housing provision, of a scale and location appropriate to the settlement’s built form. What constitutes small-scale is not stated. On other developments, the policy is silent. With regard to the present appeal, Policy SP4 therefore gives no specific support, but neither does it preclude such development. Again I find not conflict.”

He stated his “[conclusions] on policy compliance” in this way (in paragraph 37):

“37.

Of the key policies on which the Council relies, there are none in either the adopted or the emerging LPs that directly support the proposed development, but nor are there any with which the present outline proposal is in conflict. All of these policies are therefore essentially neutral, weighing neither for nor against the development. It follows therefore, that in this case the final planning balance will turn on other material considerations.”

In his “Conclusions” at the end of his decision letter the inspector said (in paragraph 119):

“119.

For the reasons set out in this decision, I find that there is only one Development Plan policy that is directly relevant to the principle of development on the appeal site, and that is adopted Policy H9. However, the development now proposed does not conflict with that policy. On the key issue of development in the countryside, adjoining larger villages such as Blean, the Adopted LP is otherwise silent or absent. So too is the emerging Draft LP.”

He went on to say that “[the] development would be in scale with the existing settlement …”, that its “benefits would be substantial” (paragraph 120), and that it would have “no significant adverse consequences” (paragraph 121). He concluded (in paragraph 123):

“123.

Having regard to NPPF paragraph 14, no material adverse impacts would arise. It follows that the likely adverse impacts would not significantly or demonstrably outweigh the development’s benefits; indeed the reverse is true. No specific NPPF policies indicate that permission should be restricted. The presumption in favour of sustainable development therefore applies, and this is a consideration that weighs heavily in support of the appeal.”

None of the “matters raised”, he thought, pointed to any other conclusion than that planning permission should be granted. He therefore allowed the appeal (paragraph 124).

The relevant legal principles are well settled and need not be revisited here.

The correct approach to determining an application for planning permission has been considered several times at the highest level, and this court has amplified the principles involved. Section 38(6) of the 2004 Act requires the determination to be made “in accordance with the [development] plan unless material considerations indicate otherwise”. The development plan thus has statutory primacy, and a statutory presumption in its favour – which government policy in the NPPF does not. Under the statutory scheme, the policies of the plan operate to ensure consistency in decision-making. If the section 38(6) duty is to be performed properly, the decision-maker must identify and understand the relevant policies, and must establish whether or not the proposal accords with the plan, read as a whole. A failure to comprehend the relevant policies is liable to be fatal to the decision (see the speech of Lord Clyde in City of Edinburgh Council v Secretary of State for Scotland [1997] 1 W.L.R. 1447, at pp.1450, and 1458 to 1460; the judgment of Lord Reed in Tesco Stores Ltd. v Dundee City Council [2012] P.T.S.R. 983, at paragraphs 18, and 21 to 23; the judgment of Lord Carnwath in

If the relevant policies of the plan have been properly understood in the making of the decision, the application of those policies is a matter for the decision-maker, whose reasonable exercise of planning judgment on the relevant considerations the court will not disturb (see the speech of Lord Hoffmann in Tesco Stores Ltd. v Secretary of State for the Environment [1995] 1 W.L.R. 759, at p.780H). The interpretation of development plan policy, however, is ultimately a matter of law for the court. The court does not approach that task with the same linguistic rigour as it applies to the construction of a statute or contract. It must seek to discern from the language used in formulating the plan the sensible meaning of the policies in question, in their full context, and thus their true effect. The context includes the objectives to which the policies are directed, other relevant policies in the plan, and the relevant supporting text. The court will always keep in mind that the creation of development plan policy by a local planning authority is not an end in itself, but a means to the end of coherent and reasonably predictable decisionmaking, in the public interest (see the judgment of Lord Reed in Tesco v Dundee City Council, at paragraphs 18 and 19; the judgment of Lord Gill in Hopkins Homes, at paragraphs 72 and 73; the judgment of Richards L.J. in Ashburton Trading Ltd. v Secretary of State for

Dove J.’s understanding of Policy H1 and Policy H9 of the adopted local plan was that they identified the locations where housing development would be acceptable and, by necessary implication, where it would not. He said (in paragraph 33 of his judgment):

“33.

… I am satisfied that the Inspector was in error when he interpreted policies H1 and H9 as being silent in relation to housing development which was not on previously developed land within urban areas and therefore concluded that there was no conflict with either of those policies in principle. Taking the language of the policy itself, and without reference to any of the explanatory text, it is clear that the purpose of the policy is to identify, for the purposes of housing development, the types of location where the plan required housing development to take place. In essence, the locations which are identified for the permission of residential development are those allocated in the plan, or non-identified sites on previously developed land within urban areas (if other criteria unrelated to location are met). It follows that if housing development is proposed in a location which does not accord with the types of locations specified in the policy, that proposal will be inconsistent with and unsupported by the policy and therefore not in accordance with it and in conflict with it. The interpretation is simple: policies H1 and H9 identify the types of location where housing development will be permitted; if housing development is proposed in other types of location it is not supported by the policy and therefore in conflict with it and, to the extent of that policy (as part of the exercise of assessing compliance with the development plan taken as a whole), not in accordance with the development plan. Whether it is described as a “negative corollary”, or a necessary inference, or an obvious implication, what matters is that it is clear that the purpose of the policy is to identify those types of location where housing development is to be permitted and if an application is made outside one of those identified types of location then that is clearly not in accordance with the policy.”

The judge’s conclusion, “based solely on the texts of policy H1 and H9”, was that the inspector had not “correctly interpreted them”, and so had not “correctly approached the application of section 38(6) of the 2004 Act”. The “correct interpretation”, in the judge’s view, did “not involve either elevating the explanatory text in relation to countryside protection … to the status of policy”, nor “the resurrection of the extinct structure plan policies protecting the countryside for its own sake”. It “simply seeks to understand where, in principle, by virtue of policies H1 and H9, it is intended that housing should occur, and then concluding that when housing development is proposed outside those supported locations it is not in accordance with the plan”. The local plan was therefore not “silent” on proposals such as Gladman’s here (paragraph 35). The judge also concluded that “[once] the scope of the interpretative exercise is expanded to include the explanatory text supporting both the strategy of the plan, and the policies in particular, then the position as to the correct interpretation arising from the words of the policies themselves is further reinforced”. He referred, in particular, to the text in paragraphs 1.7, 1.24(a), 2.11, 2.13, and 2.53 to 2.55 (paragraph 36). He saw support for these conclusions in the judgment of Sales L.J. in Gladman Developments Ltd. v Daventry District Council, and the first instance judgment in Crane v Secretary of State for Communities and Local Government [2015] EWHC 425 (Admin) (paragraphs 37 to 42). He went on to conclude that Policy H9 was in “perfectly clear” terms. It applied to proposals “in excess of minor development, on previously developed sites within villages”. There was, he said, “nothing either in the text of the policy, or in the explanatory text at paragraph 2.56, which justifies the Inspector’s contrary conclusion in paragraph 23 of the decision letter” (paragraph 45).

The judge also accepted that the inspector had misconstrued Policy SP4 of the draft replacement local plan in paragraph 34 of his decision letter, though this added little to his conclusions on the policies of the adopted local plan (paragraph 46).

For Gladman, Mr John Barrett submitted that the adopted local plan did not impose a “ceiling” on the amount of new residential development in the city council’s area. Policy H1 and Policy H9 did not operate as a “closed list”. Policy H1, properly understood, did not identify the types of location where the adopted local plan “required” housing development to take place, nor did it prohibit development in other locations. The language of the policy was clear and unambiguous. The policy was permissive. It supported the approval of residential development on “sites allocated for housing or mixed use” and on “previously developed land within the urban areas”, subject to the qualifications it stated. However, it did not purport to identify the only locations where housing development would be acceptable. It could not be understood as precluding such development outside existing urban areas. That was a strained and unjustifiably “purposive” interpretation. It was belied by Policy H3, even though that policy had not been saved. A policy such as Policy H3 could not have been included in the local plan if Policy H1 and Policy H9 had been intended to prevent development on large “greenfield” sites outside the urban areas. Secondly, Mr Barrett submitted, the inspector’s conclusion in paragraph 18 of his decision letter was correct. The “negative corollary” was, in this case at least, a mistaken concept, inconsistent with the basic principles bearing on the interpretation of planning policy to which Lord Reed referred in his judgment in Tesco v Dundee City Council (in paragraphs 18 and 19). Thirdly, Mr Barrett contended that the relevant explanatory text was in any case consistent with the interpretation adopted by the inspector. He pointed out that in several places, including paragraphs 1.7, 2.17, 2.54 and 2.58, the adopted local plan had acknowledged the need for some release of “greenfield” sites to ensure the housing requirement was fully met.

As for Policy H9, Mr Barrett acknowledged that the inspector’s interpretation was a

“generous” one. Again, however, he submitted that the language of the policy was permissive. Subject to its four criteria being met, it would be satisfied. The inspector had applied those criteria to Gladman’s proposal, and, in the exercise of his planning judgment, found that the proposal complied with them. In doing so, he made no error of law. Here too the city council’s “negative corollary” argument should be rejected. Whether taken on its own or when read with the relevant explanatory text, the policy did not implicitly preclude development outside villages.

On behalf of the city council, Ms Isabella Tafur supported the judge’s interpretation of the local plan policies in their context, and his conclusion that the inspector had failed to interpret them correctly. Policy H1 and Policy H9 were components of a complete spatial strategy for housing development in the local plan. As a result of the Secretary of State’s saving direction, that strategy no longer included Policy H3, and did not countenance development in locations such as the appeal site here. The inspector was wrong to construe Policy H1 as being permissive towards housing development in the locations to which it referred but “silent” on proposals elsewhere. In his conclusions on Policy H9 the inspector was wrong to interpret that policy as permissive towards proposals for development, such as Gladman’s, that were not merely “minor”, on sites that were not “previously developed”, and not “within villages”. Having misinterpreted the policies, the inspector had misapplied them in concluding that Gladman’s proposal was not in conflict with the local plan. This prevented him from deciding the appeal consistently with his duty under section 38(6) of the 2004 Act.

I cannot accept Mr Barrett’s argument. I think Ms Tafur’s is basically right. I agree with Dove J.’s conclusions on the meaning and effect of these two policies of the local plan, and on the errors in the inspector’s understanding and application of them.

Mr Barrett rightly conceded in the course of argument that Gladman’s proposed development could not be said to be in accordance with any specific policy of the adopted local plan. It did not find any explicit support in any of the saved policies for housing development. It was not development of a kind or in a location identified either in Policy H1 or Policy H9. It might once have found support in Policy H3, but that policy had not been saved by the Secretary of State.

As Ms Tafur submitted, at the time of the inspector’s decision Policy H1 and Policy H9 belonged to a comprehensive local plan strategy for housing development in the city council’s area – which, as she put it, contained a “spatial vision” for the delivery of housing in that area and “policies to deliver that vision”. The strategy did not merely include specific allocations of land for such development in the local plan period. It established, in explicit terms, a clear and complete hierarchy of locations in which proposals for new housing would or might be acceptable and consistent with the plan. Although each of the individual policies composing the strategy was in permissive terms, it is necessary to consider their true effect in combination. Together they formed a suite of policies for housing development, which left out none of the locations where such development might be expected to receive planning permission, subject to relevant criteria being met. Their effect was to identify, in addition to the allocated sites, the whole range of potentially acceptable locations for housing development, the type and scale of development suitable in each location, and the circumstances in which such development was likely to be approved.

At the top of the hierarchy, under Policy H1, were “sites allocated for housing or mixed use

In my view, therefore, this not a case in which it could properly be said that the development plan was “silent” in the sense of paragraph 14 of the NPPF (as published in March 2012). There was not an absence of relevant policy. As the judge recognized, the saved policies of the local plan continued to be a corpus of relevant policy sufficient to enable the proposal to be judged acceptable or unacceptable in principle (see the judgment of Lord Carnwath in Hopkins Homes, at paragraph 54; the judgment of Stephen Richards L.J. in Cherkley, at paragraph 18; and the first instance judgment in Bloor Homes East Midlands Ltd. v Secretary of State for Communities and Local Government [2014] EWHC 754 (Admin), at paragraphs 49 to 64).

There was no implication in Chapter 2 of the local plan that housing development outside the identified hierarchy of locations in the saved policies would or might be acceptable. A housing proposal with no explicit support in any of those policies was not to be treated as favourably as a proposal within the hierarchy – or even more so. On the contrary, I think the judge was right to conclude that the natural and necessary inference here was that housing development of a kind or in a location other than those explicitly supported under the saved policies, including Policy H1 and Policy H9, could not be regarded as being in accordance with the development plan. Indeed, it would be in conflict with the plan, because it would be contrary to the comprehensive strategy for housing development embodied in the surviving policies. This, in my view, is plain from the policies in their own terms, read together, and without recourse to their objectives and the explanation given for them in their supporting text. The simple point here is that if it had been the intention of the city council that housing development outside the locations identified in Policy H1 would generally be acceptable, a policy such as Policy H9 would not have been necessary, and would not have been cast as it was.

The policies themselves were perfectly clear. The judge’s conclusion to that effect was right. As he recognized, the fact that the polices were expressed in permissive terms does not exclude the obvious corollary that proposals without their explicit support were not in accordance with them or with the plan’s comprehensive strategy for housing development. As he also recognized, however, this necessary inference is only reinforced by the policy objectives and the supporting text, which emphasized the city council’s intention to steer housing development to the existing urban areas and previously developed land and away from undeveloped sites in the countryside. The inference, therefore, is not neutral or positive towards development without specific support in the policies, but negative.

In that respect, as the judge also concluded, this case bears some similarity to Crane, where the housing policies of a neighbourhood plan were found to “connect to each other” and to “form a coherent whole” (see the first instance judgment, at paragraphs 11 to 13, 37, and 42 to 47). The relevant policies of the neighbourhood plan and their context in that case were, of course, different from the local plan strategy here, and it would not be right to take from the reasoning in the judgment any general principle of wide application beyond cases truly analogous on their facts. However, the parallel with this case seems clear. In both cases, the relevant policies made a unified strategy, which governed proposals for housing development in the area covered by the plan and implicitly excluded proposals other than those with express support.

This understanding of the meaning and effect of Policy H1 and Policy H9 is not undone by the fact that Policy H3, which did not survive in the Secretary of State’s saving direction, had envisaged development in an additional category, namely “the development of large sites … not identified in the plan” – so long as it did not prejudice the plan’s “environmental and sustainability strategy” and was acceptable “in sequential terms compared with other available sites”, or was “required to meet a quantitative or qualitative need”. When Policy H3 ceased to be part of the local plan’s housing strategy, development of that kind was no longer supported by any policy in the plan. This did not mean, however, that the effect of the saved policies – in particular, Policy H1 and Policy H9 – was now to imply that such development was acceptable, in spite the Secretary of State’s decision not to save Policy H3. Neither Policy H1 nor Policy H9 changed its own meaning. Policy H1 continued to refer to Policy H3, though not with the effect of incorporating it into the strategy again as a policy generally supporting the development of unidentified large sites, but only to indicate the approach that would be taken to proposals for “previously developed land within the urban area” within the scope of Policy H1 itself. The fact that Policy H3 was included in the plan at its adoption, and later not saved, does not negate the inference that housing development without explicit support in the policies comprising the strategy, both before and after the saving direction, was not in accordance with the plan. The strategy was altered, but the inference remained.

Turning now to the interpretation specifically of Policy H9, again I agree with the judge. On a straightforward interpretation of it, the policy related only to proposals for “new residential development, in excess of minor development”, on “previously developed sites” that were “within villages”. It did not relate to proposals such as Gladman’s, for the development of new housing on land that was not previously developed and outside a village. The four criteria in the policy refined its support for development of the kind to which it related. Their effect was only to qualify the policy. They did not enlarge its reach to embrace locations for development to which it did not refer. There was no implication in the policy that development outside its scope was acceptable in principle so long as it met the four criteria. That is not what the policy said, and not what it meant. Nor can it sensibly be suggested that proposals such as Gladman’s, though clearly outside the scope of the policy, merited a more liberal approach than those within its scope – because they escaped the application of its four criteria. That concept cannot be reconciled with Policy H9 as it was framed, and the obvious intention behind it. Indeed, it would make a nonsense both of the policy itself and of the hierarchy of locations for housing development set out in the plan.

In my view, therefore, the inspector was in error when, in paragraph 23 of his decision letter, he construed Policy H9, read with the text in paragraph 2.56, as meaning that its requirements “are to be applied to any residential proposal which is either outside a village, or exceeds minor development, or is not on brownfield land”. He was also wrong, in paragraph 24, to read the policy as if it was permissive towards proposals such as Gladman’s, subject only to the four criteria being met. The policy was relevant to this proposal, but only in the sense that it excluded such development – the opposite sense to the inspector’s conclusion. He was in my view wrong then to apply the policy criteria to the proposal (in paragraphs 91 to 96) as if that is what the policy required, and wrong (in paragraph 119) to conclude, in effect, that the proposed development was in accordance with the policy. Thus he both misinterpreted and then

The judge’s assessment was, in my opinion, correct. Contrary to the inspector’s analysis and the conclusion he stated in paragraph 37 of his decision letter, there was not only no support for the proposed development – either explicit or implicit – in the saved policies of the adopted local plan, but the policies were not “essentially neutral, weighing neither for nor against the development”, and the proposal was “in conflict” with them. In the circumstances I cannot see how the inspector could properly avoid the conclusion that a decision to grant planning permission here would not be a decision “in accordance with the plan”. It would be a decision incompatible with the plan. This is not to say of course that the inspector could not lawfully have granted planning permission for the proposal, but only that he had first to confront the question of whether there were material considerations of sufficient force, including relevant government policy in the NPPF, to outweigh the “presumption in favour of the development plan” arising from section 38(6). That, however, is not the way he went about the balancing exercise in his “Conclusions”. Instead of applying the statutory “presumption in favour of the development plan” as he should have done, he applied only the policy “presumption in favour of sustainable development”. His decision therefore cannot stand.

As the judge concluded, this ground of the city council’s challenge added little to the previous one, given his conclusions on that. My conclusions on the saved policies of the adopted local plan are the same as his, and if they are correct the appeal must fail. I shall therefore deal with this ground very briefly.

Mr Barrett submitted that, as the inspector concluded in paragraph 34 of his decision letter, Policy SP4 of the draft replacement local plan, like Policy H1 and Policy H9 of the adopted local plan, was a permissive policy, in that it supported development of the kinds to which it referred. It did not have a “negative corollary” for other forms of housing development outside the specified categories. It was neutral, not negative, towards such development. And in any event by the time the inspector came to determine Gladman’s appeal the draft policy no longer referred to “small-scale” development in the rural service centres, including Blean. On that basis, the inspector’s conclusions that on “other developments” the draft policy was “silent”, and that it gave “no specific support” to Gladman’s proposal “but neither did it preclude such development”, were sound.

Ms Tafur submitted that the inspector fell into a similar error in interpreting Policy SP4 as he had in construing the two policies of the adopted local plan. He had erred in concluding that the policy was to be regarded as “silent” on housing development other than “small-scale” development at rural service centres such as Blean. Once again, Ms Tafur submitted, a wrong interpretation of policy led to an incorrect application of it, which rendered the inspector’s decision unlawful.

It is enough to say that I see force in Ms Tafur’s submissions here, for similar reasons to those I have given for rejecting the previous ground, but that even if Mr Barrett’s argument is right, and indeed might have been stronger if the inspector had considered the policy in its modified form, the appeal would still have failed. The inspector’s misinterpretation and misapplication of the two policies of the local plan and his consequent failure to perform his duty under section 38(6) were enough, without more, to sustain the city council’s challenge.

For those reasons I would dismiss the appeal.

I agree with both judgments.

I am grateful to Lindblom LJ for his comprehensive analysis. I agree with his analysis and with his conclusion that the decision of Dove J was correct. I add a brief concurring judgment of my own as I consider that the proper interpretation of the relevant policies of the local plan is clear.

At the heart of Mr Barrett’s submissions was the proposition that, although Policy H3 was not saved, its presence in the original local plan colours the proper interpretation of Policy H1 and Policy H9, and that, properly interpreted, they are not restrictive or exclusive but permissive. In advancing that proposition, he urged on us that, in accordance with Hopkins Homes Ltd v Secretary of State for Communities and Local Government [2017] UKSC 37, [2017] 1 WLR 1865 at [22] and [72] the local plan should be interpreted objectively, in accordance with its language read in its proper context. Ms Tafur presented a more nuanced approach to interpretation, and Lindblom LJ has summarised the interpretative principles in [22] above. Whether an objective and literal interpretation is adopted or one in which the wider planning context colours the literal wording, I consider it is plain that the appellant’s interpretation of Policy H1 and Policy H9 is not correct.

Policy H1, Policy H3 and Policy H9 are set out in full in the judgment of Lindblom LJ.

The appellant accepts that its proposed development does not fall within Policy H1 or Policy H9.

I have no difficulty in accepting Mr Barrett’s submission that the meaning of Policy H1 and Policy H9 cannot change merely because Policy H3 was not saved.

Policy H1 is addressing residential development within the urban areas. Its wording is rather clumsy in part. On its literal wording it specifies two situations in which residential development is permitted: (1) on sites allocated for housing or mixed use as shown on the Proposals Map; and (2) on non-identified sites on previously developed land within the urban areas, provided (a) there is unlikely to be an excessive supply of new housing development coming forward within the Plan period and (b) the particular site does not make an identifiable contribution to the economic, environmental or social well-being of the town or District. Policy H1 then provides that in situation (2) Policy H3 applies. That makes perfect sense, as Policy H3 is addressing the development of housing sites of 5 dwellings or more on unidentified sites. On its literal wording, it provides that planning permission for such large development sites will not be permitted if they prejudice the Local Plan’s environmental and sustainability strategy, and will only be permitted if they are acceptable in sequential terms compared with other available sites or are required to meet a quantitative or qualitative need.

Accordingly, the reference in Policy H1 to Policy H3 in the original draft Local Plan, did not colour Policy H1 so as to make Policy H1, on its literal and objective meaning, non-exclusive as regards its subject matter. On the contrary, it further restricted what was expressly permitted under Policy H1.

Turning to Policy H9, which appears under the heading “New Housing in Villages”, it did not, unlike Policy H1, make any express or implicit reference to Policy H3 in the original Local Plan. Lindblom LJ has referred to, and quoted in whole or in part, paragraphs 2.55, 2.56 and 2.57 of the supporting text for Policy H9. In summary, that supporting text makes clear that there is potential for some limited minor housing development or infill development in some villages (paragraph 2.55), that certain specified villages are no longer seen as suitable for development in excess of minor development (2.56), and that there will be some instances where brownfield land (that is, previously developed land) within villages might be appropriate in all the circumstances for more than minor residential development, with Policy H9 defining the circumstances in which such development in excess of minor development will be permitted. Policy H9 states that such development will only be permitted where four specified conditions are satisfied.

It is clear that Policy H9 and its supporting text were always intended to be a comprehensive policy regarding new housing in villages. It would be perverse and make a nonsense of Policy H9 if it were possible to obtain planning permission for more than minor development to take place in villages without having to comply with the conditions specified in H9, that is pursuant to a more liberal regime than that specified in Policy H9, because the development does not fall within Policy H9 itself or the supporting text.

Turning to Policy H3, paragraphs 2.19-2.21, which were its supporting text, show that it was intended to apply to (1) new greenfield sites identified through the LDF process in due course (paragraph 2.19), and (2) large unidentified sites which might come forward (paragraph 2.20), especially where the proposals were for housing as part of a comprehensive redevelopment to regenerate a designated area in the Plan such as regeneration zones or town centres (2.21). Those unidentified sites are not expressly restricted to developed or undeveloped land or to land within or outside urban areas.

For the reasons I have given, Policy H1 deals exclusively with residential development on previously developed non-identified sites within the urban areas, and Policy H9 deals exclusively with new residential development, in excess of minor development, within villages. To that extent H3 must be read subject to Policy H1 and Policy H9. Insofar as Policy H3 contemplated the possibility of housing development outside Policy H1 and Policy H9 on unidentified large sites, that is to say not being previously developed land within the urban areas and not being land in a village – such as the appellant’s proposed development, H3 was a stand-alone provision. When it was not saved, that category of permitted development was removed from the Local Plan. That did not alter the meaning or scope of Policy H1 or Policy H9.

The presence of Policy H3 in the original Local Plan, expressly permitting particular types of development outside Policy H1 and Policy H9, including the type of development which the appellant wishes to carry out, undermines rather than supports the appellant’s case that, even if Policy H3 was never there, that type of development would still have been permissible under the Local Plan. The obvious meaning of the Local Plan is that only those types of development expressly authorised by the various stated Policies are permitted under the Local Plan.

The above interpretation of the Local Plan is reinforced having regard to the wider planning context. Lindblom LJ has referred to many relevant provisions of the Local Plan. I would place particular emphasis on the following statements. “The City Council is aiming to locate the maximum amount of new housing possible on previously developed land” (paragraph 0.4). “The Local Plan sets out a spatial strategy and vision for the District for the period to 2011, and in the longer term.” (paragraph 0.7). “It is the City Council’s objective to provide a choice of good quality and affordable housing in the District with the aim to locate the maximum amount of new housing possible on previously developed land” (paragraph 0.9.2). Further, chapter 2 which has the title “Providing Decent Housing” makes clear (in paragraph 2.1) that the objective is to “maximise housing development on land that has previously been developed, is derelict or underused (brownfield land) within the urban areas”. Paragraphs 2.7, 2.8, 2.9 (dealing with housing development within the urban areas), are to similar effect. Paragraph

Also of particular note is paragraph 2.13, which deals with sites outside urban areas but within villages that are previously developed, used, underused or derelict. It says that such sites could come forward as large windfall sites if they do not have an adverse impact on the social and physical infrastructure of the villages and surrounding areas and are acceptable in all other respects. It said that these will be assessed against Policy H9, and that housing development on previously developed land outside the villages will not be acceptable unless there are exceptional circumstances, and where it is sustainable. This makes perfectly clear that there was never any scope for the application of H3 so as to establish some kind of liberal and unspecified general policy applicable to land in villages not covered by paragraphs 2.55-2.58 and Policy H9.

It follows that the Local Plan was not “silent” on the issue of the appellant’s proposed development, as that notion was explained by Lindblom J in Bloor Homes East Midlands Ltd v Secretary of State for Communities and Local Government [2014] EWHC 754 (Admin) at [50] as follows:

“50.

The answer to the question “Is the plan silent?” will sometimes be obvious, because the plan simply fails to provide any relevant policy at all. But often it may not be quite so clear-cut. The term “silent” in this context does not convey some universal and immutable meaning. The NPPF does not itself explain what the Government had in mind when it used that word. But silence in this context must surely mean an absence of relevant policy. I do not think a plan can be regarded as “silent” if it contains a body of policy relevant to the proposal being considered and sufficient to enable the development to be judged acceptable or unacceptable in principle.”

Section 38(6) of the Planning and Compulsory Purchase Act 2004 requires the determination of planning permission to be made in accordance with the development plan unless material considerations indicate otherwise. For the reasons which I have given, and which have been given more fully by Lindblom LJ, far from the Local Plan adopting the kind of neutrality (i.e. neither express permission nor express prohibition) for which Mr Barrett contended (and even if, which it is unnecessary to decide on this appeal, such “neutrality” was sufficient to satisfy section 38(6)), the Local Plan is clearly and unambiguously against permission for the appellant’s proposed development.