Doncaster Local Plan Viability Testing

DONCASTER LOCAL PLAN VIABILITY TESTING

Completed on behalf of Doncaster Metropolitan Borough Council

District Valuer Services (DVS) August 2016

EXECUTIVE SUMMARY

1. We have been commissioned by Doncaster Metropolitan Borough Council (“the Council”) to undertake an area wide viability study, specifically to consider the impact Council policies, as well as market fluctuations, have on scheme viability.

2. We have been instructed to assess a sample of hypothetical sites across the Borough, to include both residential schemes and non-residential developments (i.e. office, industrial, leisure developments etc.).

3. This assessment has been undertaken specifically for the purposes of a Whole Plan Viability Study. Its intention is to consider broad average viability appraisal inputs across the area of the study, and not set precedents for individual site viability assessments. In other words, the high level findings of this study should not be used to inform individual viability appraisals of ‘real’ development sites, which (as per the guidance) will need to be undertaken on a site by site basis reflecting the specific nature of the land in question.

4. This assessment has been undertaken in the context of the National Planning Policy Framework (“NPPF”) and Planning Practice Guidance (“PPG”), as well as the relevant professional guidance; “Viability Testing Local Plans” June 2012 by the Local Housing Delivery Group (“The Harman Review”) and “Financial viability in planning” August 2012 by the Royal Institution of Chartered Surveyors (RICS).

5. As part of the process, in line with the guidance, we have actively engaged with stakeholders to help ensure the assumptions adopted within the appraisal are realistic. In this case, we arranged a Stakeholder Workshop to allow an open forum discussion on viability matters. This was attended by a variety of key stakeholders; including land owners, agents, planning consultants, house builders, various representatives from different Council departments, as well as external public sector bodies. Following this workshop, a questionnaire was circulated to all identified stakeholders (including those unable to attend the workshop) seeking further details on their views on viability matters. The workshop and returned questionnaires formed part of the evidence base of the conclusions reached.

6. In accordance with the guidance, we have adopted the residual approach to site testing, which involves identifying the sales revenue for the completed scheme, and from this deducting the relevant costs of delivering the project (including the site value and developer’s profit).

7. As part of our review, we have adopted ‘sensitivity analysis’. This involves running a number of appraisal scenarios, varying key appraisal inputs to determine the impact these changes could have on the overall viability. This iteration process allows a more robust assessment of viability and is recommended within the guidance.

8. For the residential sites, we have concluded that land located within ‘high’ value areas are comfortably viable with the Council’s proposed affordable housing provision of 25%, together with various other draft S106 policies. However, for sites located within ‘medium’ and ‘low’ value areas, the viability pressure is greater. Having run various scenarios, for sites in medium and low value areas we have concluded that it is appropriate to adopt a reduced affordable housing provision of 15%.

9. For non-residential sites, we have concluded that supermarket, strategic warehouse, hotel and town centre shop schemes are all viable, even with the application of the Council’s draft policies. However, we have concluded that industrial and office schemes are currently unviable, even if the Council policies are removed. This is due to macro-economic factors affecting these market sectors. The only scheme type that sees a significant benefit in reducing the Council policies is non-strategic warehousing, which improves from being marginally viable to viable when the policies are reduced (or removed).

Summary Schedule – Key ‘Basic’ Viability Assumptions (Residential)

Appraisal input Assumptions Gross to net ratio < 0.5 Ha 100%

0.5 – 2.0 Ha 85% 2.0 – 5.0 Ha 80% > 0.5 Ha 75% Scheme density 35 dwellings per net Ha Average house size 92.90 sq. m (1,000 sq. ft.) Affordable rent transfer values 45% of market value

Shared ownership transfer values 67.5% of market value

Starter homes discount 80% of market value

Average ‘basic’ build cost BCIS lower quartile – £798 per sq. m BCIS median – £900 per sq. m External / site infrastructure costs 15% of the basic build cost

Contingency Greenfield – 3% of basic build costs Brownfield – 5% of basic build costs

‘Abnormal’ development costs Greenfield – £100,000 per net Ha Brownfield (cleared) – £200,000 per net Ha Brownfield (occupied) – £300,000 per net Ha Professional fees Sub 20 dwellings – 8% of basic build costs / externals Over 20 dwellings – 6% of basic build costs / externals

Marketing costs Sub 10 dwellings – 1.5% of sales revenue Over 10 dwellings – 3% of sales revenue

Plus additional allowance for legal costs at £500 per dwelling

Developer’s return Sub 10 dwellings – Market Value / Starter Homes 15% of sales revenue, Affordable rent / Shared ownership 8% of cost

Over 10 dwellings – Market Value / Starter Homes 18.5% of sales revenue, Affordable rent / Shared ownership 8% of cost

Finance costs Sub 10 dwellings – 7% debit Over 10 dwellings – 6% debit

Average sales values Low value area – £1,500 per sq. m Medium value area – £1,750 per sq. m High value area – £2,250 per sq. m

Threshold Land Values Greenfield Low value area – £197,680 / Ha (£80k / acre) Greenfield Medium value area – £271,810 / Ha (£110k / acre)

Greenfield High value area – £345,940 / Ha (£140k / acre) Brownfield (cleared) – £185,325 / Ha (£75k / acre) Brownfield (occupied) – £370,650 / Ha (£150k / acre)

Summary Schedule – Key ‘Basic’ Viability Assumptions (Non-Residential)

Appraisal input Assumptions Gross to net ratio Hotel / town centre shop 100%

Strategic warehouse 90% Industrial / non-strategic warehouse 80% Supermarket / offices 75% Type of disposal Speculative – Industrial / office / town centre shop Pre-let – Hotel / warehousing / supermarket Lease length for tenant 15 years Rent free period 12 months Average ‘basic’ build cost - Strategic warehouse £446 per sq. m - Non-strategic warehouse £446 per sq. m - Industrial (2,000 to 10,000 sq. m) £446 per sq. m - Supermarket (small and large) £951 per sq. m - Office 500 sq. m £1,228 per sq. m - Office 200 sq. m £1,052 per sq. m - Town centre retail £644 per sq. m - Hotel £1,405 per sq. m External / site infrastructure costs 10% of the basic build cost

Contingency 3% of basic build costs

Professional fees 8% of basic build costs / externals

Marketing costs Letting – 15% of the annual Market Rent Investment sale – 1.5% of the sale price agreed Both include a legal fee

Developer’s return Speculative – 15% on cost Pre-let – 12.5% on cost

Finance costs 6% debit

Average sales values Rent and yield approach, as detailed within the main body of the report below

Threshold Land Values Brownfield (cleared) – £185,325 / Ha (£75k / acre) Greenfield – £271,810 / Ha (£110k / acre)

CONTENTS

1. INTRODUCTION Pg. 8

Instruction 8

2. VIABILITY METHODOLOGY 9

National Planning Policy Framework (“NPPF”) 9 Planning Practice Guidance (“PPG”) 9 Professional Guidance for Viability Assessments 12 The Financial Appraisal Model / The 'Residual' Method 16 Stakeholder engagement 20 Summary 21

3. RESIDENTIAL - VIABILITY ASSUMPTIONS 22

Introduction 22 Site Types 22 Gross and Net Developable Areas 23 Capacity / Density 25 Dwelling Mix and Sizes 26 Specification 29 Affordable Rented Assumptions 29 Intermediate / Shared Ownership Assumptions 32 Starter Homes 32 Market Value Sales Revenue 33 'Basic' Build Costs 34 Externals / Infrastructure 37 Contingency 40 Abnormal Development Costs 42 Professional Fees 44 Marketing 45 Developer's Profit 47 Finance 50 Threshold Land Value 51 Site Acquisition and Disposal Costs 52 Section 106 Contributions / Emerging Policy Aspirations 52 Scenario testing / sensitivity analysis 54

4. RESIDENTIAL – APPRAISAL RESULTS 56

Introduction 56 Test 1 – Council policies (bar affordable housing), BCIS lower quartile 56 Test 2 – Council policies (bar affordable housing), BCIS median 59 Test 3 – Estimated costs linked to council policy CCMRE3 are removed 61 Test 4 – Sales revenue increased by 5% 64 Test 5 – Starter Homes 66 Test 6 – All draft Council planning policies removed 69 Conclusions 71

5. NON-RESIDENTIAL – VIABILITY ASSUMPTIONS 72

Introduction 72 Site type 72 Methodology 73 Evidence 73 Scenario testing / sensitivity analysis 78

6. NON-RESIDENTIAL – APPRAISAL RESULTS 80

Introduction 80 Test 1 – Council policies 80 Test 2 – Excludes all Council policies 81 Test 3 – Excludes all Council policies except for flood risk mitigation costs 82 Conclusions 83

7. FINAL COMMENTS 84

Table A - Settlement Value Areas 23 Table B - Development type and viability tests 23 Table 1 – Minimum gross internal floor areas and storage (sq. m) 28 Table 2 – Sample of rented modern houses & affordable rent calculation 31 Table 3 – Test 1 Urban extension viability results 57 Table 4 – Test 1 Urban settlement viability results 58 Table 5 – Test 2 Urban extension viability results 59 Table 6 – Test 2 Urban settlement viability results 60 Table 7 – Test 3 Urban extension viability results 62 Table 8 – Test 3 Urban settlement viability results 63 Table 9 – Test 4 Urban extension viability results 64 Table 10 – Test 4 Urban settlement viability results 65 Table 11 – Test 5 Urban extension viability results 67 Table 12 – Test 5 Urban settlement viability results 68 Table 13 – Test 6 Urban extension viability results 69 Table 14 – Test 6 Urban settlement viability results 70 Table 15 – Non-residential Market Rent and yield ranges 74 Table 16 – Non-residential key appraisal inputs 75 Table 17 – Test 1 Council policy appraisal results 80 Table 18 – Test 2 No Council policy appraisal results 81 Table 19 – Test 3 No Council policy (except flood risk mitigation) appraisal results 82

APPENDIX 1 – Doncaster Housing Market Review 86 APPENDIX 2 – Threshold Land Value Assessment 104 APPENDIX 3 - Map to Illustrate High, Medium and Low Value Areas 127

1. INTRODUCTION

1.1 Instruction

1.1.1 Doncaster Metropolitan Borough Council (“the Council”) is currently in the process of developing a Local Plan. As part of this process the Council is looking to ensure, through a robust evidence assessment, that the Local Plan will be deliverable and viable.

1.1.2 In the context of the above District Valuer Services (“DVS”), part of the Valuation Office Agency, has been commissioned by the Council to undertake viability appraisals of a sample of hypothetical sites across the Borough, to include both residential schemes and non-residential developments (i.e. office, industrial, leisure developments etc.). Please note, the scope of this instruction specifically excludes large urban extension housing schemes 400+ units, which will be dealt with separately if any such allocations are proposed through the Local Plan. DVS provides valuation advice to public bodies throughout the UK. It has extensive experience in undertaking development appraisals and employs specialists in development work.

1.1.3 Please note, this document has been prepared specifically for the purposes of a Whole Plan Viability Study. Its intention is to consider broad average viability appraisal inputs across the area of the study, and not set precedents for individual site viability assessments. In other words, the high level findings of this study should not be used to inform individual viability appraisals of ‘real’ development sites, which (as per the guidance) will need to be undertaken on a site by site basis reflecting the specific nature of the land in question.

2. VIABILITY METHODOLOGY

2.1 National Planning Policy Framework (“NPPF”)

2.1.1 Whole Plan Viability Assessments should be undertaken in the context of the National Planning Policy Framework (“NPPF”) March 2012. This has a section entitled “Ensuring viability and deliverability”:

“Pursuing sustainable development requires careful attention to viability and costs in plan-making and decision-taking. Plans should be deliverable. Therefore, the sites and the scale of development identified in the plan should not be subject to such a scale of obligations and policy burdens that their ability to be developed viably is threatened. To ensure viability, the costs of any requirements likely to be applied to development, such as requirements for affordable housing, standards, infrastructure contributions or other requirements should, when taking account of the normal cost of development and mitigation, provide competitive returns to a willing land owner and willing developer to enable the development to be deliverable”. (Paragraph 173)

“Local planning authorities should set out their policy on local standards in the Local Plan, including requirements for affordable housing. They should assess the likely cumulative impacts on development in their area of all existing and proposed local standards, supplementary planning documents and policies that support the development plan, when added to nationally required standards. In order to be appropriate, the cumulative impact of these standards and policies should not put implementation of the plan at serious risk, and should facilitate development throughout the economic cycle. Evidence supporting the assessment should be proportionate, using only appropriate available evidence.” (Paragraph 174)

2.2 Planning Practice Guidance (“PPG”)

2.2.1 Furthermore, a Whole Plan Viability Assessment should also have regard to Planning Practice Guidance, including specifically:

Stakeholder engagement

“Collaboration: a collaborative approach involving the local planning authority, business community, developers, landowners and other interested parties will improve understanding of deliverability and viability.” (Paragraph 004, as revised March 2014)

Viability in plan-making

“Local Plans and neighbourhood plans should be based on a clear and deliverable vision of the area. Viability assessment should be considered as a tool that can assist with the development of plans and plan policies. It should not compromise the quality of development but should ensure that the Local Plan vision and policies are realistic and provide high level assurance that plan policies are viable.

Development of plan policies should be iterative – with draft policies tested against evidence of the likely ability of the market to deliver the plan’s policies, and revised as part of a dynamic process.

Evidence should be proportionate to ensure plans are underpinned by a broad understanding of viability. Greater detail may be necessary in areas of known marginal viability or where the evidence suggests that viability might be an issue – for example in relation to policies for strategic sites which require high infrastructure investment.” (Paragraph 005, as revised March 2014)

Approach to testing

“Assessing the viability of plans does not require individual testing of every site or assurance that individual sites are viable; site typologies may be used to determine viability at policy level. Assessment of samples of sites may be helpful to support evidence and more detailed assessment may be necessary for particular areas or key sites on which the delivery of the plan relies.” (Paragraph 006, as revised March 2014)

Approach to costing

Plan makers should consider the range of costs on development. This can include costs imposed through national and local standards, local policies and the Community Infrastructure Levy, as well as a realistic understanding of the likely cost of Section 106 planning obligations and Section 278 agreements for highways works. Their cumulative cost should not cause development types or strategic sites to be unviable. Emerging policy requirements may need to be adjusted to ensure that the plan is able to deliver sustainable development.” (Paragraph 007, as revised March 2014)

Approach to market changes

Plan makers should not plan to the margin of viability but should allow for a buffer to respond to changing markets and to avoid the need for frequent plan updating. Current costs and values should be considered when assessing the viability of plan policy. Policies should be deliverable and should not be based on an expectation of future rises in values at least for the first five years of the plan period. This will help to ensure realism and avoid complicating the assessment with uncertain judgements about the future. Where any relevant future change to regulation or policy (either national or local) is known, any likely impact on current costs should be considered”. (Paragraph 008, as revised March 2014)

Competitive return

The National Planning Policy Framework states that viability should consider “competitive returns to a willing landowner and willing developer to enable the development to be deliverable.” This return will vary significantly between projects to reflect the size and risk profile of the development and the risks to the project. A rigid approach to assumed profit levels should be avoided and comparable schemes or data sources reflected wherever possible.

A competitive return for the land owner is the price at which a reasonable land owner would be willing to sell their land for the development. The price will need to provide an incentive for the land owner to sell in comparison with the other options available. Those options may include the current use value of the land or its value for a realistic alternative use that complies with planning policy.” (Paragraph 015, as revised March 2014)

2.2.2 In undertaking a Whole Plan Viability Assessment it is therefore vital to adopt an evidence based approach, which seeks to ensure that any Council policies do not undermine viability.

2.3 Professional Guidance for Viability Assessments

2.3.1 Surveyors are now assisted by two relatively recent publications, although the guidance between each is somewhat contradictory. We have commented on each publication as follows:

“Financial viability in planning” August 2012 by the Royal Institution of Chartered Surveyors (RICS):

Para 2.5.2, Box 10, “…nature of the applicant should normally be disregarded as should the benefits or dis-benefits that are unique to the applicant.”

2.3.2 Thus, appraisals should be done assuming hypothetical, typical landowners and developers and the views and aspirations of the actual owner are not relevant if these views differ from general market practice.

Para 2.3.2, Box 7, “Site value should equate to the market value subject to the following assumption: that the value has regard to the development plan policies and all other material planning considerations and disregards that which is contrary to the development plan.”

2.3.3 As indicated above, this refers to the site value as usually being assessed by means of a residual development appraisal. However, the suggestion seems to be that planning policies should be fixed and land value subject to change (which contradicts the view of the landowner having a minimum land value below which they would sell).

Para 2.1.2 “It follows, for example, that the land value is flexible and not a fixed figure to the extent that Site Value has to be determined as part of the viability assessment.”

2.3.4 This appears to support the above view that it is the Council’s policy which drives the land value, not the other way round. However, the RICS document does acknowledge that the flexibility in land value cannot result in the value going below the Current Use Value (“CUV”), stating:

Para 3.4.4 “The return to the landowner will be in the form of a land value in excess of current use value but it would be inappropriate to assume an uplift based on set percentages.”

2.3.5 This appears to support the view of setting a land value (often referred to as the “Threshold Land Value” or “TLV”) for development appraisals, which is to somehow be linked to the Current Use Value (“CUV”). However, no guidance is given as to how to determine the link between the CUV and the TLV.

2.3.6 Furthermore, in particular no guidance is given to assessing greenfield land, where the CUV may only be £12,500 - £25,000 per Ha and clearly a TLV only slightly above the CUV would not represent a sufficient incentive for a landowner to sell for development.

“Viability Testing Local Plans” June 2012 by the Local Housing Delivery Group (“The Harman Review”).

Pg. 29 “We recommend that the Threshold Land Value is based on a premium over current use values and credible alternative use value (noting the exceptions below)”.

2.3.7 This therefore contradicts the guidance provided by the RICS, where adopting a percentage uplift above the CUV is not recommended.

2.3.8 One of the exceptions referred to relates to “non-urban” and “greenfield” sites.

Pg. 30 “It is widely recognised that this approach [i.e. a percentage increase over CUV] can be less straight forward for non-urban sites or urban extensions, where land owners are rarely forced or distressed sellers…This is particularly the case in relation to large greenfield sites…Accordingly, the uplift to current use value sought by the landowners will invariably be significantly higher than in an urban context and requires very careful consideration”.

2.3.9 This does not mean that an assessment of the CUV has no part to play in the process of assessing greenfield sites. A typical landowner will still want to know what the value of his/her site is without the planning permission applied for, and then judge by how much, if at all, the CUV increases when planning consent is granted. The difference is that, for urban brownfield sites a premium uplift of circa 25% – 50% of the CUV may be deemed sufficient to incentivise a landowner to sell (e.g. if the CUV is £200,000 per Ha, applying a 50% uplift would mean a TLV of £300,000 per Ha, which would be attractive to a landowner). For a greenfield site, if the CUV is only say £10,000 per Ha then a 50% uplift (i.e. a TLV of £15,000 per Ha) would clearly not incentivise a landowner to release the land for development. In reality, the ‘uplift’ would need to be more like 15 – 25 times (or more) the CUV.

2.3.10 In terms of how to evidence the approach to greenfield sites the document goes on to say:

Pg. 30 “…local sources should be used to provide a view on market values (the ‘going rate’), as a means of giving a further sense check on the outcome of the current use plus premium calculation”.

Pg. 30 “…for sites of this nature [i.e. greenfield], it will be necessary to make greater use of benchmarks, taking into account of local partner views on market data and information on typical minimum price provisions used within developer / site promoter agreements for sites of this nature”.

2.3.11 This therefore seems to advocate using evidence of TLVs identified as part of the viability process, as well as using market transactions as a general ‘sense check’.

2.3.12 However, it is stressed that there are limitations of assessing land sales. Assessing actual land sales for the purposes of identifying a TLV is not straight forward, as the price someone is willing to pay for a piece of development land (and indeed accept for a piece of development land) is subject to many factors, which includes:

- The type of development that could be brought forward. - The gross to net ratio (it may be that a large section of the site is constrained and cannot be developed). - The potential density of any proposed scheme. - Whether any third parties benefit from a ransom position preventing access to the site. - Whether there are any title constraints. - The abnormal costs associated with developing the site (i.e. any untypical cost, such as deep pile foundations to mitigate ground concerns, flooding mitigation works etc.). - The planning policies that relate to a specific type of scheme. - Whether a purchaser benefits from synergistic value with any neighbouring land they already own or will own in the future. - Whether a vendor is under financial pressure to sell. - Whether a house-builder is keen to have a presence in a particular location etc.

2.3.13 There are therefore a number of factors which impact the price someone is willing to pay for development land, because ultimately each development site is unique. For example, you could have 2 sites next to each other sold at the same time, each being 5 Ha and the same shape. However, one may have significant flooding issues and a poor access route, whereas the other may have no concerns. The price paid for the land affected by ‘abnormal’ development costs (i.e. in this case flooding and a poor access route) would therefore most likely be significantly less than the unaffected site. The reasons for the difference in value, though, would not be identifiable by simply looking at the price paid for the land on a ‘per Ha’ basis.

2.3.14 This means it is extremely difficult to compare two land transactions because in reality only some of the factors outlined above (which is not an exhaustive list) will be known to the analysing surveyor.

2.3.15 In this respect, land transactions are useful in providing a ‘sense check’ but they should not be regarded as providing a definitive view on threshold land values, particularly on a ‘price per Ha’ basis, because in most cases the full details of the transaction (and the factors which impact value) will not be known. Land sales should therefore be considered after the other sources of evidence are identified, and provide a ‘sense check’ only.

2.4 The Financial Appraisal Model / The ‘Residual’ Method

2.4.1 The professional guidance advocates adopting the ‘residual’ method when considering viability assessments. This is an established valuation approach, which as a concept is relatively straight forward and can be illustrated by the following equation:

Gross Development Value (i.e. Total Revenue)

Less

Development Costs (Land Acquisition + Construction + Fees + Finance)

Equals

Residue for Developer’s Profit and Risk

2.4.2 However, please note it is not a requirement of an appraisal that the residue is always equal to the developer’s profit. The model can be amended so that developer’s profit is a fixed input and say the land value is shown as the residue. If this were the case the model would be amended to:

Completed Development Value (i.e. Total Revenue)

Less

Development Costs (Developer’s Profit and Risk + Construction + Fees + Finance)

Equals

Residue for Land Acquisition

2.4.3 Equally, the land value and developer’s profit can be inputted as fixed figures, representing the minimum profit and minimum land value considered necessary for the scheme to be implemented. The subsequent residue could therefore be used to show the monies available for planning policy contributions (for example).

2.4.4 Whilst a simple concept, it is stressed that in reality the residual method often becomes a complicated and detailed approach. This is because the methodology inherently requires a wide variety of inputs to be factored into the assessment, all of which are subject to variance (e.g. sales values, build costs, professional fees, abnormal works, Council policies, profit, marketing, finance etc.). All of these inputs need to be considered carefully, as potentially relatively small variances to one or two inputs could have a significant impact on the results of the assessment. This inherent flaw in the methodology is recognised by the RICS and wider industry, and as a result ‘sensitivity’ testing is recommended to try and minimise the impact of these potential variances. Nevertheless, the industry still considers this to be the most appropriate methodology for assessing development sites and appraising viability.

2.4.5 Notwithstanding the flaws in the methodology, for the purposes of a Whole Plan Viability Assessment, and based on our experience, we would advocate the approach of fixing the land value and the developer’s profit at average levels considered appropriate for the scheme to be implemented, which would then be inputted into an appraisal together with all the other general costs of development (including build costs, professional fees, finance etc.). However, at this point any affordable housing contributions or CIL charges would be excluded from the appraisal. The costs would then be deducted from the Gross Development Value (or total revenue). If the appraisal produces a surplus, this surplus could then be shared out as Planning Policy contributions. If the scheme produces a deficit the scheme would be regarded as being unviable, even without any Planning Policies applied.

2.4.6 The advantage of approaching the viability assessment in this way is that the appraisal would clearly show whether a scheme was viable or not (by demonstrating a surplus or a deficit). If the scheme did produce a surplus, and was subsequently regarded as being viable, then at this point Planning Policies could be built into the appraisal and the viability re-tested on a ‘trial and error’ basis until an agreed point.

2.4.7 The negative side to this approach is that there would have to be a pre- determined ‘level’ at which a scheme was regarded as being acceptably viable. The most obvious level in this regard would be if the scheme resulted in a ‘zero’ return (or thereabouts), in other words the scheme neither produced a surplus nor produced a deficit.

2.4.8 However, some may argue that this is at the ‘extremes’ or ‘breaking point’ of viability, and therefore is not in line with the sentiment of the NPPF or PPG. In this respect, setting a pre-determined level of surplus above a ‘zero’ return would provide a suitable ‘buffer’, minimising the risk of the Planning Policies undermining viability.

2.4.9 That said, in our experience, if a scheme produced a relatively small deficit then this would not undermine implementation, as a developer would look to potentially renegotiate the land value, consider value engineering or simply accept a slightly reduced profit margin as a means of ensuring the scheme was still undertaken. In other words, and based on our experience, we do not believe that a developer would ‘walk away’ from a scheme if it ultimately produced a small deficit. In this regard, it can be considered that there is already an implicit buffer allowance in a viability appraisal, therefore a scheme producing a zero return would not be at the extremes or breaking point of viability, as the reality is a developer would allow a small amount of ‘flex’ in their profit margin / build costs to cover such eventualities.

2.4.10 Scenario testing / sensitivity analysis can also be used to ensure robust conclusions are reached (an approach advocated in the PPG and the relevant viability guidance), helping to minimise the risk of the Council’s policies undermining viability and limiting deliverability. This involves running a number of scenarios based on variables in key inputs. For example,

(i) Variances in costs (ii) Variances to developer’s profit (iii) Variances in Council Policies (iv) Changes in time – inflation rates applied to certain inputs, such as houses and sales and build costs

2.4.11 The purpose of the scenario testing / sensitivity analysis is an attempt to consider likely variations within the appraisal model, essentially with a view to identifying the best and worst case positions of viability. From this more robust conclusions can be reached in terms of viability.

2.4.12 That said, we would stress that ‘over testing’ can produce excessive data which can prove more difficult to analyse (making it difficult to reach meaningful conclusions). Scenario testing / sensitivity analysis is a useful tool in Whole Plan testing, however in our view this should be limited to a handful of key scenarios.

2.5 Stakeholder engagement

2.5.1 The NPPF and PPG, as well as the Harman Review, advocate stakeholder engagement in the preparation of a Local Plan and the process of undertaking a Whole Plan viability assessment. This is an important step in helping ensure planning policies are set at realistic levels which do not undermine the viability and deliverability of sites.

2.5.2 Based on past experience of undertaking Whole Plan viability assessments, and having considered the approach other Local Authorities have taken to stakeholder engagement (particularly paying attention to examples of Local Plans which have been examined and approved by the Planning Inspectorate), we have adopted the following approach:

- Stakeholder Workshop 1: a meeting was held at the Council offices on 1st June 2016. Stakeholders (including developers, house builders, planning consultants, agents, landowners, public sector bodies etc.) were invited to attend. The invitation was also published on the Council’s website as an open invite for any other interested stakeholders to attend the workshop. The format was an open forum discussion, albeit DVS used a power-point presentation as a prompt for discussion on key viability appraisal inputs. This gave stakeholders the opportunity to provide some initial thoughts on key viability assumptions, allowing them to challenge and be challenged on their views.

- Questionnaire: after the workshop those who attended (and those who were unable to attend) were sent a “Viability Questionnaire and Evidence Trawl” seeking the views of the stakeholders on key appraisal inputs. This gave an opportunity to stakeholders to provide more considered comments on key viability appraisals and submit supporting evidence. 3 completed questionnaires were sent back to DVS.

- Stakeholder Workshop 2: the initial intention was to hold 2 Workshops, with the second workshop an opportunity to discuss initial testing results and also any outstanding queries on appraisal inputs. However, there was a lack of response from the stakeholders to the option of a second workshop. This,

combined with the limited response to the questionnaire, meant the decision was taken not to hold a second workshop.

2.5.3 We would stress that we have not simply looked to follow the views of the stakeholders, and have instead assessed all of the evidence identified with a view to reaching an independent view. The stakeholder workshops / questionnaire was an important part of the evidence base identified.

2.6 Summary

2.6.1 The NPPF and PPG is clear that Council Policies should not be set at levels which could potentially undermine the viability and deliverability of development projects. Whole Plan viability assessments should be undertaken in this context.

2.6.2 The relevant professional guidance advocates the use of the residual method for the assessment of viability. This is an established and RICS approved valuation methodology.

2.6.3 That said, there are inherent flaws in the residual method, due to the wide number of inputs often required, each of which are potentially subject to variance. To minimise this impact the professional guidance recommends scenario testing / sensitivity analysis is adopted, which is also advocated in Whole Plan viability assessments.

2.6.4 The residual method can be presented in different ways, with the land value, profit or even Council policies set as the ‘residual’ or outcome of the appraisal. Based on our experience we would advocate fixing the land value and developer’s profit at pre-agreed levels, and therefore the ‘residual’ (if any) could be used to meet Council policies. We consider this to be the simplest way of approaching Whole Plan viability assessments.

2.6.5 As part of the process, it is vital that stakeholders are appropriately engaged, to help ensure planning policies are set at realistic levels.

3. RESIDENTIAL – VIABILITY ASSUMPTIONS

3.1 Introduction

3.1.1 This section of the report sets out the assumptions and appraisal inputs used in the appraisal testing of residential development sites.

3.2 Site Types

3.2.1 DVS have undertaken various area wide studies, involving both ‘real’ site assessments (i.e. appraising a sample of actual development sites) and hypothetical sites.

3.2.2 Both methods are considered to be reasonable approaches for the purposes of a Whole Plan viability assessment.

3.2.3 However, in this case, and having discussed both options through the Stakeholder Workshop and the subsequent questionnaire, we decided to adopt hypothetical site testing as this enables the same site to be tested across different locations, which clearly demonstrates changes in viability across local housing markets and different value areas.

3.2.4 Having discussed the nature of the hypothetical sites with the Council and the stakeholders, and having researched the local market dynamics, we have adopted the following hypothetical development sites:

Residential – Urban Extension

- Number of dwellings: 50, 100 and 400. - Value areas: Low, Low/medium, Medium and High (as detailed in Table A below). - Locations: Doncaster Main Urban Area, 7 Main Towns, and 10 Service Towns/Villages. - Please note, as agreed with the Council and Stakeholders not all of the above variances were tested. However, through a combination of the above variances we established 28 hypothetical site types to test.

Table A. Settlement Value Areas Area/ Site Value Corresponding Settlement(s) Total Location number of site types to test Service Towns/ High Auckley-Hayfield Green, Finningley, 2 Villages: Bawtry, Tickhill, Sprotbrough, Barnburgh-Harlington Doncaster Main Low Bentley 2 Urban Area Doncaster Main Medium Scawsby 3 Urban Area Main Town Medium Mexborough 3 Main Town Low / Conisbrough & Denaby 3 Medium Main Towns Low Thorne-Moorends Rossington Adwick- 8 Woodlands Main Towns, Medium Hatfield-Stainforth & Armthorpe, 3 Doncaster Main Edenthorpe & Barnby Dun Urban Area & Service Village Service Towns Medium Askern, Skellow & Carcroft 2 Service Town Low Edlington 2 TOTAL 28

Residential – Urban Settlement

- Number of dwellings: 1, 5, 14, 50 and 100. - Value areas: Low, Medium and High (as detailed in Appendix 3). - Type: greenfield, brownfield (cleared) and brownfield (occupied). - Please note, as agreed with the Council and Stakeholders not all of the above variances were tested. However, through a combination of the above variances we established 27 hypothetical site types to test.

3.2.5 Please see Appendix 3 for a map of the high, medium and low value areas. Each development type has been tested in the three value areas, as summarised in Table B below. Table B. Development type and viability tests Development Type 3 x value areas (high, medium Total number of & low) site types to test Single dwelling greenfield infill plot H,M,L 3 5 units cleared site H,M,L 3 5 units occupied site H,M,L 3 14 units cleared site H,M,L 3 14 units occupied site H,M,L 3 50 units cleared site H,M,L 3 50 units occupied site H,M,L 3 100 units cleared site H,M,L 3 100 units occupied site H,M,L 3 TOTAL 27

3.2.6 The above site types are considered to be reflective of the type of sites likely to come forward during the lifetime of the Local Plan.

3.2.7 However, please note we have not tested larger ‘urban extension' sites (i.e. larger than 400 dwellings), which may be delivered during the life of the Local Plan. We understand such sites will be assessed by way of a separate assessment, and would act as an addendum to this study.

3.3 Gross And Net Developable Areas

3.3.1 For the purposes of this study we have assumed the net developable area of a site is defined as follows:

Net developable area refers to the total area of land available for development, not necessarily the total area of a property itself. It does not include open space, drainage land, regional roads and land used for other public facilities.

In relation to housing sites, net developable area excludes main roads, buffer zones, structural landscaping, other uses such as local shops, school sites where required, and general open space and, wherever possible, features of natural heritage interest. Net developable area includes local access roads, parking areas, footpaths and local open space such as children’s play areas and amenity space.

3.3.2 At the Stakeholder Workshop initial gross to net site areas were discussed. It was agreed that smaller sites will require less of a reduction from gross to net when compared to larger sites (where there are greater needs for infrastructure, open space provisions etc. all of which serve to reduce the net developable area of a site).

3.3.3 At the Workshop, and in the subsequent 3 completed questionnaires, reference was also made to the gross to net areas taking into account policy asks with regard to public open space provisions.

From our conversations with the Council, in respect to the precise requirement it is understood the Open Space Provision requirement will be site specific and depend on exact location of site, site size, and green space deficiency/need in the area. For the purpose of whole plan testing and using hypothetical sites DVS has had regard to Draft Policy 43: Open Space Provision in New Developments (see below 3.21 for further details) which requires the following:

All developments 10 units+ will be expected to provide between 7% - 18% of the site area as open space or provide an equivalent commuted sum in lieu (18% as open space or amenity space and at least 7% children's play and sport or by commuted sum). Also management and maintenance will be required. As the provision of open space is site specific, the appraisals assume the following:

- A commuted sum of 13% of the land value for all schemes providing 50 or less dwellings - Schemes providing 51 – 100 dwellings require an on-site provision, equivalent to 13% of the gross site area. - Schemes providing 101 or more dwellings require an on-site provision, equivalent to 18% of the gross site area.

3.3.4 In terms of evidence of actual sites appraised by DVS, we have reviewed a number of individual viability assessments (33 in total) appraised by DVS in recent years where we were able to identify the gross and net areas of the site. The size of the schemes range from 0.71 Ha to 18.68 Ha and show the following average gross to net site areas:

0.5 – 2Ha 88.49% 2.0 – 5Ha 85.32% > 5 Ha 74.32%

3.3.5 Taking into account the identified evidence, but also making specific allowances for the Council’s draft open space policy, we have subsequently adopted the following gross to net site areas:

< 0.5Ha 100% 0.5 – 2Ha 85% 2.0 – 5Ha 80% > 5 Ha 75%

3.4 Capacity / Density

3.4.1 Capacity is typically measured on the basis of dwellings per net developable hectare.

3.4.2 On a site by site basis the number of dwellings per net developable hectare will vary depending on the nature of the scheme, the product being provided and the location. For the purposes of this study we have therefore looked to adopt a reasonable average to apply to the hypothetical site types.

3.4.3 By way of evidence, we have reviewed a number of individual viability assessments (a reduced sample of 24 in total) appraised by DVS in recent years. Please note, the sample was smaller than the gross to net area assessment as we limited the sample to schemes providing 25 or more dwellings. The average across the sample equated to 34.92 dwellings per net developable Ha.

3.4.4 At the Stakeholder workshop DVS suggested a density range of 35 – 40 dwellings per net Ha. The general consensus at the workshop was that 35 dwellings per net developable Ha was an appropriate average to use in the site testing. It was noted that the Council’s recent Housing & Economic Land Availability Assessment evidence base also had a stakeholder group agreed appropriate density ranging between 30-40 dwellings per net developable Ha.

3.4.5 Of the 3 questionnaires completed, 1 indicated a range of 30 – 35 dwellings per net developable Ha was appropriate, whereas the other 2 indicated a range of 35 – 40 was reasonable.

3.4.6 Having considered all of the above we have adopted an average capacity equivalent to 35 dwellings per net developable Ha.

3.4.7 With regards to density, this is expressed as total square metres per net developable Ha. Again, this will fluctuate depending on the nature of the site and in particular the type of dwellings that are being constructed (i.e. 3 storey townhouses will have a different density rate to single storey bungalows).

3.4.8 At the Stakeholder Workshop the general view was that density should fall within the range of circa 2,750 to 3,350 sq. m per net developable Ha.

3.4.9 In terms of the 3 questionnaire responses: 1 did not comment on this point, 1 indicated that 3,350 sq. m per net developable Ha appeared “on the high side”, whilst the other stated that 3,350 sq. m per net developable Ha was “a reasonable assumption to reflect national housebuilder expectations”.

3.4.10 Of the sample of 24 viability appraisals received by DVS across the general region, the analysis shows an average equivalent to 2,975 sq. m per net developable Ha. However, it is noted that there are a couple of anomalous entries in the sample, which has served to reduce the overall average. In this regard, it is noted that the median equates to 3,127 sq. m per net developable Ha.

3.4.11 Having taken all into consideration, we have taken the view that a ‘ceiling’ of circa 3,350 sq. m per net developable Ha is appropriate for the purposes of this study. However, in trying to maximize return whilst achieving a balance between detached, semi-detached, terraces etc. we believe that house builders would look to achieve as close as possible to this ceiling level, particularly given the increased use of 2.5 storey house products (i.e. dwellings that provide additional usable space, typically bedrooms, in the void under the roof). As such have appraised the sites on this basis.

3.5 Dwelling Mix And Sizes

3.5.1 We have considered a variety of housing types to include traditional detached, semi-detached and terraced 2 storey housing, 2.5 / 3 storey variations of these house types, bungalows, muse houses and apartments.

3.5.2 Prior to the market ‘crash’ in 2008 developers were regularly looking to apartments and 3 storey variations of semi-detached and terraced dwellings as a way to increase scheme densities and maximise revenues. 3 storey town houses were particularly popular as a way of maximising the space available to the family market, without having to increase the plot sizes (enabling builders to maximize the amount of usable space on a scheme). They typically involved relatively narrow structures where the layout of the traditional house was re- imagined (e.g. kitchens placed on 1st floors, bedrooms on ground floors etc.).

3.5.3 However, since the downturn in the market there has been a general shift in demand, with more traditional 2 storey products proving significantly more popular in the market place than the 3 storey variations (which in turn has resulted in a sharp general fall in the values achieved for the 3 storey house types). Recognising this, house-builders have largely turned away from 3 storey products seen at the height of the market between 2004 and 2007. Instead, the majority of house builders are now delivering ‘tried and tested’ traditional 2 storey products.

3.5.4 That said, there has been a recent increased use of 2.5 storey products (i.e. extra bedrooms are built into the roof void). The benefit of this option is that an increased floor space is achieved without having to re-work the layout of the traditional house i.e. the kitchen and the living room remain at ground floor level with bedrooms on the upper floors, which generally remains more popular with purchasers.

3.5.5 For the above reasons, we have not therefore included any 3 storey house types. Instead, we have assumed there would be a greater focus on traditional 2 storey house types, mixed with some 2.5 storey products described above.

3.5.6 As for apartments, these were equally popular with house builders prior to 2008, with demand primarily being driven by buy-to-let investors, who were keen to take advantage of favourable buy-to-let mortgage products and strong capital growth in the residential market. The result was an increase in apartments outside of the traditional city / town centre locations, with apartments becoming a regular fixture in suburban / edge of settlement housing developments. However, in the wake of the crash the buy-to-let market suffered a sharp decline. As such demand for apartments fell, which in turn meant values decreased significantly. In many cases developers were left with apartments that they were unable to sell unless heavily discounted. In light of this recent market experience, in the current climate house builders are still taking a more cautious approach to the apartment sector.

3.5.7 For the majority of the sites across the Doncaster area, given their locations it is therefore considered unlikely that apartments would make up a significant proportion of the accommodation provided (if any). For the purposes of this study we have not subsequently looked to factor in any apartments into the appraisals.

3.5.8 In summary, for all of our appraisals we have looked to include a mix of 2/2.5 storey traditional detached, semi-detached and terraced housing, as these are the types of dwellings we anticipate housebuilders will continue to promote going forward.

3.5.9 In terms of an appropriate average dwelling size, the sample of 24 appraisals used above to determine capacity / density, indicates an average house size of 85 sq. m (or a median of 88 sq. m).

3.5.10 However, the purposes of this study, we have adopted a slightly higher average house size of 92.90 sq. m (or 1,000 sq. ft.), which allows for the use of more 2.5 storey products, which would increase the overall average size of the dwellings.

3.5.11 This slightly higher figure has also been chosen to ensure the dwellings built meet minimum requirements, as detailed in the Department for Communities & Local Government’s (“DCLG”) “Technical housing standards – nationally described space standard” March 2015, which can be summarised below in

Table 1 – Minimum gross internal floor areas and storage (sq. m)

Number of Number of bed spaces 1 2 3 beds (persons) storey storey storey 1b 1p 39 2p 50 58 2b 3p 61 70 4p 70 79 3b 4p 74 84 90 5p 86 93 99 6p 95 102 108 4b 5p 90 97 103 6p 99 106 112 7p 108 115 121 8p 117 124 130 5b 6p 103 110 116 7p 112 119 125 8p 121 128 134 6b 7p 116 123 129 8p 125 132 138

3.5.12 Having considered the above minimum space standards an average of 92.90 sq. m (or 1,000 sq. ft.) was considered appropriate in undertaking this study.

3.6 Specification

3.6.1 In the market place, there will be some variation in the specification of the final dwellings; and in the degree of aspiration for high quality design. For the purposes of a Whole Plan viability assessment we recommend an assumption that the sites to be developed reflect an average design specification for that particular location (therefore in high value areas a better specification is allowed for than in lower value areas).

3.6.2 It is also recognised that draft Policy 67 Housing design Standards seeks to ensure all new dwellings are built to Part M4 (2) of the building regulations (which relates to the accessibility and adaptability of dwellings).

3.7 Affordable Rented Assumptions

3.7.1 For the purposes of this assessment we have assumed any affordable housing provision is to be provided ‘on-site’.

3.7.2 The Home and Communities Agency (“HCA”) publication “Rent Standard Guidance” April 2015 defines Affordable Rent as follows:

“Homes let on Affordable Rent terms fall within the definition of social housing but are exempt from the full requirements of rent restructuring” Pg. 12 Paragraph 4.2

“Homes let on Affordable Rent terms should be made available at a rent level of up to 80% of gross market rents (inclusive of service charges where applicable)”. Pg. 12 Paragraph 4.4

“the maximum annual rent increase on an Affordable Rent property will be the Consumer Price Index (CPI) + 1.0%. CPI will be taken as at September of the previous year. This figure is a ceiling, not a target. It is open to providers to increase rents by a lower figure where circumstances justify doing so. On each occasion that an Affordable Rent tenancy is issued for a property -whether it is

let to a new tenant or an existing tenancy is re-issued, providers are required to reset the rent based on a new valuation, to ensure that it remains at no more than 80% of the relevant market rent. This requirement overrides the CPI + 1.0% limit. Pg. 13 and 14 Para 4.13 & 4.24

3.7.3 In short, Affordable Rented units are therefore calculated in relation to the private sector, with a maximum charge of 80% of the Market Rent. This is combined with an annual increase cap of CPI + 1%, albeit whenever a new tenancy is entered into the affordable rent reverts to a maximum of 80% of the Market Rent, overriding the annual increase.

3.7.4 In terms of identifying appropriate transfer values for Affordable Rented units the following methodologies can be adopted:

(i) Rent and yield – arrive at a transfer value by identifying the net rental income to the Registered Provider and capitalising this using an appropriate yield.

(ii) Comparable – transfer values submitted by applicants as part of their own viability appraisals.

3.7.5 For method (i) we assume that a Registered Provider would look to charge the maximum allowed, being 80% of the Market Rent. We also assume that any service charge would only cover costs incurred (i.e. no profit is made from the service charge). We would not therefore include service charges in our assessment.

3.7.6 We have then assessed Market Rents across Doncaster. Once an average Market Rent was identified for each housing type in each location we calculated 80% of this figure (reflecting the maximum chargeable rent outlined above, being 80% of Market Rent).

3.7.7 From this we have looked to ‘net’ down the gross Affordable Rent by making allowances for management, bad debts, voids and repairs / maintenance. We consider fixed costs of £500 per annum for management and £600 per annum for general repairs / maintenance to be reasonable. We have also made an allowance of 3% on the gross Affordable Rent to reflect bad / debts and voids.

3.7.8 We have then applied a capitalisation rate (yield) of circa 6% to arrive at the transfer value (rounded up or down to nearest £5k).

3.7.9 We have looked at a sample of 20 modern houses across Doncaster, which are either currently available to let or have recently been let, as follows:

Table 2 – Sample of rented modern houses & affordable rent calculation

Market Afford Rent Rent Transfer Street Locality Type Bed p.c.m p.a. Value Langer St DN4 Terr 2 440 4,224 49,955 Ansult Court DN5 Terr 2 475 4,560 55,387 Kirkby Avenue DN5 Terr 2 495 4,752 58,491 Field View DN12 Terr 2 495 4,752 58,491 Hams Road DN3 Terr 2 575 5,520 70,907 Twigg Crescent DN3 Terr 3 575 5,520 70,907 Walstow Crescent DN3 Terr 3 595 5,712 74,011 Branchcroft Drive DN4 Terr 3 625 6,000 78,667 Cavalier Court DN4 Terr 3 645 6,192 81,771 Harris Rd DN3 Terr 3 650 6,240 82,547 Ellers Road DN4 Terr 3 695 6,672 89,531 Harden Mews DN3 Terr 3 700 6,720 90,307 Huxterwell Drive DN4 Semi 2 525 5,040 63,147 Aidans Close DN2 Semi 2 550 5,280 67,027 Sunningdale Drive DN12 Semi 3 520 4,992 62,371 Wood Court DN3 Semi 3 725 6,960 94,187 Honeysuckle Close DN4 Semi 3 750 7,200 98,067 Longfield Drive DN3 Det 3 675 6,480 86,427 Wellingley Rd DN4 Det 4 850 8,160 113,587 Apple Tree Way DN4 Det 4 1,100 10,560 152,387

3.7.10 Having identified a broad transfer value for each house we have then looked to identify this as a ‘rate per sq. m’. To do this, we have divided the estimated transfer value by the average house value used in this study (i.e. an average house size of 92.90 sq. m). This produces a transfer value range of circa £535 to £1,640 per sq. m, with an average of £860 per sq. m. This average equates to circa 49% of the medium sales value rate of £1,750 per sq. m.

3.7.11 In terms of method (ii), typically we see affordable rented units in appraisals received from developers / house builders roughly equating to circa 50% – 55% of market value.

3.7.12 At the Stakeholder workshop the general discussion was around the current uncertainty in the affordable housing sector, owing to changes in funding, proposed introduction of Starter Homes etc. The general consensus was to err on the side of caution when identifying average transfer values in the appraisals.

3.7.13 Having considered all the above, for the purposes of the appraisal testing we have adopted average Affordable Rent transfer values at 45% of the equivalent Market Value dwelling.

3.8 Intermediate / Shared Ownership Assumptions

3.8.1 As part of the Government’s Help to Buy initiative, the Government defines Intermediate/Shared Ownership as being a scheme provided through housing associations whereby the purchaser buys a share of the home (between 25% – 75% of the home’s value) and pays a rent on the remaining share.

3.8.2 Typically we see intermediate/shared ownership units in appraisals received from developers/house builders equating to circa 67.5%/70% of market value.

3.8.3 Again, the general view at the Stakeholder workshop was for a cautious approach to be adopted given current uncertainties in the affordable housing sector.

3.8.4 Having considered all the above, for the purposes of the appraisal testing we have adopted average Intermediate / Shared Ownership transfer values at 67.5% of the equivalent Market Value dwelling.

3.9 Starter Homes

3.9.1 The Housing and Planning Act 2016 imposes a requirement for Starter Homes, stating under Chapter 1, Section 4 of the legislation, “An English planning authority must carry out its relevant planning functions with a view to promoting the supply of starter homes in England”.

3.9.2 A Starter Home is defined under Section 2 as being:

(a) a new dwelling, (b) available for purchase by qualifying first-time buyers only,

(c) to be sold at a discount of at least 20% of the market value, (d) to be sold for less than the price cap, and (e) subject to any restrictions on sale or letting specified in regulations made by the Secretary of State

3.9.3 However, at this stage there is no regulation confirming how Starter Homes are to be introduced into the planning system (e.g. whether this will replace existing affordable housing, be part of the mix of affordable housing, or be in addition to traditional affordable units).

3.9.4 For the purposes of this assessment we have incorporated Starter Homes into a separate scenario test, on the assumption that it would replace traditional affordable housing. This is seen as a useful way of assessing how Starter Homes will impact on the overall viability (i.e. positively or negatively) and enable a direct comparison with the equivalent scenarios that incorporate traditional affordable housing.

3.10 Market Value Sales Revenue

3.10.1 Please see Appendix 1 for a detailed review of the local housing market.

3.10.2 It is important to stress that a series of factors will influence values and that, although development schemes do have similarities, every site is unique. Consequently, whilst market conditions in general will broadly reflect national economic circumstances and local supply / demand factors, within an area there will be particular localities and site-specific factors that generate different values and costs. The range of sites tested in this study allows assessment of viability across varying localities for this reason.

3.10.3 For the purposes of our appraisal testing we have subsequently adopted the following average sales values:

Area Average Sales value (£ per sq. m)

Low value area £1,500 Medium value area £1,750 - £2,000 High value area Over £2,250

3.11 ‘Basic’ Build Costs

3.11.1 A frequently used source of data regarding build costs is the Building Cost Information Service (BCIS) of the RICS. This is a national database, which, for different locations across the country and through sample analysis, seeks to provide indicative costs for constructing a dwelling, to include the foundation works. This is expressed as a rate (£) per square metre of the proposed dwelling and includes scheme preliminaries, but excludes any external costs associated with the project (i.e. fencing, gardens, site infrastructure such as services and road ways, drains, street lights etc.).

3.11.2 For the following reasons, we are of the view that the BCIS has limitations as evidence of residential construction prices, particularly when assessing larger schemes (40/50 dwellings or more):

- In Jan 2016 DVS undertook an analysis of the BCIS data recorded since Jan 2011. Of the 106 housing schemes submitted during this period, 68% related to schemes comprising 20 or less units. Only 7.55% of the data related to schemes over 50 homes.

- It is also stressed that volume house-builders (who would most likely develop large schemes) do not contribute to the data. It is generally accepted that volume house-builders are able to construct houses at a cheaper rate (owing to their ability to bulk-buy materials, offer more regular work and negotiate cheaper contracts with sub-contractors etc.). As the cheaper volume house-builder costs are not reflected by the BCIS, the data can be regarded as being inherently high, at least when trying to determine the construction costs for a large scheme.

- BCIS tracks tender prices, not actual costs. The reality is that developers will typically look to negotiate down tenders. In this regard the BCIS figures are inherently high.

3.11.3 The BCIS data should, therefore, be considered in the context of the above.

3.11.4 In July 2016 BCIS data for general housing costs, rebased to South Yorkshire, shows the following rates:

- A lower quartile figure of £798 per sq. m - Median figure of £900 per sq. m - Upper quartile of £1,017 per sq. m.

3.11.5 In terms of other data sources, we have also considered ‘live’ tender information from the Homes and Communities Agency (“HCA”). The HCA has a tender framework called the Delivery Partner Panel 2 (“DPP2”), which was created primarily to speed up the disposal of surplus public sector land to enable residential construction to proceed. The panel includes mainly national and regional house builders, and therefore accesses information not available to the BCIS. As part of the tender process panel members are invited to submit appraisals on individual sites, with the intention being that by ‘bidding’ against one another the land returns will be maximised. This is therefore considered to be a strong source of information as it gives a clear indication of what house builders are willing / able to build houses for in a competitive situation.

3.11.6 In Jan 2016 the HCA provided average figures for 81 developments (with an average 231 residences) the median build cost ascertained from tender bids is £831 per sq. m. As this data is derived primarily from volume house builders it is considered to be appropriate when assessing larger schemes.

3.11.7 However, the figure of £831 per sq. m has been attached a “weighting” of 100, which is in line with the approach the BCIS takes to show regional variances. Different areas are given different weightings by the BCIS, for example I note South Yorkshire currently has a weighting of 93. Whilst it is unclear on what basis these weightings are assessed, in this instance we consider it reasonable to apply the current weighting to the DPP2 data, to help ensure the data provides a more accurate assessment of the local area. Applying a weighting of £93 to £831 per sq. m would decrease the build costs to circa £773 per sq. m.

3.11.8 In terms of drawing a conclusion from the above information we would highlight two keys factors which we consider to be significant:

(i) For the larger schemes (say 25 – 50 units plus) it is assumed these would most likely attract volume house-builders.

(ii) Volume house builders are able to deliver schemes at reduced costs when compared to smaller local / regional builders.

3.11.9 At the Stakeholder workshop the general view was that BCIS remains a valid source of data on build costs. Reference was made to other viability studies prepared on behalf of other local authorities, which had adopted the BCIS and ultimately received approval from the Planning Inspectorate. However, it was agreed that the data did have its limitations. The 3 completed questionnaires also advocated the use of BCIS.

3.11.10 We also note other area wide studies have tended to favour the use of the BCIS. However, there are example of other studies which also appear to conclude that using the BCIS median figure is not appropriate for an area wide viability assessment. For example, the Calderdale Council Local Plan and CIL Viability Evidence undertaken by GVA, dated Oct 2015, concluded that construction costs based on the BCIS lower quartile figures was appropriate for the purposes of the study.

3.11.11 In the context of the above, we have subsequently looked to use the BCIS in determining the ‘basic’ build costs of each dwelling.

3.11.12 However, in light of the DPP2 information, which demonstrates there is a saving on build costs for larger developments, we have looked to run 2 scenarios when testing each hypothetical site type; the first being based on the BCIS lower quartile figure of £798 per sq. m, the second being based on the median of £900 per sq. m. This approach is therefore designed to capture, within the testing process, the differential identified in the evidence to ensure the results are as accurate as possible.

3.12 Externals / Infrastructure

3.12.1 In addition to the per sq. m build costs described above, allowance needs to be made for a range of infrastructure costs – roads, drainage, and services within the site; parking, footpaths, landscaping and other external costs; as well as off- site costs for drainage (including SUDs) and other services.

3.12.2 Many of these items will depend upon individual site circumstances and can only be estimated following a detailed assessment of each site. This is not practical within the scope of this study and therefore, based upon the experience of our Quantity Surveyors, a general allowance in relation to the build costs should be made.

3.12.3 Based on this experience, the standard approach to determining suitable infrastructure / external works as part of a viability assessment, and one that is commonly used in the industry, is to apply a percentage allowance to the basic build cost. Our experience is that, as a percentage of basic build cost, this tends to be in the region of 15%, albeit it is acknowledged that this will fluctuate depending on the nature of the site.

3.12.4 At the Stakeholder workshop the general view was that a 15% allowance for external costs was a reasonable assumption for the purposes of Whole Plan viability study. Further to this, 2 of the completed questionnaires also advocated the use of a 15% allowance. The other completed questionnaire did not indicate an appropriate percentage allowance, although the stakeholder did state that it was appropriate to adopt a single percentage allowance across all site types.

3.12.5 We have also reviewed a number of area wide viability studies from the wider region. Please note a number of the area wide studies date back to 2013. This reflects the studies that were available in the public domain at the time of writing this report. This ‘lag’ is likely to be because of the time it takes, for example, for a CIL viability study to be undertaken and a CIL charge being implemented, such as the GVA Leeds viability CIL study is dated January 2013, however CIL was not introduced within Leeds until April 2015. However, some of the key principles are still considered to be relevant, despite dating back to 2013.

3.12.6 Furthermore, please note we have not referenced all of the studies identified when considering external costs. This is because in some of the studies we were unable to identify what had been allowed for in the external costs. For example, the GVA Leeds viability study adopted an ‘all in’ build cost on a rate per sq. m, which included the construction cost of the houses, preliminaries and external costs. We were therefore unable to extrapolate what had been specifically allowed for in terms of external costs.

3.12.7 Notwithstanding the comments above, we have identified the following external costs as shown within area wide studies:

 Gedling Borough Council Local Plan Viability Assessment – undertaken by NCS, dated Mar 2016. Externals / infrastructure inherently included within adopted build cost, therefore a percentage allowance cannot be provided.

 Bradford City Council CIL Viability Evidence – undertaken by DTZ, dated June 2015. Externals / infrastructure 15% of basic build cost.

 Calderdale Council Local Plan and CIL Viability Evidence – undertaken by GVA, dated Oct 2015. Externals / infrastructure 15% of basic build cost.

 Mansfield District Council Local Plan: Viability Assessment – undertaken by DSP Planning and Development Viability Consultants, dated Nov 2015. Externals / infrastructure 5% - 20% of basic build cost (depending on the nature of the site type).

 Wakefield Metropolitan District Council CIL Viability Study – undertaken by DTZ, dated Feb 2014. Externals / infrastructure 15% of basic build cost.

 Sheffield City Council CIL Viability Study – undertaken by BNP Paribas, dated Feb 2014. Externals / infrastructure 15% of basic build cost.

 Leicester, Leicestershire & Rutland CIL Viability Study – undertaken by HDH Planning, dated January 2013. Externals / infrastructure 10% of basic build cost for the smallest sites, increasing to 20% for the larger greenfield schemes.

 North York Moors National Park Authority CIL Economic Viability Assessment – undertaken by Peter Brett Associates, dated November 2013. Externals / infrastructure 10% of basic build cost.

 North East Lincolnshire Local Plan and CIL Viability Assessment – undertaken by GVA, dated September 2013. Externals / infrastructure 20% of basic build cost.

 Rotherham Metropolitan Borough Council Community Infrastructure Levy Study – undertaken by Peter Brett Associates, dated July 2013. Externals / infrastructure 15% of basic build cost.

 Harrogate CIL Economic Viability Assessment, dated March 2013. Externals / infrastructure 10% of basic build cost.

 Stafford Borough Council Report on Viability and Deliverability, dated July 2013. Externals / infrastructure 10% of basic build cost.

 Selby District Council Community Infrastructure Levy Study – undertaken by Peter Brett Associates, dated Sept 2013. Externals / infrastructure 10% of basic build cost.

3.12.8 The external costs included across the 13 area wide studies show an average of 12.5% to circa 14.6%.

3.12.9 Taking all into consideration we have taken the view that a single figure of 15% of the basic build cost is considered fair and reasonable.

3.13 Contingency

3.13.1 In addition to basic build costs and external / infrastructure works described above, it is common practice to include a contingency allowance in the event that any unforeseen costs arise once the development proceeds. We consider this to be fair and reasonable for the purposes of a viability appraisal.

3.13.2 Our experience is that, as a percentage of basic build cost and externals / infrastructure, this tends to range from circa 2% to 5%, depending on the nature of the site.

3.13.3 At the Stakeholder workshop the general view was that 3% - 5% was appropriate. 2 of the completed questionnaires suggested 3% for greenfield and 5% for brownfield, whereas the other suggested 3.5% for greenfield and 5% for brownfield.

3.13.4 As further evidence we have again looked at viability appraisals received from applicants, received by DVS. We have again used the same sample of 33 appraisals (used above for the assessment of gross and net areas), which have been received during the last 2 years for mixed housing sites, covering a wide range of sites (from 28 to 600 dwellings). Please note as these cases contain sensitive commercial information we have not provided the full address details or the parties involved. The contingency costs identified range from 0% to 7.15%. The average across the sample for greenfield sites equates to 2.96% and brownfield sites equates to 3.07%. This suggests that in practice there is little differential between greenfield and brownfield contingency allowances.

3.13.5 In addition, we have also considered the DPP2 evidence referenced above (in the consideration of basic build costs). For 81 projects in 2015, being a mix of brownfield and greenfield sites, the mean contingency equated to 2.7%.

3.13.6 As per external costs, we have also reviewed a number of area wide viability studies from the wider region. Please note we have not referenced all of the studies identified when considering contingency. This is because in some of the studies we were unable to identify what had been allowed for in the contingency.

3.13.7 Notwithstanding the comments above, we have identified the following contingency as shown within area wide studies:

 Gedling Borough Council Local Plan Viability Assessment – undertaken by NCS, dated Mar 2016. Contingency 5% of basic build cost.

 Bradford City Council CIL Viability Evidence – undertaken by DTZ, dated June 2015. Contingency 5% of basic build cost.

 Calderdale Council Local Plan and CIL Viability Evidence – undertaken by GVA, dated Oct 2015. Contingency 3% of basic build costs for greenfield sites, increasing to 5% for brownfield sites.

 Mansfield District Council Local Plan: Viability Assessment – undertaken by DSP Planning and Development Viability Consultants, dated Nov 2015. Contingency 5% of basic build cost.

 Wakefield Metropolitan District Council CIL Viability Study – undertaken by DTZ, dated Feb 2014. Contingency 5% of basic build cost.

 Sheffield City Council CIL Viability Study – undertaken by BNP Paribas, dated Feb 2014. Contingency 5% of basic build cost.

 Leicester, Leicestershire & Rutland CIL Viability Study – undertaken by HDH Planning, dated January 2013. Contingency 2.5% for the greenfield sites, increasing to 5% for the larger greenfield schemes.

 North York Moors National Park Authority CIL Economic Viability Assessment – undertaken by Peter Brett Associates, dated November 2013. Contingency 5% of basic build cost.

 North East Lincolnshire Local Plan and CIL Viability Assessment – undertaken by GVA, dated September 2013. Contingency 3% of basic build cost.

 Rotherham Metropolitan Borough Council Community Infrastructure Levy Study – undertaken by Peter Brett Associates, dated July 2013. Contingency 5% of basic build cost.

 Harrogate CIL Economic Viability Assessment – undertaken by Tym & Partners (Part of Peter Brett Associates), dated March 2013. Contingency 5% of basic build cost.

 Stafford Borough Council Report on Viability and Deliverability – undertaken by Levvel Ltd, dated July 2013. No contingency has been included.

 Selby District Council Community Infrastructure Levy Study – undertaken by Peter Brett Associates, dated Sept 2013. Contingency 5% of basic build costs.

3.13.8 9 of the 13 studies identified show a single contingency allowance of 5%. However, it should be noted that 4 of these studies adopt external / infrastructure costs at only 10% (therefore below the 15% concluded above for the purposes of this study). This may explain why the contingency has been set at 5%, which in our experience is towards the top end of the range.

3.13.9 Having assessed all of the available evidence we have taken the view that a single figure of 3% of the basic build cost is considered fair and reasonable for the purposes of this area wide study.

3.14 Abnormal Development Costs

3.14.1 Abnormal costs relate to issues such as decontamination, adverse geo technical conditions, off-site highway works, demolition of existing buildings etc. (i.e. works which would not be associated with a ‘standard’ scheme). Abnormal costs will vary significantly depending on the nature of the scheme, ranging from zero to potentially several million pounds.

3.14.2 Often abnormal costs are not revealed until a full scheme design is completed and the relevant due diligence undertaken. It is therefore impossible to provide a robust assessment of the likely abnormal costs that could be associated with the sites. For this reason, in undertaking district wide viability studies for some other Councils we have adopted a zero (nil) cost for abnormal works.

3.14.3 However, based on our experience we are of the view that it is appropriate to include a ‘spot allowance’ for abnormal costs (as it is felt that most sites within the area have in the past and are likely in the future to attract additional abnormal costs).

3.14.4 Our approach is to consider abnormal costs on a ‘rate per Ha’ basis.

3.14.5 In our appraisals we have subsequently adopted a figure of £100,000 per gross Ha for the greenfield sites, £200,000 per gross Ha for cleared brownfield sites and £300,000 per Ha for occupied brownfield sites. Whilst arbitrary, this at least acknowledges that most sites tend to attract some form of abnormal costs.

3.15 Professional Fees

3.15.1 Our experience is that circa 6% of build costs is appropriate (expressed as a percentage of the basic build costs plus the external works).

3.15.2 At the Stakeholder workshop a range of 5% - 8% was discussed. 2 of the completed questionnaires suggested 8%, whilst the other suggested a range of 8% to 10% depending on the nature of the site type.

3.15.3 As further evidence we have again looked at viability appraisals received from applicants, received by DVS. We have again used the same sample of 33 appraisals (used above for the assessment of gross and net areas), which have been received during the last 2 years for mixed housing sites, covering a wide range of sites (from 28 to 600 dwellings). Please note as these cases contain sensitive commercial information we have not provided the full address details or the parties involved. The professional fees identified range from 0% to 13%. The average across the sample equates to 5.77%.

3.15.4 In addition, we have also considered the DPP2 evidence referenced above. For 81 projects in 2015, the mean professional fees equated to 5.10%.

3.15.5 As per external costs and contingency, we have also reviewed a number of area wide viability studies from the wider region. We have identified the following professional fees as shown within area wide studies (please note a couple of studies are not included in this sample, as they did not explicitly state what allowance had been made for professional costs):

 Gedling Borough Council Local Plan Viability Assessment – undertaken by NCS, dated Mar 2016. Professional fee 8% of build cost.

 Bradford City Council CIL Viability Evidence – undertaken by DTZ, dated June 2015. Professional fee 6% of build cost.

 Calderdale Council Local Plan and CIL Viability Evidence – undertaken by GVA, dated Oct 2015. Professional fee 5% of basic build cost for sites larger than 50 dwellings, 8% for sub 50 dwellings.

 Mansfield District Council Local Plan: Viability Assessment – undertaken by DSP Planning and Development Viability Consultants, dated Nov 2015. Professional fee 10% of build cost.

 Wakefield Metropolitan District Council CIL Viability Study – undertaken by DTZ, dated Feb 2014. Professional fee 6% of build cost.

 Sheffield City Council CIL Viability Study – undertaken by BNP Paribas, dated Feb 2014. Professional fee 10% of build cost.

 Leicester, Leicestershire & Rutland CIL Viability Study – undertaken by HDH Planning, dated January 2013. Professional fee 10% of build cost.

 North York Moors National Park Authority CIL Economic Viability Assessment – undertaken by Peter Brett Associates, dated November 2013. Professional fee 10% of build cost.

 North East Lincolnshire Local Plan and CIL Assessment – undertaken by GVA, dated September 2013. Professional fee 8% of build cost.

 Harrogate CIL Economic Viability Assessment – undertaken by Tym & Partners (Part of Peter Brett Associates), dated March 2013. Professional fee 10% of build cost.

 Selby District Council CIL Study – undertaken by Peter Brett Associates, dated Sept 2013. Professional fee 10% of build cost.

3.15.6 6 out of the 11 studies therefore include professional fees of 10%. Again, though, a number of these studies showed what we considered to be extremely low external costs (at 10% of basic build costs, compared with our figure of 15%) and therefore it was unclear whether some of the costs that we would regard as external works were actually being factored within the ‘professional fees’ section of the appraisal.

3.15.7 On balance, and given our experience in the market place, we have concluded that for schemes that are most likely to attract a volume house builder we consider 6% of the basic build / external cost to be appropriate, due to the limited

requirement for ‘original’ thought in the design process. For smaller schemes (which we have assumed as being sub 20 dwellings) that would attract say local builders we would increase this figure to 8%, principally to reflect the added professional input required at the design stage.

3.16 Marketing

3.16.1 Our experience is that circa 3% of sales revenue is appropriate.

3.16.2 At the Stakeholder workshop the general consensus was that 3% was appropriate, albeit excluding legal fees. All 3 of the completed questionnaires agreed with 3%, but on the basis that legal fees would be allowed for separately within the appraisal.

3.16.3 In addition, we have also considered the DPP2 evidence referenced above. For 81 projects in 2015, the mean marketing equated to 3.5% (inclusive of legal fees).

3.16.4 As per external costs and contingency, we have also reviewed a number of area wide viability studies from the wider region. We have identified the following marketing and legal fees as shown within area wide studies (please note a couple of studies are not included in this sample, as they did not explicitly state what allowance had been made for professional costs):

 Gedling Borough Council Local Plan Viability Assessment – undertaken by NCS, dated Mar 2016. Marketing (including legal fees) 2.5% of sales.

 Bradford City Council CIL Viability Evidence – undertaken by DTZ, dated June 2015. Marketing (including legal fees) 3.5% of sales.

 Calderdale Council Local Plan and CIL Viability Evidence – undertaken by GVA, dated Oct 2015. Marketing (excluding legal fees) 3% of sales.

 Mansfield District Council Local Plan: Viability Assessment – undertaken by DSP Planning and Development Viability Consultants, dated Nov 2015. Marketing (excluding legal fees) 3% of sales.

 Wakefield Metropolitan District Council CIL Viability Study – undertaken by DTZ, dated Feb 2014. Marketing (including legal fees) 3.5% of sales.

 Leicester, Leicestershire & Rutland CIL Viability Study – undertaken by HDH Planning, dated January 2013. Marketing (including legal fees) 2.5% of sales.

 North York Moors National Park Authority CIL Economic Viability Assessment – undertaken by Peter Brett Associates, dated November 2013. Marketing (including legal fees) 3% of sales.

 North East Lincolnshire Local Plan and CIL Viability Assessment – undertaken by GVA, dated September 2013. Marketing (including legal fees) 3% of sales.

 Harrogate CIL Economic Viability Assessment – undertaken by Tym & Partners (Part of Peter Brett Associates), dated March 2013. Marketing (including legal fees) 3% of sales.

 Selby District Council Community Infrastructure Levy Study – undertaken by Peter Brett Associates, dated Sept 2013. Marketing (including legal fees) 3% of sales.

3.16.5 2 out of the 10 studies therefore include marketing fee of 3% of sales values (excluding legal costs), whilst 4 of the 10 show marketing fees of 3% inclusive of legal costs.

3.16.6 On balance, and given our experience in the market place, we have concluded that 3% of sales value is appropriate, plus an additional allowance for legal costs (which we have set at £500 per dwelling, which is considered to be, if anything, on the generous side based on individual viability appraisals my office receives from applicants).

3.17 Developer’s Profit

3.17.1 When considering profit margins for the purposes of a Whole Plan viability assessment we have looked to apply what is considered to be the ‘minimum’ return a developer would require to develop each site.

3.17.2 From our experience, we consider the two tiered approach advocated by the HCA to be appropriate. This involves applying a profit level to the Market Value homes (normally 15 – 20% of revenue) and a lower contractor’s margin (5 – 10% on cost) to the affordable dwellings. When the two are factored together this gives an overall ‘blended’ profit level.

3.17.3 At the Stakeholder workshop views were expressed that for the market value units a rate of 17.5% of sales revenue was appropriate. Equally, others suggested a figure of 20% of sales revenue. 1 of the completed questionnaires supported a rate of 17.5% of sales revenue on market value dwellings, whilst the other 2 suggested a rate of 20% for the market value dwellings. 2 of the completed questionnaires also supported a reduced rate of 8% to 10% for the affordable dwellings.

3.17.4 The DPP2 evidence, for 81 projects in 2015, the mean developer’s profit equates to 19% on GDV for the open market dwellings and 8.1% on cost for the affordable dwellings.

3.17.5 As further evidence we have again looked at viability appraisals received from applicants, received by DVS. However, of the sample of 33 appraisals referenced above, 5 did not ‘fix’ the profit margin in the appraisal (and showed negative profit margins). In assessing profit the sample is therefore reduced to 28 appraisals. Of this reduced sample, 10 show a minimum developer’s profit of 17.5% (or lower), 3 show a return of 17.5% to 18.5%, and the remaining 15 show a minimum profit of 20% (or higher). The average profit across the sample equates to 18.77% (for market value dwellings).

3.17.6 We have also reviewed a number of area wide viability studies from the wider region. We have identified the following profit margins as shown within area wide studies (please note a couple of studies are not included in this sample, as they did not explicitly state what allowance had been made for professional costs):

 Gedling Borough Council Local Plan Viability Assessment – undertaken by NCS, dated Mar 2016. 15% to 20% return for market value dwellings, 6% to 10% for affordable units.

 Bradford City Council CIL Viability Evidence – undertaken by DTZ, dated June 2015. 20% return for market value dwellings, 6% for affordable units.

 Calderdale Council Local Plan and CIL Viability Evidence – undertaken by GVA, dated Oct 2015. 18% return for market value dwellings.

 Mansfield District Council Local Plan: Viability Assessment – undertaken by DSP Planning and Development Viability Consultants, dated Nov 2015. 20% return for market value dwellings, 6% for affordable units.

 Wakefield Metropolitan District Council CIL Viability Study – undertaken by DTZ, dated Feb 2014. 20% return for market value dwellings, 6% for affordable units.

 Leicester, Leicestershire & Rutland CIL Viability Study – undertaken by HDH Planning, dated January 2013. 20% return when applied to costs not revenue, which typically equates to circa 17.5% to 18% on revenue.

 North York Moors National Park Authority CIL Economic Viability Assessment – undertaken by Peter Brett Associates, dated November 2013. 20% return when applied to costs not revenue, which typically equates to circa 17.5% to 18% on revenue.

 North East Lincolnshire Local Plan and CIL Viability Assessment – undertaken by GVA, dated September 2013. 25% return when applied to costs not revenue, which typically equates to circa 22% to 23% on revenue.

 Harrogate CIL Economic Viability Assessment – undertaken by Tym & Partners (Part of Peter Brett Associates), dated March 2013. 20% return when applied to costs not revenue, which typically equates to circa 17.5% to 18% on revenue.

3.17.7 Based on the evidence above there is clearly an element of ‘flex’ in profit levels depending on the nature of each particular scheme. The data suggests that a general range of 15% – 20% on GDV for market value houses is fair and reasonable at the current time, with the higher end of the range tending to be applied to larger schemes in lower value areas (and perhaps with higher associated abnormal / infrastructure costs). For affordable housing a profit margin of 6% – 10% of cost appears to be broadly reasonable.

3.17.8 For clarity, when undertaking individual viability assessments it is therefore necessary to assess the appropriate level of profit within the context of the above parameters. Each site should be taken on its individual merits before a conclusion is reached on a suitable level of profit. We do not therefore consider a fixed figure to be appropriate when undertaking individual viability appraisals.

3.17.9 However, for the purposes of this study we consider a developer’s profit equivalent to 18.5% of revenue to be appropriate, reducing to 8% on cost for the affordable units. However, please note for the smallest site types (sub 10 dwellings) we have reduced the profit margin on market value dwellings to 15% on revenue (please note affordable units does apply to residential schemes under 10 dwellings).

3.18 Finance

3.18.1 It has been assumed throughout this study that VAT either does not arise or that its effects can be ignored.

3.18.2 There are various approaches that can be taken to assessing finances, including more complicated methods such as assessing the “Internal Rate of Return” or an assessment of the “Return on Capital Employed”. However, having discussed this at the stakeholder workshops we considered the general consensus to be that area wide studies should adopt a relatively simple approach to assessing finances, where interest rates (and credit rates) at

perceived market levels are inputted into an industry approved toolkit (in this case the HCA DAT).

3.18.3 Our experience of other areas in the North of England (including specifically the North East of England) suggests that finance costs have dropped slightly in the last 18 – 24 months, and that a range of 5.5% – 6.5% is now more common place. The DPP2 evidence referenced above actually shows an average interest rates of 3.7%.

3.18.4 In this regard, we consider a debit rate at 6% to be reasonable in the current climate (albeit it is stressed this is likely to change if Bank of England interests rates increase significantly). However, for small schemes (sub 10 dwellings) we have increased this rate to 7%.

3.18.5 We would also advocate the use of a credit rate of 3% (which tends to be relevant to larger, multi phased developments completed over a number of years). Again, this is considered to be in line with schemes we are appraising across the region and also guidance provided by the Homes and Communities Agency (“HCA”).

3.19 Threshold Land Value (“TLV”)

3.19.1 Please see Appendix 2 for a detailed review of the concept of threshold land value and the methodology used in identifying suitable values.

3.19.2 As detailed in the appendix, for the purposes of this study the following TLVs for greenfield sites across Doncaster have been adopted:

Area Definition (£ per sq. m) TLV range (£ per gross Ha)

Low value area £1,500 £200,000 Medium value area £1,750 - £2,000 £270,000 High value area Over £2,250 £350,000

3.19.3 However, please note for significantly larger urban extension sites (400 units plus), and any strategic sites, we would expect a level of discount from the above figures to reflect quantum.

3.19.4 As for brownfield site types, for the purposes of the viability testing we have looked to differentiate between cleared, brownfield sites and occupied, brownfield sites, an approach which was supported by the stakeholder engagement. For each, we have therefore adopted a different TLV, as follows:

Cleared, brownfield - £185,000 per Ha

Occupied, brownfield - £370,000 per Ha

3.20 Site Acquisition and Disposal Costs

3.20.1 The developments are assumed to proceed immediately and so no allowance should be made for holding costs, or indeed any income arising from ownership of the site prior to implementation. Acquisition Costs include stamp duty at the prevalent rate and an allowance of 0.5% for acquisition legal fees, plus an allowance for land registry fees. On larger, more complex sites an agent fee (circa 1%) can be deemed appropriate.

3.21 Section 106 Contributions / Emerging Policy Aspirations

3.21.1 Emerging policies should be considered as part of the testing. In accordance with the NPPF / PPG the Council policies should not undermine viability.

3.21.2 The Council have confirmed the following information on emerging draft policies that have the potential to impact on viability, and the associated cost (which may be subject to change):

 Policy 15: Delivering the necessary range of housing – for sites providing in excess of 15 units or 0.5 ha at 25% affordable provision (on-site) o Tenure split should reflect the latest SHMA.

 Policy 23: Promoting sustainable transport within new developments – New developments may be required to make contributions towards transport / highways infrastructure. o For sites in excess of 50 dwellings the appraisals incorporate a cost of £1,000 per dwelling.

 Policy 43: Open space provision in new developments – schemes providing in excess of 10 dwellings will be asked to provide a percentage of the gross site area as open space or provide a commuted sum in lieu. For schemes providing 10 to 50 dwellings the requirement will be 13% of the gross site area, but as a commuted sum. For schemes of 100 dwellings or more the requirement will be an on-site provision equivalent to 18% of the gross site area.

 Policy 46: Valuing biodiversity and geodiversity - a site specific policy o For the purposes of the study this has been applied to sites providing in excess of 50 dwellings, the appraisals incorporate this cost, at a rate of £250 per dwelling.

 Policy 67: Housing design standards – all homes will be required to meet the Nationally Described Space Standards as a minimum. All new housing should also be built to Part M4(2) of the current building regulations.

 Policy 77: New education facilities. May apply to sites in excess of 20 dwellings. o For the purposes of the study this has been applied to sites providing in excess of 50 dwellings, the appraisals incorporate this cost, at a rate equivalent to £1,500 per dwelling.

 Policy 80: Contamination and unstable land – some brownfield sites may incur contaminated land / unstable land remediation costs. o For the purposes of the study an allowance of £2,000 per unit has been incorporated into the appraisals of the brownfield sites.

 Policy 82: Flood risk management, the policy explains proposals in Flood Risk Zones 2 & 3 will be required to mitigate residual risks. o For the purposes of the study an allowance of £4,000 per unit has been incorporated into the appraisals.

3.22 Scenario testing / sensitivity analysis

3.22.1 As indicated above, the residual method used in viability testing is not without its flaws, owing to the wide variety of inputs used in the approach (all of which could be subject to variance, which can potentially lead to a wide range of outputs in the appraisals making it more difficult to reach robust conclusions from the data). Recognising this limitation, the PPG and the relevant professional guidance for viability assessments advocates adopting scenario testing / sensitivity analysis approach to Whole Plan viability studies.

3.22.2 At the stakeholder workshop, and from the completed questionnaires received, there was a general agreement that scenario testing / sensitivity analysis should consider fluctuations in sales values and build costs. There was also a general view that testing should be focused on likely fluctuations in inputs and not test changes which were unlikely to occur.

3.22.3 Having considered the above we have adopted the following approach to scenario testing / sensitivity analysis:

Test 1 Adopts all draft Council policies detailed above in 3.21, except for Policy 15 (i.e. 25% affordable housing). Instead, the level of affordable housing provision is adjusted up to a point where the scheme is considered to be viable (if possible, otherwise a nil provision is adopted).

For ‘basic’ build costs, the BCIS lower quartile is adopted at £798 per sq. m.

Test 2 As Test 1, except the ‘basic’ build costs has been adjusted to the BCIS median figure of £900 per sq. m.

Test 3 As Test 1, except Council Policy 82 relating to flood risk mitigation (i.e. a cost in the appraisals of £4,000 per dwelling) is removed.

Test 4 As Test 1, except the sales revenue is increased by 5%.

Test 5 As Test 1, except the affordable dwellings (if any are demonstrated) are provided through the Starter Homes tenure basis only.

Test 6 As Test 1, except all draft Council planning policies (including affordable housing) are removed.

3.22.4 As indicated above, as part of the testing process it is important to set at what ‘point’ a scheme is deemed to be viable or unviable. However, based on our experience, viability is not necessarily ‘black and white’ and often reference is made to schemes being ‘marginally viable’ (i.e. the return generated from the scheme may be sufficient to attract some developers, but not necessarily all). There is no specific guidance on how this should be considered within Whole Plan testing, however the PPG is clear that Council policies should not be set at ‘extremes’ of viability.

3.22.5 In this regard, and based on our experience, we have set the following parameters for determining whether a scheme is viable or not:

Viable A scheme is deemed to be viable if the appraisal generates a surplus which is equivalent to 2.50% (or higher) of the sales revenue. Schemes which meet this criteria are marked in green.

Marginally A scheme is deemed to be marginally viable if the output viable from the appraisal falls within – (minus) 2.49% up to + 2.49% of the sales revenue. Scheme which meet this criteria are marked in orange.

Unviable A scheme is deemed to be unviable if the appraisal generates a deficit which is equivalent to – (minus) 2.50% (or lower) of the sales revenue. Schemes which meet this criteria are marked in red.

4. RESIDENTIAL – APPRAISAL RESULTS

4.1 Introduction

4.1.1 In undertaking our appraisals we have used the HCA Development Appraisal Tool (“DAT”). This is an industry approved, free to download, cash flow toolkit. The appraisal assumptions are inputted into the program, with the results calculated through a cash flow. The program requires a fixed land value to be inputted (in this case the identified TLV), as well as a fixed profit margin. The toolkit then shows either a surplus or a deficit, based on these parameters.

4.1.2 As indicated above in 3.22.5, whether the scheme is deemed to be viable, marginally viable, or unviable depends on the level of surplus / deficit generated by the appraisal.

4.1.3 For ease of reference the typical locations attributed to the hypothetical value areas are summarised below:

Urban Extension Site Testing

No Area/Site Location Value Corresponding Settlement(s) Auckley-Hayfield Green, Service Towns/ Finningley, Bawtry, Tickhill, 1 & 2 High Villages: Sprotbrough, Barnburgh- Harlington)

Doncaster Main Urban 3 & 4 Low Bentley Area

Doncaster Main Urban 5,6 &7 Medium Scawsby Area

8,9 &10 Main Town Medium Mexborough

Low / 11,12 &13 Main Town Conisbrough & Denaby Medium

14,15,16,17, Thorne-Moorends Rossington 18,19,20 & Main Towns Low Adwick-Woodlands 21 Main Towns, Hatfield-Stainforth & Armthorpe, 22,23 & 24 Doncaster Main Urban Medium Edenthorpe & Barnby Dun Area & Service Village

25 & 26 Service Towns Medium Askern, Skellow & Carcroft

27 & 28 Service Town Low Edlington

Urban Settlement Site Testing No Development Type 3 x value Total number of areas (high, site types to test medium & low) 1, 2 & 3 Single dwelling greenfield infill plot H,M,L 3

4, 5 & 6 5 units cleared site H,M,L 3

7, 8 & 9 5 units occupied site H,M,L 3

10,11&12 14 units cleared site H,M,L 3

13,14 &15 14 units occupied site H,M,L 3

16,17 &18 50 units cleared site H,M,L 3

19, 20 & 21 50 units occupied site H,M,L 3

22, 23 & 24 100 units cleared site H,M,L 3

25, 26 & 27 100 units occupied site H,M,L 3

TOTAL 27

4.2 Test 1 – Council policies (bar affordable housing), BCIS lower quartile

4.2.1 Test 1 adopted a ‘basic’ build cost in line with the BCIS lower quartile, equivalent to £798 per sq. m. The level of affordable housing was adjusted, initially to a point where viability could be demonstrated (i.e. the surplus exceeded 2.5% of the sales revenue). Where this was not possible the affordable housing level was adjusted to zero.

The results have been split into separate tables for urban extension sites and urban settlement sites, as follows:

Table 3 – Test 1 Urban extension viability results

No Afford Surplus / % of No Site Value band Viable units (%) deficit GDV

1 Town & village High 50 24.00% £1,322,296 12.65% Viable 2 Town & village High 100 24.00% £2,471,563 11.82% Viable 3 Doncaster main urban Low 50 0.00% -£176,472 -2.53% Unviable 4 Doncaster main urban Low 100 0.00% -£332,462 -2.39% Marginal 5 Doncaster main urban Medium 50 12.00% £234,294 2.88% Viable 6 Doncaster main urban Medium 100 12.00% £424,557 2.61% Viable 7 Doncaster main urban Medium 400 8.50% £1,865,374 2.87% Viable 8 Main town Medium 50 14.00% £236,738 2.91% Viable 9 Main town Medium 100 12.00% £424,557 2.61% Viable 10 Main town Medium 400 10.00% £1,784,332 2.74% Viable 11 Main town Low / med 50 0.00% £197,786 2.62% Viable 12 Main town Low / med 100 0.00% £329,602 2.18% Marginal 13 Main town Low / med 400 0.00% £1,034,489 1.71% Marginal 14 Main town Low 50 0.00% -£176,472 -2.53% Unviable 15 Main town Low 100 0.00% -£332,462 -2.39% Marginal 16 Main town Low 400 0.00% -£1,218,617 -2.19% Marginal 17 Main town Low 50 0.00% -£176,472 -2.53% Unviable 18 Main town Low 100 0.00% -£332,462 -2.39% Marginal 19 Main town Low 50 0.00% -£176,472 -2.53% Unviable 20 Main town Low 100 0.00% -£332,462 -2.39% Marginal 21 Main town Low 400 0.00% -£1,218,617 -2.19% Marginal 22 Town, Urban, Village Medium 50 24.00% £265,293 3.26% Viable 23 Town, Urban, Village Medium 100 12.00% £424,557 2.61% Viable 24 Town, Urban, Village Medium 400 9.75% £1,631,876 2.51% Viable 25 Service town Medium 50 12.00% £234,294 2.88% Viable 26 Service town Medium 100 12.00% £424,557 2.61% Viable 27 Service town Low 50 0.00% -£176,472 -2.53% Unviable 28 Service town Low 100 0.00% -£332,462 -2.39% Marginal

 50 / 100 dwellings in high value areas are comfortably viable, providing 25% affordable.

 Medium value schemes are all viable, broadly supporting a 12% affordable provision.

 Low value scheme (or low / medium value areas) are at best only marginally viable even with a 0% affordable housing provision.

 In the medium value areas, generally speaking schemes of 50 units return the strongest viability results when compared to schemes of 100 or 400.

Table 4 – Test 1 Urban settlement viability results

Value No Afford Surplus / % of No Site Viable band units (%) deficit GDV

1 Green infill High 1 0.00% £57,145 27.34% Viable 2 Green infill Medium 1 0.00% £21,683 13.34% Viable 3 Green infill Low 1 0.00% £5,031 3.61% Viable 4 5 units cleared High 5 0.00% £291,185 27.86% Viable 5 5 units cleared Medium 5 0.00% £104,582 12.87% Viable 6 5 units cleared Low 5 0.00% £11,280 1.62% Marginal 7 5 units occupied High 5 0.00% £250,250 23.94% Viable 8 5 units occupied Medium 5 0.00% £63,647 7.83% Viable 9 5 units occupied Low 5 0.00% -£29,725 -4.27% Unviable 10 14 units cleared High 14 28.58% £449,857 15.37% Viable 11 14 units cleared Medium 14 14.28% £93,935 4.13% Viable 12 14 units cleared Low 14 0.00% -£59,591 -3.05% Unviable 13 14 units occupied High 14 28.58% £335,540 11.47% Viable 14 14 units occupied Medium 14 0.00% £70,076 3.08% Viable 15 14 units occupied Low 14 0.00% -£174,433 -8.94% Unviable 16 50 units cleared High 50 24.00% £1,460,789 13.98% Viable 17 50 units cleared Medium 50 14.00% £214,711 2.64% Viable 18 50 units cleared Low 50 0.00% -£306,987 -4.41% Unviable 19 50 units occupied High 50 24.00% £1,015,190 9.71% Viable 20 50 units occupied Medium 50 0.00% £82,049 1.01% Marginal 21 50 units occupied Low 50 0.00% -£757,883 -10.88% Unviable 22 100 units cleared High 100 25.00% £2,806,113 13.42% Viable 23 100 units cleared Medium 100 12.00% £470,676 2.90% Viable 24 100 units cleared Low 100 0.00% -£571,781 -4.10% Unviable 25 100 units occupied High 100 25.00% £1,727,587 8.26% Viable 26 100 units occupied Medium 100 0.00% £127,105 0.78% Marginal 27 100 units occupied Low 100 0.00% -£1,500,747 -10.77% Unviable

 Single plots schemes are viable, regardless of value band. Affordable contributions do not apply.

 Schemes comprising 5 dwellings are viable in high / medium value areas (again on the basis that an affordable housing contribution is not applicable).

 Schemes in high value areas providing 14, 50 or 100 dwellings are viable, with a 25% provision.

 For cleared, brownfield sites in medium value areas, an affordable provision of circa 12% – 14% is shown to be viable. Occupied brownfield sites, though, show a 0% affordable provision.

 For sites within low value areas, regardless of the nature of the brownfield land, the results show an unviable scheme for development comprising 14, 50 and 100 dwellings.

4.3 Test 2 – Council policies (bar affordable housing), BCIS median

4.3.1 Test 2 adopted a ‘basic’ build cost in line with the BCIS median, equivalent to £900 per sq. m. The level of affordable housing was adjusted, initially to a point where viability could be demonstrated (i.e. the surplus exceeded 2.5% of the sales revenue). Where this was not possible the affordable housing level was adjusted to zero.

4.3.2 Please note, for ease of reference we have split the results for urban extension sites and urban settlement sites into separate tables, as follows:

Table 5 – Test 2 Urban extension viability results

No Afford Surplus / No Site Value band % of GDV Viable units (%) deficit

1 Town & village High 50 24.00% £771,670 7.38% Viable 2 Town & village High 100 24.00% £1,427,927 6.83% Viable 3 Doncaster main urban Low 50 0.00% -£724,014 -10.39% Unviable 4 Doncaster main urban Low 100 0.00% -£1,376,995 -9.88% Unviable 5 Doncaster main urban Medium 50 0.00% -£15,214 -0.19% Marginal 6 Doncaster main urban Medium 100 0.00% -£41,234 -0.25% Marginal 7 Doncaster main urban Medium 400 0.00% -£1,763,747 -2.71% Unviable 8 Main town Medium 50 0.00% £9,278 0.11% Marginal 9 Main town Medium 100 0.00% -£41,234 -0.25% Marginal 10 Main town Medium 400 0.00% -£315,266 -0.48% Marginal 11 Main town Low / med 50 0.00% -£348,103 -4.61% Unviable 12 Main town Low / med 100 0.00% -£705,824 -4.68% Unviable 13 Main town Low / med 400 0.00% -£2,582,439 -4.28% Unviable 14 Main town Low 50 0.00% -£724,014 -10.39% Unviable 15 Main town Low 100 0.00% -£1,376,995 -9.88% Unviable 16 Main town Low 400 0.00% -£4,890,539 -8.77% Unviable 17 Main town Low 50 0.00% -£724,014 -10.39% Unviable 18 Main town Low 100 0.00% -£1,376,995 -9.88% Unviable 19 Main town Low 50 0.00% -£724,014 -10.39% Unviable 20 Main town Low 100 0.00% -£1,376,995 -9.88% Unviable 21 Main town Low 400 0.00% -£4,890,539 -8.77% Unviable 22 Town, Urban, Village Medium 50 0.00% -£15,214 -0.19% Marginal 23 Town, Urban, Village Medium 100 0.00% -£41,234 -0.25% Marginal 24 Town, Urban, Village Medium 400 0.00% -£315,266 -0.48% Marginal 25 Service town Medium 50 0.00% -£15,214 -0.19% Marginal 26 Service town Medium 100 0.00% -£41,234 -0.25% Marginal 27 Service town Low 50 0.00% -£724,014 -10.39% Unviable 28 Service town Low 100 0.00% -£1,376,995 -9.88% Unviable

 Only schemes in high value areas are able to viably support any affordable housing provision (at 25% or thereabouts).

 All other schemes demonstrate, at best, a marginally viable scheme (even with a nil affordable housing provision).

Table 6 – Test 2 Urban settlement viability results

Value No Afford Surplus / No Site % of GDV Viable band units (%) deficit

1 Green infill High 1 0.00% £45,308 21.68% Viable 2 Green infill Medium 1 0.00% £9,846 6.06% Viable 3 Green infill Low 1 0.00% -£6,822 -4.90% Unviable 4 5 units cleared High 5 0.00% £232,319 22.23% Viable 5 5 units cleared Medium 5 0.00% £45,716 5.62% Viable 6 5 units cleared Low 5 0.00% -£47,698 -6.85% Unviable 7 5 units occupied High 5 0.00% £191,384 18.31% Viable 8 5 units occupied Medium 5 0.00% £4,781 0.59% Marginal 9 5 units occupied Low 5 0.00% -£88,730 -12.73% Unviable 10 14 units cleared High 14 28.58% £281,924 9.63% Viable 11 14 units cleared Medium 14 0.00% £19,748 0.87% Marginal 12 14 units cleared Low 14 0.00% -£224,927 -11.53% Unviable 13 14 units occupied High 14 28.58% £167,493 5.72% Viable 14 14 units occupied Medium 14 0.00% -£95,028 -4.18% Unviable 15 14 units occupied Low 14 0.00% -£339,769 -17.42% Unviable 16 50 units cleared High 50 24.00% £905,145 8.66% Viable 17 50 units cleared Medium 50 0.00% -£19,486 -0.24% Marginal 18 50 units cleared Low 50 0.00% -£859,983 -12.34% Unviable 19 50 units occupied High 50 24.00% £457,909 4.38% Viable 20 50 units occupied Medium 50 0.00% -£469,719 -5.78% Unviable 21 50 units occupied Low 50 0.00% -£1,311,877 -18.83% Unviable 22 100 units cleared High 100 25.00% £1,750,721 8.38% Viable 23 100 units cleared Medium 100 0.00% -£7,365 -0.05% Marginal 24 100 units cleared Low 100 0.00% -£1,637,292 -11.75% Unviable 25 100 units occupied High 100 25.00% £666,379 3.19% Viable 26 100 units occupied Medium 100 0.00% -£932,460 -5.74% Unviable 27 100 units occupied Low 100 0.00% -£2,570,578 -18.45% Unviable

 Only schemes in high value locations are able to support an affordable housing provision (at 25% or thereabouts). No other scheme can support an on-site affordable housing provision.

 For single plot and 5 dwellings schemes, the medium value areas show a positive position on viability, albeit on the basis that an affordable housing provision is not applicable. Schemes of a similar size in low value areas are shown as being unviable.

 For schemes comprising 14, 50 or 100 dwellings in medium or low value areas, only developments of cleared brownfield sites in medium value areas show marginally viable outputs. All the rest show an unviable return.

4.4 Test 3 – Estimated costs linked to Policy 83 Flood risk management are removed

4.4.1 Test 3 adopts the same assumptions as Test 1, except for the removal of anticipated costs associated with Council Policy 83, which relates to flood risk mitigation allowances. 4.4.2 Please note, for ease of reference we have split the results for urban extension sites and urban settlement sites into separate tables, as follows:

Table 7 – Test 3 Urban extension viability results

No Afford Surplus / No Site Value band % of GDV Viable units (%) deficit

1 Town & village High 50 24.00% £1,508,209 14.43% Viable 2 Town & village High 100 24.00% £2,821,559 13.50% Viable 3 Doncaster main urban Low 50 0.00% £11,490 0.16% Marginal 4 Doncaster main urban Low 100 0.00% £20,994 0.15% Marginal 5 Doncaster main urban Medium 50 22.00% £230,125 2.83% Viable 6 Doncaster main urban Medium 100 21.00% £429,486 2.64% Viable 7 Doncaster main urban Medium 400 17.50% £1,774,505 2.73% Viable 8 Main town Medium 50 24.00% £232,553 2.86% Viable 9 Main town Medium 100 20.00% £450,903 2.77% Viable 10 Main town Medium 400 18.75% £1,756,186 2.70% Viable 11 Main town Low / med 50 8.00% £188,312 2.49% Viable 12 Main town Low / med 100 8.00% £399,152 2.64% Viable 13 Main town Low / med 400 5.00% £1,614,998 2.67% Viable 14 Main town Low 50 0.00% £11,490 0.16% Marginal 15 Main town Low 100 0.00% £20,994 0.15% Marginal 16 Main town Low 400 0.00% £18,067 0.03% Marginal 17 Main town Low 50 0.00% £11,490 0.16% Marginal 18 Main town Low 100 0.00% £20,994 0.15% Marginal 19 Main town Low 50 0.00% £11,490 0.16% Marginal 20 Main town Low 100 0.00% £20,994 0.15% Marginal 21 Main town Low 400 0.00% £18,067 0.03% Marginal 22 Town, Urban, Village Medium 50 24.00% £451,999 5.56% Viable 23 Town, Urban, Village Medium 100 20.00% £412,374 2.54% Viable 24 Town, Urban, Village Medium 400 18.00% £1,634,178 2.51% Viable 25 Service town Medium 50 22.00% £230,125 2.83% Viable 26 Service town Medium 100 20.00% £450,903 2.77% Viable 27 Service town Low 50 0.00% £11,490 0.16% Marginal 28 Service town Low 100 0.00% £20,994 0.15% Marginal

 50 or 100 dwelling schemes in high value areas are comfortably viable with 25% affordable.

 Schemes in medium value areas all return viable schemes, with an improved affordable housing provision (when compared to Test 1) of circa 18% – 25%

 Schemes in low / medium value areas are viable with an affordable housing provision of 5% - 8%.

 Schemes in low value areas are unable to deliver any affordable housing, however the viability results are improved when compared to Test 1 (with all of the low value site tests showing a marginally viable result, whereas in Test 1 a number of schemes were unviable).

Table 8 – Test 3 Urban settlement viability results

No Afford Surplus / % of No Site Value band Viable units (%) deficit GDV

1 Green infill High 1 0.00% £61,080 29.22% Viable 2 Green infill Medium 1 0.00% £25,618 15.76% Viable 3 Green infill Low 1 0.00% £8,966 6.43% Viable 4 5 units cleared High 5 0.00% £310,749 29.73% Viable 5 5 units cleared Medium 5 0.00% £124,146 15.27% Viable 6 5 units cleared Low 5 0.00% £30,845 4.43% Viable 7 5 units occupied High 5 0.00% £269,814 25.82% Viable 8 5 units occupied Medium 5 0.00% £83,211 10.24% Viable 9 5 units occupied Low 5 0.00% -£10,114 -1.45% Marginal 10 14 units cleared High 14 28.58% £504,552 17.24% Viable 11 14 units cleared Medium 14 21.43% £81,742 3.59% Viable 12 14 units cleared Low 14 0.00% -£4,664 -0.24% Marginal 13 14 units occupied High 14 28.58% £390,235 13.34% Viable 14 14 units occupied Medium 14 7.14% £57,882 2.54% Viable 15 14 units occupied Low 14 0.00% -£119,505 -6.13% Unviable 16 50 units cleared High 50 24.00% £1,646,089 15.75% Viable 17 50 units cleared Medium 50 22.00% £208,834 2.57% Viable 18 50 units cleared Low 50 0.00% -£119,691 -1.72% Unviable 19 50 units occupied High 50 24.00% £1,200,885 11.49% Viable 20 50 units occupied Medium 50 0.00% £212,704 2.62% Viable 21 50 units occupied Low 50 0.00% -£570,271 -8.18% Unviable 22 100 units cleared High 100 25.00% £3,157,447 15.11% Viable 23 100 units cleared Medium 100 22.00% £415,813 2.56% Viable 24 100 units cleared Low 100 0.00% -£212,695 -1.53% Marginal 25 100 units occupied High 100 25.00% £2,080,239 9.95% Viable 26 100 units occupied Medium 100 2.00% £406,217 2.50% Viable 27 100 units occupied Low 100 0.00% -£1,140,091 -8.18% Unviable

 Single plots schemes are viable, regardless of value band. Affordable contributions do not apply.

 5 dwelling scheme are mostly viable. Again, affordable contributions do not apply.

 Schemes in high value areas providing 14, 50 or 100 dwellings are viable, with a 25% provision.

 For cleared, brownfield sites (50 or 100 units) in medium value areas, an affordable provision of circa 20% is demonstrated. For occupied brownfield sites, this reduces to a nominal 0% – 2%.

 For schemes comprising 50 and 100 dwellings in low value areas generally the results show an unviable scheme (even without affordable housing), and at best only marginally viable.

4.5 Test 4 – Sales revenue increased by 5%

4.5.1 Test 4 adopts the same assumptions as Test 1, except the sales revenues are increased by 5%. Please note, the HCA DAT’s sensitivity function does not adjust the level of affordable housing when sales revenues are increased. For the purposes of the analysis, we have therefore shown the surplus / deficit of the scheme based on the same affordable housing levels shown in Test 1.

4.5.2 Please note, for ease of reference we have split the results for urban extension sites and urban settlement sites into separate tables, as follows:

Table 9 – Test 4 Urban extension viability results

Change No Afford Test 1 % % of No Site Value band Viable from Test units (%) of GDV GDV 1 1 Town & village High 50 25.00% 12.65% 15.36% Viable 2.71% 2 Town & village High 100 24.00% 11.82% 14.40% Viable 2.58% 3 Doncaster main urban Low 50 0.00% -2.53% 1.06% Marginal 3.59% 4 Doncaster main urban Low 100 0.00% -2.39% 1.04% Marginal 3.43% 5 Doncaster main urban Medium 50 12.00% 2.88% 6.03% Viable 3.15% 6 Doncaster main urban Medium 100 12.00% 2.61% 5.61% Viable 3.00% 7 Doncaster main urban Medium 400 8.50% 2.87% 5.62% Viable 2.75% 8 Main town Medium 50 12.00% 2.91% 5.98% Viable 3.07% 9 Main town Medium 100 12.00% 2.61% 5.61% Viable 3.00% 10 Main town Medium 400 0.00% 2.74% 5.45% Viable 2.71% 11 Main town Low / med 50 0.00% 2.62% 6.19% Viable 3.57% 12 Main town Low / med 100 0.00% 2.18% 5.59% Viable 3.41% 13 Main town Low / med 400 0.00% 1.71% 4.72% Viable 3.01% 14 Main town Low 50 0.00% -2.53% 1.06% Marginal 3.59% 15 Main town Low 100 0.00% -2.39% 1.04% Marginal 3.43% 16 Main town Low 400 0.00% -2.19% 0.84% Marginal 3.03%

17 Main town Low 50 0.00% -2.53% 1.06% Marginal 3.59% 18 Main town Low 100 0.00% -2.39% 1.04% Marginal 3.43% 19 Main town Low 50 0.00% -2.53% 1.06% Marginal 3.59% 20 Main town Low 100 0.00% -2.39% 1.04% Marginal 3.43% 21 Main town Low 400 0.00% -2.19% 0.84% Marginal 3.03% 22 Town, Urban, Village Medium 50 24.00% 3.26% 5.98% Viable 2.72% 23 Town, Urban, Village Medium 100 13.00% 2.61% 5.61% Viable 3.00% 24 Town, Urban, Village Medium 400 9.75% 2.51% 5.23% Viable 2.72% 25 Service town Medium 50 12.00% 2.88% 6.03% Viable 3.15% 26 Service town Medium 100 13.00% 2.61% 5.61% Viable 3.00% 27 Service town Low 50 0.00% -2.53% 1.06% Marginal 3.59% 28 Service town Low 100 0.00% -2.39% 1.04% Marginal 3.43%

 Increasing the GDV by 5%, as a broad average, improves the overall surplus generated by circa 3% (when expressed as a percentage of GDV). However, only 2 schemes improve from being previously marginally viable to viable (and therefore could provide some affordable housing).

 5 schemes improve from previously being unviable to showing a marginally viable return. However, none have improved sufficiently to provide affordable housing.

 In conclusion, the increase of the GDV by 5% (when build costs remain static) does not have a significant impact on the viability conclusions, as it is not sufficient to change unviable / marginally viable schemes into viable developments that can provide affordable housing.

Table 10 – Test 4 Urban settlement viability results

Change No Afford Test 1 % % of No Site Value band Viable from units (%) of GDV GDV Test 1 1 Green infill High 1 0.00% 27.34% 31.39% Viable 4.05% 2 Green infill Medium 1 0.00% 13.34% 17.39% Viable 4.05% 3 Green infill Low 1 0.00% 3.61% 7.66% Viable 4.05% 4 5 units cleared High 5 0.00% 27.86% 31.88% Viable 4.02% 5 5 units cleared Medium 5 0.00% 12.87% 16.88% Viable 4.01% 6 5 units cleared Low 5 0.00% 1.62% 5.64% Viable 4.02% 7 5 units occupied High 5 0.00% 23.94% 27.96% Viable 4.02% 8 5 units occupied Medium 5 0.00% 7.83% 11.85% Viable 4.02% 9 5 units occupied Low 5 0.00% -4.27% -0.24% Marginal 4.03% 10 14 units cleared High 14 28.58% 15.37% 18.09% Viable 2.72% 11 14 units cleared Medium 14 14.28% 4.13% 7.39% Viable 3.26% 12 14 units cleared Low 14 0.00% -3.05% 0.76% Marginal 3.81% 13 14 units occupied High 14 28.58% 11.47% 14.18% Viable 2.71% 14 14 units occupied Medium 14 0.00% 3.08% 6.89% Viable 3.81%

15 14 units occupied Low 14 0.00% -8.94% -5.12% Unviable 3.82% 16 50 units cleared High 50 24.00% 13.98% 16.70% Viable 2.72% 17 50 units cleared Medium 50 12.00% 2.64% 6.08% Viable 3.44% 18 50 units cleared Low 50 0.00% -4.41% -0.79% Marginal 3.62% 19 50 units occupied High 50 20.00% 9.71% 12.44% Viable 2.73% 20 50 units occupied Medium 50 0.00% 1.01% 4.61% Viable 3.60% 21 50 units occupied Low 50 0.00% -10.88% -7.25% Unviable 3.63% 22 100 units cleared High 100 24.00% 13.42% 16.03% Viable 2.61% 23 100 units cleared Medium 100 12.00% 2.90% 5.93% Viable 3.03% 24 100 units cleared Low 100 0.00% -4.10% -0.61% Marginal 3.49% 25 100 units occupied High 100 25.00% 8.26% 10.84% Viable 2.58% 26 100 units occupied Medium 100 0.00% 0.78% 4.24% Viable 3.46% 27 100 units occupied Low 100 0.00% -10.77% -7.25% Unviable 3.52%

 Again, increasing the GDV by 5%, as a broad average, improves the overall surplus generated by circa 3% (when expressed as a percentage of GDV). However, only 3 schemes improve from being previously marginally viable to viable.

 4 schemes improve from previously being unviable to showing a marginally viable return. However, none have improved sufficiently to provide affordable housing.

4.6 Test 5 – Starter Homes

4.6.1 Test 5 adopts the same assumptions as Test 1, except the affordable houses (if demonstrated) are replaced (in full) by Starter Homes.

4.6.2 Please note, for ease of reference we have split the results for urban extension sites and urban settlement sites into separate tables, as follows:

Table 11 – Test 5 Urban extension viability results

Starter No Surplus / % of No Site Value band Homes Viable units deficit GDV (%) 1 Town & village High 50 24.00% £1,919,098 18.36% Viable 2 Town & village High 100 25.00% £3,596,703 17.21% Viable 3 Doncaster main urban Low 50 0.00% -£176,472 -2.53% Marginal 4 Doncaster main urban Low 100 0.00% -£332,462 -2.39% Marginal 5 Doncaster main urban Medium 50 24.00% £384,486 4.73% Viable 6 Doncaster main urban Medium 100 25.00% £696,532 4.28% Viable 7 Doncaster main urban Medium 400 25.00% £2,180,455 3.35% Viable 8 Main town Medium 50 24.00% £408,850 5.03% Viable 9 Main town Medium 100 25.00% £696,532 4.28% Viable 10 Main town Medium 400 25.00% £2,180,455 3.35% Viable 11 Main town Low / medium 50 0.00% £197,786 2.62% Viable 12 Main town Low / medium 100 0.00% £329,602 2.18% Marginal 13 Main town Low / medium 400 0.00% £1,034,489 1.71% Marginal 14 Main town Low 50 0.00% -£176,472 -2.53% Unviable 15 Main town Low 100 0.00% -£332,462 -2.39% Marginal 16 Main town Low 400 0.00% -£1,218,617 -2.19% Marginal 17 Main town Low 50 0.00% -£176,472 -2.53% Unviable 18 Main town Low 100 0.00% -£332,462 -2.39% Marginal 19 Main town Low 50 0.00% -£176,472 -2.53% Unviable 20 Main town Low 100 0.00% -£332,462 -2.39% Marginal 21 Main town Low 400 0.00% -£1,218,617 -2.19% Marginal 22 Town, Urban, Village Medium 50 24.00% £797,534 9.81% Viable 23 Town, Urban, Village Medium 100 25.00% £696,532 4.28% Viable 24 Town, Urban, Village Medium 400 25.00% £2,180,455 3.35% Viable 25 Service town Medium 50 24.00% £478,589 5.89% Viable 26 Service town Medium 100 25.00% £877,046 5.39% Viable 27 Service town Low 50 0.00% -£176,472 -2.53% Unviable 28 Service town Low 100 0.00% -£332,462 -2.39% Marginal

 With the inclusion of starter homes, all developments in medium and high value areas are able to support an affordable housing provision of 25%.

 The adoption of starter homes does not impact on sites in low / low to medium value areas, which are all still unable to provide any affordable housing.

Table 12 – Test 5 Urban settlement viability results

Starter No Surplus / % of No Site Value band Homes Viable units deficit GDV (%) 1 Green infill High 1 0.00% £57,145 27.34% Viable 2 Green infill Medium 1 0.00% £21,683 13.34% Viable 3 Green infill Low 1 0.00% £5,031 3.61% Viable 4 5 units cleared High 5 0.00% £291,185 27.86% Viable 5 5 units cleared Medium 5 0.00% £104,582 12.87% Viable 6 5 units cleared Low 5 0.00% £11,280 1.62% Marginal 7 5 units occupied High 5 0.00% £250,250 23.94% Viable 8 5 units occupied Medium 5 0.00% £63,647 7.83% Viable 9 5 units occupied Low 5 0.00% -£29,725 -4.27% Unviable 10 14 units cleared High 14 28.57% £662,413 22.64% Viable 11 14 units cleared Medium 14 28.57% £168,588 7.41% Viable 12 14 units cleared Low 14 0.00% -£59,591 -3.05% Unviable 13 14 units occupied High 14 28.57% £546,676 18.68% Viable 14 14 units occupied Medium 14 0.00% £70,076 3.08% Viable 15 14 units occupied Low 14 0.00% -£174,433 -8.94% Unviable 16 50 units cleared High 50 26.00% £2,147,287 20.55% Viable 17 50 units cleared Medium 50 26.00% £475,546 5.85% Viable 18 50 units cleared Low 50 0.00% -£306,987 -4.41% Unviable 19 50 units occupied High 50 24.00% £1,705,717 16.32% Viable 20 50 units occupied Medium 50 0.00% £82,049 1.01% Marginal 21 50 units occupied Low 50 0.00% -£757,883 -10.88% Unviable 22 100 units cleared High 100 25.00% £4,110,075 19.66% Viable 23 100 units cleared Medium 100 25.00% £742,715 4.57% Viable 24 100 units cleared Low 100 0.00% -£571,781 -4.10% Unviable 25 100 units occupied High 100 25.00% £3,209,329 15.35% Viable 26 100 units occupied Medium 100 0.00% £127,105 0.78% Marginal 27 100 units occupied Low 100 0.00% -£1,500,747 -10.77% Unviable

 The results for sites 1 – 9 (where affordable housing is not applicable) remain unchanged.

 For cleared, brownfield sites those located in medium and high value areas are able to support a 25% affordable provision. Sites in low value areas are unable to support any starter homes.

 For occupied, brownfield sites only developments in high value areas can support a 25% provision. Sites located in medium and low value areas are unable to provide any starter homes.

4.7 Test 6 – All draft Council planning policies removed

4.7.1 Test 6 adopts the same assumptions as Test 1, except the S106 obligations (including affordable housing) are removed. The results therefore simply show whether the scheme is viable, marginally viable or unviable with no S106 obligations.

4.7.2 Please note, for ease of reference we have split the results for urban extension sites and urban settlement sites into separate tables, as follows:

Table 13 – Test 6 Urban extension viability results

No Surplus / No Site Value band % of GDV Viable units deficit

1 Town & village High 50 £2,181,000 20.87% Viable 2 Town & village High 100 £4,117,141 19.70% Viable 3 Doncaster main urban Low 50 -£51,472 -0.74% Marginal 4 Doncaster main urban Low 100 -£95,632 -0.69% Marginal 5 Doncaster main urban Medium 50 £652,620 8.03% Viable 6 Doncaster main urban Medium 100 £1,224,493 7.53% Viable 7 Doncaster main urban Medium 400 £4,092,697 6.29% Viable 8 Main town Medium 50 £676,945 8.33% Viable 9 Main town Medium 100 £1,224,493 7.53% Viable 10 Main town Medium 400 £4,092,697 6.29% Viable 11 Main town Low / medium 50 £322,143 4.27% Viable 12 Main town Low / medium 100 £565,272 3.74% Viable 13 Main town Low / medium 400 £1,853,252 3.07% Viable 14 Main town Low 50 -£51,472 -0.74% Marginal 15 Main town Low 100 £256,918 1.84% Marginal 16 Main town Low 400 -£394,051 -0.71% Marginal 17 Main town Low 50 -£51,472 -0.74% Marginal 18 Main town Low 100 -£95,632 -0.69% Marginal 19 Main town Low 50 -£51,472 -0.74% Marginal 20 Main town Low 100 -£95,632 -0.69% Marginal 21 Main town Low 400 -£394,051 -0.71% Marginal 22 Town, Urban, Village Medium 50 £652,620 8.03% Viable 23 Town, Urban, Village Medium 100 £1,224,493 7.53% Viable 24 Town, Urban, Village Medium 400 £4,092,697 6.29% Viable 25 Service town Medium 50 £652,620 8.03% Viable 26 Service town Medium 100 £1,224,493 7.53% Viable 27 Service town Low 50 -£51,472 -0.74% Marginal 28 Service town Low 100 -£95,632 -0.69% Marginal

 All schemes in low / medium, medium and high value areas show a viable scheme when all draft Council Planning Policies are removed.

 Schemes in low value areas are still only marginally viable, even if draft Council policies are removed.

Table 14 – Test 6 Urban settlement viability results

No Surplus / No Site Value band % of GDV Viable units deficit

1 Green infill High 1 £57,145 27.34% Viable 2 Green infill Medium 1 £21,683 13.34% Viable 3 Green infill Low 1 £5,031 3.61% Viable 4 5 units cleared High 5 £291,185 27.86% Viable 5 5 units cleared Medium 5 £104,582 12.87% Viable 6 5 units cleared Low 5 £11,280 1.62% Marginal 7 5 units occupied High 5 £250,250 23.94% Viable 8 5 units occupied Medium 5 £63,647 7.83% Viable 9 5 units occupied Low 5 -£29,725 -4.27% Unviable 10 14 units cleared High 14 £693,014 23.68% Viable 11 14 units cleared Medium 14 £198,250 8.71% Viable 12 14 units cleared Low 14 -£49,607 -2.54% Unviable 13 14 units occupied High 14 £578,760 19.78% Viable 14 14 units occupied Medium 14 £83,765 3.68% Viable 15 14 units occupied Low 14 -£164,448 -8.43% Unviable 16 50 units cleared High 50 £2,319,697 22.20% Viable 17 50 units cleared Medium 50 £655,715 8.07% Viable 18 50 units cleared Low 50 -£181,324 -2.60% Unviable 19 50 units occupied High 50 £1,875,015 17.94% Viable 20 50 units occupied Medium 50 £207,347 2.55% Viable 21 50 units occupied Low 50 -£632,043 -9.07% Unviable 22 100 units cleared High 100 £4,459,872 21.34% Viable 23 100 units cleared Medium 100 £1,276,449 7.85% Viable 24 100 units cleared Low 100 -£330,298 -2.37% Marginal 25 100 units occupied High 100 £3,561,149 17.04% Viable 26 100 units occupied Medium 100 £366,352 2.25% Marginal 27 100 units occupied Low 100 -£1,258,238 -9.03% Unviable

 All but one of the scenarios in medium or high value areas show a viable scheme.

 There is still viability pressure on schemes in low value areas, even if all of the draft Council policies are removed.

4.8 Conclusions

4.8.1 Schemes in low value areas, regardless of size and scenario test, were shown to be unable to support any level of affordable housing.

4.8.2 Schemes in high value areas, regardless of size and scenario test, were shown to comfortably support an affordable housing provision of 25% (whether this includes starter homes or not).

4.8.3 Schemes within medium value areas showed a fluctuation in the level of affordable housing provision that could be viably supported, depending on the size of the scheme, nature of the land and the specific scenario test (ranging from 0% – 25%). However, taking into account the various tests, we would suggest circa 10% – 15% as being a fair representation as to the average level of affordable housing scheme in medium value areas could support.

4.8.4 Taking into account the above, we conclude that a policy ‘starting’ point of 25% appears out of kilter for most schemes likely to be brought forward across Doncaster (which will be located in low and medium value areas).

4.8.5 For schemes within low to medium value areas we would suggest 15% as being a more appropriate aspiration (with 25% retained for higher value locations).

5. NON-RESIDENTIAL – VIABILITY ASSUMPTIONS

5.1 Introduction

5.1.1 This section of the report sets out the assumptions and appraisal inputs used in the appraisal testing of non-residential development sites.

5.2 Site Types

5.2.1 In line with the residential testing, and having discussed testing options through the Stakeholder Workshop and the subsequent questionnaire, we again decided to adopt hypothetical site testing for the assessment of non-residential development.

5.2.2 Having discussed the nature of the hypothetical sites with the Council and the stakeholders, and having researched the local market dynamics, we have adopted the following hypothetical development sites for the purposes of the appraisal testing:

- Strategic warehouse, total gross internal area 50,000 sq. m. - Non-strategic warehouse, total gross internal are 20,000 sq. m - Industrial and manufacturing, total gross internal area 10,000 sq. m - Industrial and manufacturing, total gross internal area 5,000 sq. m - Industrial and manufacturing, total gross internal area 2,000 sq. m - Supermarket, total gross internal area 4,000 sq. m, greenfield site - Supermarket, total gross internal area 4,000 sq. m, cleared brownfield site - Supermarket, total gross internal area 1,500 sq. m, greenfield site - Supermarket, total gross internal area 1,500 sq. m, cleared brownfield site - Office, total net internal area 500 sq. m - Office, total net internal area 200 sq. m - Town centre retail, net internal area 150 sq. m - Hotel (Travel Lodge, Premier Inn etc.), gross internal area 3,000 sq. m

5.3 Methodology

5.3.1 In undertaking our assessment we have again adopted an appraisal approach (outlined above in 2.4). This involves identifying the value of the completed scheme (gross development value), and from this netting of the costs of the development. In line with the residential testing we have adopted a fixed level of profit within each appraisal, as well as a fixed land value (the Threshold Land Value). If the residual amount left over is a surplus which exceeds 2.5% of the gross development value, then the scheme is deemed to be viable. If the scheme produces a deficit, which falls below – (minus) 2.5% of the gross development value the scheme is considered to be unviable. If the residual amount falls within these two parameters the scheme is considered to be marginally viable.

5.4 Evidence

5.4.1 For non-residential (or commercial) property, where direct capital evidence is less abundant than the residential market, a slightly different approach can be adopted. This involves identifying what is considered to be the annual Market Rent for the property. Once this is established, this is then capitalised using an investment yield (essentially a market multiplier) to arrive at a capital value. For example, for a 1,000 sq. m industrial unit the Market Rent identified, based on comparable evidence, may equate to say £50 per sq. m. This therefore gives a Market Rent (essentially a gross annual income) of £50,000 per annum. To arrive at a capital value an appropriate investment yield is identified, again established by considering comparable evidence. If that gross initial yield is say 8% (12.5 as a market multiplier) the ‘end value’ of the industrial unit is 12.5 multiplied by £50,000, which equals £625,000. This is then inputted into the viability appraisal as being the ‘GDV’ of the site.

5.4.2 As part of HMRC / VOA we have detailed information on all commercial transactions (sales and lettings). This is used primarily in the assessment of business rates for non-residential property, but can be equally used in the assessment of Market Rents and investment yields for commercial property. For the purposes of this study we have subsequently drawn upon this information to identify rental and capital values for the various property types. We have also used public access websites (available through licenses), including Costar SUITE and EGi to ensure a robust evidence base.

5.4.3 We have identified the following rental and yield ranges for the different property types, which vary due to locational factors:

Table 15 – Non-residential Market Rent and yield ranges

Development Type Unit size Rental range Gross Initial (sq. m) (£ per sq. m) Yield range (%) Strategic warehouse 50,000 £43 - £51 5.75% – 8.4% Non-strategic warehouse 20,000 £40 - £51 5.75% – 8.4% Industrial & manufacturing 10,000 £30 - £62 6% – 8.4% Industrial & manufacturing 5,000 £32 - £48 6% – 8.4% Industrial & manufacturing 2,000 £20 - £53 6% – 8.4% Supermarket 4,000 £135 - £230 5.5% – 8% Supermarket 4,000 £135 - £230 5.5% – 8% Supermarket 1,500 £115 - £170 5.5% – 6.5% Supermarket 1,500 £115 - £170 5.5% – 6.5% Office 500 £53 - £129 7%– 11% Office 200 £59 - £183 7%– 11% Town centre retail 150 £61 - £375 6% – 8.5% Hotel 3,000 £59 - £87 5.5% – 6.5%

Please note, the above ranges reflect comparables from properties broadly similar in size to the non-residential site typologies. The data is also derived from non-residential property across the Doncaster Borough, and therefore reflects market fluctuations based on locational factors.

5.4.4 From the range of identified evidence we have adopted appropriate rental rates and yields for the various hypothetical sites tests.

5.4.5 Having identified appropriate GDV’s for the various development types we have then considered other key appraisal inputs, drawing upon other non-residential schemes we have appraised:

Table 16 – Non-residential key appraisal inputs

Appraisal input Comment

Threshold Land For the purposes of the testing we have assumed Value (“TLV”) cleared brownfield sites, with a TLV equivalent to £185,325 per Ha (£75k / acre). The only exception is supermarket site types, where we have assumed a scenario both brownfield (cleared) and separately greenfield, the latter at £271,810 per Ha (£110k / acre) Basic build cost For the residential sites we were able to draw on a variety of information sources regarding build costs. For non-residential sites that information is more limited. We have subsequently placed a greater reliance on data from the BCIS, which suggests the following:

- Strategic warehouse £446 per sq. m - Non-strategic warehouse £446 per sq. m - Industrial (2,000 to 10,000 sq. m) £446 per sq. m - Supermarket (small and large) £951 per sq. m - Office 500 sq. m £1,228 per sq. m - Office 200 sq. m £1,052 per sq. m - Town centre retail £644 per sq. m - Hotel £1,405 per sq. m

Externals This will vary depending on the nature of the scheme. For example, a small retail unit ‘in town’ will require significantly less space than say a supermarket, which would require a large amount of car parking. For the purposes of our assessment we consider a range of 5% – 10% to be appropriate, again dependent on the specific scheme.

Contingency From our experience, we consider 3% to be reasonable for the assessment of non-residential development sites.

Professional fees Given that non-residential schemes tend to require more individual thought and design than standard housing products delivered by volume house building, we have adopted professional fees of 8%.

Draft Policy 23: This draft policy is intended to apply for larger Promoting development sites. Having discussed this with the sustainable Council we have made an allowance of £100,000 per transport within new Ha for the strategic warehouse, non-strategic developments warehouse, all industrial and supermarket site types.

Draft Policy 46: Having discussed this with the Council we have Valuing biodiversity made an allowance of £10,000 per Ha for each site and geodiversity type.

Draft Policy 70: Major proposals are expected to meet BREEAM Design of non- rating of ‘very good’ and 10% renewables. This is residential, assumed to have been allowed for within the BCIS commercial and rate applied. employment developments

Draft Policy 80: Considered applicable to brownfield sites only. Contamination and Having discussed this with the Council a rate of unstable land £35,000 per Ha is considered reasonable for inclusion in the appraisal testing.

Draft Policy 83: For the purposes of the initial testing, and having Flood risk discussed this with the Council, each site type has management been assumed to require flood risk mitigation measures, at a cost equivalent to £140,000 per Ha. Agency / legal fees Based on our experience, and other non-residential schemes we have appraised, for the letting of the accommodation we have adopted a 10% (of the annual Market Rent) agent letting fee together with a 5% legal fee. For the investment sale of the accommodation when fully let we have allowed a 1% agency fee (of the sale price agreed) and a 0.5% legal fee.

Finance We have adopted a debit interest rate of 6%.

Developer’s Profit For non-residential accommodation the level of profit is often gauged against the cost of delivering the scheme (rather than the GDV, which tends to be the convention with residential property). Furthermore, the level of profit on cost will vary dependent on whether the scheme is speculative or not. In other words, if the property has been ‘pre-let’ (i.e. a tenant has signed up to a tenancy) before the construction works have commenced, then the risk associated with the scheme is considered to be significantly lower. From our experience, for a ‘pre-let’ development a profit margin of 12.5% of cost is generally acceptable. If, though, the scheme is entirely speculative (i.e. the tenants will be identified after the accommodation has been constructed) then the level of profit on cost, in our experience,

increases to 15% on cost. This reflects the potential for voids that would be incurred whilst a tenant is identified.

For the purposes of this assessment, we have assumed that the hotel, strategic warehouse, non- strategic warehouse and supermarkets would be built on a ‘pre-let’ basis, and have therefore applied a profit margin of 12.5% on cost. For all other development types we have assumed the developments would be speculative and have subsequently increased the profit margin to 15% on cost.

5.5 Scenario testing / sensitivity analysis

5.5.1 As indicated above, the residual method used in viability testing is not without its flaws, owing to the wide variety of inputs used in the approach (all of which could be subject to variance, which can potentially lead to a wide range of outputs in the appraisals making it more difficult to reach robust conclusions from the data). Recognising this limitation, the PPG and the relevant professional guidance for viability assessments advocates adopting scenario testing / sensitivity analysis approach to Whole Plan viability studies.

5.5.2 At the stakeholder workshop, and from the completed questionnaires received, there was a general agreement that scenario testing / sensitivity analysis should consider fluctuations in sales values and build costs. There was also a general view that testing should be focused on likely fluctuations in inputs and not test changes which were unlikely to occur.

5.5.3 Having considered the above we have adopted the following approach to scenario testing / sensitivity analysis:

Test 1 Adopts all applicable draft Council policies

Test 2 Excludes all draft Council policies

Test 3 Excludes flood risk mitigation costs only

5.5.4 As indicated above, as part of the testing process it is important to set a ‘point’ in which a scheme is deemed to be viable or unviable. However, based on our experience, viability is not necessarily ‘black and white’ and often reference is made to schemes being ‘marginally viable’ (i.e. the return generated from the scheme may be sufficient to attract some developers, but not necessarily all). There is no specific guidance on how this should be considered within Whole Plan testing, however the PPG is clear that Council policies should not be set at ‘extremes’ of viability.

5.5.5 In this regard, and based on our experience, we have set the following parameters for determining whether a scheme is viable or not:

Viable A scheme is deemed to be viable if the appraisal generates a surplus which is equivalent to 2.50% (or higher) of the sales revenue. Schemes which meet this criteria are marked in green.

6. NON-RESIDENTIAL – VIABILITY ASSUMPTIONS

6.1 Introduction

6.1.1 In undertaking our appraisals we have used ARGUS. This is an industry approved, cash flow toolkit, designed primarily for commercial development. The appraisal assumptions are inputted into the program, with the results calculated through a cash flow. The program requires a fixed land value to be inputted (in this case the identified TLV), as well as a fixed profit margin. The toolkit then shows either a surplus or a deficit, based on these parameters.

6.1.2 Whether the scheme is deemed to be viable, marginally viable, or unviable depends on the level of surplus / deficit generated by the appraisal.

6.2 Test 1 – Council policies

6.2.1 This tests each commercial scenario site type on the assumption that all draft Council policies are included within the assessment.

Table 17 – Test 1 Council policy appraisal results

Green Land Flood Surplus / Type Transport infra remed risk deficit % of GDV Viable Hotel £0 £1,000 £3,500 £14,000 £454,327 6.76% Viable Strategic Warehouse £600,000 £60,000 £210,000 £840,000 £4,471,192 10.66% Viable Supermarket - green £180,000 £18,000 £0 £252,000 £4,342,576 32.47% Viable Supermarket - brown £180,000 £18,000 £63,000 £252,000 £4,440,741 33.21% Viable Supermarket - green £0 £5,000 £0 £70,000 £691,480 19.81% Viable Supermarket - brown £0 £5,000 £17,500 £70,000 £718,747 20.60% Viable Industrial £150,000 £15,000 £52,500 £210,000 -£1,961,875 -37.20% Unviable Industrial £75,000 £7,500 £26,250 £105,000 -£980,937 -37.20% Unviable Industrial £35,000 £3,500 £12,250 £49,000 -£462,277 -41.64% Unviable Large office £0 £1,000 £3,500 £14,000 -£539,273 -102.40% Unviable Small office £0 £1,000 £3,500 £14,000 -£178,625 -71.45% Unviable Non-strategic warehouse £300,000 £30,000 £105,000 £420,000 £61,376 0.41% Marginal Town centre shop £0 £170 £595 £2,380 £21,206 6.46% Viable

 7 of the 13 commercial site types tested return a viable result. This includes all iterations of supermarket development, a hotel scheme, a strategic warehouse scheme and a town centre shop. To the most part these are considered to be comfortably viable, and would still produce a viable results even if some costs were to increase marginally.

 5 of the remaining schemes produced an unviable result. It is stressed that the losses generated by these schemes were significant (and include industrial and office development).

 The only marginally viable scheme was non-strategic warehousing.

6.3 Test 2 – Excludes all Council policies

6.3.1 This tests each commercial scenario site type on the assumption that all draft Council policies are excluded within the assessment.

Table 18 – Test 2 No Council policy appraisal results

Surplus / Type deficit GDV % of GDV Viable Hotel £472,118 £6,723,748 7.02% Viable Strategic Warehouse £6,076,452 £41,936,479 14.49% Viable Supermarket - green £4,764,676 £13,373,546 35.63% Viable Supermarket - brown £4,920,349 £13,373,546 36.79% Viable Supermarket - green £763,650 £3,489,875 21.88% Viable Supermarket - brown £806,893 £3,489,875 23.12% Viable Industrial -£1,531,391 £5,273,658 -29.04% Unviable Industrial -£765,696 £2,636,829 -29.04% Unviable Industrial -£361,504 £1,110,244 -32.56% Unviable Large office -£520,259 £526,620 -98.79% Unviable Small office -£159,630 £250,000 -63.85% Unviable Non-strategic warehouse £854,639 £14,801,110 5.77% Viable Town centre shop £24,249 £328,312 7.39% Viable

 The non-strategic warehouse scheme changes from being previously unviable (when the Council policies were applied), to be being viable (when the Council policies were removed). For this particular development type the level of Council policies is therefore a key determinant of viability.

 The 5 commercial site types that were unviable when the Council policies were applied (industrial and office) remain unviable when the Council policies are removed. In this respect, the level of Council policy is not a determining factor in viability, because the schemes remain unviable regardless of the level of Council policies. The lack of viability is therefore down to market factors (primarily that the cost of building industrial and office schemes significantly outweighs the ‘end values’ achievable).

 The remaining site types remain comfortably viable.

6.4 Test 3 – Excludes all Council policies except for flood risk mitigation costs

6.4.1 This tests each commercial scenario site type on the assumption that all draft Council policies are excluded within the assessment, except for flood risk mitigation costs.

Table 19 – Test 3 No Council policy (except flood risk mitigation) appraisal results

Surplus / Type Flood risk deficit GDV % of GDV Viable Hotel £14,000 £458,495 £6,723,748 6.82% Viable Strategic Warehouse £840,000 £5,271,637 £41,936,479 12.57% Viable Supermarket - green £252,000 £4,523,822 £13,373,546 33.83% Viable Supermarket - brown £252,000 £4,679,495 £13,373,546 34.99% Viable Supermarket - green £70,000 £696,182 £3,489,875 19.95% Viable Supermarket - brown £70,000 £739,424 £3,489,875 21.19% Viable Industrial £210,000 -£1,748,660 £5,273,658 -33.16% Unviable Industrial £105,000 -£874,330 £2,636,829 -33.16% Unviable Industrial £49,000 -£412,526 £1,110,244 -37.16% Unviable Large office £14,000 -£534,862 £526,620 -101.57% Unviable Small office £14,000 -£174,213 £250,000 -69.69% Unviable Non-strategic warehouse £420,000 £452,236 £14,801,110 3.06% Viable Town centre shop £2,380 £21,915 £328,312 6.68% Viable

 Again, the non-strategic warehouse scheme changes from being previously unviable (when the Council policies were applied), to be being viable (when just the flood mitigation costs are applied).

 The 5 commercial site types that were unviable when the Council policies were applied (industrial and office) remain unviable when only the flood mitigation costs are included. Again, in this respect, the level of Council policy is not a determining factor in viability.

 The remaining site types remain comfortably viable.

6.5 Conclusions

6.5.1 Based on the above results, we conclude that the draft Council policies would not serve to undermine viability for the following site types:

- Hotel - Supermarket (regardless of size or site type) - Strategic warehouse - Town centre shop

6.5.2 However, there is currently a fundamental lack of viability for industrial and office development sites across the Borough. This is to the extent that, even if all Council policies are removed, the schemes are still likely to face significant viability pressures owing to macro market forces (i.e. the ‘end values’ in building industrial and office schemes are not sufficient to outweigh the costs of development).

6.5.3 Based on the testing undertaken, the only scheme that would have a significant benefit from a removal of Council policies would be non-strategic warehousing. When the Council policies are applied this returns only a marginally viable scheme. Reducing these policies therefore serves to create a viable project.

7. FINAL COMMENTS

7.1 The assessment of viability has been undertaken in accordance with the requirements the NPPF / PPG, adopting a collaborative approach where we have actively sought to engage with stakeholders.

7.2 The approach has been the same for testing residential and non-residential development sites; assessing hypothetical sites through a residual appraisal approach. Assumptions within these appraisals has been based on robust evidence, which has been presented to stakeholders inviting challenge.

7.3 In assessing a sites viability status we have adopted a 3 tier approach; concluding that each site is either viable, marginally viable or unviable (in accordance with a pre- determined criteria based on the surplus / deficit as a percentage of the overall scheme revenue).

7.4 It is acknowledged that the residual approach has its limitations (principally that the wide level of inputs within an appraisal can mean that small changes to some or many of these inputs can significantly impact on the overall appraisal outcome). Recognising these limitations, and in accordance with viability guidance, as well as feedback from stakeholders, we have subsequently used sensitivity analysis, which involves running various scenario tests, varying certain key inputs to examine the impact these have on the overall conclusions (such as sales revenue, build costs, Council policies etc.).

7.5 For the residential testing, based on the sensitivity analysis approach, we have concluded that the policy aspiration of 25% is appropriate for schemes located in high value areas, as well as the other Council policy asks (including flood risk mitigation works, education contributions, highways etc.). However, for schemes in medium to low value areas we believe that this aspirational figure is unrealistic. Instead, and as a ‘starting point’, we believe a 15% affordable housing provision would be more appropriate (albeit viability will still need to be determined on a site by site basis), in addition to the other Council policy asks. Please note, in arriving at this conclusion we have factored in the forthcoming introduction of Starter Homes, which is likely to have a positive impact (in viability terms) on residential schemes located in medium value areas.

7.6 For the commercial testing, again based on the sensitivity analysis approach, we have concluded that hotels, strategic warehousing, all supermarket types and town centre shops are showing viable schemes, even when the Council’s draft policies are applied.

7.7 However, for industrial and office schemes, due to current difficult market conditions, the appraisals are showing unviable projects returning significant deficits (even when the Council’s policies are removed). In this regard, even if the Council removes its policies, the current market forces ensure that industrial and office schemes are likely to remain unviable, and therefore are likely to face pressure on deliverability.

7.8 The only commercial scheme where the outcome is marginally viable is non-strategic warehousing. The application of the full Council policies means that schemes of this nature are likely to only be marginally viable. Viability is subsequently improved if these policy asks are reduced.

David Newham RICS Registered Valuer, Principal Surveyor DVS Draft report and study August 2016 -

Cecilia Reed MRICS RICS Registered Valuer, Sector Leader Viability DVS Final report June 2017

Appendix 1

Doncaster Housing Market Review

APPENDIX 1 HOUSING MARKET ANALYSIS - LOCAL MARKET CONDITIONS

1.0 Introduction

1.1 This section provides an assessment of local market conditions and the evidence base for the assumptions on house prices used in our financial appraisals.

1.2 We have considered average values across the Doncaster Borough. It is important to stress that a series of factors will influence values and that, although development schemes do have similarities, every site is unique. Consequently, whilst market conditions in general will broadly reflect national economic circumstances and local supply/demand factors, within an area there will be particular localities and site-specific factors that generate different values and costs. The range of sites tested in this study allows assessment of viability across varying localities for this reason.

1.3 The comments below relate to prevailing market conditions at the valuation date (August 2016). It should be stressed that values fluctuate and that assessments of viability will alter over relatively short periods of time.

2.0 Market Overview

2.1 In recent years there has been a general feeling of improvement in the residential market across the UK. This is supported by anecdotal evidence from both estate agents and house builders, who in some cases have indicated that demand levels have returned to ‘pre-crash’ levels (i.e. pre 2008).

2.2 That said, in certain areas there have been signs of values levelling / contracting during the last 3 – 6 months. The EU Referendum decision in June 2016 has undoubtedly created uncertainty across the property market and may be a factor in this short term trend. However, the medium term impact of the decision is still as yet unclear, therefore for the purposes of this study we are unable robustly conclude whether the impact will be positive or negative. For this reason we have not made a specific allowance in house values (up or down) to reflect the EU Referendum decision, but have instead relied purely on existing sales evidence.

2.3 As a general comment, there is still considered to be an under-supply of housing across country, and therefore strong demand (particularly from first time buyers) will continue to underpin house price growth. Furthermore, the Government’s Help to Buy scheme has been extended until 2020 (which is seen as major benefit to first time buyers looking to purchase new build homes). Equally, the introduction of the Housing and Planning Act 2016 includes various initiatives designed to increase house delivery, and as part of this the introduction of Starter Homes (available at an 80% discount of Market Value) will help first time buyers enter the market.

3.0 Postcode Areas and New Build Sales Evidence

3.1 As part of our research we have looked at average value trends across the different postcode locations of Doncaster Borough.

3.2 Furthermore, as part of HMRC / VOA we have access to detailed information on all residential transactions (submitted through Stamp Duty Land Tax returns), as well as additional details held on individual properties used in determining Council Tax bands. For the purposes of this study we have subsequently looked to identify average values (on a per sq m basis) of dwellings within modern residential estates. We have limited the evidence to estates that fall within (or at least close to) the boundaries of Doncaster Borough and also transactions which have occurred since January 2014. Please note the average sizes given reflect the approximate net sales area, as defined by the RICS Property Measurement 1st Edition:

3.3 For ease of reference we have presented the evidence in postcode areas. It is acknowledged that within postcode areas there will be fluctuations in values achievable, which has been taken into consideration in our assessment.

DN1

3.4 The comprises of the central area of Doncaster town plus an area to the north, broadly covering dwellings around the A18 to the south of the town centre, up to the A19 / A630 to the north.

3.5 In terms of residential values, at the current time the average house value within DH1 for both new build and second hand sales equates to £95,326 (data taken from Zoopla.co.uk). This is a slight fall of circa -0.74% during the last 12 months.

3.6 However, the general average can be a little misleading as it depends on that house types are available in a location (if there is a high proportion of terraced houses, for example, this can artificially decrease the average figure). In this respect, it is also useful to look at the average values of each particular house type (for both new build and secondary homes). In DN1 these are currently showing (data from Zoopla.co.uk) the following averages:

Detached - £285,398 (£1,830 per sq m) Semi-detached - £148,392 (£1,346 per sq m) Terrace - £83,663 (£1,195 per sq m) Flats - £88,934

3.7 In DN1 we note new build house sales achieved at the following:

- Muse Developments, The Gables: close to Doncaster town centre. Modern design 1, 2 and 3 bed housing. 6 x 1/2 bed semi-detached houses sold in 2014 and 2015, at an average of £1,780 per sq m. 35 x 1/2 bed terraced houses sold between Apr 14 and Mar 16, achieving an average of circa £1,805 per sq m. For the terraced houses, there was an average increase in price of circa 1.7% between 2015 and 2016.

3.8 The Muse Developments scheme suggests that new build dwellings will command a premium above the average for the locality (which includes second hand house sales). For example, the average across DN1 for semi-detached houses currently stands at £1,346 per sq m, compared with £1,780 per sq m achieved by Muse Developments (uplift of circa 25%). Further the evidence also contradicts the general trend of falling house values during the last 12 months, as in fact house prices within the scheme have increased in 2016. However, it is acknowledged that this is only one development, and could reflect the nature of the location, product etc, rather than being a tangible sign of a general market trend.

DN2

3.9 This broadly covers an area to the north east of Doncaster town centre up to Edenthorpe, comprising houses around the A630, A18, Thorne Road, Barnby Dun Road and Armthorpe Road.

3.10 The average house value within DN2 for both new build and second hand sales equates to £122,666 (data taken from Zoopla.co.uk). This is a fall of circa -3.61% during the last 12 months.

3.11 More specifically, the average values for each house type in DN2 are currently showing the following (data from Zoopla.co.uk):

Detached - £192,332 (£1,453 per sq m) Semi-detached - £120,991 (£1,335 per sq m) Terrace - £101,160 (£1,055 per sq m) Flats - £105,381 (£1,733 per sq m)

3.12 In DN2 we note new build house sales achieved at the following:

- Johal Homes, Golden Fleece Court: small scheme located in the Wheatley Hills area of Doncaster, located within an ex-Council owned estate. 7 x 3 bed detached houses sold in 2014 and 2015, at an average of £1,422 per sq m.

3.13 Given the limited data in this postcode area it is difficult to draw any firm conclusions in terms of average prices achieved.

DN3

3.14 This broadly covers a large area to the north east and east of Doncaster town centre, and includes part of the M18 corridor, Armthorpe, Branton, Edenthorpe, Kirk Sandall and Barnby Dun.

3.15 The average house value within DN3 for both new build and second hand sales equates to £147,284 (data taken from Zoopla.co.uk). This is a fall of circa -3.49% during the last 12 months.

3.16 More specifically, the average values for each house type in DN3 are currently showing the following (data from Zoopla.co.uk):

Detached - £195,641 (£1,744 per sq m) Semi-detached - £122,001 (£1,410 per sq m) Terrace - £107,939 (£1,335 per sq m)

3.17 In DN3 we note new build house sales achieved at the following:

- Harron Homes, Beaumont Gardens: located in Edenthorpe, comprising mainly detached 2 storey houses. Between April 2014 and Feb 2016 47 x 4 bed detached houses sold, at an average of £2,052 per sq m. There was an average increase in price of circa 2.6% between 2015 and 2016 (although the data only includes a small number of sales in 2016).

3.18 In terms of houses built post 2000 and sold after Apr 2014, and have identified 59 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,733 per sq m, 28 x 2, 3 and 4 bed semi-detached houses, which since Apr 2014 have sold at an average of £1,535 per sq m, as well as 45 x 2, 3 and 4 bed terraced houses, which since Apr 2014 have sold at an average of £1,456 per sq m.

3.19 The Harron Homes scheme therefore suggests that new build schemes will command a premium above the average for the locality (which includes second hand house sales). Further the evidence also contradicts the general trend of falling house values during the last 12 months, as in fact house prices within the scheme have increased in 2016. However, it is acknowledged that this is only one development, and could reflect the nature of the location, product etc, rather than being a tangible sign of a general market trend.

DN4

3.20 This broadly covers a large area to the south of Doncaster town centre, and includes Warmsworth, dwellings around the A1 (M), A630 and A6182 (Balby Carr), the Lakeside Shopping Outlet and housing along the A638 corridor to the south east of Doncaster.

3.21 The average house value within DN4 for both new build and second hand sales equates to £146,629 (data taken from Zoopla.co.uk). This is a fall of circa -2.75% during the last 12 months.

3.22 More specifically, the average values for each house type in DN4 are currently showing the following (data from Zoopla.co.uk):

Detached - £229,939 (£1,755 per sq m) Semi-detached - £125,119 (£1,496 per sq m) Terrace - £87,742 (£1,130 per sq m)

3.23 In DN4 we note new build house sales achieved at the following:

- Keepmoat Homes, Dominion: located in Balby, close to Junction 3 of the M18. Comprising a mix of detached, semi-detached and terraced houses. Between June 2014 and Feb 2016 9 x 3 bed detached houses sold, at an average of £1,898 per sq m. Between Dec 2014 and Mar 2016 41 x 2, 3 and 4 bed semi- detached houses sold, at an average of £1,735 per sq m. Between June 2014 and Jan 2016 16 x 2 and 3 bed terraced houses sold, at an average of £1,819 per sq m.

- Strata Homes, Dominion: adjacent to the Keepmoat housing detailed above, forming part of the same scheme. Between Dec 2014 and Jan 2016 21 x 4 bed 2 storey detached houses sold, at an average of £1,992 per sq m. Between Aug 2014 and Apr 2016 4 bed 3 storey semi-detached houses sold, at an average of £1,550 per sq m (3 storey houses command a lower rate per sq m than 2 storey equivalents).

- Persimmon Homes, Warren Park: located in Bessacar. Between May 2015 and Mar 2016 22 x 3, 4 and 5 bed detached houses sold, at an average of £2,216 per sq m. Between Jun 2015 and Dec 2015 13 x 2, 3 and 4 bed semi- detached houses sold, at an average of £1,993 per sq m. In Dec 2015 7 x 2 bed terraced houses sold, at an average of £2,000 per sq m.

- David Wilson Homes, Serenity: located at Lakeside to the south east of Doncaster town centre. Between Apr 2014 and Mar 2016 12 x 3, 4 and 5 bed detached houses sold, at an average of £2,166 per sq m. Between May 2014 and Feb 2016 22 x 3 bed semi-detached houses sold, at an average of £2,141 per sq m. Between Apr 2014 and Mar 2016 18 x 3 bed terraced houses sold, at an average of £1,911 per sq m.

- Lovell Homes, Lincoln Homes: located at Lakeside, adjacent to the David Wilson Homes scheme detailed above. Between Apr 2014 and Mar 2016 6 x 3 bed detached houses sold, at an average of £1,578 per sq m. Between Apr 2014 and Mar 2016 23 x 2 and 3 bed semi-detached houses sold, at an average of £1,497 per sq m. Between Apr 2014 and Aug 2015 8 x 2 bed terraced houses sold, at an average of £1,707 per sq m.

3.24 The Balby and Lakeside developments detailed above highlight that different values can be achieved in neighbouring locations due to the quality / nature of the housing product being offered.

3.25 In terms of houses built post 2000 and sold after Apr 2014, and have identified 73 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,856 per sq m, 30 x 2, 3 and 4 bed semi-detached houses, which since Apr 2014 have sold at an average of £1,572 per sq m, as well as 61 x 2, 3 and 4 bed terraced houses, which since Apr 2014 have sold at an average of £1,420 per sq m.

DN5

3.26 This covers a wide geographical area to the north, north-west and west of Doncaster Town, and includes (amongst other settlements) Sprotbrough, Barnburgh, Hooton Pagnell, Marr, Bentley, Toll Bar and Arksey. The main routes through the area include the A1 (M), A625, A19 and A638 York Road.

3.27 The average house value within DN5 for both new build and second hand sales equates to £141,465 (data taken from Zoopla.co.uk). This is a slight fall of circa - 0.79% during the last 12 months.

3.28 More specifically, the average values for each house type in DN5 are currently showing the following (data from Zoopla.co.uk):

Detached - £241,565 (£1,884 per sq m) Semi-detached - £121,850 (£1,529 per sq m) Terrace - £86,724 (£1,152 per sq m)

3.29 In DN5 we note new build house sales achieved at the following:

- Keepmoat Homes, Evergreen: located in Bentley, close to Junction 3 of the M18. Comprising a mix of 3 and 4 bedroom dwellings. Between June 2014 and Mar 2016 4 x 3 and 4 bed detached houses sold, at an average of £1,348 per sq m. Between Jul 2014 and Mar 2016 26 x 3 and 4 bed semi-detached houses sold, at an average of £1,278 per sq m. It is noted that houses prices grew by circa 4% from 2015 to 2016. Between May 2014 and Jan 2016 24 x 3 bed terraced houses sold, at an average of £1,316 per sq m.

3.30 In terms of houses built post 2000 and sold after Apr 2014, and have identified 12 x 3, 4 and 5 bed detached houses, with an average price achieved of £2,020 per sq m, 7 x 2, 3 and 4 bed semi-detached houses, which since Apr 2014 have sold at an average of £1,499 per sq m, as well as 21 x 2, 3 and 4 bed terraced houses, which since Apr 2014 have sold at an average of £1,642 per sq m.

DN6

3.31 This covers a wide geographical area to the north of Doncaster Town, mainly along the A1, A638 and A19 corridors. This area includes (amongst other settlements) Norton, Askern, Carcroft, Adwick le Street and Moss.

3.32 Average house values in DN6 for new build and second hand sales equates to £125,016 (data Zoopla.co.uk), a fall of circa -2.34% during the last 12 months. More specifically, the average values for each house type in DN6 are currently showing the following (data from Zoopla.co.uk):

Detached - £221,415 (£1,658 per sq m) Semi-detached - £104,196 (£1,302 per sq m) Terrace - £76,864 (£990 per sq m)

3.33 Since 2014, new build house sales are limited in DN6 and as we such we have been unable to identify any large estates implemented by national / regional housebuilders. We have extended our research to include houses built post 2000 and sold after Apr 2014, and have identified 16 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,505 per sq m, as well as 6 x 2, 3 and 4 bed terraced houses, which since Apr 2014 have sold at an average of £1,479 per sq m. Please note, as these house sales are older and include second-hand sales it is assumed contemporary, brand new homes would command a premium above these averages.

DN7

3.34 This covers an area to the north east of Doncaster Town, which includes part of the M18 and M180 corridors. This area includes (amongst other settlements) Hatfield, Stainforth and Fishlake.

3.35 The average house value within DN7 for both new build and second hand sales equates to £130,411 (data taken from Zoopla.co.uk). This is a fall of circa -1.56% during the last 12 months.

3.36 More specifically, the average values for each house type in DN7 are currently showing the following (data from Zoopla.co.uk):

Detached - £190,567 (£1,529 per sq m) Semi-detached - £105,202 (£1,249 per sq m) Terrace - £82,993 (£1,249 per sq m)

3.37 Since 2014, new build house sales are limited in DN7 and as we such we have been unable to identify any large estates implemented by national / regional housebuilders. We have subsequently extended our research to include houses built post 2000 and sold after Jan 2014, and have identified 31 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,575 per sq m, as well as 13 x 2, 3 and 4 bed terraced houses, which since Feb 2014 have sold at an average of £1,405 per sq m. Please note, as these house sales are older and include second-hand sales it is assumed contemporary, brand new homes would command a premium above these averages.

DN8

3.38 This covers an area immediately to the north east of DN7, and also includes part of the M18 and M180 corridors. This area includes (amongst other settlements) Moorends, Thorne and Sandtoft.

3.39 The average house value within DN8 for both new build and second hand sales equates to £111,020 (data taken from Zoopla.co.uk). This is a fall of circa -1.83% during the last 12 months.

3.40 More specifically, the average values for each house type in DN8 are currently showing the following (data from Zoopla.co.uk):

Detached - £165,592 (£1,410 per sq m) Semi-detached - £95,209 (£1,141per sq m) Terrace - £83,893 (£1,033 per sq m)

3.41 Since 2014, new build house sales are limited in DN8 and as we such we have been unable to identify any large estates implemented by national / regional housebuilders. We have subsequently extended our research to include houses built post 2000 and sold after Jan 2014, and have identified 17 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,611 per sq m, as well as 10 x 2 and 3 bed terraced houses, which since Jul 2014 have sold at an average of £1,459 per sq m. Please note, as these house sales are older and include second-hand sales it is assumed contemporary, brand new homes would command a premium above these averages.

DN9

3.42 This covers a large geographical area located to the east and south of the M18 and M180 corridors. This area includes (amongst other settlements) Finningley, Wroot, Haxey, Epworth, Belton and Owston Ferry.

3.43 The average house value within DN9 for both new build and second hand sales equates to £207,027(data taken from Zoopla.co.uk). This is a fall of circa -2.26% during the last 12 months.

3.44 More specifically, the average values for each house type in DN9 are currently showing the following (data from Zoopla.co.uk):

Detached - £243,271 (£1,755 per sq m) Semi-detached - £141,762 (£1,550per sq m) Terrace - £115,010 (£1,346 per sq m)

3.45 In DN9 we note new build house sales achieved at the following:

- David Wilson Homes, Pembridge Park: located in Auckley. Comprising a mix of 3 and 4 bedroom dwellings. Between Apr 2014 and Mar 2016 58 x 3 and 4 bed detached houses sold, at an average of £2,129 per sq m. From 2015 and 2016 the house values have remained broadly static. Between Dec 2014 and Dec 2015 8 x 3 bed semi-detached houses sold, at an average of £2,102 per sq m. Between Dec 2014 and Jul 2015 5 x 2 and 3 bed terraced houses sold, at an average of £2,095 per sq m.

- Taylor Wimpey Homes, Westlands: located in Auckley, adjacent to the David Wilson Homes detailed above. Comprising a mix of 3, 4 and 5 bedroom dwellings. Between Apr 2014 and Apr 2016 28 x 4 bed detached houses sold, at an average of £1,929 per sq m. From 2015 and 2016 the house values have remained broadly static. Between Apr 2014 and Mar 2016 28 x 3 bed semi- detached houses sold, at an average of £1,747 per sq m.

3.46 In terms of houses built post 2000 and sold after Apr 2014, and have identified 85 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,668 per sq m, 4 x 3 and 4 bed semi-detached houses, which since Apr 2014 have sold at an average of £1,532 per sq m, as well as 10 x 2 and 3 bed terraced houses, which since Apr 2014 have sold at an average of £1,776 per sq m.

DN10

3.47 Located to the south east of Doncaster town centre, this area includes (amongst other settlements) Bawtry, Austerfield, Gringley on the hill and Misterton. These settlements are largely located along the A631, A614 and A638 trunk roads.

3.48 The average house value within DN10 for both new build and second hand sales equates to £242,873 (data taken from Zoopla.co.uk). This is a fall of circa -1.14% during the last 12 months.

3.49 More specifically, the average values for each house type in DN10 are currently showing the following (data from Zoopla.co.uk):

Detached - £296,030 (£2,002 per sq m) Semi-detached - £160,323 (£1,658 per sq m) Terrace - £147,381 (£1,895 per sq m)

3.50 Since 2014, new build house sales are limited in DN10 and as we such we have been unable to identify any large estates implemented by national / regional housebuilders. We have subsequently extended our research to include houses built post 2000 and sold after Apr 2014, and have identified 41 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,846 per sq m, as well as 6 x 2, 3 and 4 bed terraced houses, which since Aug 2014 have sold at an average of £1,846 per sq m. Please note, as these house sales are older and include second-hand sales it is assumed contemporary, brand new homes would command a premium above these averages.

DN11

3.51 Located to the south of Doncaster town centre, this area includes (amongst other settlements) Tickhill, Loversall, Rossington and Harworth. The area is mainly formed around the A1 (M) and M18 corridors.

3.52 The average house value within DN11 for both new build and second hand sales equates to £155,252 (data taken from Zoopla.co.uk). This is a fall of circa -1.94% during the last 12 months.

3.53 More specifically, the average values for each house type in DN11 are currently showing the following (data from Zoopla.co.uk):

Detached - £235,254 (£1,948 per sq m) Semi-detached - £118,447 (£1,421 per sq m) Terrace - £193711 (£1,206 per sq m)

3.54 Since 2014, new build house sales are limited in DN11 and as we such we have been unable to identify any large estates implemented by national / regional housebuilders. We have subsequently extended our research to include houses built post 2000 and sold after Apr 2014, and have identified 39 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,691 per sq m, 16 x 2, 3 and 4 bed semi-detached houses, which since Apr 2014 have sold at an average of £1,596 per sq m, and 39 x 3, 4 and 5 bed detached houses, with an average price achieved of £1,691 per sq m, 17 x 2, 3 and 4 bed terraced houses, which since May 2014 have sold at an average of £1,464 per sq m. Please note, as these house sales are older and include second-hand sales it is assumed contemporary, brand new homes would command a premium above these averages.

DN12

3.55 Located to the south west of Doncaster town centre, this area includes (amongst other settlements) Conisbrough, Old Denaby and New Edlington. The area is mainly served by the A630 and A1 (M).

3.56 The average house value within DN12 for both new build and second hand sales equates to £106,462 (data taken from Zoopla.co.uk). This is a fall of circa -3.77% during the last 12 months.

3.57 More specifically, the average values for each house type in DN12 are currently showing the following (data from Zoopla.co.uk):

Detached - £185,365 (£1,496 per sq m) Semi-detached - £98,707 (£1,216 per sq m) Terrace - £75,291 (£936 per sq m)

3.58 In DN12 we note new build house sales achieved at the following:

- Keepmoat Homes, Yew Gardens: located in New Edlington. Between Dec 2014 and Jul 2015 2 x 3 bed detached houses sold, at an average of £1,605 per sq m. Between Apr 2014 and Mar 2016 29 x 2 and 3 bed semi-detached houses sold, at an average of £1,469 per sq m. From 2015 and 2016 the house values increased by circa 2%. Between Aug 2014 and Nov 2015 10 x 2 and 3 bed terraced houses sold, at an average of £1,366 per sq m.

3.59 In terms of houses built post 2000 and sold after Apr 2014, and have identified 7 x 3 and 4 bed detached houses, with an average price achieved of £1,725 per sq m, 10 x 2 and 3 bed semi-detached houses, which since Apr 2014 have sold at an average of £1,481 per sq m, as well as 20 x 2, 3 and 4 bed terraced houses, which since Apr 2014 have sold at an average of £1,331 per sq m.

Summary

3.60 To aid analysis, below we have summarised the average house prices achieved in each postcode since April 2014, splitting the data into houses built between 2000 and 2009, and houses built post 2009.

Table 1 – Postcode house price averages

Pcde 2000-2009 Sample Post 2009 Sample

DN1 £ - 0 £ 1,802 42 DN2 £ - 0 £ 1,530 13 DN3 £ 1,597 132 £ 1,992 152 DN4 £ 1,642 164 £ 1,931 276 DN5 £ 1,731 40 £ 1,615 97 DN6 £ 1,511 26 £ 1,759 3 DN7 £ 1,477 53 £ 1,526 25 DN8 £ 1,529 30 £ 1,202 13 DN9 £ 1,667 100 £ 1,950 177 DN10 £ 1,846 47 £ 1,807 41 DN11 £ 1,595 76 £ 1,781 49 DN12 £ 1,417 38 £ 1,511 72

3.61 We have also shown the average house prices achieved for various settlements across Doncaster Borough. Again we have differentiated between houses built between 2000 and 2009 and the prices achieved for houses build post 2009. The data is summarised below in Table 2.

Table 2 – Settlement house price averages

2000 to Settlement Sample Post 2009 Sample 2009 Adwick & Woodlands £ 1,226 4 £ - 0 Thorne-Moorends £ 1,529 30 £ 1,202 13 Bentley £ 1,492 8 £ 1,305 61 Edlington £ 1,444 10 £ 1,452 45 Rossington £ 1,627 45 £ 1,480 6 Askern £ 1,538 8 £ 1,432 1 Carcroft & Skellow £ 1,517 5 £ - 0 Barnby Dun & Kirk Sandall £ 1,620 31 £ 1,293 11 Mexborough £ 1,414 10 £ 1,516 92 Stainforth & Hatfield £ 1,477 53 £ 1,526 25 Conisbrough & Denaby £ 1,500 24 £ 1,610 27 Armthorpe £ 1,548 88 £ 1,653 11 Scawsby / Scawthorpe £ 1,570 12 £ 1,685 18 Lakeside / edge of Doncaster £ 1,593 12 £ 1,887 123 Balby / SW Doncaster £ 1,596 102 £ 1,912 103 Auckley & Hayfield Green & Finningley £ 1,639 30 £ 2,001 152 Edenthorpe £ 1,372 3 £ 2,057 48 Bessacarr / SE Doncaster £ 1,869 130 £ 2,083 51 Tickhill £ 1,991 4 £ 2,150 2 Sprotbrough / Barnburgh & Harlington £ 2,001 17 £ 2,597 18 Bawtry £ 2,007 31 £ 2,602 1

100

3.62 Based on all of the above considerations we have subsequently allocated the following settlements into the value bands, as follows:

Table 3 – Settlement value banding

Value Band Postcode Settlements (broad)

High DN9 3 Auckley & Hayfield Green DN9 3 Finningley DN10 6 Bawtry DN11 9 Tickhill DN5 7 Sprotbrough DN5 7 Barnburgh & Harlington Medium DN5 8 Scawsby / Scawthorpe DN4 7 Bessacarr / SE Doncaster S64 0 Mexborough DN7 6 Hatfield DN7 5 Stainforth DN3 3 Armthorpe DN3 2 Edenthorpe DN3 1 Barnby Dun / Kirk Sandall DN6 0 Askern DN6 8 Carcroft & Skellow DN4 / DN1 Lakeside / edge of Doncaster DN4 / DN1 Balby / SW Doncaster Low / medium DN12 2 Conisbrough & Denaby Low DN5 0 Bentley DN8 4 / DN8 5 Thorne-Moorends DN11 0 Rossington S64 0 / DN6 7 Adwick & Woodlands DN12 1 Edlington

4.0 Conclusions

4.1 The average postcode analysis (data derived from Zoopla), which incorporates all house types and ages of construction, shows a general trend that house prices have fallen slightly during the last 12 months.

4.2 However, it is unclear whether this is a long term trend, or whether this is temporary (reflecting external factors, such as the EU referendum decision). There are a number of examples of new build schemes where house values have in fact increased during the last 12 months, which suggests the data shown from Zoopla may be misleading and may not necessarily apply to new build housing.

101

4.3 Furthermore, the general consensus in the market place is that there remains a high level of demand for housing across the country, and with the continuation of Government-led initiatives such as ‘Help to Buy’ and the introduction of Starter Homes, it remains likely that new build houses will, generally speaking, continue to attract strong levels of demand in the short to medium (which will ultimately serve to underpin values).

4.4 In this respect, and whilst we have been sensitive to the indicators which point to a slight slowing in house values during the last 12 months, we do not believe it is appropriate to adopt an overly negative approach in setting house prices within this study. In our view, new build housing will continue to attract strong levels of demand, aided by Government led initiatives.

4.5 In terms of setting appropriate values for testing in the viability study, based on our experience and having discussed this at the Stakeholder workshop, we have adopted the approach of setting a single average figure for 3 categories of value areas, being: low, medium and high.

4.6 The average figure for each value area is based on the analysis of the data highlighted above, being an average across all house types (i.e. detached, semi- detached, terraced). The figures are taken within the context of the local market.

4.7 The broad average values range for each area as follows:

Low: Houses in low value areas are defined as generating average sales values for new build dwellings in the region of £1,500 per sq m.

Medium: Houses in medium value areas are defined as generating average sales values for new build dwellings in the region of £1,750 to £2,000 per sq m.

High: Houses in high value areas are defined as generating average sales values for new build dwellings in the region of £2,250 per sq m (or higher).

102

4.8 Based on these broad bandings, for the purposes of the viability testing, we have subsequently adopted the following average sales values for each area:

Area Average Sales value (£ per sq m)

Low value area £1,500 Medium value area £1,750 High value area £2,250

103

Appendix 2

Threshold Land Value Assessment

104

APPENDIX 2 THRESHOLD LAND VALUE ASSESSMENT

1.0 Introduction

1.1 This section focuses on Threshold Land Values (“TLVs”), and includes a definition of the concept, a review of the relevant viability guidance and the approach adopted for the purposes of this study.

1.2 Land value is a key component of a development appraisal, albeit it can often be the ‘outcome’ of the appraisal rather than being a fixed figure (hence why appraisals are often referred to as being ‘residual’, because once all the inputs are included the ‘residue’, if there is any, is used as the sum which a developer could pay for the site).

1.3 However, the ‘residue’ from the appraisal (as a land value) does not always meet the expectations of the landowner. If the developer is only able to pay a significantly reduced sum below the landowner’s expectations then the outcome is fairly straight forward; the land would not be sold for development. In undertaking a viability assessment a minimum land value is therefore identified (sometimes referred to as the “benchmark” or “threshold” land value - TLV).

1.4 This slightly changes the focus of a development appraisal when used for testing viability, because rather than the land value being freely subject to change there becomes a minimum figure below which a landowner would not release the land for development. If this minimum figure is reached, other inputs within the appraisal would need to be subject to change to ensure viability. As the majority of development costs (like build costs, professional fees, funding etc) are relatively fixed the only ‘flex’ could therefore be in the developer’s profit or the Council’s policy. Paragraph 173 of the National Planning Policy Framework (“NPPF”) indicates that the Council’s policy should be at a level which provides the developer and landowner with a competitive return, which is often taken to mean that the Council’s policies should be an input subject to change if the minimum land value has not been met but the Council’s full policies have been achieved.

105

2.0 Viability Guidance

2.1 Identifying the level of an appropriate TLV is itself not straight forward. Surveyors are now assisted by two relatively recent publications, although the guidance between each is somewhat contradictory. We have commented on each publication as follows:

“Financial viability in planning” August 2012 by the Royal Institution of Chartered Surveyors (RICS):

Para 2.5.2, Box 10, “…nature of the applicant should normally be disregarded as should the benefits or dis-benefits that are unique to the applicant.”

2.2 Thus, appraisals should be done assuming hypothetical, typical landowners and developers and the views and aspirations of the actual owner are not relevant if these views differ from general market practice.

2.3 As indicated above, this refers to the site value as usually being assessed by means of a residual development appraisal. However, the suggestion seems to be that planning policies should be fixed and land value subject to change (which contradicts the view of the landowner having a minimum land value below which they would sell).

Para 2.1.2 “It follows, for example, that the land value is flexible and not a fixed figure to the extent that Site Value has to be determined as part of the viability assessment.”

106

2.4 This appears to support the above view that it is the Council’s policy which drives the land value, not the other way round. However, the RICS document does acknowledge that the flexibility in land value cannot result in the value going below the Current Use Value (“CUV”), stating:

Para 3.4.4 “The return to the landowner will be in the form of a land value in excess of current use value but it would be inappropriate to assume an uplift based on set percentages.”

2.5 This appears to support the view of setting a TLV for development appraisals, which is to somehow be linked to the CUV However, no guidance is given as to how to determine the link between the CUV and the TLV.

2.6 Furthermore, in particular no guidance is given to assessing greenfield land, where the CUV may only be £12,500 - £25,000 per Ha and clearly a TLV only slightly above the CUV would not represent a sufficient incentive for a landowner to sell for development.

“Viability Testing Local Plans” June 2012 by the Local Housing Delivery Group (“The Harman Review”).

Pg 29 “We recommend that the Threshold Land Value is based on a premium over current use values and credible alternative use value (noting the exceptions below)”

2.7 This therefore contradicts the guidance provided by the RICS, where adopting a percentage uplift above the CUV is not recommended.

2.8 One of the exceptions referred to relates to “non-urban” and “greenfield” sites.

Pg 30 “ It is widely recognized that this approach [i.e. a percentage increase over CUV] can be less straight forward for non urban sites or urban extensions, where land owners are rarely forced or distressed sellers…This is particularly the case in relation to large greenfield sites…Accordingly, the uplift to current use value sought by the landowners will invariably be significantly higher than in an urban context and requires very careful consideration”.

107

2.9 This does not mean that an assessment of the CUV has no part to play in the process of assessing greenfield sites. A typical landowner will still want to know what the value of his/her site is without the planning permission applied for, and then judge by how much, if at all, the CUV increases when planning consent is granted. The difference is that, for urban brownfield sites a premium uplift of circa 25 – 50% of the CUV may be deemed sufficient to incentivise a landowner to sell (e.g. if the CUV is £200,000 per Ha, applying a 50% uplift would mean a TLV of £300,000 per Ha, which would be attractive to a landowner). For a greenfield site, if the CUV is only say £10,000 per Ha then a 50% uplift (i.e. a TLV of £15,000 per Ha) would clearly not incentivise a landowner to release the land for development. In reality, the ‘uplift’ would need to be more like 15 – 25 times (or more) the CUV.

2.10 In terms of how to evidence the approach to greenfield sites the document goes on to say:

Pg 30 “…local sources should be used to provide a view on market values (the ‘going rate’), as a means of giving a further sense check on the outcome of the current use plus premium calculation”.

Pg 30 “…for sites of this nature [i.e. greenfield], it will be necessary to make greater use of benchmarks, taking into account of local partner views on market data and information on typical minimum price provisions used within developer / site promoter agreements for sites of this nature”.

2.11 This therefore seems to advocate using evidence of TLVs identified as part of the viability process, as well as using market transactions as a general ‘sense check’. However, there are limitations of assessing land sales (please see below section 2.3 DVS approach: evidence types identified).

108

3.0 Methodology

3.1 In summary, a TLV can therefore be regarded as being effectively the average price that an average developer / house builder would be willing to pay for a site, being at a level which would incentivize an average landowner to release the site for development. A TLV does not therefore seek to reflect excessive demands from unreasonable parties, but instead looks to reflect a reasonable price for all parties concerned. It therefore follows that the price paid for a specific site by a specific developer may not necessarily be appropriate as the TLV when assessing viability, because the price paid may be considered to be too low or excessive when compared with the wider market.

3.2 The valuation process to identify this ‘reasonable’ price involves the surveyor judging where the value of the site would be if the respective costs of applying all the Council’s existing and emerging planning policies and undertaking abnormal works (if applicable) were fully reflected. This is then viewed alongside the price at which a reasonable, hypothetical, commercially-minded landowner would dispose of the land having regard to the site’s Current Use Value (CUV) or any Alternative Use Value (AUV), should one be available.

3.3 Settling on this ‘reasonable’ land value in an appraisal is not therefore straightforward and the guidance is contradictory and can be interpreted in different ways. Landowners naturally want as high a price as they can achieve and some of them are not prepared to recognise how the impact of the cost of planning gain and abnormals drives down net land values materially.

3.4 As indicated above, to complicate matters the approach to assessing an appropriate TLV for greenfield sites is also slightly different to brownfield land, because the ‘premium uplift’ on a greenfield site should be significantly higher than that of brownfield land.

3.5 Furthermore, abnormal costs should also be a key consideration when assessing an appropriate TLV for a particular site. If there are 2 identical sites next door to one another, but one has significant abnormal works and the other doesn’t, then the TLV for the site affected by the abnormal works should have a lower TLV than the unaffected site.

3.6 In accordance with the Harman Review, we concur that market transactions alone do not provide an adequate evidence base on which to consider TLVs.

109

3.7 When assessing TLVs we look to assess a variety of evidence sources, including:

 TLVs submitted by developers / house builders in their own viability appraisals.  TLVs as agreed with developers / house builders as part of negotiations over individual viability appraisals.  TLVs determined as part of the planning appeal decisions.  TLVs assessed for the purposes of area wide studies.

3.8 The above therefore provides direct evidence sources on actual TLVs, and therefore it is easier to make a direct comparison. The Harman Guidance suggests this should be the ‘first step’ when looking to identify appropriate TLVs.

3.9 The ‘second step’ is to then consider market transactions / land sales. However, assessing actual land sales for the purposes of identifying a TLV is not straight forward, as the price someone is willing to pay for a piece of development land (and indeed accept for a piece of development land) is subject to many factors, which includes:

- The type of development that could be brought forward. - The gross to net ratio (it may be that a large section of the site is constrained and cannot be developed). - The potential density any of proposed scheme. - Whether any third parties benefit from a ransom position preventing access to the site. - Whether there are any title constraints. - The abnormal costs associated with developing the site (i.e. any untypical cost, such as deep pile foundations to mitigate ground concerns, flooding mitigation works etc). - The planning policies that relate to a specific type of scheme. - Whether a purchaser benefits from synergistic value (formerly known as marriage value) with any neighbouring land they already own or will own in the future. - Whether a vendor is under financial pressure to sell. - Whether a house-builder is keen to have a presence in a particular location etc etc.

110

3.10 There are therefore a number of factors which impact upon the price someone is willing to pay for development land, because ultimately each development site is unique. For example, you could have 2 sites next to each other sold at the same time, each being 5 Ha and the same shape. However, one may have significant flooding issues and a poor access route, whereas the other may have no concerns. The price paid for the land affected by ‘abnormal’ development costs (i.e. in this case flooding and a poor access route) would therefore most likely be significantly less than the unaffected site. The reasons for the difference in value, though, would not be identifiable by simply looking at the price paid for the land on a ‘per Ha’ basis.

3.11 This means it is extremely difficult to compare two land transactions because in reality only some of the factors outlined above (which is not an exhaustive list) will be known to the analysing surveyor.

3.12 In this respect, land transactions are useful in providing a ‘sense check’ but they should not be regarded as providing a definitive view on threshold land values, particularly on a ‘price per Ha’ basis, because in most cases the full details of the transaction (and the factors which impact upon value) will not be known. Land sales should be therefore considered after the other sources of evidence identified, and provide a ‘sense check’ only.

3.13 Please note, when assessing the evidence and considering appropriate TLVs we have looked to distinguish between greenfield and brownfield sites, for the reasons outlined above in 2.9 onwards).

4.0 Evidence – greenfield

4.1 A ‘greenfield’ site is considered to be a site which has not previously been developed before and essentially comprises grassland or agricultural land.

4.2 In terms of direct TLV evidence (not transactional evidence), we have identified the following TLVs for greenfield sites, identified from viability appraisals received from applicants. For the purposes of this exercise we have looked at TLVs for greenfield sites across the North East of England, Yorkshire and the East Midlands. Whilst a large geographical area this gives a good indication of how TLVs for greenfield sites remain relatively consistent across regions (please note for confidentiality reasons we are unable to provide the full details of each case):

111

 Medium value area near to Leeds, West Yorkshire – greenfield site, net area 6.47 Ha, proposal for 181 dwellings. Abnormals circa £333,500 per net Ha. Average house price £1,888 per sq m. January 14 a regional developer submitted a viability appraisal, indicating a TLV equivalent to £680,000 per net Ha (£573,000 per gross Ha).

 Medium value area North Yorkshire, commutable to Leeds – greenfield site, net area 8.07 Ha, proposal for 179 dwellings. Abnormals circa £415,000 per net Ha. Average house price £1,977 per sq m April 14 a national house builder submitted a viability appraisal, indicating a TLV equivalent to £494,000 per net Ha (£369,500 per gross Ha).

 Low value area South Yorkshire – greenfield site, net area 2.76 Ha, proposal for 97 dwellings. Abnormals circa £54,000 per net Ha. Average house price £1,391 per sq m. June 14 a national house builder submitted a viability appraisal, indicating a TLV equivalent to £217,500 per net Ha (£199,000 per gross Ha).

 Medium value area North Yorkshire, commutable to Leeds – greenfield site, net area 3.57 Ha, proposal for 103 dwellings. Abnormals circa £408,000 per net Ha. Average house price £1,842 per sq m. June 14 a regional house builder submitted a viability appraisal, indicating a TLV equivalent to £284,000 per net Ha (£247,000 per gross Ha).

 Medium value area West Yorkshire – greenfield site, net area 9.19 Ha, proposal for 166 dwellings. Abnormals circa £590,500 per net Ha. Average house price £1,923 per sq m Sept 14 a national house builder submitted a viability appraisal, indicating a TLV equivalent to £631,000 per net Ha (£474,500 per gross Ha).

 Low value area West Yorkshire – greenfield site, net area 8.47 Ha, proposal for 283 dwellings. Abnormals circa £173,000 per net Ha. Average house price £1,587 per sq m. Nov 14 a national house builder submitted a viability appraisal, indicating a TLV equivalent to £294,000 per net Ha (£267,000 per gross Ha).

112

 Medium value in West Yorkshire - greenfield site, net area 16.08 Ha, proposal for 560 dwellings. Abnormals circa £148,000 per net Ha. Average house price £2,099 per sq m. April 15 a national firm of chartered surveyors submitted a viability appraisal, indicating a TLV equivalent to £618,000 per net Ha (£390,000 per gross Ha).

 Medium value in West Yorkshire - greenfield site, net area 1.27 Ha, proposal for 42 dwellings. Abnormals circa £709,000 per net Ha. Average house price £2,152 per sq m. Mar 15 a national house builder submitted a viability appraisal, indicating a TLV equivalent to £363,000 per net Ha (£310,500 per gross Ha).

 High value area in West Yorkshire – greenfield site, net area 3.50 Ha, proposal for 84 dwellings. Abnormals circa £217,500 per gross Ha. Average house price circa £1,958 per sq m (please note DVS argued this should have been in excess of £2,500 per sq m given the high value area). May 15 a national house builder submitted a viability appraisal, indicating a TLV equivalent to £445,000 per gross Ha (net size unknown).

 Low value area in Lincolnshire – greenfield site, net area 3.20 Ha, proposal for 108 dwellings. Abnormals circa £506,500 per net Ha. Average house price £1,629 per sq m. June 15 a regional house builder submitted a viability appraisal, indicating a TLV equivalent to £309,000 per net Ha (£277,000 per gross Ha).

 Medium value area in West Yorkshire – greenfield site, net area 2.16 Ha, proposal for 66 dwellings. Abnormals circa £414,503 per net Ha. Average house price £1,920 per sq m. Oct 15 a regional planning consultant submitted a viability appraisal, indicating a TLV equivalent to £531,265 per net Ha (£466,604 per gross Ha).

 Medium value area in Derbyshire – greenfield site, net area 7.53 Ha, proposal for 201 dwellings. Abnormals circa £70,279 per net Ha. Average house price £1,880 per sq m. Dec 15 a national surveying firm submitted a viability appraisal, indicating a TLV equivalent to £420,070 per net Ha (£302,000 per gross Ha).

113

 Medium value area in Lincolnshire – greenfield site, net area 4.46 Ha, proposal for 137 dwellings. Abnormals circa £78,998 per net Ha. Average house price £2,033 per sq m. Dec 15 a national surveying firm submitted a viability appraisal, indicating a TLV equivalent to £262,838 per net Ha (£235,000 per gross Ha).

 Medium value area in Nottinghamshire – greenfield site, net area 2.31 Ha, proposal for 91 dwellings. Abnormals nil. Average house price £1,939 per sq m. Apr 16 a national housebuilder submitted a viability appraisal, indicating a TLV equivalent to £363,237 per net Ha (£236,000 per gross Ha).

 Medium value area in North Yorkshire – greenfield site, net area 7.57 Ha, proposal for 276 dwellings. Abnormals nil. Average house price £1,616 per sq m. Apr 16 a planning consultant submitted a viability appraisal, indicating a TLV equivalent to £535,573 per net Ha (£377,500 per gross Ha).

 High value area in Derbyshire – greenfield site, gross site area 10.70 Ha, proposal for 246 dwellings. Abnormals £500,854 per gross Ha. Average house price £2,271 per sq m. Jun 16 a local surveying firm submitted a viability appraisal, indicating a TLV equivalent to £430,319 per gross Ha.

 Low value area in South Yorkshire – greenfield site, gross site area 2.48 Ha, proposal for 80 dwellings. Abnormals £40,376 per net Ha. Average house price £1,597 per sq m. Jun 16 a local planning consultant submitted a viability appraisal, indicating a TLV equivalent to £201,879 per net Ha (£154,500 per gross Ha).

4.3 Based on the above sample of 17 identified TLVs the average equates to circa £338,495 per gross Ha, ranging from £154,500 per gross Ha to £573,000 per gross Ha.

4.4 More specifically, in the areas considered to be ‘low value’, the average across the sample of TLVs equates to circa £224,000 per gross Ha. For the ‘medium value’ sites the average increases to £362,000 per gross Ha. The ‘high value’ site sample is small and therefore is considered to be less reliable (albeit shows an average of circa £437,500 per gross hectare).

114

4.5 Although not a definitive source of information, this at least gives a general indication of the levels of TLVs being applied by developers / house builders to greenfield sites across the wider regions in viability appraisals (albeit with the acknowledgement that these figures are naturally on the high side). It therefore stands to reason that TLVs within an area such as Doncaster (not considered to be vastly different in terms of the types of houses being provided and the values achieved across Yorkshire, the North East and the East Midlands) should certainly not exceed the upper end of this range.

4.6 In addition to the above we have also considered TLVs identified by private practice chartered surveyors in CIL / general area wide studies undertaken on behalf of local authorities. Again, we have considered these from a broader region, to include Yorkshire and the East Midlands, as follows (please note some studies were not explicit with regard to what they considered the appropriate TLV to be, and instead simply indicated whether the residual land value return was broadly reasonable or not – these have subsequently been excluded from the sample. Also, we have limited the sample to reports published post 2014):

 Sheffield City Council CIL Viability Study – undertaken by BNP Paribas, dated Feb 2014. The report concludes “the approach of using current use values is a more reliable indicator of viability than using market values or prices paid for sites” and for greenfield sites adopts a TLV of £247,000 per Ha.

 Wakefield Metropolitan District Council CIL Viability Study – undertaken by DTZ, dated Feb 2014. Adopts the following benchmark land values: low value area £679,537 per net Ha, medium value area £834,000 per net Ha and high value area £988,400 per net Ha. It is stressed that these figures are based on net areas, therefore is difficult to compare with our study which assumes gross areas. Also, DTZ do not differentiate between greenfield or brownfield values.

 Selby CIL Addendum Report – undertaken by Peter Brett Associates in April 2014. The report doesn’t appear to distinguish between greenfield and brownfield sites. For low value areas a TLV equivalent to £450,000 per net developable Ha (so likely to be sub £370,000 on a gross Ha basis). For medium value areas this increases to £650,000 per net Ha

115

(so again on a gross basis more likely to be £494,000 - £555,000 per gross Ha). For high value areas this increases to £900,000 per net Ha (so more like £740,000 - £800,000 on a gross basis per Ha).

 Leicester City Council CIL Viability Study Update – undertaken by HDH Planning in December 2014. For greenfield sites assumed a TLV range of £280,000 per gross Ha to £310,000 per gross Ha.

 Mansfield District Council Local Plan: Viability Assessment – undertaken by DSP Planning and Development Viability Consultants, dated Nov 2015. TLV range of £350,000 to £500,000 per Ha. However, DSP do not differentiate between greenfield or brownfield values.

 Calderdale Council Local Plan and CIL Viability Evidence – undertaken by GVA, dated Oct 2015. Greenfield TLV of £383,005 per Ha.

 Gedling Borough Council Local Plan Viability Assessment – undertaken by NCS, dated Mar 2016. Greenfield TLV £786,238 per Ha.

4.7 Of the 7 studies referenced, the suggested TLV ranges from £247,000 per Ha up to £988,400 per Ha (albeit at the high end of the range reflecting a net Ha and being only applicable to a high value area). Given the broad range identified there is therefore little discernible pattern on which to draw any firm conclusions.

4.8 That said, what is clear is that the studies agree that the best approach is to consider the Current Use Value (“CUV”) of the land (which for greenfield is typically an agricultural use) and then from this apply a level of premium to incentivize the landowner to release the land from development. This, in line with the Harman Review guidance, is deemed a more appropriate approach than basing TLV’s on actual land transactions, due to the wide fluctuations that can occur in land deals (transactional evidence is therefore deemed more appropriate to use as a sense check). The major disagreement across the sample is therefore what the level of premium should be above the CUV, which fluctuates significantly between the studies.

116

4.9 In conclusion, we would stress that each site should still be taken on its merits when looking to identify a suitable TLV, as this will fluctuate on a site by site basis depending on the abnormal costs of development, impact of Council policies etc. However there are some broadly accepted ‘norms’ or range of norms.

4.10 It should also be noted that quantum is likely to play a role on larger schemes (i.e. to reflect the fact a developer would effectively be buying in ‘bulk’, the TLV for large sites (in particular strategic sites) should be discounted on a price per acre basis). Furthermore, for the larger sites the developer / house builder is likely to draw down the land as and when needed, which would have a positive impact on the finance costs.

4.11 Having taken into account all of the above we are of the view that it is appropriate to adopt a range of TLVs for the purposes of this report. In this case, we have looked to apply a different range depending on whether the site is located within what is perceived to be a low, medium or high value area.

4.12 In our analysis we have assumed a low value area supports values of £1,500 per sq m, a medium value area values of £1,750 per sq m, and a high value area £2,250 per sq m.

4.13 Based on these adjusted definitions, and in light of the evidence identified we have adopted the following TLVs for greenfield sites across Doncaster:

Area Definition (£ per sq m) TLV range (£ per gross Ha)

Low value area £1,500 £200,000 Medium value area £1,750 £270,000 High value area £2,250 £350,000

4.14 However, please note for significantly larger strategic sites, as indicated above, we would expect a level of discount from the above figures to reflect quantum.

4.15 On the basis that agricultural land typically attracts values in the region of £10,000 to £25,000 per Ha (and is less sensitive to location than other site types), the above TLV range represents, broadly, an uplift in value of circa 15 to 20 times the current use value of an assumed agricultural use. When considered in this context the above TLV’s are considered to provide an attractive uplift in value for the landowner.

117

4.16 As a ‘sense check’ only we have also subsequently reviewed market transactions to determine whether these suggested figures are appropriate, albeit acknowledging that market transactions have their limitations when being assessed in the context of benchmark land values, as per the comments above (see above 3.9 onwards).

4.17 We have identified the following sample of greenfield transactions which have taken place after Jan 2014, for sites in excess of 0.25 Ha

Table 1 – Greenfield transactional evidence

Date Address Price Gross £ per Ha (Ha)

Jun-14 Athelstane Crescent, Edenthorpe DN3 £1,988,387 2.09 £951,381 Apr-15 Manor Farm, Bawtry Road, Bessacarr DN4 £3,800,000 6.02 £631,229 Jul-15 Land at Duftons Close, Conisbrough DN12 £95,000 0.47 £202,128 Jul-15 Land at Ridge Bank Lane, Woodlands DN6 £135,000 0.67 £201,493 Sep-15 Dixon Rd/Thompson Av, Edlington DN12 £600,000 2.47 £242,915 Nov-15 Land to rear 88 Thorne Rd, Edenthorpe £175,000 0.44 £579,545 DN3 Nov-15 Doncaster Rd, Denaby Main DN12 £2,352,000 6.75 £348,444 Dec-15 Land to west of Hurst Lane, Auckley DN9 £1,100,000 11.86 £92,749 Jan-16 Land at Chase Park, Woodlands DN6 £1,848,000 2.50 £739,200 Apr-16 Spa Terrace,Askern DN6 (UNSOLD at Ask.£110,000 0.40 £275,000 auction Apr 2016)

4.18 Please note, when researching greenfield sites we initially attempted to identify the price paid by volume housebuilders for current ‘live’ housing schemes across Doncaster Borough. However, this proved extremely difficult as often the price paid was not available, or the deal was complicated (with phased payments over a number of years, for example) to the extent where we were unable to identify a single ‘rate per Hectare’ paid.

4.19 Of the 10 transactions identified, there is a price range of £92,749 to £951,381 per gross Ha, with an average of £426,408 per gross Ha. This wide range highlights the difficulties associated with assessing land transactions, as there is no discernible pattern to the prices paid.

4.20 That said, the adopted TLV’s, as shown above in 4.13, do fall within the broad identified range. In this regard, they are not considered to be significantly out of kilter with the market and therefore are appropriate for the purposes of the appraisal testing.

118

5.0 Evidence – brownfield

5.1 ‘Brownfield’ refers to sites that either currently have existing buildings, or previously had buildings on the site (and have now been cleared). In other words, a brownfield site is classed as land that is either currently developed or has been developed in the past.

5.2 At the Stakeholder workshop, and in the 3 subsequent completed questionnaires received, a distinction was made between:

- Brownfield sites which are cleared and immediately ready to develop.

- Brownfield sites which have existing buildings on-site which require demolition (and may require vacating if occupied).

5.3 The general view given by the Stakeholders was that brownfield sites with existing buildings in situ may have a significantly higher TLV than brownfield sites which are already cleared. We have subsequently considered this when analysing evidence.

5.4 We have identified the following TLVs for cleared brownfield sites, identified from viability appraisals received from applicants. For the purposes of this exercise we have looked at TLVs for brownfield sites again across the North East of England, Yorkshire and the East Midlands. Whilst a large geographical area this gives a good indication of how TLVs for brownfield sites are assessed in other regions.

 Low value area South Yorkshire – former industrial works, gross size 1.64 Ha, proposal for 60 dwellings. Abnormals circa £714,000 per net Ha. Average house price £1,562 per sq m. June 14 a national developer submitted a viability appraisal, indicating a TLV equivalent to £274,556 per gross Ha.

 Medium value area West Yorkshire – former industrial facility. Gross area 1.93 Ha, proposal for 68 dwellings. Abnormals circa £609,000 per net Ha. Average house price £1,725 per sq m. Jan 15 a local agent submitted a viability appraisal, indicating a TLV equivalent to £839,302 per gross acre.

119

 Medium value area Lincolnshire – former industrial site. Gross area 0.40 Ha, proposal for 12 dwellings. Abnormals circa £447,000 per net Ha. Average house price £1,748 per sq m. Mar 16 a regional development consultant submitted a viability appraisal, indicating a TLV equivalent to £599,030 per gross Ha.

 High value area South Yorkshire – cleared former school. Gross area 2.23 Ha, proposal for 52 dwellings. Abnormals circa £228,000 per net Ha. Average house price £2,107 per sq m. May 14 a national housebuilder submitted a viability appraisal, indicating a TLV equivalent to £1,345,372 per gross Ha.

 High value area Derbyshire – former airfield. Gross site area 39.35 Ha, proposal for 367 dwellings. Abnormals circa £259,000 per net Ha. Average house price £2,115 per sq m. May 14 a regional development consultant submitted a viability appraisal, indicating a TLV equivalent to £170,632 per gross Ha.

 Medium value area Lincolnshire – cleared former industrial. Gross area 1.18 Ha, proposal for 23 dwellings. Abnormals circa £115,000 per net Ha. Average house price £1,763 per sq m. Apr 16 a regional firm of chartered surveyors submitted a viability appraisal, indicating a TLV equivalent to £220,866 per gross Ha.

5.5 As the sample is relatively small it is difficult to draw any firm conclusions. However, what the data appears to show is a wide variance (£170,000 to £1,345,000) in the level of TLV for cleared brownfield sites, depending on the nature of the site (e.g. level of abnormals, size of site, former use, housing values achievable etc).

5.6 For brownfield sites with existing buildings in situ (some occupied, some not), we have identified the following TLVs:

 Low value area South Yorkshire – industrial buildings on site, gross size 1.50 Ha, proposal for 17 dwellings. Abnormals circa £43,000 per net Ha. Average house price £1,423 per sq m. Nov 15 a local firm of architects submitted a viability appraisal, indicating a TLV equivalent to £26,642 per gross Ha.

120

 Low value area West Yorkshire – industrial buildings. Gross area 0.30 Ha, proposal for 14 dwellings. Abnormals circa £791,000 per net Ha. Average house price £1,557 per sq m. Jun 16 a local agent submitted a viability appraisal, indicating a TLV equivalent to £1,455,521 per gross acre.

 Medium value area Derbyshire – industrial estate. Gross area 18 Ha, proposal for 600 dwellings. Abnormals circa £164,000 per net Ha. Average house price £1,665 per sq m. Mar 15 a national firm of chartered surveyors submitted a viability appraisal, indicating a TLV equivalent to £511,215 per gross Ha.

 Medium value area Nottinghamshire – petrol station. Gross area 0.57 Ha, proposal for 16 dwellings. Abnormals not stated. Average house price £1,882 per sq m. Jul 15 a local developer submitted a viability appraisal, indicating a TLV equivalent to £480,279 per gross Ha.

 Medium value area West Yorkshire – industrial estate. Gross site area 1.43 Ha, proposal for 65 dwellings. Abnormals circa £537,000 per net Ha. Average house price £1,948 per sq m. Jan 15 a local chartered surveyor submitted a viability appraisal, indicating a TLV equivalent to £665,000 per gross Ha.

 Medium value area Lincolnshire – public house. Gross site area 0.34 Ha, proposal for 32 dwellings. Abnormals not stated. Average house price £2,000 per sq m. Feb 15 a local chartered surveyor submitted a viability appraisal, indicating a TLV equivalent to £1,853,250 per gross Ha.

 Medium value area West Yorkshire – public house. Gross site area 0.32 Ha, proposal for 20 dwellings. Abnormals circa £50,000 per net Ha. Average house price £2,022 per sq m. Jan 15 a regional planning consultant submitted a viability appraisal, indicating a TLV equivalent to £879,000 per gross Ha.

 Medium value area West Yorkshire – industrial estate. Gross site area 1.20 Ha, proposal for 48 dwellings. Abnormals circa £857,000 per net Ha. Average house price £2,133 per sq m. Mar 16 a local chartered surveyor submitted a viability appraisal, indicating a TLV equivalent to £603,000 per gross Ha.

121

 High value area Nottinghamshire – public house. Gross site area 0.12 Ha, proposal for 16 dwellings. Abnormals not stated. Average house price £2,181 per sq m. Mar 16 a local chartered surveyor submitted a viability appraisal, indicating a TLV equivalent to £1,676,574 per gross Ha.

 High value area West Yorkshire – office buildings. Gross site area 0.68 Ha, proposal for 22 dwellings. Abnormals circa £264,000 per net Ha. Average house price £2,530 per sq m. Mar 16 a national firm of chartered surveyors submitted a viability appraisal, indicating a TLV equivalent to £1,630,000 per gross Ha.

5.7 Again, the evidence points to a large variance in TLV depending on the specific nature of the site, and in particular the nature of the existing buildings (in this case the range is £26,642 to £1,853,250 per gross Ha). There is therefore no discernible pattern in the TLVs for brownfield sites.

5.8 In short, we therefore conclude that the variance between CUVs for brownfield sites across different locations is considered to be higher than for greenfield sites. This may be partly due to agricultural land values remain relatively consistent across regions, therefore the underlying CUV of a greenfield site will not be subject to more minimal change across low, medium and high value areas. In contrast, the CUV’s for brownfield sites are likely to vary more significantly. For example, a prime serviced industrial site (with good links to the motorway network) may have a CUV of £750,000 - £1,000,000 per Ha. A tertiary industrial site, with poor access to the motorways, may only have a CUV of sub £250,000 per Ha. Whilst perhaps an extreme example, it highlights the potential for variance in brownfield site TLVs.

5.9 Equally, our experience is that the AUV is likely to play a bigger role on brownfield rather than greenfield sites. For example, a brownfield site in an old industrial area may be viewed as having potential for long term regeneration, therefore other employment uses (offices, retail, leisure etc) may need to be factored into the TLV (which may have a significantly higher value). This, in some cases, this may significantly increase the TLV for a brownfield site.

122

5.10 For these reasons, the method of establishing a CUV and then adding some level of incentive uplift (which in our experience tends to be an uplift of between 10% and 30%), alongside an assessment of any credible AUV, can produce a wide range of TLVs for brownfield sites. It is therefore difficult to provide 1 or 2 overall averages across an area for brownfield sites, because the CUV / AUV of each site will need to be rigorously assessed before any meaningful conclusion is made.

5.11 In this regard, we have considered the conclusions drawn on suitable brownfield TLVs from other area wide studies (please note we have excluded the Peter Brett reports referenced above, on the basis that these reports do not give explicit figures for brownfield sites, they only provide a combined average for greenfield and brownfield sites. Furthermore, we have not included the GVA reports as it is unclear what TLVs have been applied to brownfield sites):

 Wakefield Metropolitan District Council CIL Viability Study – undertaken by DTZ, dated Feb 2014. Adopts the following benchmark land values: low value area £679,537 per net Ha, medium value area £834,000 per net Ha and high value area £988,400 per net Ha. It is stressed that these figures are based on net areas, therefore is difficult to compare with our study which assumes gross areas. Also, the DTZ does not differentiate between greenfield or brownfield values.

 Calderdale Council Local Plan and CIL Viability Evidence – undertaken by GVA, dated Oct 2015. Greenfield TLV of £383,005 per Ha.

123

 Gedling Borough Council Local Plan Viability Assessment – undertaken by NCS, dated Mar 2016. Brownfield TLV £991,238 per Ha.

 Leicester City Council CIL Viability Study Update – undertaken by HDH Planning in December 2014. For brownfield sites, the report assumes an average TLV range (based on the CUV + 20%) of £420,000 to £528,000 per gross Ha (depending on the nature of the current use and location).

5.12 Taking the mid-point of the ranges (where applicable) the above shows an average TLV for a brownfield site of £600,207 per gross Ha.

5.13 It is difficult to draw any firm conclusions from the data assessed and would stress it is less reliable to establish an average TLV for brownfield sites than greenfield sites due to the potential variance in CUVs and greater impact of locational factors.

5.14 Please note, the evidence identified is related to other market locations, and as indicated above when assessing brownfield sites locational factors are even more important when assessing appropriate TLVs. We have subsequently assessed market transactions to provide a further insight into the local market.

5.15 As indicated above, in addition to the ‘direct’ TLV evidence identified above, we have also looked to analyse actual land transactions as part of our considerations, albeit using this only as a general ‘sense check’.

5.16 We have identified the following sample of brownfield transactions which have taken place after Jan 2014, for sites in excess of 0.25 Ha:

Table 2 – Brownfield transactional evidence

DATE ADDRESS PRICE GROSS £ PER HA (Ha) Apr-14 Land at Carr House Road, Doncaster DN1 £450,000 3.32 £135,542 Jun-14 Sandtoft Indust Est, Belton DN10 £200,000 0.81 £247,219 Jun-14 Belton Road, Sandtoft DN8 £900,000 5.10 £71,417 Jul-14 Former Youth Club, Kirkby Avenue DN5 £85,000 0.25 £340,000 Nov-14 Kirk Sandall Indust Est, Doncaster Rd DN3 £1,100,000 2.13 £516,432 Dec-14 Holme Wood Lane, Armthorpe £9,018,960 15.70 £574,090 Dec-14 Ancient Lane, Hatfield Woodhouse DN7 £290,000 0.53 £547,170 Dec-14 Thorne Hall & Depot, Ellison St, Thorne DN8 £151,000 0.57 £264,912 Feb-15 Former highway depot,Station Rd,Misterton DN10 £227,000 0.35 £648,571 Feb-15 Almholme Grange, Almholme Lane, Arksey £260,000 0.81 £320,988 (UNSOLD at auction Feb 2015) Mar-15 Park Drive, Sprotbrough £150,000 0.96 £63,234 Mar-15 Lakside Boulevard, Doncaster DN4 £4,350,000 5.55 £783,784

124

Apr-15 Former Rossington Colliery,New RossingtonDN11 £2,773,438 2.91 £953,071 Jul-15 North side Stevens Rd, Balby DN4 £1,400,000 2.51 £557,769 Aug-15 Land west Carr House Road, Doncaster DN2 £325,000 0.45 £722,222 Aug-15 Former Belle Vue Stadium, Bawtry Rd DN4 £2,135,200 3.58 £596,425 Aug-15 Ashmount Club 50, High Road, Balby DN4 £200,000 0.30 £666,667 Sep-15 Snape Lane, Harworth DN11 £288,000 0.81 £355,556 Nov-15 Westwood Business Park, Belton Rd DN8 £570,000 1.21 £471,074 Nov-15 Briars Lane, Stainforth DN7 £1,200,000 1.80 £666,667 Feb-16 Glebe House, Haynes Road, Thorne DN8 £200,000 0.40 £500,000 Mar-16 Former Reservoir, Green Lane, Scawthorpe DN5 £736,619 0.86 £856,534 Apr-16 Sandall Lane, Kirk Sandall DN3 £240,000 0.60 £400,000

5.17 As per our greenfield land assessment, when researching brownfield sites we initially attempted to identify the price paid by volume housebuilders for current ‘live’ housing schemes across Doncaster Borough. Again, this proved extremely difficult as often the price paid was not available, or the deal was complicated to the extent where we were unable to identify a single ‘rate per Hectare’ paid.

5.18 Of the 23 transactions a range of £63,234 to £953,071 per gross Ha is shown (with an average of £489,537 per gross Ha). Again, it is difficult to draw any robust conclusions given the wide variance in price paid within the sample.

5.19 Again, and as stressed above, the added complication for brownfield sites is that there is likely to be a significant differential in the price paid linked to whether the site is currently cleared or currently occupied with existing buildings (this is not an issue for greenfield sites, which are assumed to be undeveloped). This is because the ‘current use value’ of the land must be factored into the TLV. For example, a cleared, redundant former industrial site may effectively have a nil current use value because it is no longer considered to be economic for industrial purposes (any value would be based purely on development potential). Alternatively, a brownfield site which has an existing, occupied office building in situ will clearly have a different value to a cleared site. Any development potential would be subject to generating a land value which exceeds the current office use. In this respect, when testing brownfield sites there is the potential for a significant variance in the TLV (particularly when compared to greenfield sites, which are assumed to have the same underlying agricultural existing use).

5.20 Taking all of the above into account, for the purposes of the viability testing we have looked to differentiate between cleared, brownfield sites and occupied, brownfield sites, an approach which was supported by the stakeholder engagement. For each, we have therefore adopted a different TLV, as follows:

125

Cleared, brownfield - £185,000 per Ha

Occupied, brownfield - £370,000 per Ha

5.21 It is acknowledged that the reality is there would likely be variations to the adopted figures, depending on the site’s location, current use, size, condition etc. In this regard, and particularly for brownfield sites, it is stressed that a ‘site by site’ assessment of the TLV is still appropriate when considering viability. In this respect the above figures should not be automatically applied to a brownfield site when determining its viability. However, for the purposes of a high-level, area wide study the above figures are considered to be appropriate for appraisal testing.

126

Appendix 3

Map of the high, medium and low value areas

127

Appendix 3. Map of the high, medium and low value areas