One of the most challenging issues at present for Essex and local authorities is the provision of sufficient homes. One solution has been the proposal of new towns in the North of Essex. Please find copied below the Inspectors report on the proposed town to the West of Colchester.
NORTH ESSEX AUTHORITIES
Strategic (Section 1) Plan
Inspector: Mr Roger Clews
Programme Officer: Andrea Copsey
Tel: 07842 643988
Address: Examination Office, Longcroft Cottage, Bentley Road, Clacton-on-Sea, Essex CO16 9BX
Emma Goodings, Head of Planning Policy & Economic Development, Braintree District Council
Karen Syrett, Place Strategy Manager, Colchester Borough Council
Gary Guiver, Planning Manager, Tendring District Council
8 June 2018
Dear Ms Goodings, Ms Syrett and Mr Guiver
EXAMINATION OF THE STRATEGIC SECTION 1 PLAN ADVICE ON THE NEXT STEPS IN THE EXAMINATION
1. Now that the hearing sessions have concluded I am able to advise you about the further steps that I consider are necessary in order for the Section 1 Plan to be made sound and legally-compliant. I shall also deal, as far as I can, with your question as to whether the Section 1 Plan [hereafter, “the Plan”] could be adopted by each of the three North Essex Authorities [NEAs], separately from and in advance of their Section 2 Plans.
2. My letter focusses on those aspects of the Plan and its evidence base which I consider require significant further work on the part of the NEAs. It also advises on specific changes that are needed to some of the Plan’s policies. More detailed matters, and aspects of the Plan that do not require significant further work at this stage, are not dealt with here but may be considered in the report I will produce at the end of the examination.
3. At this point my letter does not deal with chapter 4 and policy SP3, which cover the Plan’s housing requirements. I will write separately about this topic once I have considered any implications the recently-published 2016 based sub-national population projections may have for the issue of Unattributable Population Change [UPC] in Tendring.
4. In document SD002a, the NEAs have suggested modifications to address some of the issues of soundness that have been identified during the examination. However, it will be clear from this letter that further main modifications will need to be made in order for the Plan to be capable of adoption. All the main modifications that are eventually proposed will of course be subject to full public consultation, and I will consider all the consultation responses before I produce my report.
5. I should make it clear that the views expressed in this letter are based on the evidence currently before me. I reserve the right to modify these views in the light of any further evidence that may come forward before the examination ends.
6. My letter deals first with legal compliance matters, then with Plan chapter 8 on the proposed Garden Communities [GCs], followed by chapters 5 and 6 dealing with employment and infrastructure provision, and then more briefly with the rest of the Plan.
Legal compliance, including compliance with the duty to co-operate
Duty to co-operate
7. Each of the NEAs has published a Duty to Co-operate [DtC] Statement setting out the steps taken to fulfil the duty in the preparation of the Plan. The DtC Statements are supported by Statements of Common Ground with neighbouring LPAs, infrastructure providers, statutory consultees and others.
8. It is apparent from the DtC Statements that substantial and effective cooperation took place, both between the NEAs themselves and with neighbouring authorities and other prescribed bodies, during the preparation of the Plan. This co-operation involved meetings, memoranda of co-operation and joint evidence preparation. The strategic, cross boundary matters addressed included assessments of need for housing, gypsy and traveller accommodation and employment land; strategic infrastructure, including improvements to the trunk and local road networks and the railway network, education, healthcare and broadband provision; and the environmental and other cross-boundary impacts of the Plan’s proposals.
9. Given the distance between the administrative area of Basildon Council and those of the NEAs, it would be unrealistic to expect the latter to play any significant role in accommodating unmet need for gypsy and traveller sites in Basildon. I find no failure of the DtC in this respect.
10. Failure of the DtC was also alleged over the NEAs’, and more specifically BDC’s, handling of the proposals by Lightwood Strategic for a GC at Pattiswick, to be known as “Monks Wood”. It seems that the first proposal for this site, in the context of the Plan, was made to BDC by Sworders on 9 March 2016. I have no reason to disbelieve the NEAs’ account that it was made known to CBC, TDC and Essex County Council [ECC] the following day. BDC then responded to Sworders on 11 March 2016 advising that an earlier call for sites period was closed but that the Pattiswick site could be considered as an objection to the Preferred Options Plan, consultation on which was due to begin in June 2016.
11. Given the stage of preparation that the Preferred Options Plan had reached by March 2016, I consider that was a reasonable position for BDC to take. The alternative would have been to assess the new site in the same way as the other proposed GC sites had already been assessed, before publishing the Preferred Options Plan. But that would have delayed the Plan preparation process, with no guarantee that other sites would not then have come forward, creating further delays. Cut-off dates have to be set if the planning process is to move forward.
12. In August 2016 Lightwood Strategic made representations on the Preferred Options Plan, enclosing a site submission form for the Monks Wood site along with supporting material. In due course, Concept Feasibility Studies for Colchester Metro Town (April 2017) and Monks Wood (May 2017) were prepared by the NEAs’ consultants, AECOM, along similar lines to those already published in June 2016 for other potential GC sites. The latter included another rejected alternative at North Colchester as well as the three allocated sites.
13. North Colchester, the Metro Town proposals and the Monks Wood site were also assessed as alternatives to the allocated GCs in the Sustainability
Appraisal [SA] for the Publication Draft Plan, published in June 2017.
(I consider the SA separately below.) All this is evidence of effective cooperation between the NEAs in the assessment of alternative sites for GCs, including Monks Wood.
14. I see no great significance in the fact that the BDC Local Plan SubCommittee resolved on 31 October 2016 to agree a Vision for its Local Plan that included GCs west of Braintree and west of Colchester, but not at Monks Wood. The relevant recommendation contained a clear caveat to the effect that any subsequent changes to the spatial strategy would be reflected in the Vision. That reflected the fact that consideration by the NEAs of the spatial strategy – of which the GCs are a major component – was still continuing. No final decisions on the Section 1 Plan, its spatial strategy and the GCs allocated in it were taken until the NEAs formally approved the Publication Draft Plan for consultation beginning in June 2017, and then resolved to submit the Plan for examination in October 2017.
15. I see nothing in legislation or national guidance to indicate that the DtC requires local planning authorities to co-operate with prescribed bodies over the potential cross-boundary impacts of sites that are considered, but rejected, for inclusion in a plan. Consequently, I see no reason to conclude that the DtC required co-operation between BDC (or the NEAs) and other external bodies in respect of Monks Wood and the other rejected GC sites. That is also the view of the NEAs and of Chelmsford Borough and Uttlesford District Councils, which adjoin the BDC area. The cross-boundary impacts of the Plan as a whole were the subject of effective co-operation, as paragraph 8 above makes clear.
16. None of the evidence I heard and read pointed to a failure in any other respect on the part of the NEAs to co-operate with each other or with prescribed bodies on any strategic matter. I find that each of the NEAs met the duty to co-operate in the preparation of the Section 1 Plan.
Failure to register representations
17. Through an unfortunate error, the NEAs failed to register the representations submitted by five respondents at Regulation 19 stage. The representations from one of those respondents, Lightwood Strategic, also contained a request to appear before and be heard by the Inspector under section 20(6) of the 2004 Act. Document IED008 sets out, at question 7(a), the elements of legislation that were breached as a result of that failure.
18. The failure to register the five sets of representations did not come to my or the NEAs’ attention until Thursday 18 January 2018, the third day of the originally-scheduled hearing sessions. The missing representations were provided to me and published on the consultation portal the next day. Arrangements were made for Lightwood to submit statements to and appear at the fourth, fifth and sixth hearing days, dealing with Matters 6, 7 and 8, the following week.
19. Lightwood would also have been entitled to appear at the Matter 1 hearing session, which had already taken place when their representations came to light. Consequently an additional hearing session for Matter 1 was held on Wednesday 9 May 2018. Lightwood were invited to submit statements to and attend that additional session, together with all the invitees to the original Matter 1 hearing session, and representatives of parish councils and community organisations in the area affected by Lightwood’s proposals for a GC at Pattiswick.
20. Lightwood consider that, notwithstanding the steps that were taken to overcome the consequences of the NEAs’ failure to register their representations, they and others are subject to prejudice in the following respects:
• They had only a few days to prepare for the Matters 6, 7 & 8 sessions, placing them at a material disadvantage compared to the other participants;
• Their not attending the original Matter 1 hearing session meant that I heard contributions without Lightwood being able to respond to, rebut or reinforce those comments, and without them being supported by others in their own submissions;
• The NEAs’ failure to submit an accurate statement of representations and to submit all representations to the SoS led to the examination proceeding and being framed and formulated by myself without reference to or benefitting from Lightwood’s case and evidence;
• The failure to make all representations publicly available prevented fellow objectors from formulating their cases and representations with reference to or benefitting from Lightwood’s case and evidence.
21. Lightwood maintain that a failure to comply with Regulation 22 cannot be cured subsequently. It is not possible after the event, they say, to gauge how the examination, evidence and representations would have altered as a result of their representations being available or how parties would have conducted themselves. Lightwood contend that the Plan should therefore return to the stage before the breach.
22. Evidently the NEAs’ failure to register the five sets of representations was a regrettable error, for which they have apologised. The question for me is whether Lightwood’s interests, or those of any other party, have been prejudiced as a result.
23. In this regard, steps were taken to ensure that Lightwood were able to appear and be heard before me on all the Matters to which their representations related. Those steps included arranging an additional hearing session for Matter 1, as explained above. While it is true that Lightwood had only a short time to prepare for the Matters 6, 7 and 8 hearing sessions, it was they who originally suggested that they should attend those hearings. Their suggestion, which I accepted, was extremely helpful in minimising delay to the examination. Lightwood were able to prepare brief hearing statements for Matters 6, 7 and 8, and I and the other participants had the opportunity to read and consider all their representations in advance of the hearing sessions.
24. As a result of all the steps taken, my view is that Lightwood and the other participants were provided with adequate opportunities to appear before and be heard by the Inspector, as the legislation requires. I consider it unlikely that any significant additional matters, issues and questions would have been discussed at the hearings had Lightwood’s representations, and the other unregistered representations, been before me at the outset of the examination. Taking all this into account, I am satisfied that the hearing sessions enabled me gain a full understanding of the views of all participants, including on Matters 1, 6, 7 & 8.
25. Overall, therefore, I find that that effective arrangements were put in place to minimise the effects of the failure to register certain representations at the right time, and that no substantial prejudice to any party resulted from that failure.
Habitats Regulations Assessment [HRA]
26. On 12 April 2018 the Court of Justice of the European Union [CJEU] issued a judgment which ruled that Article 6(3) of the Habitats Directive must be interpreted as meaning that mitigation measures (referred to in the judgment as measures which are intended to avoid or reduce effects) should be assessed within the framework of an appropriate assessment [AA], and that it is not permissible to take account of measures intended to avoid or reduce the harmful effects of the plan or project on a European site at the screening stage.
27. The HRA report on the pre-submission Plan contained both a screening stage assessment and an AA. The NEAs will need to ensure that the screening stage assessment of that report, and any future HRA reports, is compatible with the CJEU’s judgment.
Other legal issues
28. I find no evidence that the NEAs failed to consult on the Plan in accordance with their Statements of Community Involvement, as required by section 19(3) of the 2004 Act. I shall consider any relevant implications of the legislation on state aid and compulsory purchase in the next section dealing with the GCs.
Cross-Boundary Garden Communities (chapter 8)
29. Three proposed GCs, providing between 29,000 and 43,000 homes in total, are a central element in the Plan’s spatial strategy for North Essex. I have no doubt that the NEAs are sincere in their aspirations for three highquality, sustainable communities, based on the principles outlined in their Garden Communities Charter [the NEGC Charter]. Their proposed approach is innovative and ambitious, and if carried out successfully it has the potential to provide for housing and employment needs not just in the current Plan period but well beyond it.
30. The GCs are identified as broad locations on the submission policies map. But it is clear from the content of policies SP7, 8, 9, & 10 [hereafter: “the GC policies”] that the submitted Section 1 Plan, if adopted, would establish both the in-principle acceptability of, and many of the specific requirements for, the proposed GC developments. Follow-on plans are intended to set out the principles of design, development and phasing for each GC, but it is this examination which must determine whether or not the GC proposals are properly justified and realistically developable. This is of more than usual importance given the large scale and long-term nature of the GC proposals, two of which will take around 30 years to complete and the other at least 40 years.
31. In my view the evidence provided to support the GC policies in the submitted Plan is lacking in a number of respects. I consider the main shortcomings in turn below. References to the individual GCs are as follows: CBBGC: Colchester/Braintree Borders GC; TCBGC: Tendring/Colchester Borders GC; WoBGC: West of Braintree GC.
Trunk road improvements
32. Policy SP5 includes two major trunk road schemes in its list of strategic infrastructure priorities: the A12 Chelmsford to A120 widening scheme which is included as a committed scheme in Highways England’s RIS1 programme, and the A120 Braintree to A12 dualling scheme which is currently under consideration for inclusion in RIS2. Both schemes are intended to relieve existing congestion problems and support economic growth in North Essex.
33. In this context, the scale of the GC proposals means that they could not be developed in full without the additional strategic road capacity provided by these schemes. In particular, WoBGC would be reliant on the A120 for eastward strategic road connections to Colchester and beyond, and both the A120 and A12 (which currently meet at Marks Tey) would provide essential strategic highway links for CBBGC.
34. I understand that decisions on what is included in the RIS2 programme are due to be made in 2019. No firm view on the feasibility of either WoBGC or CBBGC can be taken until it is known whether or not the A120 dualling scheme is included in that programme (or can be otherwise fully funded). While the GCs would contribute to the cost of the scheme, I have seen no evidence that it could be fully funded if it is not included in RIS2. It may be possible to devise interim solutions to accommodate a proportion of the generated traffic, and thereby enable early phases of one or both GCs to proceed, but that would not justify an in-principle endorsement of the GC proposals as a whole.
35. Moreover, the two alternative alignments currently under consideration for the widened A12 in the Marks Tey area are not compatible with the proposed layout of CBBGC as set out in the Concept Framework. In order to avoid having an unacceptable severance effect, the improved A12 would need to take a line some distance to the south-east of those existing alternatives. The NEAs have made a bid to Government for funds to facilitate that further alternative alignment, but the outcome is not yet known.
36. In addition, a decision has yet to be made on the alignment for the dualled A120. The alternative alignments still being considered have quite different implications for the A120’s relationship with CBBGC.
37. I appreciate that the NEAs, ECC and Highways England are working together constructively to resolve these issues. Nonetheless, greater certainty over the funding and alignment of the A120 dualling scheme and the feasibility of realigning the widened A12 at Marks Tey is necessary to demonstrate that the GC proposals are deliverable in full.
Rapid transit system
38. A rapid transit system [RTS] for North Essex is an integral part of the GC proposals. Policy SP7 requires the new communities to be planned around a “step change” in integrated and sustainable transport systems. The Concept Frameworks for each GC all include a RTS as a key element of the movement and access framework. And the Jacobs Movement and Access
Study [MAS] sets a target for 30% of all journeys to, from and within the GCs to be made by rapid transit, rising to 38% for journeys with an external origin or destination.
39. It is unlikely that those extremely ambitious targets would be achieved or even approached unless rapid transit services to key destinations are available early on in the lifetime of the GCs. That is evident from section 1.3 of the MAS, which advises that the priority is to provide high-quality infrastructure for active modes and rapid transit that is integrated with immediate and future land use. It must have a directness, journey time and convenience benefit over the private car from the very beginning to realise this potential.
40. However, planning of the proposed RTS has reached only a very early stage. The North Essex Rapid Transit Study [NERTS] is a high-level assessment of the costs and benefits of a RTS. It assesses demand, and outlines route options and a range of costs, for an extensive network linking the three GCs to Colchester, Braintree and Stansted. But it is not a feasibility study which investigates whether such a network could actually be delivered on the ground. Nor does it recommend which of the modal options (bus, guided bus, tram, etc) should be taken forward, or identify a timescale for delivery.
41. The cost of the RTS, even in broad terms, cannot be determined until these decisions have been made. While the Technical Note on bus rapid transport prepared by Iceni Projects provides alternative indicative costings it does not resolve these points. Further work is needed before it can be shown in both practical and financial terms that a RTS could be delivered.
42. In order to demonstrate that the RTS is deliverable at the time it is needed, further work needs to be done:
• to determine which modal option is to be used and its capital cost implications;
• to establish the feasibility and capital cost of its route(s) on the ground, including its alignment outside the GCs themselves;
• to refine passenger and revenue forecasts; and to establish a timescale for its delivery in stages.
43. On the basis of this work, both a realistic range of costs for the RTS, and the sources from which those costs will be met, need to be identified. Discussions also need to be held with potential operators so that they are involved in the process of developing the proposals.
Marks Tey station
44. The existing Marks Tey railway station, on the Great Eastern Main Line [GEML] between London and Norwich, is within, but close to the eastern edge of, the indicative boundary of CBBGC. In principle, the station would be a considerable asset for CBBGC. However, its current peripheral position would integrate poorly with the structure of the GC. The CBBGC Concept Framework proposes its relocation some 2km to the south-west, where it would form part of a transport interchange in the new town centre.
45. Discussions with the railway infrastructure providers on this proposal are at an early stage, and no firm commitments to it have been made. Moreover, at present there are no clear proposals on how to maintain interchange between the GEML and the Sudbury branch line, which currently occurs at Marks Tey. Adequate interchange arrangements would be essential to the acceptability of the relocation scheme.
46. Both the Concept Framework and policy SP9 make it clear that they do not see the relocation of Marks Tey station as essential to the success of the GC. Nonetheless it would be a significant missed opportunity, in my view, if a GC on the scale currently proposed in this area were to proceed with the station on its periphery. As the Concept Framework points out, a station in a town centre generates a focus of activity, supporting higher density development and helping to create an active and vibrant centre.
47. The Hyas viability appraisal for CBBGC allows £50M towards the cost of relocating the station. While work will need to be done to refine that figure and to identify other sources of funding, it is a reasonable allowance to make at this stage. However, it appears in the spreadsheet in 2057/58, 30 years into the proposed build period. That is far too late to enable the station to be integrated into the planning of the new town centre, and for it to have the beneficial effects envisaged by the Concept Framework. If the relocation of Marks Tey station is to form part of a proposed GC, the allocation of funding for it must be made much earlier in the build period.
Delivery of market and affordable housing
48. The NEAs’ own publications envisage each of the three proposed GCs starting to deliver housing in 2023/24. WoBGC is expected to deliver 250 dwellings in that first year and in each subsequent year to the end of the Plan period (2033). The other two GCs would build up more gradually to rates of 300 dwellings per annum [dpa] for TCBGC from 2027/28 onwards and 350dpa for CBBGC from 2031/32 onwards. The Hyas appraisal envisages slightly different delivery rates.
49. Credible research by NLP indicates that sites over 2,000 dwellings take an average of around seven years from the submission of the first planning application to the delivery of the first dwellings on site. However, it also shows that planning approval for greenfield sites tends to take somewhat less time than for brownfield. Moreover, the work already done by the NEAs and others to develop concept frameworks and masterplans for each GC would help shorten that time further.
50. On this basis I consider it reasonable to assume that the planning approval process would allow housing delivery at any GC(s) to start within four or five years from the adoption date of the plan (or plan revision) which establishes the GC(s) in principle. However, that timescale could alter depending on how long it takes to put the necessary infrastructure in place, as discussed above.
51. The NLP research found that greenfield sites providing more than 2,000 dwellings deliver around 170dpa on average, with substantial variation around that mean figure. Factors supporting a higher delivery rate include the market strength of the area, the size of the site, public sector involvement in infrastructure provision, and the proportion of affordable housing.
52. All these factors suggest that GCs in North Essex could achieve build-out rates higher than the NLP average. Nonetheless, out of the 13 sites in this category NLP identified only one large site which achieved average delivery of more than 300dpa, and the data for that site cover a period of only three years. Moreover, their analysis of the few sites for which data is available over 10 years or more revealed pronounced peaks and troughs in the annual delivery figures.
53. All this leads me to the view that, while it is not impossible that one or more of the GCs could deliver at rates of around 300dpa, it would be more prudent to plan, and carry out viability appraisal, on the basis of an annual average of 250dpa.
54. The way in which the numbers of dwellings delivered at the GCs would be allocated to the individual NEAs for monitoring purposes is set out in Plan paragraph 8.15. I find nothing to indicate any in-principle conflicts between the proposed approach and national policy or guidance. However, the Plan also needs to make it clear how the allocation would be carried out in the event of a shortfall in planned delivery – the current approach of deferring that question to a future memorandum of understanding is not an effective one.
55. The GC policies seek 30% affordable housing as part of the overall housing provision in each GC. Achieving that proportion is necessary both to meet the demonstrated need for affordable housing in the Plan area and to achieve the NEGC Charter’s goal of creating mixed and balanced communities. Because of the shortcomings in the Hyas viability assessment outlined below, its conclusions over the deliverability of affordable housing at each of the three allocated GCs cannot be relied upon. The further viability work that needs to be undertaken to correct those shortcomings will, therefore, also need to demonstrate that 30% affordable housing can be delivered at any GC that may be proposed.
56. The NEGC Charter’s Principle 3 seeks to provide access to one job per household within each new GC or within a short distance by public transport. It states that the employment function will be a key component of creating character and identity and sustainable communities. Policy SP7 describes the GCs as incorporating a range of homes, employment and other facilities, thereby reducing the need for outward commuting.
57. In this context, it is surprising that the GC policies contain no specific figures for the amount of employment land or floor-space to be provided at each of the GCs. Instead there are only general requirements to provide and promote opportunities for employment and a wide range of jobs, skills and training opportunities, and suggested locations for different types of employment use. This is in contrast to the figures (expressed as a range) in each policy setting out the expected level of housing development.
58. I recognise that setting employment land requirements for different uses and allocating land to meet them is a complex process, involving forecasts of market demand across different employment sectors. If the sites provided do not match the demands of the market, the jobs will not come. To that extent I agree with the NEAs that it is not possible to predict accurately the exact mix of employment space that will be required this far in advance of development. But that would not preclude setting indicative requirements for the overall amount of employment land or floorspace at each GC.
59. It would be inappropriate to delegate this role to the individual DPDs, as the NEAs suggest. The role of the DPDs is to take forward the design, development and phasing of the GCs based on the principles established in this Plan. It is difficult to see how they could perform that role without an indication of the amount of employment the GCs are expected to provide. Setting indicative requirements in this Plan would not prevent them changing in future: indeed they should be reviewed each time the Plan itself is reviewed, to ensure that they continue to reflect economic realities.
60. Section 5 of the report North Essex GCs Employment & Demographic Studies [E&DS] sets a range of future employment estimates and associated floorspace requirements for each GC. These are derived from a series of demographic and employment projections based on analysis of existing local conditions and potential future scenarios.
61. However, both the Hyas report and the Concept Frameworks that have been developed for each GC include alternative employment land and floorspace allocations which are apparently more ambitious than those based on the E&DS scenarios. If the NEAs wish to set indicative requirements for the GCs at those levels, they would need to be supported by evidence at least as robust as that provided in the ED&S.
62. The most recent assessment of the GCs’ financial viability before me is the April 2017 Viability Assessment by Hyas [“the Hyas report”]. The assessment was conducted at a strategic level, appropriate to the relatively early stage of evolution of the GC proposals. It follows the residual valuation method, in which all the costs of undertaking the development – apart from the land cost – are subtracted from the development’s total sale value. The resulting figure is the residual value. If the residual value is at least equal to the cost of acquiring the land needed for the development, then the development can be said to be viable.
63. For reliance to be placed on the outcome of the assessment, well-founded assumptions need to have been made about both the likely costs and value of the development, and about the cost of acquiring the land.
64. In terms of costs and value, the Hyas report makes generally reasonable assumptions about development mix and value, and about land preparation, construction and utilities costs, and developer profit. However, as explained below it does not deal adequately with transport infrastructure costs, land purchase and interest, or contingency allowances.
Transport infrastructure costs
65. The evidence provided to support the Hyas report – including additional information from the AECOM Social Infrastructure Model – shows that costings for most items of infrastructure were arrived at in a consistent and logical manner and are generally reasonable.
66. However, as noted above the proposals for a rapid transit system, the provision of which is essential to the successful development of the GCs, are still at a very early stage. According to the NERTS, the capital costs of the scheme range between £249m and £1,672m (including a prudent 64% optimism bias allowance) depending on which option is eventually chosen. The direct and indirect RTS contributions allowed for in Hyas’s baseline appraisals for the three GCs appear unlikely to meet even the lowest of those figures. Nor has any clear evidence been provided to show that the balance of the RTS’s capital costs could be funded from other sources.
67. Consequently, it is by no means clear that adequate allowances for the costs of necessary transport infrastructure have been built into the viability assessment. To ensure that the viability assessment reflects the actual cost as closely as possible, the relevant figures should be reviewed when the rapid transit system proposal is further advanced and more accurate information is available on its likely cost.
68. If any additional contributions from the GCs, apart from those already included, are expected towards the A12 widening or the A120 dualling scheme, they would also need to be allowed for in the viability appraisal.
Land purchase and interest
69. The Hyas report uses a financial model, developed by ATLAS, based on a “master-developer” model of delivery. In this model the master developer acquires the development land and undertakes strategic investment in enabling works and strategic infrastructure, before selling on the serviced plots to individual housebuilders or commercial developers to build them out. Interest on borrowing to fund the strategic investment, and a financial return to the master developer on that investment, are built into Hyas’s viability assessment.
70. It is unclear whether the 6% interest figure assumed for strategic investment borrowing is justified, having regard to the legislation on state aid as highlighted in the advice to the NEAs by PwC. Further clarification on this point is necessary.
71. More importantly, however, no allowance is made in the Hyas appraisal for interest on borrowing to fund land purchase by the master developer. The Harman report Viability Testing Local Plans (June 2012) specifically warns against overlooking interest costs on land purchase. Given the scale and duration of the GC development programme, those costs will be substantial. In their response to Government on the New Towns Act 1981 (Local Authority Oversight) Regulations, the NEAs themselves refer to “significant land costs which will be largely debt-funded in advance of land receipts”.
72. In order to take account of land purchase interest costs, the residual values shown in Hyas’s summary tables 5.3.1, 6.3.1 and 7.3.1 would need to be discounted by an appropriate amount. That would require assumptions to be made about the timing of land purchase and disposal. For example, the earlier GC viability work by AECOM assumed that land would be purchased in tranches two years before it was required for development.
73. Until Hyas’s residual values have been adjusted to take account of the substantial cost of interest on land purchase, no reliance can be placed on them as an indication of the viability of the proposed GCs.
Contingencies and sensitivity testing
74. The Hyas report modelled a range of different scenarios for each GC. The variables used were: various proportions of market and affordable housing and starter homes; uplifts of 0%, 5% and 10% on overall infrastructure costs; and uplifts of 0%, 5% and 10% on development value (to reflect a “Garden Community premium”).
75. 10% would be an unusually low figure if it was intended to represent the sole contingency allowance on infrastructure costs. The NEAs produced further evidence12 setting out what they claimed amounted to a total 42% contingency allowance for CBBGC, as an example of the approach taken for all three GCs. Over a third of that amount, however, is the 15% profit allowance intended as an incentive to perform the master-developer role referred to above.
76. A 15% profit allowance is not excessive given that, as the NEAs accept, the Plan needs to be neutral as to whether the master-developer role is played by a public or private sector body. Even if the oversight role is retained in the public sector, it is quite possible that many of the master-developer functions would need to be outsourced. Consequently, the master developer profit allowance should not be counted as part of the overall contingency allowance.
77. The other additional element which the NEAs identified as part of the total contingency allowance was what they termed “in-built contingency” of around 24% on certain capital sums for infrastructure. Tracing these figures back to their source documents shows that most do indeed represent an uplift of around 20% on the minimum cost identified for each item. However, as was demonstrated at the hearing sessions, 20% or 24% is a low contingency figure for major capital projects. A contingency allowance of at least 40% would align better with the approach taken, for example, by Highways England when costing large-scale infrastructure schemes.
78. I recommend therefore that alongside the generic cost uplift figure of up to 10% used in the Hyas report, sensitivity appraisals are carried out based on additional contingency allowances of 20% and 40% on relevant infrastructure schemes for each GC, such as road improvements, park-and-ride and rapid transit. That would give an adequate range of possible costs to inform the overall viability assessment.
79. On the income side, my comments above on the likely rate of housing delivery at the GCs will need to be taken into account when calculating receipts from development value. It is important also that realistic assumptions are made about the income generated by commercial floor-space. I have commented above on the discrepancies between the employment land and floor-space allocations used in the Hyas report and those identified elsewhere in the evidence base.
80. I recognise that the aim of bringing forward homes rapidly at the GCs may conflict with the ability to achieve a GC premium on house prices. That does not mean that Hyas were unjustified in sensitivity-testing a 5% and 10% premium, in order to appraise a range of possible outcomes. However, it is inconsistent with this approach to regard the £3,000 per unit uplift applied to site preparation and enabling costs as a contingency allowance, as identified in EB/13(2/2a). Given that the avowed purpose of the uplift is to create a high-quality public realm and sense of place, it would seem to be essential if any GC price premium is to be achieved.
81. I share the NEAs’ view that it would not be helpful to attempt to include an allowance for inflation in the residual valuation appraisal. Predicting movements in house prices in particular would be difficult over such a long period, and allowing for cost inflation would be meaningless without a corresponding adjustment for development value.
Price of land
82. There is a difference between the headline value paid for a fully-serviced development site, and the net value which takes account of the costs of enabling works and strategic infrastructure, and of policy requirements such as the provision of affordable housing. The net land value is the appropriate comparator with the residual value that emerges from a valuation model such as that used by Hyas. In other words, it is quite appropriate to take account of up-front enabling and infrastructure costs (which in the Hyas/ATLAS model are incurred by the master developer) and policy requirements, when negotiating to purchase land for development.
83. However, as the Harman report points out, what ultimately matters for housing delivery is whether the value received by the landowner is sufficient to persuade him or her to sell the land for development. I consider it unlikely that most landowners would sell their land for development without at least a reasonable uplift on its existing use value. This has clear implications for the deliverability of the GCs.
84. That does not necessarily mean that a price of £100k per acre would need to be paid, as is suggested in Volume 3 of the GC Concept Feasibility Study. Ultimately, of course, the actual land price will emerge from negotiations with individual landowners. But in order to demonstrate that the GC proposals can be delivered, the NEAs will need to show through viability assessment that a reasonable uplift on current use values can be achieved.
85. Alternatively, if the NEAs intend to use compulsory purchase or other powers to acquire development land at a lower value than could be achieved through negotiation, clear evidence would need to be provided that such a course of action is capable of achieving that outcome (and is also compatible with human rights legislation). That has not been demonstrated by the evidence currently before me.
Conclusions on viability
86. For the foregoing reasons, it has not been demonstrated that the GCs proposed in the submitted Plan are financially viable. Further viability assessment, taking account of all the points above, will need to be carried out on any GC proposals that the NEAs bring forward. Because of the GCs’ long development timescales, it would be advantageous for the residual valuation appraisal to be supplemented with a discounted cashflow assessment in order to provide a more complete analysis.
87. The NEGC Charter envisages that Local Delivery Vehicle(s) [LDVs], accountable to the NEAs with both private and public sector representation, will be responsible for delivering the GCs. Three LDVs together with a holding company called NEGC Ltd have been incorporated in readiness to perform this role. Subsequently, in response to consultation on the proposed New Towns Act 1981 [Local Authority Oversight] Regulations, the NEAs have indicated an interest in the formation of a locally-led development corporation, overseen by the NEAs, to deliver the GCs.
88. The Charter also envisages a private-public sector partnership funding arrangement for the GCs involving the sharing of project risk and reward. Public sector investment in the funding and delivery process, it is said, will help to facilitate the timely and co-ordinated provision of infrastructure and services.
89. The Hyas report envisages that the LDVs will perform the role of master developer for each GC. Similarly, the NEAs’ response to consultation on the draft Local Authority Oversight Regulations suggests that the locally-led development corporation would act as master developer. As the Charter makes clear, there are likely to be advantages in terms of public engagement, long-term democratic oversight and access to public financial support if the master developer is a public-sector entity. However, this is not a legal or practical requirement. In principle the role could also be performed by a private-sector body.
90. In its paragraph (ii), policy SP7 seeks to encapsulate the principles that the delivery model for the GCs should follow and the objectives it should seek to achieve. The requirements it places on landowners and promoters to secure high-quality place-making, to fund the infrastructure necessary to address the impacts of development, and to manage and maintain the onsite infrastructure are generally compatible with relevant guidance in the NPPF and PPG. The final sentence of the paragraph defines the tasks the delivery model will need to perform, taking an appropriately neutral stance on who will perform them.
91. However, the specific reference in the first sentence to “sharing risk and reward” between the public and private sector conflicts with the longestablished legal principle that revenue or profit may not be appropriated by a public-sector body without explicit Parliamentary sanction. The reference may have been intended by the NEAs as a statement of aspiration, but its inclusion in SP7 as one of the principles with which the GCs “will conform” makes it an unlawful policy requirement. It is therefore necessary to remove it from the policy, as the NEAs now propose.
92. In the same sentence, it is also necessary for soundness to remove the reference to “deploying new models of delivery” as a policy requirement. It may be a legitimate aspiration of the NEAs but there is no substantial evidence to show that only (unspecified) new models of delivery are capable of achieving the policy’s objectives.
93. SA of the Section 1 Plan was carried out by ECC’s Place Services at both the Preferred Options and the Draft Publication stage. The resulting reports were published for consultation alongside the Plan in June 2016 and June 2017 respectively.
94. The 2016 SA report contains an assessment of the preferred spatial strategy and four alternatives to it, and an assessment of eleven GC options, of which three were selected for inclusion in the Preferred Options version of the Plan. By comparison, the 2017 report assesses six alternatives to the chosen spatial strategy, and thirteen GC options. In the later report there is also an appraisal of three different approaches to strategic growth, and an assessment of the cumulative impacts of the three allocated GCs and of nine alternative combinations. The significantly wider scope and content of the 2017 report is evidence that account was taken of the responses to consultation in 2016.
95. It may be that the NEAs had decided, before the 2016 report was complete, which GCs they wished to include in the Preferred Options version of the Plan. That in itself is not unlawful, provided that the SA is approached with an open mind, and that its results and the consultation responses on it are taken into account in the ongoing preparation of the Plan. Similarly, the fact that the spatial strategy and the three allocated GCs remained essentially unchanged between the Preferred Options and the submitted versions of the Plan is not necessarily evidence of a closed-minded approach to plan preparation. The important question is whether the SA and the related plan preparation processes were carried out lawfully and with due regard to national policy and guidance.
96. In my view there are three principal shortcomings in these respects concerning, first, the objectivity of the assessment of the chosen spatial strategy and the alternatives to it, secondly, the clarity of the descriptions of those alternatives and of the reasons for selecting them, and thirdly, the selection of alternative GCs and combinations of GCs for assessment. I shall consider each in turn.
Objectivity of assessment
97. As noted above, four alternatives to the chosen spatial strategy were assessed in the 2016 report, and six alternatives in the 2017 report. In both reports the short- and medium-term results are identical for the chosen spatial strategy (which includes the three allocated GCs) and all the alternatives. That is to be expected, since there would be no substantial development at the GCs until later in the Plan period. The key comparison is of the long-term results, which are intended to show effects in the latter stages of the Plan period and, where relevant, beyond.
98. In the long term the chosen spatial strategy is assessed in the 2017 report as having a strong prospect of significant positive impacts on six sustainability objectives relating to: housing, health, vitality and viability of centres, the economy, sustainable travel behaviour, and accessibility and infrastructure provision. By contrast, Alternative 4, which involves growth at existing settlements without the allocation of any GCs, is assessed as having strong or minor negative effects on all those objectives except for sustainable travel behaviour, where its effects are said to be uncertain.
99. Taking into account my findings above on the GC proposals, it is not possible to see the objective basis for many of the widely divergent assessments of these two scenarios. Without more evidence to show that the necessary transport infrastructure for the GCs could be provided viably and in a timely fashion, the strong positive scores for the chosen strategy in respect of sustainable travel behaviour and accessibility are unwarranted. The lack of any quantitative employment land or floorspace requirements for the GCs undermines the strong positive score given to its economic benefits. There is no substantial evidence to show that the chosen spatial strategy would have strong benefits in terms of health and the vitality and viability of centres, or that Alternative 4 would detract from these objectives.
100. The narrative on page 83 of the 2017 report explains the reasons for rejecting Alternative 4. It says that if no GCs were to be allocated, existing settlements would have to respond to the need for growth by allowing higher densities and the development of more marginal peripheral land. This could lead to the over-expansion of some settlements and would not offer a sustainable distribution across the wider area. While this goes some way towards explaining the negative score given to Alternative 4 in terms of its landscape impact, it does not account for the strong negative impact it is seen as having on the objective of housing provision.
101. Similar comments apply to the analysis at pages 171-184 of the 2017 report, where the GC approach to strategic scale growth is compared with what are described as “New Towns” and “Traditional Approaches”. Traditional Approaches appear from their description to correspond quite closely to Alternative 4 as described above.
102. In this analysis, Traditional Approaches receive negative scores for their ability to provide well designed and sustainable housing, for their effects on designated nature conservation sites, and for their ability to provide for adequate school places, recreational facilities and open space, without any clear evidential basis for these judgments. GCs again receive positive scores for sustainable transport provision, employment opportunities, and the viability of existing centres, which I regard as unwarranted for the reasons given above.
103. As a result, I consider that in assessing the chosen spatial strategy against alternatives that do not include GCs, the authors of the SA report have generally made optimistic assumptions about the benefits of GCs, and correspondingly negative assumptions about the alternatives, without evidence to support many of those assumptions. As a result these assessments lack the necessary degree of objectivity and are therefore unreliable.
Clarity of descriptions of alternatives and reasons for selection
104. Two of the alternatives to the chosen spatial strategy are described in the
2017 SA report as follows: A focus on allocating all of the explored Garden Community options proposed in the Strategic Area at smaller individual scales, and A focus on stimulating infrastructure and investment opportunities across the Strategic Area. In response to my questions at the 9 May hearing session, the NEAs explained that the first of these involved the allocation of five GC options for 2,500 dwellings each within the Plan period, and that the second involved the allocation of three GCs in areas where there was an evidenced need for regeneration.
105. However, it would have been difficult to understand from the descriptions given in the report that this is what they involved. Indeed, the reference to “smaller individual scales” in the first option is actually misleading, since the three GCs in the chosen spatial strategy are also intended to deliver 2,500 dwellings each within the Plan period. And the lack of reference to GCs in the second option obscures the fact that it involves allocating three of them.
106. There is a similar lack of clarity in the reasons given for selecting the alternatives for assessment. The paragraphs on pages 79-80 of the 2017 SA report which introduce the alternatives do little more than provide descriptions of them. There is no substantial account of the rationale for choosing those particular alternatives.
107. I appreciate that a somewhat fuller description is given of the “New Towns” and “Traditional Approaches” which are assessed as alternatives to GCs on pages 171-184 of the 2017 SA report, and of the reasons for their selection. But that is a different level of analysis, assessing the relative benefits of GCs in general terms. It is the analysis at pages 76-84 which is intended to appraise the particular spatial strategy proposed in the Plan and reasonable alternatives to it, as the legislation requires.
108. Reasons are given on page 82 of the SA report for rejecting Alternatives 2 and 3, involving the allocation of one or two GCs only. It is apparent from the reasons given that the assessment was conducted on the basis that, in each of these alternatives, the GC(s) were assumed to provide all the 7,500 dwellings within the Plan period that would be provided by the three GCs in the chosen spatial strategy. But that is not explained clearly in the description of Alternatives 2 and 3. Nor is it explained why these alternatives could not also have been assessed on the more reasonable basis that each GC would provide 2,500 dwellings in the Plan period, with the rest of the 7,500 dwellings provided at or around existing settlements in a similar fashion to Alternative 4.
109. I consider that the lack of clarity I have identified in the descriptions of some of the alternatives to the chosen spatial strategy, and in the reasons for selecting them, is likely to breach the legal requirements for the SA report to provide an outline of the reasons for selecting the alternatives dealt with, and for the public to be given an effective opportunity to express their opinion on the report before the plan is adopted.
Selection of GCs and combinations for assessment
110. The thirteen GC options assessed in Appendix 1 of the 2017 SA report include the original eleven from the 2016 report plus the Colchester Metro Plan proposed by CAUSE and Lightwood’s proposed Monks Wood site.
111. There is some confusion over the basis on which Monks Wood was assessed as a GC option. On pages 188-199 of the 2017 SA report it is stated that it was assessed as providing up to 15,000 dwellings, including 5,151 in the Plan period. That figure of 15,000 is at odds with the published AECOM evaluation of Monks Wood (May 2017) which on page 32 refers to its development capacity as 5,151 dwellings in total.
112. The source for the 15,000 dwellings figure used in the SA report appears to be a March 2017 draft of the AECOM assessment. It is unclear how that figure was derived, but it is not reflected in any of the material submitted by Sworders or Lightwood in support of their proposals for Monks Wood. Lightwood did assess options providing up to 13,600 dwellings in a study provided to BDC on 31 March 2017. However, their position now is that its maximum capacity is 7,000 dwellings.
113. No blame necessarily attaches to the authors of the SA report for assessing Monks Wood on the basis of 15,000 dwellings, as it seems they were working with the figure given to them by AECOM at the time. That is consistent with the approach they took to the other alternative GC sites. However, as there is no clear evidence to support that figure, the assessment cannot be relied upon. I do not accept that it would have made no difference if Monks Wood had been assessed on the basis of 7,000 or 5,000 dwellings rather than 15,000. It is clear from the assessments of the other GC options that there are some variations in scoring that can only be explained by similar differences in scale.
114. The assessment of alternative combinations of GC sites is at pages 226-244 of the 2017 SA report. The NEAs’ explanation that the results of the assessment of Option 5 (WoBGC, Monks Wood & CBBGC) also justify rejection of a combination of Monks Wood, CBBGC & TCBGC is unconvincing given the very different relationships between the three locations in each of those scenarios. It is difficult to see the logic of assessing Monks Wood as an alternative to CBBGC and to TCBGC, but not to WoBGC, when appraising combinations of three GCs. Moreover, the Option 5 assessment is likely to have been influenced by an inaccurate understanding of the scale of the Monks Wood scheme, as already discussed.
115. In order to demonstrate that all the alternatives had been assessed on an equivalent basis, Monks Wood would need to have been assessed as a GC option at a scale of around 5,000 dwellings corresponding to the published AECOM evaluation, and an additional three-GC combination of Monks Wood, CBBGC & TCBGC would need to have been assessed. The absence of such assessments is a further shortcoming of the SA.
Other SA points
116. On page 185 the SA makes it clear that a minimum threshold of 5,000 dwellings was set when selecting GC options for assessment. That is substantially higher than the minimum size of 1,500 dwellings set by the Government for garden village proposals. It is also higher than the thresholds of 3,000 houses or 4,000 dwellings (houses and flats) requiring a new secondary school, according to ECC’s Developers Guide to
Infrastructure Contributions (2016). However, the latter thresholds would support only a four-form entry secondary school, the minimum size that ECC regard as financially viable.
117. In setting the GC threshold it was legitimate, in my view, for the NEAs to take account of the increased financial viability, curriculum choice and range of facilities that a larger secondary school could provide. It was logical also to take into account the greater range of employment opportunities, healthcare and other community facilities that could be supported by a GC of 5,000 dwellings compared with a smaller settlement.
118. It is not feasible to test every possible option through SA. Reasonable planning judgments have to be made on what to include. That is recognised in the legal requirement for reasons to be given for the selection of alternatives for assessment. In my view the SA report provides adequate reasons for setting a threshold of 5,000 dwellings for the GC options.
Conclusions on SA
119. I have considered the SA at length as it is the principal evidence document that seeks to justify the NEAs’ choice of a spatial strategy involving three GCs, and their choice of the three allocated GCs themselves. Because of the shortcomings I have identified, I consider that the SA fails to justify those choices. As a result, it has not been demonstrated that the chosen spatial strategy is the most appropriate one when considered against the reasonable alternatives, as the tests of soundness require.
120. It may be helpful for me to set out some suggestions as to how the shortcomings in the SA might be rectified. I stress that these are suggestions only, and are intended to provide no more than an outline of the further work required. I would be happy to consider any alternative SA proposals the NEAs might wish to make, provided they address the shortcomings I have identified. In either case it would be advisablel if I were to agree the proposals before the SA work is begun.
121. In making these suggestions I rely on the principle that deficiencies in SA may be rectified, or “cured”, by later SA work, established in the Cogent Land case and restated by the Court of Appeal in No Adastral New Town Ltd. I do not agree that the other caselaw drawn to my attention indicates that the scale of the GC proposals would preclude such an approach here. My suggestions also assume that the NEAs will wish to continue to include GCs among the options in any future SA work.
122. Before embarking on further SA work the NEAs will need to re-examine the evidence base for any GC proposals they wish to assess, especially with regard to viability, the provision of transport infrastructure and employment opportunities, in order to ensure that they have a sound basis on which to score them against the SA objectives.
123. The first stage in the further SA work should then be an objective comparison of individual GC site options at a range of different sizes. My comments above on the way that GC sites were selected for assessment in the 2017 SA report should be taken into account at this stage. In particular, if Monks Wood is included as an option it would be sensible – unless further evidence to the contrary emerges – to assess it on the basis of both 7,000 dwellings, as now favoured by Lightwood, and 5,000 dwellings as in the published AECOM report. If WoBGC is included, account should be taken of the effects on it of overflying aircraft to and from Stansted airport, and of its impact on the Andrewsfield airfield, in order to address legitimate concerns raised at the Matter 8 hearing.
124. Adequate reasons will need to be given for taking forward or rejecting each of the GC options assessed. Assessing the GC options first, with the benefit of an updated evidence base and before the spatial strategy options, should help to ensure that the assessment of the latter is appropriately realistic.
125. The second stage of the further work should be an assessment of alternative spatial strategies for the Plan area. The alternatives considered, and the reasons for selecting them, will need to be set out more clearly than the alternatives on pages 79-80 of the 2017 SA report. I suggest that the alternatives should include, as a minimum, the following:
• Proportionate growth at and around existing settlements
• CAUSE’s Metro Town proposal
• One, two or more GCs (depending on the outcomes of the first-stage assessment)
126. Explicit assumptions should be made about the amount of development each option would involve, both at GCs and elsewhere, and the broad locations for that development. For the options involving GCs, each of the individual site options that survives the first-stage assessment, and each feasible combination of those surviving site options, should be assessed. To address my point above on Alternatives 2 and 3, options including one or two GCs should also include appropriate corresponding levels of proportionate growth at existing settlements. There should be liaison with CAUSE to ensure that their Metro Town proposal is fully understood and assessed appropriately, and similar liaison with the promoters of the GC site options where necessary.
127. Provided that the alternative spatial strategies are assessed objectively and with due regard to the evidence base, the second stage assessment should provide a sound basis for the selection of a preferred spatial strategy for the Plan (which may or may not include GCs).
128. While it is for the NEAs to decide who should carry out the further SA work, it might be advisable to consider appointing different consultants from those who conducted the 2016 and 2017 SA reports. This would help ensure that the further work is free from any earlier influence and is therefore fully objective.
129. The NEAs will also need to give consideration to the relationship between SA of their Section 1 and Section 2 Plans, to ensure that between them they provide an adequate basis for the SA adoption statement that will be required for each of their Local Plans.
Conclusions on Cross-Boundary Garden Communities
130. It will be evident from the foregoing discussion that I consider that the Garden Community proposals contained in the Plan are not adequately justified and have not been shown to have a reasonable prospect of being viably developed. As submitted, they are therefore unsound. I consider the resulting implications for the examination of the Section 1 and Section 2 Plans towards the end of this letter.
131. However, this is not to say that GCs may not have a role to play in meeting development needs in North Essex. I recognise that substantial time, effort and resources have already been invested in developing the GC proposals, not only by the NEAs but also by the Government, landowners, potential developers, infrastructure providers and others. It is possible that when the necessary additional work I have outlined is completed, it will provide justification for proceeding with one or more GC proposals – although any such justification would of course be subject to further testing at examination.
132. Having said that, on the basis of the evidence I have considered so far I would advise that simultaneously bringing forward three GCs on the scale proposed in the submitted Plan is likely to be difficult to justify. This is mainly because of the difficulty of co-ordinating the provision of infrastructure, particularly large-scale transport infrastructure, with the development of the GCs. In particular it is very unlikely, in my view, that the whole of the rapid transit system as proposed in the NERTS could be provided quickly enough to support commencement of development at all three GCs in the timescale envisaged in the submitted Plan. A more workable way of proceeding would be to lay out the rapid transit system in discrete stages, with the development of any proposed GC(s) taking place sequentially alongside it.
133. On this point I would endorse the advice in the North Essex Garden Communities Peer Review, led by Lord Kerslake [the Kerslake Review], that the NEAs should be prepared to differentiate their delivery strategy and timetable for each of the proposed GC locations, and need to be clear on the phasing of the infrastructure necessary to unlock the development potential at each location. When they have carried out the additional work outlined above, the NEAs should be in a position to set out a clear strategy and timetable for delivering any GCs that are proposed, in step with the major road and public transport infrastructure that is needed to support them.
134. My view that any GC proposals must be clearly shown to be financially viable also reflects advice in the Kerslake Review. The NEAs have, quite rightly, set high aspirations for the quality of their GC proposals and for the provision of affordable housing, open space, and social and community facilities in them. Clarity is needed at the outset over the affordability and deliverability of those aspirations, to ensure that they are not compromised during the development process because of unclear or conflicting expectations.
Providing for Employment (chapter 5)
135. Drawing on studies carried out for each council area, policy SP4 sets out employment land requirements for the Plan period. These are expressed as a range between a baseline figure and a higher-growth scenario figure. That is an appropriate approach, reflecting the inherent uncertainty in economic forecasting and the consequent need for flexibility.
136. For Braintree, the requirements are derived from the East of England economic forecasting model [EEFM], with adjustments made for local factors and drivers of economic change. The resulting figures in submitted policy SP4 reasonably reflect likely future economic conditions in the district, subject to the modification proposed in SD002a which corrects an arithmetical error in the baseline figure.
137. I saw no clear explanation for the baseline figure for Tendring set out in the submitted policy. However, a credible baseline figure has now been derived based on the Experian economic forecasting model, and is proposed in SD002a as a modification to the policy. The submitted higher-growth scenario figure was based on a misinterpretation of the relevant study, and a further modification is suggested to correct it. Provided that the modified figures are in SD002a are adopted, policy SP4 will reflect the evidence on likely future demand for employment land in Tendring.
138. As submitted, the range of requirements for Colchester is derived from the Colchester Employment Land Needs Assessment [ELNA]. That study developed four scenarios for employment growth based respectively on EEFM forecasts, past completion rates between 2006 and 2011 (actual and adjusted), and labour supply based on population projections. The labour supply scenario provides an appropriate baseline figure for policy SP4.
139. Actual past completion rates are assessed by ELNA as a negative figure, largely due to the relocation of a single firm which resulted in the loss of 120k sqm of industrial floorspace. It seems clear that this single event skewed the completion figures, and that this effect was especially strong in view of the relatively short trend period over which they were assessed. However, ELNA’s compensatory adjustment has the effect of transforming a net annual loss of some 10,500sqm of industrial floorspace into a net gain of around 6,500sqm. That is an unusually big adjustment and it results in an industrial land requirement which is nearly four times that of the EEFMbased scenario, and some seven times greater than the scenario based on labour supply. Such a level of industrial demand is also much greater than anything revealed in the studies for Braintree and Tendring.
140. ELNA itself advises that its scenarios based on past completion rates provide a less robust basis for understanding need than its other two scenarios. It is surprising, therefore, that the adjusted “higher past completion rate” scenario provides the basis for the policy SP4 highergrowth scenario requirement figure for Colchester. In my view the latter is unrealistically high and needs to be replaced.
141. I advise replacing it with the requirement figure of about 30ha derived from ELNA’s EEFM-based scenario. In my view the latter is a robustly-justified figure which would allow adequate headroom for future economic growth. According to ELNA, it would imply growth of 341 jobs per annum in Colchester over the Plan period, an increase of around 25% on both the annual average growth rate from 1991-2014 and on the rate implied by the policy SP4 baseline figure.
142. Alternatively, the NEAs may wish to undertake further work to derive a robust higher-growth scenario for Colchester, which would require further testing at examination.
Infrastructure and Connectivity (chapter 6)
143. Policy SP5 lists what are said to be strategic priorities for infrastructure in North Essex. As submitted, however, the list contains only a small number of specific infrastructure schemes. Most of the items in it read as policy objectives or statements of intent, rather than as identifiable projects. Modifications proposed by the NEAs go a little way towards addressing this shortcoming, by identifying that particular major road improvements and a rapid transit scheme are required for the GCs. However, the reference to the rapid transit scheme is still couched in general terms, no doubt reflecting the early stage of development that the scheme has reached.
144. The further work outlined above on transport infrastructure provision, particularly of the rapid transport scheme, should make it possible to refine policy SP5 and the related provisions of the GC policies in order to provide a clear strategy for delivering any GCs that are proposed in step with the necessary supporting infrastructure.
Remainder of the Plan: chapters 1, 2, 3, 7 & 9
145. The modifications proposed by the NEAs to these chapters of the Plan and the policies they contain largely address the issues of unsoundness that had previously been identified. However, it is likely that further modifications to some of them will need to be made in the light of my conclusions on the GC policies. This applies especially to policy SP2 (Spatial Strategy).
Adoption of the Section 1 Plan in advance of Section 2?
146. The Section 1 Plan was not prepared as a joint local development document under section 28 of the 2004 Act. Instead, each of the NEAs submitted a separate Local Plan, containing a Section 1 and a Section 2, for examination – albeit that the content of Section 1 is identical in each Local Plan.
147. I can see nothing in the relevant legislation that would allow part of a submitted Local Plan to be adopted separately from the rest of it. However, I am not qualified to give a legal opinion on the point, and moreover section 23 of the 2004 Act makes it clear that the decision whether or not to adopt a Local Plan is one that the LPAs must make themselves. I would therefore recommend that the NEAs seek their own legal advice on this question.
148. Nonetheless, it may be helpful for me to set out the options available to the NEAs, as I see them, on the assumption that Section 1 cannot be adopted in advance of Section 2. In deciding how to proceed the NEAs will evidently need to take into account my views, as set out above, on the scope of the main modifications and further work that are needed to make the Section 1 Plan sound and legally-compliant. Essentially it seems to me that they have three main options.
149. Option 1 would be for the NEAs to agree to remove the GC proposals from the Section 1 Plan at this stage, and commit to submitting a partial revision of Section 1 for examination by a defined time, for example within two or three years. This would involve drawing up main modifications to remove the current GC proposals and address the other soundness issues identified above. The NEAs would also need to amend their Local Development Schemes [LDS] to include the proposed partial revision to Section 1.
150. These steps should enable the Section 2 examinations to proceed, and subject to the findings of those examinations and to consultation on the main modifications to Section 1 and (potentially) to Section 2, each Local Plan should then be able to proceed to adoption. In preparing for the Section 2 examinations the NEAs would, of course, need to consider any implications of the removal of the current GC proposals – and any implications of my forthcoming findings on policy SP3 – for housing land supply in each NEA in the years before the partial revision comes forward.
151. Following the Section 2 examinations, under Option 1 the NEAs would then carry out further work on the evidence base and Sustainability Appraisal, as outlined in my comments above on the GC proposals. That further work would provide the basis for revised strategic proposals to be brought forward for examination as a partial revision to the Section 1 Plan, within the timescale identified in the revised LDS. The revised strategic proposals could in principle include one or more GC(s), if justified by the further evidence and SA work.
152. Option 2 would involve the NEAs carrying out the necessary further work on the evidence base and Sustainability Appraisal, and bringing forward any resulting revised strategic proposals, before the commencement of the Section 2 examinations. Due to the considerable length of time this is likely to take, it would be necessary to suspend the examination of Section 1 while the work is carried out and consultation on the SA and any revised strategic proposals takes place. Following the suspension, further Section 1 hearings would need to be held to consider the revised strategic proposals.
153. It seems to me that in this option the Section 2 examinations could not sensibly proceed before the additional Section 1 hearings had taken place and the Inspector’s initial views on the revised proposals were known, as any significant revisions to Section 1 would have consequences for the examination of Section 2.
154. It is also possible under Option 2 that other parts of the evidence base for both Section 1 and Section 2 might become out of date or overtaken by changes in national policy. Should this occur, there would be a risk of additional delay to the examination of both parts of the Plan while the relevant evidence is updated and any necessary modifications are brought forward.
155. All this means that even in the most favourable circumstances the adoption of the NEAs’ Local Plans would be substantially delayed under Option 2, compared with Option 1. In turn this could give rise to continuity problems for all participants in the examinations of the plans.
156. Option 3 would be to withdraw the Section 1 and Section 2 Plans from examination and to resubmit them with any necessary revisions, after carrying out the required further work on the evidence base and SA, and the relevant consultation and other procedures required by legislation.
157. I expect that this letter will come as a disappointment to the NEAs after all the hard work and resources they have committed to bringing the Section 1 Plan forward for examination. Nonetheless, I hope it will be appreciated that my findings do not necessarily represent a rejection of their commendable ambitions for high-quality, strategic-scale development in North Essex. Equally, however, the scale of those ambitions, and the long timescale over which any GC proposals would come forward, require that adequate time and care are taken now to ensure that any proposals are realistic and robust.
158. I am not inviting comments on the contents of this letter. But I will assist the NEAs with any queries, and with any further advice they may need on taking forward the necessary further work and changes to the Plan I have identified. I would appreciate it if you would let me know, as soon as you are able to, which of the options outlined in paragraphs 148 to 156 above, or any alternative course of action, the NEAs wish to pursue. This will enable an outline timescale for the remainder of the examination to be devised. Please contact me through the Programme Officer, with a copy to the PINS case officer.
Roger Clews Inspector
 The three NEAs in the context of this letter are Braintree District Council [BDC],
Colchester Borough Council [CBC], and Tendring District Council [TDC]
 Suggested Modifications to the Publication Draft Braintree, Colchester and Tendring Local Plans: Section One (Feb 2018)
 Prepared by CAUSE in 2015 as a potential alternative strategy for growth in North Essex
 Email from Richard Walker of Lightwood to the Programme Officer, 18 January 2018
 People over Wind, Peter Sweetman v Coillte Teoranta [CJEU Case C-323/17]
 Referred to in the Plan as Development Plan Documents or DPDs
 See document EB/065 for references
 Start to Finish – How Quickly do Large-scale Housing Sites Deliver? (Nov 2016)
 At least one other viability appraisal has been carried out on behalf of GC promoters, but as it was not disclosed to the examination I cannot place reliance upon it.
 The Garden Cities and Large Sites Financial Model
 PwC, North Essex Garden Communities Final Report (14 Dec 2016) 12 EB/13(2/2a)
 See the next section on delivery mechanisms
 See, for example, Attorney-General v Wilts United Dairies Ltd  37 TLR 884, and Congreve v Home Secretary  2 WLR 291
 Cogent Land LLP v Rochford DC  EWHC 2542 (Admin) and No Adastral New
Town Ltd v Suffolk Coastal DC & SSCLG  EWCA Civ 88