affordable housing provision and developer contributions

Braintree Final Report – November 2009

Affordable Housing Provision and Developer Contributions in the District of Braintree

A Report to Braintree District Council,

Final Report

November 2009

THREE DRAGONS


CONTENTS

Executive Summary p3

Introduction p6

Methodology p11

High Level Testing p14

Land supply, small sites and the use of commuted sums p27

Case Study viability analysis p33

Main findings and conclusions p39

Appendices

Notes of workshop p42

Three Dragons Model Key assumptions p 46

Rural and urban areas p 50


EXECUTIVE SUMMARY

1. Three Dragons was appointed by Braintree District Council to undertake an affordable housing viability study across the district to assess the viability of market schemes against a range of affordable housing options and economic scenarios. The following report suggests maximum deliverable affordable housing targets for the district and tests the sensitivity of those targets against a range of development mixes, thresholds and build cost factors. The study was overseen by a team of officers from the Council’s Planning and Housing Departments and drew on data on site supply and planning policy provided by the authority. Key assumptions and methodology were verified at a stakeholder workshop attended by developers, landowners and RSLs.

Methodology

2. Viability was assessed using a residual development appraisal model. The model deems the value of a site to be the scheme’s gross revenue minus the cost of development. In the first stage, build costs, including professional fees, marketing, finance costs and overhead are deducted from the total scheme revenue. A developer margin is also applied, leaving a gross residual site value. S106 contributions are deducted from the gross residual site value to give a net residual value.

3. The next stage of assessment considers existing or alternative use values for land. As build costs, S106 contributions and affordable housing levels on a site increase, the net residual value of a scheme falls. If the net residual value falls below a site’s alternative use value, it is unlikely to come forward.

4. Using a notional 1 hectare site, a range of scenarios for affordable housing delivery were tested against a number of residential development mixes and densities. Seven sub-markets within Braintree District were identified. There was significant variation in residual value between housing sub-market areas, with urban areas consistently producing lower values than rural areas in the district.

Findings

5. There is a significant variation in residual value between the market value areas

6. Values in the urban areas of Braintree, Halstead and Hedingham and Witham are consistently lower than those in rural areas elsewhere in the district

7. At 100% market housing 60 dph produces the highest residual values, ranging from £5.23m per hectare in Three Fields to £2.09m per hectare in Witham. As the proportion of affordable housing increases the most profitable density falls, to 45 dph at 25-40% affordable housing and to 35 dph at 50% affordable housing.

At 30% affordable housing without grant at 45 dph a per hectare residual value of £2.75m is achieved in Three Fields, falling to £750,000 in Witham.

At 35% affordable housing Three Fields, Eastern Fringe and Great Notley and Rayne achieve values per hectare ranging from £1.39-2.41m at 45 dph , but residuals are below £1m in Northern Fringe, Braintree, Halstead and the Hedinghams and Witham.


At 40% affordable housing only in Three Fields, Eastern Fringe and Great Notley and Rayne are residuals of over £1m achieved.

At 50% affordable housing Three Fields achieves a residual of £1.39m per hectare. Nowhere else achieves more than £1m per hectare.

8. In the higher value market areas of Three Fields, Eastern Fringe, and Great Notley and Rayne residual values of £1m were achieved for 30% affordable housing when an increased S106 package (£20,000) was introduced, but this was not the case in the lower value areas. However, assuming a combined cost of £40,000 per unit for S106 and Code for Sustainable Homes Level 6, the three higher value areas only produced residual values over £1m at 20% affordable housing.

9. While lower value sub market areas produced positive residual values at 10% and 20% affordable housing in the same scenarios, it is unlikely that those sites could compete with existing/alternative use values.

10. The availability of grant was found to have a significant beneficial impact on scheme viability. Sites were tested assuming a grant of £45,000 per social rented unit and £20,000 for New Build Homebuy. Grant funding had a greater impact on viability in lower value areas.

Site supply

11. In Braintree District 50% of new dwellings supplied in 2009 were developed on sites smaller than the 15 unit threshold. Small sites were found to be particularly important in rural areas where there are currently no outstanding planning consents for schemes of 15 or more dwellings, In urban areas 81% of all outstanding planning consents are on schemes of 15 or more units and a further 10% are on sites of 10-14 units.

12. The economics of development of smaller sites was explored through a series of case studies which demonstrated that in high value rural areas residual values of upto £1m per hectare were being achieved even on very small sites at upto 35% affordable housing. Residuals per hectare are lower in lower value areas but there is no clear evidence to show that they decrease on a per hectare basis with size of site.

Policy conclusions

13. Target percentage of affordable housing; The report proposes two possible approaches:

Option one - a single target for the whole district. Given the relatively low values in some of the urban areas, this would need to be set relatively conservatively and a target of 30% would seem appropriate;

Option two - a ‘split target’ which recognises the very significant variation in house prices and residual values between urban and rural areas. The split target would be 40% in rural areas and 30% in the urban areas of Braintree, Halstead, the Hedinghams and Witham


Under both options, the proposed SUEs at Braintree and Witham should be planned for on the basis of a 30% target, subject to up to date scheme specific viability appraisals

14. Thresholds: In the urban parts of the district, the evidence on site supply indicates that the national indicative minimum threshold of 15 dwellings is appropriate. However, the profile of site supply in the rural areas (coupled with the high levels of need) indicates that adopting a low threshold is justified.

15. A threshold of 5 dwellings would capture about a third of the supply in rural areas but to really increase delivery of affordable housing would mean introducing a zero threshold, Of course, for schemes of 1 and 2 dwellings, on site provision is not possible (simply because of the numbers involved) and the operation of a zero threshold would need to be linked to the collection of commuted sums and a strategy in place to ensure the money collected was spent to increase the supply of rural affordable housing.


1 INTRODUCTION

Review of project aims

1.1 Braintree District Council appointed Three Dragons to undertake an affordable housing and residential economic viability study covering the local authority area. The work was overseen by a team drawn from the Planning and Housing Departments of the Council.

1.2 The broad aims of the study were to:

Undertake an assessment of viability for market housing schemes set against a range of potential affordable housing options and a range of growth scenarios.

Recommend and suggest maximum viable and deliverable affordable housing targets

and test the sensitivity of these targets to tenure splits and a range of development thresholds.

Need for Affordable Housing

The 2008 Strategic Housing Market Assessment carried out by Fordhams found a net annual need for 697 affordable units using the established model for estimating need. The report suggested that the Council should move to 40% of the units being for intermediate tenure if the overall target for affordable housing changed from 30% to 40%An update to the SHMA was carried out in 2009 and this identified problems with mortgage availability which meant that, whilst the downturn in the housing market has improved affordability, accessibility is limited. That Update did not however propose any amendment of the affordable housing target suggested in the 2008 SHMA.

Policy context - national

1.3 This study focuses on the percentage of affordable housing sought on mixed tenure sites and the size of site from above which affordable housing is sought (the site size threshold). National planning policy, set out in PPS3 makes clear that local authorities, in setting policies for site size thresholds and the percentage of affordable housing sought, must consider development economics and should not promote policies which would make development unviable.

PPS3: Housing (November 2006) states that:

“In Local Development Documents, Local Planning Authorities should:

Set out the range of circumstances in which affordable housing will be required. The national indicative minimum site size threshold is 15 dwellings. However, Local Planning Authorities can set lower minimum thresholds, where viable and practicable, including in rural areas. This could include setting different proportions of affordable housing to be sought for a series of site-size thresholds over the plan area. Local Planning Authorities will need to undertake an informed assessment of the economic viability of any


thresholds and proportions of affordable housing proposed, including their likely impact upon overall levels of housing delivery and creating mixed communities”. (Para 29)

1.4 The companion guide to PPS31 provides a further indication of the approach which Government believes local planning authorities should take in planning for affordable housing. Paragraph 10 of the document states:

“Effective use of planning obligations to deliver affordable housing requires good negotiation skills, ambitious but realistic affordable housing targets and thresholds given site viability, funding ‘cascade’ agreements in case grant is not provided, and use of an agreement that secures standards.” (our emphasis)

Policy context – Eastern Region

1.5 The East of England Plan (May 2008) does not set a specific target for affordable housing provision at district level although a regional target of 35% is suggested in Policy H2 (below). The supporting text (para 5.8) comments that targets above 35% may be justified in more pressured areas. Across the Region 18% gross affordable housing completions were achieved in 2007/08 (Table 8.5 East of England Plan AMR 2007/08)

Policy context – District of Braintree

1.6 The most recent affordable housing policy for Braintree District is given in adopted Local Plan Policies RLP 5 and 6 and the accompanying affordable housing SPD. This seeks 30% affordable housing on sites of 15 or more units. The proposed tenure split for the affordable housing consists of 80% social rent and 20% intermediate housing.

1.7 In 2007/08 142 affordable homes were completed in Braintree, (22.6% of net completions). Performance in Braintree is therefore above the East of England average. Table 1.1 below shows performance since 2004/05. Grant funding from the Housing

1 CLG, Delivering Affordable Housing, November 2006


Corporation/HCA was required in order to deliver this level of affordable housing provision.


Table 1.1 Affordable housing provision 2004/05 to 2007/08

Braintree District Council

Number of affordable dwellings Completed

Percentage of net dwellings completed

2007/08 142 22.6

2006/07 188 28.6

2005/06 69 14.0

2004/05 91 13.1

Source Braintree District Council data (AMR)

Research undertaken

1.8 There were four main strands to the research undertaken to complete this study:

Discussions with a project group of officers from the local authority which informed

the structure of the research approach;

Analysis of information held by the local authority, including that which described

the profile of land supply;

Use of the Three Dragons Toolkit to analyse scheme viability (and described in

detail in subsequent chapters of this report);

A workshop held with developers, land owners, their agents and representatives

from a selection of Registered Social Landlords active in Braintree. A full note of the

workshop is shown in Appendix 1.

Structure of the report

1.9 The remainder of the report uses the following structure:

Chapter 2 explains the methodology we have followed in, first, identifying sub

markets and, second, undertaking the analysis of development economics. We

explain that this is based on residual value principles;

Chapter 3 provides analysis of residual values generated across a range of different

development scenarios (including alternative percentages and mixes of affordable

housing) for a notional 1 hectare site.


Chapter 4 considers options for site size thresholds. It reviews national policy and

the potential future land supply and the relative importance of small sites. The

chapter considers practical issues about on-site provision of affordable housing on

small sites and the circumstances in which collection of a financial contribution

might be appropriate (and the principles by which such contributions should be

assessed);

Chapter 5 identifies a number of case study sites. For each site type, there is an

analysis of the residual value of the sites and this is compared with their existing

use value.

Chapter 6 summarises the evidence collected through the research and provides a

set of policy recommendations.


2 METHODOLOGY

Introduction

2.1 In this chapter we explain the methodology we have followed in, first, identifying sub markets (which are based on areas with strong similarities in terms of house prices) and, second, undertaking the analysis of development economics. The chapter explains the concept of a residual value approach and the relationship between residual values and existing/alternative use values.

Viability – starting points

2.2 We use a residual development appraisal model to assess development viability. This mimics the approach of virtually all developers when purchasing land. This model assumes that the value of the site will be the difference between what the scheme generates and what it costs to develop. The model can take into account the impact on scheme residual value of affordable housing and other s106 contributions.

2.3 Figure 2.1 below shows diagrammatically the underlying principles of the approach. Scheme costs are deducted from scheme revenue to arrive at a gross residual value. Scheme costs assume a profit margin to the developer and the ‘build costs’ as shown in the diagram include such items as professional fees, finance costs, marketing fees and any overheads borne by the development company.

2.4 The gross residual value is the starting point for negotiations about the level and scope of s106 contribution. The contribution will normally be greatest in the form of affordable housing but other s106 items will also reduce the gross residual value of the site. Once the s106 contributions have been deducted, this leaves a net residual value.


Figure 2.1 Theory of the Section 106 Process

2.5 Calculating what is likely to be the value of a site given a specific planning permission, is

only one factor in deciding what is viable.

2.6 A site is extremely unlikely to proceed where the costs of a proposed scheme exceed the revenue. But simply having a positive residual value will not guarantee that development happens. The existing use value of the site, or indeed a realistic alternative use value for a site (e.g. commercial) will also play a role in the mind of the land owner in bringing the site forward and thus is a factor in deciding whether a site is likely to be brought forward for housing.

2.7 Figure 2.2 shows how this operates in theory. Residual value falls as the proportion of affordable housing increases. At some point (here ‘b’), alternative use value (or existing use value whichever is higher) will be equal to scheme value. If there is a reasonable return to the land owner at point ‘b’ (i.e ‘b’ reflects best possible current use value (alternative or existing) and there is a sufficient return, then the scheme will come forward. At point ‘c’, affordable housing will make the site unviable. At ‘a’ the scheme should be viable with affordable housing. The diagram does not assume grant. .


Figure 2.2 Affordable housing and alternative use value

2.8 The analysis we have undertaken uses a Three Dragons Viability model. The model is

explained in more detail in Appendix 2, which includes a description of the key assumptions used.


3 HIGH LEVEL TESTING

Introduction

3.1 This chapter of the report considers viability for mixed tenure residential development for a number of different proportions and types of affordable housing. The analysis is based on a notional 1 hectare site and has been undertaken for a series of market value areas across the district. This chapter reports the results of this appraisal and explores the relationship between the residual value for the scenarios tested and existing/alternative use values.

Market value areas

3.2 Variation in house prices will have a significant impact on development economics and the impact of affordable housing on scheme viability.

3.3 We undertook a broad analysis of development across the housing market, using HM Land Registry data to identify market value areas in the District. The areas are defined by reference to postcode sectors/wards and their house prices and provide the basis for a set of indicative new build values as at mid 2008. The purpose of this analysis is to help establish a broad starting point for target setting in the light of the general relationships between development revenues and development costs.

3.4 We identified 7 housing market sub-areas, Three Fields, Eastern Fringe, Great Notley and Rayne, Northern Fringe, Braintree town, Halstead and the Hedinghams and Witham. Full details of market value areas and house prices are given in Appendix 2. A map of house price areas is set out in Figure 3.1

3.5 We test at 2008 house prices, but we also model the impact of a 10% increase in house prices and a 10% and 20% reduction in house prices. In July 2009 house prices in Essex had fallen by an average of 13.4% from their mid-2008 level (source Land Registry).


Figure 3.1 Braintree District Market Value Areas 2008

Testing assumptions (notional one hectare site)

3.6 For the viability testing, we defined a number of development mix scenarios, using a range of assumptions agreed with the councils. The scenarios were based on an analysis of typical development mixes and were discussed at the stakeholder workshop.

3.7 We modelled at densities of 30 dph, 45 dph, 60 dph and 80 dph. Mixes were as given in table 3.1 below.


Table 3.1. Densities and mixes

3.8 We calculated residual site values for each of these (base mix) scenarios in line with a further set of tenure assumptions. These were 0%, 10%,15%, 25%; 30%; 35%; 40%; and 50% affordable housing. These were tested at 80% Social Rent and 20% New Build HomeBuy in each case. For the New Build HomeBuy, the share purchase was assumed to be 45% with 2.75% being charged on the unsold equity. All the above assumptions were agreed with the authority. In order to allow for land acquisition costs residual values achieved were then discounted by 10% and it is this figure which is referred to in all the following analysis.

Other S106 contributions

3.9 For the base modelling we have undertaken (and unless shown otherwise) we have assumed that other planning obligations have a total cost of £5,000 per unit. We have also tested the impact of higher levels of S106 contributions. These are likely to be particularly applicable to proposed growth locations such as at Panfield Lane Braintree Hatfield Rd Witham and Forest Rd Witham

3.10 We have also modelled an illustrative £20,000 planning obligations package. This is typical of the higher level of planning obligations required for major sites and quoted to us at developer workshops across the country (values have ranged from £15,000 per unit to £30,000 per unit.)

Code for Sustainable Homes

3.11 During the life of the Plan it is expected that developers will be required to provide housing at significantly higher Code for Sustainable Homes levels, moving to Level 4 by 2013 and to Level 6 by 2016. (CSH will be enforced through a combination of amendment to the building regulations which will be compulsory and planning


obligations which will be at the discretion of the local planning authority). We assume a base additional cost of £5,000 per dwelling to move to Code Level 4.

3.12 Costs for achieving higher CSH levels vary depending on the size and location of the site and the method of achieving carbon neutrality chosen. Research carried out by Cyril Sweett for dclg in 2008 “(Cost Analysis of the Code for Sustainable Homes Final Report”) indicated that the cost of moving to higher code levels ranges from £19,000 to £47,000 per unit. According to that report it is expected that costs of compliance will fall over time and by 2016 they could have fallen by between 16% and 25%.

3.13 We model an illustrative £20K increase in development cost per unit which at 45 dph equals £900,000 per hectare. We assume that more energy efficient new homes do not attract a “consumer premium”, ie there is no compensating increase in market value

Lifetime Homes

3.14 The cost of provision of Lifetime Homes is estimated by dclg at £500 per unit. This has been modelled within wider CSH and S106 costs.

Large greenfield sites – costs and viability issues

3.15 There are various reasons why very large sites (both green and brownfield sites) will be more costly to develop than urban sites and small rural sites (unless the latter are contaminated require the removal of existing buildings), The first of these relates to the issue of gross/nett developable area. Very large sites will normally have substantial requirements for public open space, provision of roads and supporting infrastructure and other land hungry requirements. We have checked the ratio of gross to nett developable area with both public and private sector landowners and with local planning authorities and are advised that a ratio of 50% nett developable area would not be an unreasonable assumption and individual sites could well have a lower nett developable area. Such costs are likely to be particularly applicable to sites of over 1,000 units but these issues will also need to be taken into account to a lesser extent when reviewing scheme viability of sites of between 100 and 1000 units.

3.16 The second issue relates to higher development costs for very large sites. Costs, and in particular, infrastructure costs for large greenfield sites are an issue. In examples quoted to us, these range from £100,000 to over £1m per hectare depending on a range of factors including the availability of utilities, drainage and topography. These costs will normally be over and above any Section 106 package or equivalent and hence again, it will be important to establish the precise loading of physical infrastructure on a site.

3.17 Against this must be set the potential land value for very large greenfield sites. These will typically be in agricultural use and the landowner would not expect to receive the same land value as for an urban site with either industrial or existing residential planning consent. Land value will be in part determined by the affordable housing target. We do not propose a different affordable housing target for large sites but recognise that the local authority and developer will need to take into account site specific circumstances when appraising large and complex sites.


Comparator Land Values

3.18 There is no specific guidance on the assessment of viability which is published by national government. In Section 2, we set out that we think viability should be judged against return to developer and return to land owner.

3.19 One approach is to take “current” land values for different development uses as a kind of ‘going rate’ and consider residual values achieved for the various scenarios tested against these. Table 3.2 (sourced from the Valuation Office) shows residential and industrial land values for the East of England July 2009. At the time of writing, there is no more up to date information publicly available.

Table 3.2 Land values per hectare East of England, Colchester and Chelmsford July 2009

Residential Land

Small sites Bulk Land Sites for flats and maisonettes

East of England £2.55m £2.67m £2.84m

Colchester £2.47m £2.35m £2.2m

Chelmsford £3.7m £3.7m £4.2m

Industrial Land

From To Average

East of England £0.33m £2.30m £0.936m

Colchester £0.40m £0.775m £0.625m

Chelmsford n/a n/a n/a

Figures in £millions per ha

Source: Valuation Office; Property Market Report, July 2009 3.20 The ‘benchmark’ of industrial land value can be important where land, currently in use

as industrial land, is being brought forward for residential development or where sites may be developed either for residential or employment use. A mark-up of 20-30% on industrial land value is often cited as a reasonable basis on which residential development might take place and we use this as a base figure for calculation when assessing the viability of affordable housing provision. A mark up of 30% on the average prices of East of England Industrial land brought forward for residential development gives an average value of £1.22m per hectare. The same calculation using figures from Colchester gives a land value of £810,000 per hectare. The value of industrial land in Colchester remained unchanged from January to July 2009, but the value of industrial land across the East of England fell by 9%. Feedback from Braintree District Council’s


Asset Management Team suggests that prices for industrial land in Braintree and Witham are broadly comparable with those in Colchester, but industrial land prices are slightly lower in Halstead.

3.21 It will be noted that these values are significantly below those which have been achieved for sites with residential planning consent in Colchester and Chelmsford. Published figures on land values for Braintree are not available. House prices in Braintree are broadly comparable with those in Colchester but 10-15% lower than those in Chelmsford. Residential land values in the East of England have fallen by 37% from their January 2008 peak and in July 2009 they were 7.8% below the January 2009 figure (Source Valuation Office). Residential land values in Colchester in July 2009 were an average of 9.0% below their January 2009 value.

3.22 Finally, we consider values for greenfield land, particularly important in the case or Sustainable Urban Extensions. At the recent development industry workshop land values for SUEs were estimated at £1m per hectare. For Rural Exception Sites £12,000 per plot was the maximum going rate. For conventional development land value should be returning to its historic norm of 30-40% of gross developable value. There was some evidence of mothballing of sites in the current market.

Density, value and the impact of affordable housing provision

3.23 We modelled a range of densities and affordable housing percentages. Affordable housing is modelled without grant. The results are shown in table 3.3 below Key findings from the tables are,

There is a significant variation in residual value between the market value areas

Values in the urban areas of Braintree, Halstead and Hedingham and Witham are consistently lower than those in rural areas elsewhere in the district

At 100% market housing 60 dph produces the highest residual values, ranging from £5.23m per hectare in Three Fields to £2.09m per hectare in Witham. As the proportion of affordable housing increases the most profitable density (marked in bold in table 3.3) falls, to 45 dph at 25-40% affordable housing and to 35 dph at 50% affordable housing.

At 30% affordable housing at 45 dph a per hectare residual value of £2.75m is achieved in Three Fields, falling to £750,000 in Witham.

At 35% affordable housing Three Fields, Eastern Fringe and Great Notley and Rayne achieve values per hectare ranging from £1.39-2.41m at 45 dph

Even at 50% affordable housing Three Fields achieves £1.39m per hectare for the 45 dph scenario


Table 3.3 Residual value – Base test - £s million per hectare

Affordable Housing %age Value per hectare £m

0% 20% 25% 30% 35% 40% 50%

35dph Three Fields £ 4.07

£ 3.51

£ 2.94

£ 2.38

£ 2.10

£ 1.82

£ 1.25

Eastern Fringe

£ 3.45

£ 2.94

£ 2.43

£ 1.92

£ 1.66

£ 1.40

£ 0.89

Great Notley and Rayne

£ 2.87

£ 2.41

£ 1.94

£ 1.48

£ 1.25

£ 1.02

£ 0.55

Northern Fringe

£ 2.22

£ 1.81

£ 1.40

£ 0.99

£ 0.79

£ 0.59

£ 0.18

Braintree

£ 2.09

£ 1.69

£ 1.29

£ 0.90

£ 0.70

£ 0.50

£ 0.10

Halstead and Hedinghams

£ 1.96

£ 1.58

£ 1.19

£ 0.80

£ 0.61

£ 0.41

£ 0.03

Witham

£ 1.84

£ 1.46

£ 1.08

£ 0.71

£ 0.52

£ 0.33

-£ 0.06


£ 4.11

£ 3.43

£ 2.75

£ 2.41

£ 2.07

£ 1.39

Eastern Fringe

£ 4.05

£ 3.43

£ 2.81

£ 2.19

£ 1.88

£ 1.57

£ 0.96


£ 3.35

£ 2.79

£ 2.23

£ 1.67

£ 1.39

£ 1.11

£ 0.55

Northern Fringe

£ 2.64

£ 2.14

£ 1.64

£ 1.14

£ 0.89

£ 0.64

£ 0.13

Braintree

£ 2.44

£ 1.95

£ 1.47

£ 0.98

£ 0.74

£ 0.50

£ 0.01


£ 2.27

£ 1.80

£ 1.33

£ 0.86

£ 0.63

£ 0.39

-£ 0.10

Witham

£ 2.11

£ 1.66

£ 1.20

£ 0.74

£ 0.51

£ 0.28

-£ 0.21



£ 4.43

£ 3.63

£ 2.84

£ 2.44

£ 2.04

£ 1.24

Eastern Fringe

£ 4.34

£ 3.62

£ 2.90

£ 2.17

£ 1.81

£ 1.45

£ 0.72


£ 3.54

£ 2.89

£ 2.23

£ 1.57

£ 1.24

£ 0.91

£ 0.26

Northern Fringe

£ 2.78

£ 2.18

£ 1.59

£ 1.00

£ 0.70

£ 0.40

-£ 0.23

Braintree

£ 2.52

£ 1.95

£ 1.37

£ 0.80

£ 0.52

£ 0.23

-£ 0.42


£ 2.30

£ 1.75

£ 1.19

£ 0.64

£ 0.36

£ 0.09

-£ 0.57

Witham

£ 2.09

£ 1.55

£ 1.01

£ 0.48

£ 0.21

-£ 0.07

-£ 0.72


£ 4.03

£ 3.14

£ 2.26

£ 1.81

£ 1.37

£ 0.49

Eastern Fringe

£ 3.88

£ 3.08

£ 2.28

£ 1.48

£ 1.08

£ 0.68

-£ 0.15


£ 3.04

£ 2.31

£ 1.58

£ 0.85

£ 0.49

£ 0.12

-£ 0.74

Northern Fringe

£ 2.38

£ 1.70

£ 1.03

£ 0.36

£ 0.02

-£ 0.39

-£ 1.21

Braintree

£ 2.06

£ 1.41

£ 0.77

£ 0.12

-£ 0.25

-£ 0.65

-£ 1.44


£ 1.77

£ 1.15

£ 0.53

-£ 0.12

-£ 0.50

-£ 0.88

-£ 1.64

Witham

£ 1.49

£ 0.89

£ 0.29

-£ 0.38

-£ 0.75

-£ 1.12

-£ 1.85

Sensitivity test results:

3.24 This section shows the effect of variation in house prices and in S106 and Code for Sustainable Homes requirements. Yellow shaded cells achieve a residual value of £1m or over. All examples are based on a 45 dph scheme.


House price change

Table 3.4 House price change – Residual value in £s million per hectare

Affordable Housing %

0% 10% 20% 25% 30% 35% 40% 50%

45dph +10%

Three Fields £ 5.45

£ 4.69

£ 3.93 £3.55

£ 3.17

£ 2.79

£ 2.41

£ 1.65

Eastern Fringe £ 4.63

£ 3.94

£ 3.25 £2.90

£ 2.55

£ 2.21

£ 1.86

£ 1.17


£ 3.86

£ 3.24

£ 2.61 £2.29

£ 1.98

£ 1.66

£ 1.35

£ 0.72

Northern Fringe £ 3.09

£ 2.52

£ 1.96 £1.68

£ 1.40

£ 1.11

£ 0.83

£ 0.27

Braintree £ 2.86

£ 2.32

£ 1.77 £1.50

£ 1.23

£ 0.95

£ 0.68

£ 0.14


£ 2.68

£ 2.15

£ 1.62 £1.36

£ 1.09

£ 0.83

£ 0.56

£ 0.03

Witham £ 2.51

£ 1.99

£ 1.48 £1.22

£ 0.96

£ 0.70

£ 0.44

-£ 0.09

45dph -10%

0% 10% 20% 25% 30% 35% 40% 50%


£ 3.54

£ 2.94 £2.64

£ 2.34

£ 2.04

£ 1.74

£ 1.14


£ 2.93

£ 2.38 £2.11

£ 1.84

£ 1.57

£ 1.29

£ 0.75


£ 2.84

£ 2.35

£ 1.86 £1.61

£ 1.37

£ 1.12

£ 0.87

£ 0.38


£ 1.77

£ 1.33 £1.11

£ 0.89

£ 0.67

£ 0.45

£ 0.01

Braintree £ 2.03

£ 1.60

£ 1.18 £0.97

£ 0.75

£ 0.54

£ 0.33

-£ 0.12


£ 1.88

£ 1.47

£ 1.05 £0.85

£ 0.64

£ 0.44

£ 0.23

-£ 0.22

Witham £ 1.74

£ 1.33

£ 0.94 £0.74

£ 0.54

£ 0.34

£ 0.14

-£ 0.32


45dph -20%

0% 10% 20% 25% 30% 35% 40% 50%

Three Fields £3.48 £2.96 £2.44 £2.18 £1.92 £1.66 £1.40 £0.88

Eastern Fringe £2.89 £2.42 £1.95 £1.71 £1.48 £1.24 £1.01 £0.54

Great Notley and Rayne £2.33 £1.91 £1.48 £1.27 £1.06 £0.85 £0.63 £0.21

Northern Fringe £1.76 £1.38 £1.01 £0.82 £0.63 £0.44 £0.25 £-0.15

Braintree £1.60 £1.23 £0.87 £0.69 £0.51 £0.28 £0.15 £-0.27

Halstead and Hedinghams £1.47 £1.12 £0.76 £0.59 £0.41 £0.24 £0.06 £-0.36

Witham £1.34 £1.00 £0.66 £0.49 £0.31 £0.14 £-0.03 £-0.45

3.25 A 10% increase in house prices produces residuals above £1m at 30% affordable housing in all locations except Witham.

3.26 The effect of a 10% decrease in house prices is to lower residuals. At 20% affordable housing they exceed £1m in all sub-areas.

3.27 The effect of a 20% reduction in house prices is to further reduce residual values. At 10% affordable housing they exceed £1m in all locations.

Impact of an increased S.106 requirement or Code for Sustainable Housing

3.28 In the earlier analysis, we have assumed a planning obligation package of £5,000 per dwelling. Table 3.4 shows residual values for a notional one hectare site at varying affordable housing percentages for a 45 dph scheme assuming a Section 106 contribution package of £20,000 per unit. At 45 dph this adds £675,000 to the cost of a scheme.

3.29 We have tested this level of planning obligations to assess the possible economic impact of such an approach which is not dissimilar to S106 packages quoted to us for major urban extensions. This should not be taken to indicate that the Council might wish to adopt this level of planning obligations package. The example assumes base house prices.


Table 3.5 Increased S106 obligations – Residual value in £s million per hectare

£20K s106 and CSH 0%

10% 20% 25% 30% 35% 40% 50%


£ 3.51 £ 2.82 £2.48

£ 2.14

£ 1.80

£ 1.46

£ 0.78

Eastern Fringe

£ 3.44

£ 2.82 £ 2.20 £1.89

£ 1.58

£ 1.28

£ 0.97

£ 0.35


£ 2.74

£ 2.18 £ 1.62 £1.34

£ 1.06

£ 0.78

£ 0.50

-£ 0.07


£ 1.53 £ 1.03 £0.78

£ 0.53

£ 0.28

£ 0.03

-£ 0.58

Braintree £ 1.83

£ 1.34 £ 0.86 £0.62

£ 0.38

£ 0.14

-£ 0.13

-£ 0.72


£ 1.67

£ 1.20 £ 0.73 £0.49

£ 0.26

£ 0.02

-£ 0.26

-£ 0.84

Witham £ 1.50

£ 1.05 £ 0.59 £0.36

£ 0.13

-£ 0.12

-£ 0.40

-£ 0.95

3.30 The effect of a £20,000 S106 requirement is to lower residuals. At 30% affordable housing they exceed £1m in Three Fields, Eastern Fringe and Great Notley and Rayne. At 10% affordable housing they exceed £1m in all locations.

3.31 The effect of a £20,000 additional cost per unit for development to CSH Level 6 would be comparable to that of a £20,000 per unit S106 requirement.

3.32 We also cost a £40,000 package comprised of a mix of additional S106 and CSH costs (ie a total of £1.35m)..

3.33 The effect of a combined £40,000 S106 and CSH requirement is to further lower residuals. Only in Three Fields, Eastern Fringe and Great Notley and Rayne do they achieve £1m with 20% affordable housing. In the Northern Fringe, Braintree, Halstead and Hedingham and Witham although residuals remain positive at 10% and 20% affordable housing it is unlikely that they would be competitive with existing or alternative use values.


Table 3.6 £40,000 S106 and CSH – Residual value in £s million per hectare

£40K S106/CSH 0% 10% 20% 25% 30% 35% 40% 50%


£ 2.90

£ 2.22 £1.88

£ 1.54

£ 1.20

£ 0.86

£ 0.18


£ 2.21

£ 1.60 £1.29

£ 0.98

£ 0.67

£ 0.36

-£ 0.32


£ 2.14

£ 1.58

£ 1.02 £0.74

£ 0.45

£ 0.18

-£ 0.13

-£ 0.81


£ 0.92

£ 0.42 £0.17

-£ 0.10

-£ 0.40

-£ 0.71

-£ 1.32

Braintree £ 1.22

£ 0.74

£ 0.25 £0.01

-£ 0.28

-£ 0.58

-£ 0.87

-£ 1.47


£ 1.06

£ 0.59

£ 0.12 -£0.15

-£ 0.43

-£ 0.72

-£ 1.01

-£ 1.58

Witham £ 0.90

£ 0.44

-£ 0.02 -£0.30

-£ 0.58

-£ 0.86

-£ 1.14

-£ 1.70

Impacts of potential grant funding

3.34 The availability of public subsidy (in the form of grant) can have a significant impact on scheme viability. Grant given to the affordable housing providers enables them to pay more for affordable housing units, thus increasing overall scheme revenue and therefore the residual value of a mixed tenure scheme. There are two main sources of grant which may be available: from the Homes and Communities Agency and/or the local authority (for example using money collected from development in the form of a commuted sum, through a s106 agreement).

3.35 We have assumed grant of £45,000 per Social Rented unit and £20,000 per New Build HomeBuy unit. This level of grant was agreed with the local authority as being a reasonable figure to use for viability testing purposes. At 35% affordable housing and 45 dph this adds £450,000 to scheme revenue. At 15% affordable housing it adds £200,000 to scheme revenue.


Table 3.7 Base plus grant - Residual value in £s million per hectare

Affordable Housing Share

0% 20% 25% 30% 35% 40% 50%


£ 4.26

£ 3.72

£ 3.19

£ 2.92

£ 2.65

£ 2.12


£ 3.58

£ 3.10

£ 2.63

£ 2.39

£ 2.16

£ 1.68


£ 3.35

£ 2.94

£ 2.52

£ 2.11

£ 1.90

£ 1.69

£ 1.28


£ 2.28

£ 1.93

£ 1.57

£ 1.40

£ 1.22

£ 0.86

Braintree £ 2.44

£ 2.10

£ 1.76

£ 1.42

£ 1.25

£ 1.08

£ 0.74


£ 2.27

£ 1.95

£ 1.62

£ 1.30

£ 1.14

£ 0.97

£ 0.65

Witham £ 2.11

£ 1.80

£ 1.49

£ 1.18

£ 1.02

£ 0.87

£ 0.56

3.36 Grant improves viability. Although the financial impact is the same across the district the percentage impact is greater in lower value areas.. With grant 35% affordable housing yields a residual of over £1m in all housing market sub-areas. Residuals of over £2m are achieved in Three Fields and the Eastern Fringe...


4. LAND SUPPLY, SMALL SITES AND USE OF COMMUTED SUMS

Introduction

4.1 This chapter reviews the policy context and options for identifying the size of sites above which affordable housing contributions would be sought, in the national policy context.

4.2 This chapter provides an assessment of the profile of land supply and the likely relative importance of small sites. It then considers practical issues about on-site provision of affordable housing on small sites and the circumstances in which collection of a financial contribution might be appropriate (and the principles by which such contributions should be assessed).

Small sites: purpose of the Analysis

4.3 PPS3 Housing sets out national policy on thresholds and affordable housing and states:

”The national indicative minimum site size threshold is 15 dwellings. However, Local Planning Authorities can set lower minimum thresholds, where viable and practicable, including in rural areas. This could include setting different proportions of affordable housing to be sought for a series of site-size thresholds over the plan area.” (Para 29)

4.4 Current local policies for Braintree are set out in Local Plan Policies RLP 5 and 6 and the accompanying affordable housing SPD, which call for 30%% affordable housing on sites of 15 or more units. .

4.5 By reducing site size thresholds and ‘capturing’ more sites from which affordable housing can be sought, an authority can potentially increase the amount of affordable housing delivered through the planning system.

4.6 In this section we examine the impact that varying site size thresholds would have on affordable housing supply. In order to do this we need to examine the likely future site supply profile.

Site supply analysis

4.7 We have analysed data on past permissions to consider how important sites of different sizes are likely to be to the future land supply. The chart below shows the percentage of dwellings in different sizes of sites covering sites that were granted planning permission in 2009


Chart 4.1:

2009 dwelling supply

by size of site

19.5

5 19

16.5

49

0

20

40

60

1-3

units

4 units 5 units 6-10

units

11-14

units

15 or

more

%

1-3 units 4 units 5 units

6-10 units 11-14 units 15 or more

Source: Braintree District Council Notes: Dwelling supply: ie with planning consent or proposed allocations for development as at 31.03.09 The above figures are gross numbers; some sites may have a lesser net number of dwellings as they have involved a demolition of one or more dwelling or a conversion from an existing dwelling(s).

4.8 The supply of dwellings comes from a broad mix of site sizes. Sites of less than 15 dwellings account for 50% of all dwellings. Of sites below 15 dwellings, there is a broad division in supply of the smallest sites (those of 1 to 4 dwellings) and ‘larger small sites’ (sites of 5 to 14 dwellings). Very small sites (1-4 dwellings) account for about 24.2% of the supply and sites of 5 to 14 dwellings for about 26.5% of the supply.

4.9 We undertook a further analysis which looked at historic data on completions in urban and rural areas. The results of this analysis are shown in table 4.2 below. This shows that historically small sites (of upto 10 units) have been much more important in rural areas than in the main urban areas.


Table 4.2 Housing completions in urban and rural areas 2001/02 to 2004/05

TOTAL COMPLETIONS

2001-2002

2002-2003

2003-2004

2004-2005

665 661 854 686

small rural 45 52 81 39

small urban 70 93 115 101

small rural % of total rural 31% 70% 75% 30%

small urban % of total urban 13% 16% 15% 18%

Source Braintree District Council data

4.10 Finally we looked at the 2009 housing land schedules which show outstanding planning consents by parish (with parishes divided into urban and rural parishes). This analysis showed that 68% of all rural housing is provided on sites of fewer than 15 units and 50% is provided on sites of fewer than 5 units, whilst 81% of housing in urban areas is provided on sites of 15 or more units. In urban areas 10.6% of supply arises on sites of 10-14 units and 7.8% on sites of less than 10 units. In rural areas apart from a few specific large sites (2 are shown in the 2009 housing land schedule) all housing is provided on sites under 15 units.

4.11 The Council has reported concerns about “bunching” of applications immediately below the threshold (ie at 14 units where the threshold is 15 units). There is no evidence of this in rural areas but in the urban areas of Halstead, Braintree and Witham it is worth noting that of the 12 sites in the 10-14 unit band 6 sites consist of 14 units. A schedule of urban and rural parishes is attached at Appendix 3.

Table 4.3 Outstanding planning consents

Urban sites Rural sites

Size of site No of units % of units Size of site No of units %of units

1 27 1.7% 1 96 26.4%

2 to 3 35 2.2% 2 to 3 70 19.3%

4 24 1.5% 4 16 4.4%

5 to 9 38 2.4% 5 to 9 18 5.0%

10 to 14 166 10.6% 10 to 14 49 13.5%

15 or more 1275 81.5% 15 or more 114 31.4%

Total 1565 100.0% Total 363 100%

Source Braintree District Council data

Note: The above figures are gross numbers; some sites may have a lesser net number of dwellings as they have involved a demolition of one or more dwelling or a conversion from an existing dwelling(s).


4.12 Given this profile of sites and the high dependence on small sites, particularly in rural areas, it is appropriate to consider the option of introducing a lower site size threshold than 15 dwellings. This would have particular significance in rural areas where two thirds of all outstanding planning consents are below the 15 unit threshold.

4.13 To achieve a meaningful increase in the supply of affordable housing, a threshold of zero dwellings is a realistic option and a detailed analysis of the viability of very small sites will be needed. Some small sites involve the demolition and redevelopment of an existing residential property. This is a situation that can cause particular viability difficulties. Not because the residual value is different from similar sites that do not involve demolition but because the existing dwelling makes for a high existing use value.

4.14 Chart 4.2 below shows previous uses of sites in rural areas in 2009 (Source Braintree Council’s own data). This chart shows that 87 dwellings out of 363 (24%) were provided through the demolition and replacement of existing dwellings. These 87 dwellings were provided on 58 sites and in 40 cases were a like for like replacement (1 dwelling for 1 dwelling or 2 dwellings for 2 dwellings). Development in gardens provided 34 units. By far the largest source of land was miscellaneous former uses (212 dwellings) : including a former Prisoner of War Camp (8 units) a former Foundry (82 units) and former commercial garages (52 units)

Chart 4. 2 Previous uses for sites in rural areas 2009

Sources of land in rural areas

212

87

34 19 5 3 3

0

50

100

150

200

250

Mis

c fo

rmer

use

Dem

olis

h an

d

repl

ace

dwel

lings

Bui

ld o

n

gard

en

Bar

n

conv

ersi

on

Conv

ert/

exte

nd

dwel

ling

Conv

ert

dom

esti

c

gara

ge

Infi

ll w

aste

land

Sites Dw ellings

4.15 The option of a lower threshold is tested in terms of scheme viability in the next chapter. This includes a number of case studies of smaller sites, down to 1 dwelling.

Small sites and management of affordable housing

4.16 We discussed the suitability of small sites for affordable housing at the workshop with the development industry and which included representatives from housing associations. There was limited debate about small sites and viability but the discussion


which did take place indicated that small sites are, on a systematic basis, no less viable than larger sites.

4.17 There was no comment from the housing associations present about any management issues which might arise if affordable housing is provided on small sites – down to as few as one or two units.

4.18 Our experience of development industry workshops held elsewhere in the country gives a very consistent housing association perspective. They see no reason why affordable housing cannot be provided in small numbers (within mixed tenure schemes) and one dwelling in a scheme can be acceptable. Not all associations will want small numbers (including single units) of affordable housing in every location – it will depend where the associations already have a management presence. But, as a general rule, there will be an association prepared to take on a small group (single unit) of affordable housing where it is provided – be this in an urban or rural context.

4.19 There may be circumstances where associations do not think it appropriate to provide affordable housing as part of a mixed tenure scheme e.g. in flatted blocks with high service charges or in very high price areas where the form of the affordable housing may be out of kilter with the majority development type. Associations would want to be consulted on the acceptability of affordable housing as part of mixed tenure schemes – whatever their size.

4.20 However, our advice to the Council should it want to pursue a (very) low threshold, is that the lack of comment at the recent local workshop needs to be taken one step further. We recommend that active housing associations in the area are asked directly whether they have any management concerns about small mixed tenure schemes with very low numbers of affordable homes.

Use of commuted sums

4.21 As a general principle, we recognise that seeking on-site provision of affordable housing will be the first priority and that provision of affordable housing on an alternative site or by way of a financial payment in lieu (or commuted sum) should only be used in exceptional circumstances. This position is consistent with national guidance in Paragraph 29 of PPS3 which states:

“In seeking developer contributions, the presumption is that affordable housing will be provided on the application site so that it contributes towards creating a mix of housing. However, where it can be robustly justified, off-site provision or a financial contribution in lieu of on-site provision (of broadly equivalent value) may be accepted as long as the agreed approach contributes to the creation of mixed communities in the local authority area” Para 29.

4.22 However, if affordable housing is sought from very small sites, in certain circumstances it becomes impractical to achieve on site provision e.g. seeking less than 33% on a scheme of 3 dwellings. There will also be occasions where on-site provision can only deliver a partial contribution towards the proportion of affordable housing sought e.g.


30% affordable housing in a scheme of 4 dwellings would deliver one affordable unit on site (representing 33% of provision). In the latter case, it is possible to devise a formula which mixes on-site provision with a commuted sum to ‘make up the balance.

4.23 Where commuted sums are sought as an alternative to direct on or off-site provision, PPS3 sets out the appropriate principle for assessing financial contributions - that they should be of “broadly equivalent value” (see para 29 as set out above). Our approach is that the commuted sum should be equivalent to the ‘developer/landowner contribution’ if the affordable housing was provided on site. One way of calculating this is to take the difference between the residual value of 100% market housing and the residual value of the scheme with the relevant percentage and mix of affordable housing.


Table 4.4 Specimen calculation of commuted sum – scheme of 5 (2 bed) dwellings

Percentage of affordable housing Residual land value Commuted sum required

Nil on site £340,000 £100,000

20% on site £240,000 Nil

4.24 If the ‘equivalence’ principle is adopted, then the decision of the local authority to take a commuted sum will be based on the acceptability or otherwise of on-site provision as a housing and spatial planning solution.

The Council’s current policy (as set out in the Affordable Housing SPD of May 2006) states that “where a commuted sum is sought . . . the level of payment required will be based on the level of subsidy that would be required to purchase units on the open market” (para 5.14). This formula (in the absence of grant) produces a higher commuted sum payment than that calculated above. In practice the Council has assumed that grant will be available and has normally negotiated a commuted sum of £30-40,000 per unit.


5 CASE STUDY VIABILITY ANALYSIS

Introduction

5.1 The analysis in Chapter 3 provides a good indication of the likely viability of sites in the district. The residual values can be compared with existing use values to establish whether land owners are likely to make a return over and above existing use value, taking into account a developer margin.

5.2 In this chapter we review a series of small sites, which enables us to explore the financial implications of adopting lower thresholds. We also analyse an illustrative 5,000 unit Sustainable Urban Extension which enables us to illustrate the impact of differing assumptions about costs and prices on financial viability.

Small Sites

5.3 At the development industry workshop on 18th August we discussed the economics of development of very small sites as opposed to the modelled 1 hectare site. There was no suggestion that the costs of developing sites below 15 dwellings differ from those of larger sites. Market values will be determined by location and mix of dwellings provided. This has been the case elsewhere where we have run similar workshops.

5.4 Current policy sets the threshold at 15 units. Evidence on historic site supply suggests that in Braintree smaller sites including those as small as 1-3 units also make a significant contribution to total housing supply. We therefore model sites of 1, 2,4,5, 6, 8 and 13 units.

Case study sites

5.5 In this section we review a number of case study developments which are examples of small sites for residential development. We have taken developments of 1,2, 4,5,6, 8 and 13 units. The examples modelled are shown in Table 5.1 below


Table 5.1 Case study sites

5.6 For each case study we have undertaken an analysis of residual values for three of our sub market areas and at levels of affordable housing from 0% to 50%. All the assumptions used are the same as for the base main analysis described in Chapter 3.

5.7 Detailed findings are set out in tables 5.2a and 5.2b below. These show that:

In Braintree Northern Fringe residual values for 100% market housing range from £1m per hectare for a scheme of 1 dwelling to more than £2m per hectare for a 13 unit scheme. The introduction of affordable housing reduces values and narrows the spread. At 25% affordable housing a 1 unit scheme has a value of £590,000 per hectare whilst a 13 unit scheme has a value of £900,000 per hectare. At higher levels of affordable housing (35% and over) the value per hectare of a 1 unit site exceeds that of a 13 unit site.

Actual scheme residual values range from £103,500 for a 1 unit scheme with no affordable housing to £52,200 for the same scheme with 30% affordable housing and less than £20,000 for the same scheme with 50% affordable housing. In the case of a

Case

study

Total

dwellings

Site area (ha) Density

(in dph)

Mix of dwellings

A 1 0.1 10 1 x 5 bed detached

B 2 0.15 13 2 x 4 bed detached

C 4 0.20 20 4 X 3 bed detached

D 5 0.20 25 2 x 3 bed semi

3 x 3 bed detached

E 6 0.20 30 3 x 3 bed semi

2 x 3 bed detached

F 8 0.20 40 4 X 3 bed semi

4 X 3 bed terrace

G 13 0.25 52 6 x 2bed flat

7 x 3 bed terrace


4 dwelling scheme the residual falls from £342,000 with no affordable housing to £181,800 with 30% affordable housing.

A similar pattern can be observed in Three Fields where residual values for 100% market housing range from £1.9m per hectare for a scheme of 1 dwelling to more than £4m per hectare for a 13 unit scheme. The introduction of affordable housing reduces values and narrows the spread. At 25% affordable housing a 1 unit scheme has a value of £1.28m per hectare whilst a 13 unit scheme has a value of £2.39m per hectare. At 40% affordable housing a 1 unit scheme has a residual value of £900,000 per hectare whilst a 13 unit scheme has a residual value of £1.39m per hectare

Actual scheme residual values range from £190,000 for a 1 unit scheme with no affordable housing to £117,000 for the same scheme with 30% affordable housing and less than £67,500 for the same scheme with 50% affordable housing. In the case of a 4 dwelling scheme the residual falls from £600,000 with no affordable housing to £376,000 with 30% affordable housing

Tables 5.2a and 5.2b showing residual values: the upper figure is the residual value for the

whole scheme and the lower figure is the equivalent residual value per hectare (in £ millions - £

1.78 = £1.78m)

Northern Fringe

Case Study % Affordable Housing

0 10 20 25 30 35 40 50

A £103,500 £86,400 £69,300 £59,400 £52,200 £43,200 £34,200 £17,100

1 dw £1.04 £0.86 £0.69 £0.59 £0.52 £0.43 £0.34 £0.17

B £197,100 £165,600 £134,100 £119,700 £102,600 £88,200 £72,000 £42,300

2 dw £1.31 £1.10 £0.89 £0.80 £0.68 £0.59 £0.48 £0.28

C £342,000 £288,900 £234,900 £207,900 £181,800 £154,800 £127,800 £74,700

4 dw £1.71 £1.45 £1.18 £1.04 £0.91 £0.77 £0.64 £0.37

D £348,300 £288,000 £227,700 £198,000 £167,400 £136,800 £108,000 £47,700

5 dw £1.74 £1.44 £1.14 £0.99 £0.84 £0.68 £0.54 £0.24

E £354,600 £288,000 £220,500 £188,100 £153,900 £120,600 £86,400 £20,700

6 dw £1.77 £1.44 £1.10 £0.94 £0.77 £0.60 £0.43 £0.10

F £395,100 £312,300 £230,400 £189,900 £147,600 £107,100 £66,600 -£19,800

8 dw £1.98 £1.56 £1.15 £0.95 £0.74 £0.54 £0.33 -£0.09

G £542,700 £414,900 £288,000 £224,100 £160,200 £96,300 £31,500 -

£114,400

13 dw £2.17 £1.66 £1.15 £0.90 £0.64 £0.39 £0.13 -£0.46


Three Fields Case Study % Affordable Housing

0 10 20 25 30 35 40 50

A £189,900 £165,600 £141,300 £127,800 £117,000 £104,400 £91,800 £67,500

1 dw £1.90 £1.66 £1.41 £1.28 £1.17 £1.04 £0.92 £0.68

B £341,100 £297,900 £254,700 £233,100 £210,600 £189,900 £168,300 £126,000

2 dw £2.27 £1.99 £1.70 £1.55 £1.40 £1.27 £1.12 £0.84

C £601,200 £526,500 £450,900 £413,100 £376,200 £338,400 £300,600 £225,900

4 dw £3.01 £2.63 £2.26 £2.07 £1.88 £1.69 £1.50 £1.13

D £629,100 £545,400 £460,800 £419,400 £378,900 £336,600 £295,200 £211,500

5 dw £3.15 £2.73 £2.30 £2.10 £1.90 £1.68 £1.48 £1.06

E £657,000 £564,300 £472,500 £427,500 £379,800 £334,800 £288,900 £197,100

6 dw £3.29 £2.82 £2.36 £2.14 £1.90 £1.67 £1.45 £0.99

F 726,300 616,500 505,800 451,800 396,000 342,000 287,100 177,300

8 dw £3.63 £3.08 £2.53 £2.26 £1.98 £1.71 £1.44 £0.89

G £1,014,300 £846,900 £681,300 £597,600 £513,900 £430,200 £346,500 £180,000

13 dw £4.06 £3.39 £2.73 £2.39 £2.06 £1.72 £1.39 £0.72

5.8 .For very small sites, which may take the form of the replacement of an existing dwelling by 1 or more new dwellings, it is important to consider existing residential use value (typically of a large detached house or bungalow). We set out below existing values of a 4 bed detached house in each of the sub-areas analysed above.

Existing property value 4 bed detached house

Northern Fringe £350,000

Three Fields £450,00

5.9 It will be seen from this analysis that since the actual residual value of a two unit scheme does not exceed the value of an existing property even at 100% market housing, 1 and 2 unit schemes which rely on the demolition and redevelopment of an existing larger residential property are unlikely to come forward. However we know that 40 units were provided on a 1-1 replacement basis in the year to March 2009 (see Chart 4.2 and para 4.13 above). In these instances the new property would need to be of


substantially higher value than the existing property for this transaction to be financially worthwhile.

5.10 At 20% affordable housing and above in Northern Fringe even a 13 unit scheme does not produce a residual in excess of the value of existing property on the same site. In Three Fields the same factor kicks in at 35% affordable housing.

Sustainable Urban Extensions

5.12 Sustainable Urban Extensions are normally developed over a long period – 5-10 years is not unusual – and are subject to variations in house prices, build costs and access to development finance which are not easily predictable in advance. This will impact on the amount of affordable housing (and other planning contributions) which can be sought and on residual land values.

5.13 We model an illustrative 5,000 dwelling scheme under different assumptions about movements in house prices and build costs in order to show the effect on residual value of variations in these factors.

The scheme is one of 5000 units developed over 5 years

No housing is completed or sold in the first year, but substantial expenditure on upfront infrastructure is incurred

S106 obligations of £20,000 per dwelling and Code for Sustainable Homes costs of £20,000 per dwelling are assumed

Nett developable area is 50% of gross developable area

30% affordable housing is provided split 21% social rent and 9% low cost home ownership

Grant is not available

5.14 If the development took place within one year (ie using the static model) the residual value would be £600,000 per hectare. Applying a discounted cash flow model and assuming no inflation in house prices or build costs the residual value falls to £375,000 per hectare. If house prices rise by 3% pa and there is no inflation in development costs residual value rises to £750,000 per hectare. If (in an extreme scenario) house prices rise by 10% pa and build costs rise by 7% pa residual value rises to £1.3m per hectare.

Assumption Residual value per hectare

Static model (site developed in 1 year) £600,000

DCF (devt over 5 yrs house prices and devt costs unchanged)

£375,000

DCF (devt over 5 yrs, house prices rise by 3% pa) £750,000

House prices rise by 10% pa, build costs rise by 7% pa £1.3m


5.15 None of these scenarios are suggested as being more likely than each other (or indeed more likely than others not modelled such as higher S106 and infrastructure costs). However what this series of examples illustrates is the sensitivity of the financial viability of long-term development to unknown future increases or decreases in costs and house prices and hence the difficulty of setting a long-term site specific affordable housing target which will not need to be reviewed during the life of the scheme.

5.16 It is also worth noting that very heavy initial capital costs are incurred and, although these are recouped over the life of the development, in the first two years (out of 5 in this case) the development is running at a loss.


6 MAIN FINDINGS AND CONCLUSIONS

Key findings

6.1 There is significant variation in market values across housing market sub-areas. We found that residual value is dependent not only on location but also on the density adopted.

6.2 Residual values are higher in rural areas than in urban areas.

6.3 60 dph schemes produce the highest residuals without affordable housing. The introduction of affordable housing makes it financially more attractive to develop at lower densities and 45 dph is the density which produces the highest residual values, except at very high proportions of affordable housing where even lower density development becomes more financially attractive.

6.4 Variations in S106 and Code for Sustainable Homes requirements will affect residual values and hence the percentage of affordable housing which can reasonably be sought. The same is true of rising or falling house prices.

6.5 The introduction of social housing grant significantly improves residuals. This is of more importance in low value areas where it could increase the chances that development will take place.

6.6 The analysis of the supply of sites in the District indicated that small sites (below the national indicative minimum of 15 dwellings) make up an important element of the supply particularly in rural areas – and that very small sites (of 1 – 4) are particularly important here although larger small sites (5 to 14 dwellings) are also an important source of land supply in the rural parts of the district.

6.7 Where a financial payment in lieu of on-site provision of affordable housing (or commuted sum) is to be sought, it should be of “broadly equivalent value”. This approach is, on the evidence we have considered, a reasonable one to take in policy terms.

6.8 If this ‘equivalence’ principle is adopted, then the decision of the local authority to take a commuted sum will be based on the acceptability or otherwise of on-site provision as a housing and spatial planning solution, not in response to viability issues.

Conclusions and policy recommendations

6.9 There is no detailed government guidance setting out how targets should be assessed, based on an assessment of viability. In coming to our conclusions, we have reviewed the residual values generated for the different sub markets in Braintree at the alternative levels of affordable housing tested and considered how these values compare with historic land values generally in the area.

6.10 From this review, we note that house prices are significantly higher in the rural areas of the district than in urban areas.


6.11 We now consider possible policy options and put forward two main approaches:

Option one - a single target for the whole district. Given the relatively low values in some of the urban areas, this would need to be set relatively conservatively and a target of 30% would seem appropriate;

Option two - a ‘split target’ which recognises the very significant variation in house prices and residual values between urban and rural areas. The split target would be 40% in rural areas and 30% in the urban areas of Braintree, Halstead, the Hedinghams and Witham

Under both options, the proposed SUEs at Braintree and Witham should be planned for on the basis of a 30% target, subject to up to date scheme specific viability appraisals

6.12 When reviewing target percentages for affordable housing, the Council will need to be mindful of the potential impacts of additional costs imposed in the form of a higher Section 106 planning gain package (we looked at £20,000) per unit, as well as the potential impacts of achieving a higher Code for Sustainable Homes level. The cumulative impacts of these two requirements could be upto £900,000 per hectare and hence a quite substantial viability ‘hit’.

Viability on individual sites

6.13 Our analysis has indicated that there will be site-specific circumstances where achievement of the affordable housing proportions set out above may not be possible. This should not detract from the robustness of the overall targets but the council will need to take into account specific site viability concerns when these are justified.

6.14 If there is any doubt about viability on a particular site, it will be the responsibility of the developer to make a case that applying the council’s affordable housing requirement for their scheme makes the scheme not viable. Where the council is satisfied this is the case, the council has a number of options open to it (including changing the mix of the affordable housing and supporting a bid for grant funding from the Homes and Communities Agency and/or using their own funds) before needing to consider whether a lower level of affordable housing is appropriate. In individual scheme negotiations, the council will also need to consider the balance between seeking affordable housing and its other planning obligation requirements.

Thresholds

6.15 In the urban parts of the district, the evidence indicates that the national indicative minimum threshold of 15 dwellings is appropriate, although we note the Council’s concerns about “bunching” at 14 units and about the balance of provision between sites of 1-9 units and 10-14 units. However, the profile of site supply in the rural areas (coupled with the high levels of need) indicate that adopting a low threshold is justified.


6.16 A threshold of 5 dwellings would capture about a fifth of the supply in rural areas but to really increase delivery of affordable housing would mean introducing a zero threshold, Of course, for schemes of 1 and 2 dwellings, on site provision is not possible (simply because of the numbers involved) and the operation of a zero threshold would need to be linked to the collection of commuted sums and a strategy in place to ensure the money collected was spent to increase the supply of rural affordable housing. The Council has expressed concern about the cost and benefit implications of negotiating a S106 Agreement on all sites in rural areas which could be entailed in the introduction of such a policy and we recognise that these concerns would need to be addressed.

6.17 In considering a very low/zero threshold, it is important to note that the nature of the current land use plays a particular role in the development economics of very small sites. Some sites down to 1 dwelling will be equally capable of delivering affordable housing as on much larger sites, particularly but not exclusively, in the higher value market areas. But there will be a group of sites where the current use is as a dwelling(s) where this will not be the case and the authority will need to take a flexible view in seeking affordable housing from these sites – whichever market value area they are in.

Commuted sums

6.18 Where commuted sums are collected a possible approach to calculating the appropriate sum sought is to base this on the equivalent amount which would be contributed by the developer/landowner were the affordable housing provided on site. This is expressed as follows:

RV 100% MH = Residual value with 100% market housing

RV AH = Residual value with X% affordable housing (say 40%) Equivalent commuted sum = RV 100% MH minus RV AH 6.19 Where commuted sums are collected, the council will need to have in place a strategy

to ensure the money is spent effectively and in a timely manner. Options for spending will be a matter for the council to consider but could include supporting schemes which would otherwise not be viable, increasing the amount of social rented housing in a scheme, increasing the proportion of family units in a scheme, seeking higher quality affordable housing (e.g. a higher level of the Code for Sustainable Homes).


Appendix 1 Notes of workshop BRAINTREE AFFORDABLE HOUSING VIABILITY STUDY

Workshop Notes

A workshop was held on 24th June 20009 at Braintree Town Hall. Representatives of the

development industry, landowners and agents and housing associations, were in attendance.

A full attendance list is given below.

Tim Lucas, Braintree DC Steve Heywood, Iceni Paul Gibson, Persimmon Melanie Sutton Hastoe Isabel Wright, Hastoe John Milner Baily Garner LLP , David Steel, Braintree DC Eleanor Dash, Braintree DC Adam Davis, Braintree DC Ian Harrison, Braintree DC John McCarthy, Bidworths Helen Shepherd, Flagship Andrew Golland, Three Dragons Kathleen Dunmore, Three Dragons Three Dragons and the local authority would like to thank all those in attendance for their inputs to the study. At the workshop Three Dragons gave a presentation summarising the methodology and outlining the process of higher level and detailed testing which would be carried out to determine viability targets. It was agreed that the Powerpoint presentation (attached) would be made available to all Workshop participants in conjunction with these feedback notes. Key issues 1 Basis for interpreting viability There was no objection in principle to the over-riding method for assessing viability proposed by Three Dragons. This measures viability by reference to residual scheme value less the existing or alternative use value of a site. The report by Three Dragons will enable the local authority to set broad policies. Individual schemes will be appraised on a scheme specific basis by the local authority taking account of site conditions and market viability. This is of particular importance in the present volatile market, in which house prices nationally are falling but a recovery can be anticipated during the life of the core strategy and relevant DPDs.


Viability testing for policy making purposes will be sensitivity tested at +/- 10% house prices but at a scheme specific level local authorities will need to take into account actual house prices for the particular scheme. 2 Overall methodology Three Dragons explained that the approach to the study will be two stage with the first stage focusing on testing a notional one hectare site, assuming different development mixes and different percentages of affordable housing, with the second stage looking at a range of generic site types, ranging from large green field through to smaller brown field, windfall type sites. Participants at the workshops generally supported the approach set out. Data sources (e.g. HMLR for house prices and BCIS for build costs) were explained to participants. The need for best primary data sources based on a large sample was understood and agreed. 3 Sub markets A key part of the study will involve the analysis of viability at a sub market level. Sub markets will be defined primarily by house prices. The Powerpoint presentation shows a map of draft areas although these are subject to further analysis. Comments were that Silver End should be with White Notley, Witham is higher value than Braintree (closer to London). Castle Hedingham is a high value area. Flats are only likely to be built in Witham, Kelvedon and Hatfield Peverel, based on proximity to the railway station. 2 bed flat prices too high, 1 beds more realistic. Participants generally welcomed the focus on sub markets, although there were some queries as to whether individual locations had been allocated to the correct submarkets. Participants were invited to submit comments on submarkets by email to Andrew Golland. Consideration was given to whether the use of differential affordable housing targets, responsive to house price differentials in different parts of a local authority, might be a proper policy response for some or all authorities. The Three Dragons viability study would demonstrate the effect of different AH targets in different locations but this was ultimately a policy decision for individual local authorities. It was noted that Strategic Sites had been optioned 10-15 years ago at minimum land values which might not be achievable in the present market taking into account the combination of house prices, S106 requirements and anticipated development costs. Land values for SUEs were estimated at £1m per hectare (please can attendees confirm). For Rural Exception Sites £12,000 per plot was the maximum going rate. For conventional development land value should be returning to its historic norm of 30-40% of gross developable value (GDV). Evidence of mothballing of sites – Witham Green and Maltings Lane. 4 Density and development mix


A template of development mixes was demonstrated showing proposed mixes of house types at different densities. 80 dph was suggested as maximum density. Rural Exceptions Sites 10 dp/acre with 8 as the average scheme size, It was suggested that bungalows should be added to the mix particularly in more rural locations. Higher density flatted developments, whilst they should be modelled to demonstrate viability, were considered unlikely to come forward in the present market. Full details of proposed mixes are attached in the Powerpoint and participants are invited to submit illustrative alternative mixes which are either proposed or have been recently developed. 5 Thresholds and the viability of smaller sites A range of views were expressed in relation to thresholds and the viability of small sites. The logic of a 15 dwelling threshold as in PPS3 was questioned – why is it 15?; the economics do not change at this point. There were arguments for the use of commuted sum contributions from very small sites although housing associations present did not report difficulties in managing small numbers of units on scattered sites within the same local authority area. Any policy on thresholds must be linked to overall land supply and the study would be considering the actual and anticipated supply of land by size of site 6 Calculation of commuted sums Any commuted sum should be the difference between the residual value of a scheme with 100% market housing and one with a mix of market and affordable housing. 7 Development costs Three Dragons presented the proposed page that will be used for the testing framework. This is included in the Powerpoint presentation. It was explained that the base build costs per square metre will be calculated from the BCIS data source (NB: costs in the Powerpoint presentation are illustrative and not Local Authority specific) The other development costs (professional fees, internal overheads, profit margins, etc) are however those which Three Dragons intend to use for base viability testing. Code for Sustainable Homes requirements could add significantly to costs at higher levels of the Code. Code level 4 was estimated to add between £7,000 and £35,000 per dwelling to costs, depending on how it was achieved. On site energy generation can add significantly to costs depending on how it is achieved. Re-engineering pattern book types is expensive. The Cyril Sweet research for EP/HC may have understated costs by excluding substructure and external works. Modelling will be undertaken at current build costs. This will reflect the standards currently being built to. Three Dragons will then test at Code Level 4 (if, subject to further analysis it is found that current development do not generally meet this Code) using an appropriate uplift in costs based on best available information. 8 Other Section 106 contributions


The level of planning gain package was discussed. – it can range from £5,000 per dwelling to £20,000 plus. Certainty as to planning obligations and defined affordable housing obligations is beneficial to developers in negotiation with landowners. 9. Affordable housing issues Intermediate tenure provision was changing from low cost home ownership to intermediate rent. Refer to Local Housing Allowance to establish base rent. 10 Protocols for negotiations on Section 106 Three Dragons explained that the project will provide the local authority with an Affordable Housing Toolkit to assist the process of negotiations on viability and Section 106 contributions. Experience has shown that this is used most effectively when this tool is also available to local developers and landowners.


Appendix 2 Key Assumptions

The Toolkit provides the user with an assessment of the economics of residential development. It allows the user to test the economic implications of different types and amounts of planning obligation and, in particular, the amount and mix of affordable housing. It uses a residual development appraisal approach which is the industry accepted approach in valuation practice. The Toolkit compares the potential revenue from a site with the potential costs of development before a payment for land is made. In estimating the potential revenue, the income from selling dwellings in the market and the income from producing specific forms of affordable housing are considered. The estimates involve (1) assumptions about how the development process and the subsidy system operate and (2) assumptions about the values for specific inputs such as house prices and building costs. These assumptions are made explicit in the guidance notes. If the user has reason to believe that reality in specific cases differs from the assumptions used, the user may either take account of this in interpreting the results or may use different assumptions. The main output of the Toolkit is the residual value. In practice, as shown in the

diagram below, there is a ‘gross’ residual value and a ‘net’ residual value. The gross

residual value is that value that a scheme generates before Section 106 is required.

Once Section 106 contributions have been taken into account, the scheme then has a

net residual value, which is effectively the land owner’s interest.


Key Data Assumptions

Market Value Areas and Indicative New Build Prices

Unit Types

Flats Terrace

Sub Markets 1 bed 2 bed 3 bed 2 bed 3 bed 4 bed Braintree £ 110,000 £ 160,000 £ 175,000 £ 170,000 £ 195,000 £ 220,000

Eastern Fringe £ 135,000 £ 190,000 £ 210,000 £ 205,000 £ 240,000 £ 260,000

Great Notley and Rayne £ 125,000 £ 175,000 £ 195,000 £ 190,000 £ 220,000 £ 240,000

Halstead and Hedinghams £ 105,000 £ 155,000 £ 170,000 £ 165,000 £ 190,000 £ 215,000

Northern Fringe £ 115,000 £ 165,000 £ 180,000 £ 175,000 £ 205,000 £ 225,000

Three Fields £ 145,000 £ 210,000 £ 230,000 £ 225,000 £ 260,000 £ 285,000

Witham £ 100,000 £ 150,000 £ 165,000 £ 160,000 £ 185,000 £ 210,000

Unit Types

Semi-Detached Detached Bungalow

Sub Markets 1 bed 2 bed 3 bed 2 bed 3 bed 4 bed 2 bed

Braintree £ 175,000 £ 205,000 £ 235,000 £ 290,000 £ 345,000 £ 375,000 £ 290,000

Eastern Fringe £ 210,000 £ 250,000 £ 285,000 £ 355,000 £ 420,000 £ 460,000 £ 355,000

Great Notley and Rayne £ 195,000 £ 230,000 £ 265,000 £ 330,000 £ 385,000 £ 425,000 £ 330,000

Halstead and Hedinghams £ 170,000 £ 200,000 £ 230,000 £ 285,000 £ 340,000 £ 370,000 £ 285,000 Northern Fringe £ 180,000 £ 210,000 £ 240,000 £ 295,000 £ 350,000 £ 380,000 £ 295,000

Three Fields £ 235,000 £ 270,000 £ 310,000 £ 385,000 £ 450,000 £ 500,000 £ 385,000

Witham £ 165,000 £ 195,000 £ 225,000 £ 280,000 £ 335,000 £ 365,000 £ 280,000

Development mixes

30 dph 45 dph 60 dph 80dph

Flats 1 bed 10% 20%

2 bed 5% 15% 50%

Terraces / Town Houses 2 bed 10% 15% 20% 20%

3 bed 10% 25% 20% 10%

Semi - Detached Houses 3 bed 35% 25% 20%

Detached Houses 3 bed 25% 20% 10%

4 bed 15% 10% 5%

5 bed 5%


Affordable Housing

Tenure Splits

Social Rent 80%

Homebuy 20%

Homebuy Equity Share 45%

Development costs Build costs based on RICS BCIS database: Development costs based on established Toolkit defaults as discussed at development industry workshop.

No abnormals assumed 10% finance costs assumed

Typical unit sizes adopted (m2):

Affordable(m2) Market(m2)

Flats 1 bed 46 45

2 bed 67 60

Terraces / Town Houses 2 bed 76 65

3 bed 84 80

Semi - Detached Houses 3 bed 86 90

Detached Houses 3 bed 90 110

4 bed 110 135

5 bed 125 150

Affordable Housing Targets

0%

10%

20%

30%

35%

40%

45%

Build Costs

Bungalows £1,150

Flats (6-15 storeys) £1,650

Flats (5 storeys) £1,210

Houses <= 75m2 £1,100

Houses > 75m2 £960

Development Costs

Professional Fees % 12% of build costs

Internal Overheads 5% of build costs

Finance (Market) 7% of build costs

Finance (Affordable Housing) 7% of build costs

Marketing Fees 3% of market value

Developers Return 17% of market value

Contractors Return 6% of developments costs


Other Affordable Housing Factors: Social rents

Social Rent

(£/week)

Flats 1 bed £63.53

2 bed £74.53

Terraces / Town Houses 2 bed £74.53

3 bed £81.11

Semi - Detached Houses 3bed £81.11

Detached Houses 3 bed £81.11

4 bed £94.59

5 bed £94.59

Rents are (rounded) target rents from the Dataspring source for March 2008 with a 5% increase to take to 2009 values Gross to net factors (Affordable housing revenue)

Social Rent Costs Intermediate Rent Costs Homebuy Costs

Management and Maintenance £1000 Management Costs 6.00% Operating Cost 2.75%

Maintenance Costs £500 Capitailisation 7.00%

Voids/bad debts 3.00% Voids/bad debts 5.00% Percentage Purchase 45.00%

Repairs reserve £500 Letting fee 1.00%

Capitalisation 7.00% Capitalisation 7.00%


Appendix 3 Rural and urban areas (Source: Residential Land Availability Housing Land March 2009) and map of housing market areas.

Rural areas

Alphamstone Foxearth Pentlow

Ashen

Gestingthorpe Rayne

Bardfield Saling Gosfield Ridgewell

Belchamp Otten Great Bardfield Rivenhall

Belchamp St Paul

Great & Little Henny Shalford

Belchamp Walter Great Maplestead Sible Hedingham

Birdbrook Great Saling Silver End

Black Notley Stambourne

Borley

Great Yeldham Steeple Bumpstead

Bradwell Greenstead Green &

Halstead Rural Stisted

Bulmer

Hatfield Peverel Sturmer

Bures Hamlet Helions Bumpstead Toppesfield

Castle Hedingham

Kelvedon Twinstead

Coggeshall

Lamarsh Wethersfield

Colne Engaine

Liston

Cressing

Little Maplestead White Colne

Earls Colne Little Yeldham

Fairstead Middleton White Notley

Faulkbourne Ovington Wickham St Paul

Feering Panfield

Finchingfield Pebmarsh

Urban areas

Bocking Blackwater Ward

Bocking North Ward

Bocking South Ward

Braintree and Bocking total

Braintree Central Ward

Braintree East Ward

Braintree South Ward

Braintree West Ward

Great Notley

Halstead

Halstead St Andrews Ward

Halstead Trinity Ward

Witham Central Ward

Witham Chipping Hill and Central Ward

Witham Chipping Hill Ward

Witham North Ward

Witham South Ward

Witham total

Witham West Ward