Wiltshire Council Strategic Housing Market Assessment

Wiltshire Council Strategic Housing Market Assessment December 2011 A report initiated by Fordham Research and completed by RS Drummond-Hay MRICS ACIH RS Drummond-Hay MRICS ACIH Bellgate, Casterton Kirkby Lonsdale, Cumbria. LA6 2LF 015242 76202 www.drummond-hay.co.uk

Important Notice RS Drummond-Hay MRICS ACIH has prepared this report for the sole use of Wiltshire Council in accordance with the proposal and instructions under which our services were performed. No other warranty, expressed or implied, is made as to the professional advice included in this report or any other services provided by us. This report may not be relied upon by any other party without the prior and express written agreement of RS Drummond-Hay MRICS ACIH. Some of the conclusions and recommendations contained in this report are based upon information provided by others (particularly Fordham Research) and upon the assumption that all relevant information has been provided by those parties from whom it has been requested. Information obtained from third parties has not been independently verified by RS Drummond-Hay MRICS ACIH, unless otherwise stated in the report. The recommendations contained in this report are concerned with the housing market, affordable housing and planning policy, guidance and regulations which may be subject to change. They reflect a Chartered Surveyor s perspective and do not reflect or constitute legal advice. Certain statements made in the report may constitute estimates, projections or other forward-looking statements and even though they are based on reasonable assumptions as of the date of the report, such forward-looking statements, by their nature, involve risks and uncertainties that could cause actual results to differ from the results predicted. RS Drummond-Hay MRICS ACIH does not guarantee or warrant any estimate or projections contained in this report. RS Drummond-Hay MRICS ACIH Bellgate, Casterton Kirkby Lonsdale Cumbria. LA6 2LF simon@drummond-hay.co.uk 015242 76202 / 07989 975 977 COPYRIGHT This report is the copyright of RS Drummond-Hay MRICS ACIH. Any unauthorised reproduction or usage by any person other than the addressee is strictly prohibited

Table of Contents Table of Contents... i Executive Summary... v Information base for analysis... v Character of the population... v Housing market areas in Wiltshire... vi Housing market prices and rents... viii Housing need and meeting it... ix The needs of particular groups... x Urban/rural comparisons... xii Affordable Rent... xii Balancing the housing market... xiv Overall level of newbuild... xv Policy implications and issues... xv 1. Introduction... 1 Background... 1 The Brief... 2 Range of evidence used in this study... 5 Analysis contained in this report... 6 2. Primary data collection... 9 Introduction... 9 Gathering the data... 9 Sub-area analysis... 10 Weighting the dataset... 13 Quality, accuracy and data protection... 14 3. Demographic and economic context... 15 Population... 15 Tenure profile... 17 Employment... 20 Income... 22 Dwelling stock... 24 4. The current housing market... 27 Introduction... 27 i

Relative prices... 28 The cost of housing in Wiltshire... 29 Analysis of housing market gaps... 32 Defining the Housing Market Area... 33 Interviews with Estate and Letting Agents... 34 Agents views on Wiltshire s Housing Markets... 34 Wiltshire s housing markets... 35 5. Housing need... 39 Summary... 39 Introduction... 39 Current need... 40 Future need... 41 Affordable housing supply... 44 Location of housing need... 48 Type of affordable accommodation required... 50 Size of affordable housing required... 53 Sensitivity analysis... 54 Local Housing Allowance & the impact of household benefit cap... 55 The existence of a benefit trap... 60 6. Particular household groups... 67 Summary... 67 Households with support needs... 68 Older persons... 74 Black and Minority Ethnic (BME) households... 77 Sexual orientation... 79 Families with children... 80 Young people... 84 7. Rural as compared with urban housing... 87 Summary... 87 Introduction... 87 Rural housing policy background... 87 Wiltshire s rural area... 88 Household characteristics... 91 Financial situation... 92 Travel patterns... 93 8. Affordable Rent: scope in Wiltshire... 97 Summary... 97 Introduction... 97 ii

Affordable Rent: a new tenure... 98 The impact on social rents... 100 Local Housing Allowance... 100 Understanding Affordable Rent... 101 Understanding the private rented sector in Wiltshire... 102 Estimating the Affordable Rent... 103 Affordable Rents compared to open market rents... 104 Affordability of Affordable Rent... 106 Supply of affordable housing to meet Affordable Rent... 109 What is the need for Affordable Rent?... 109 Generating new affordable dwellings vs. meeting need... 109 Current Grant Funding... 110 Development Economics of Affordable Rents... 111 Value of Affordable Rented Homes... 112 The use of relets... 115 What level should the Affordable Rent be set at?... 116 9. Improving market balance over the longer term... 117 Summary... 117 Introduction... 117 Demographic projections... 118 Relating the ONS projections to the Council s work... 119 Adequacy of the housing stock... 120 Tenure of housing required... 122 Size of housing required within each tenure... 124 Future specialist housing needs... 126 LTBHM outputs in 2016 and 2021... 129 Reconciling future estimates for affordable housing... 132 Tenure required if a 30% affordability threshold is used... 133 10: Policy Implications & Monitoring... 135 Introduction... 135 Final housing market areas for policy... 135 Housing Need... 137 Particular groups... 137 Rural issues... 137 Affordable Rent... 137 New affordable housing required... 138 Type and tenure of new housing required over the period... 138 Effect of total numbers changes... 139 Monitoring... 139 Glossary... 141 iii

Appendix 1: Housing costs by sub-area... 147 Appendix 2: Sub-area and Community Area outputs... 153 Introduction... 153 Sub-Areas... 153 Community Areas... 157 Appendix 3: HMA characteristics... 165 Introduction... 165 Patterns in accommodation type... 165 Patterns in household circumstance... 167 Patterns of work and travel... 168 Patterns in previous homes evidence... 169 Type of home sought... 172 The choice of home... 173 Overview... 175 Appendix 4. Search patterns... 177 Past moves: size and type of dwelling considered... 177 Past moves: tenure choice and outcome... 179 Past moves: maximum price and actual purchase price... 180 Past moves: maximum rent and actual rent... 181 Search pattern areas... 181 Reasons for moving and reasons for choosing... 183 Satisfaction with housing and life... 186 Planned future moves... 187 Conclusions... 190 Appendix 5: Estate Agents Observations... 191 Appendix 6: Stakeholder discussion of the report s findings... 197 Minutes of Wiltshire and Swindon Housing Market Partnership meeting: item on Wiltshire SHMA... 197 List of attendees at the Partnership Meeting... 198 Appendix 7: Questionnaire... 199 iv

Executive Summary 1. The 2011 Strategic Housing Market Assessment (SHMA) is the first study of this type to be carried out across the authority of Wiltshire. The study conforms to the key CLG Guidance (PPS3 Housing 2010 and Practice Guidance Version 2 (2007)). It also takes account of the new Coalition Government s agenda in including a detailed examination of Affordable Rent. Information base for analysis 2. This study draws on a variety of sources to provide a substantial evidence base. These include the policy framework, secondary data, primary data from a new household survey and qualitative information from several stakeholder consultations and a large number of interviews with estate and letting agents in the County. 3. The household survey was a substantial piece of new research that provides an up-to-date information source on the Wiltshire housing market. It was carried out by a postal survey. The questionnaire was designed in consultation with the Steering Group to meet the requirements of the Brief for the study and those of the CLG Guidance. Some 26,500 questionnaires were sent out, and a total of 5,948 households completed the survey. This is more than enough for a robust analysis of the Council area and relevant Community Areas and sub-areas. The dataset has been weighted so as to be representative of the household population of the County, using ONS 2008 household projections alongside other robust secondary data. Character of the population 4. The total estimated Wiltshire population and household numbers are presented in the table below. They show significant growth over the past decade. The 193,000 households are contained within just over 200,000 dwellings: there are always some vacancies due to the movement of the population around the stock. There is no significant issue of property being empty for lack of demand or unfitness. As can be seen, average household size has declined slightly since 2001, from 2.45 to 2.39. Table ES1 Household and population change, 2001 to 2008 and 2011 2001 2008 2011*** % change Households* 177,000 187,000 193,000 9.04% Population** 432,973 454,000 462,000 6.70% Average household size 2.45 2.43 2.39 - *CLG household estimates **ONS mid-year population estimates *** Data for 2011 are projections, not estimates v

5. The figure below shows the tenure profile in Wiltshire compared to the region and England as a whole. The figures show that Wiltshire s tenure profile more closely resembles that of the South West region rather than the national one. Figure ES1 Tenure (2001) Wiltshire 31.1% 40.9% 14.3% 7.9% 5.8% South West England 34.1% 29.2% 39.5% 39.0% 13.5% 19.3% 9.6% 8.8% 3.8% 3.2% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Owner-occupation (no mortgage) Owner-occupation (with mortgage) Social rented Private rented Other 6. There were 179,000 employee jobs in Wiltshire in 2008. The number of jobs has increased by over a tenth (12.6%) since 1998. The number of VAT registered businesses in the County at the end of 2007 was 18,370; this is an increase of 3,670 since 1997 (25.0%). The median earned income for full-time employed Wiltshire residents in 2010 was 25,392. Housing market areas in Wiltshire 7. Our analysis of the housing markets in existence in Wiltshire was based on existing (secondary) sources, discussions with stakeholders (mainly estate agents in this case) and analysis of the household survey. Our conclusions about housing markets in Wiltshire were at two levels (housing market area (HMA) and sub-market), as shown on the next map. The following is a list of the main HMA and sub-market areas which are mapped below: Housing Market Area The South the area to the south of Salisbury Plain. This area is centred on Salisbury and looks south. Salisbury Plain is a relatively sparsely populated area that forms a natural geographic boundary with areas to the north. Salisbury itself is sufficiently distinct from the rural areas and the towns to the west of it to justify being identified as a sub-market within this wider market. The West the area to the west of Devizes and Calne. The division of northern part of the County is broadly into two. Devizes and Calne can sit quite comfortably in either so have been defined as a single sub-market. The Northwest has strong links with the City of Bath and the major regional centre of Bristol. The M4 and links to the M5 are strong influences on this market. vi

The East the area to the east of Devizes and Calne. This part of the County looks towards the prosperous towns of Reading and Newbury and even further away to Oxford and London. Swindon in the northeast is a very strong influence on this part of the County. The areas close to Swindon have also been recognised as a separate sub-market. Distinct sub-markets Salisbury Plain This a distinct place overlapping the bottom portion of the Northeast and Northwest market areas. It is sparsely populated and those towns within the area tend to be lower priced areas than the remainder of the County. The area is not completely selfcontained however. The presence of the numerous military camps and towns and the emerging Super Garrison set the area a little apart from the rest of the County. Salisbury City is a relatively small city that it is in the heart of the southern market area that we have identified. Since much of the area is rural we have designated Salisbury a separate sub-market. Outer Swindon Swindon is a large and growing town. Swindon s boundaries are drawn closely to the edge of the built up area and so its influence on neighbouring villages outside its administrative area is strong. Devizes and Calne are similar in terms of housing market areas as they are both major towns in central Wiltshire. Devizes and Calne can sit quite comfortably in either the Northwest or the Northeast so have each been defined as a single sub-market. vii

Figure ES2 Wiltshire s Housing markets Housing market prices and rents 8. As can be seen from the figure below, Wiltshire is, on average, a high priced Council area: higher than both the England and South West averages. viii

Figure ES3 Land Registry price changes 2000-2010 (mean) Average price 300,000 250,000 200,000 150,000 100,000 Wiltshire South West England 50,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 9. This is consistent with its rural character (discussed below): the price differentials are significant between urban and rural housing. Figure ES4 Entry-level purchase price by size of dwelling 300,000 280,931 Entry-level purchase price 250,000 200,000 150,000 100,000 93,195 81,193 139,771 111,793 198,686 148,664 214,315 50,000 0 1 bed 2 bed 3 bed 4 bed Rural Urban Housing need and meeting it 10. The estimated net annual housing need across Wiltshire is 3,862 households. The largest net need is found in West Towns followed by North and Central Rural. The number of households in need would decrease from 3,862 to 2,721 if 30% of gross household income rather than 25%, were spent on housing. ix

11. The data suggests that almost half of new affordable housing should be Affordable Rent, 36.8% Shared Ownership and 14.6% social rented properties in terms of affordability. However, the actual proportion of Affordable Rent and shared ownership built should be determined by local demand. Some 46% of the net need is for one bedroom accommodation, 28% for two bedroom dwellings and a quarter for homes with three or more bedrooms. 12. There has recently been a reduction in the overall limit ( cap ) on benefit paid to help households with their rent (local Housing Allowance (LHA)) from the median private rented value to the 30th percentile. An overall benefit cap based on the average income of working households is also potentially going to be introduced. We examined the likely effect of these changes in Wiltshire. We found that the LHA cap is unlikely to affect many households. The total benefit cap may affect a few hundred households across the Council area, especially those requiring four-bed or larger dwellings. 13. We also analysed the issue of a benefit trap: whether the benefit system creates a substantial disincentive to enter employment by resulting in a fall in overall income from doing so. We found that there was a trap for very low income jobs, where the pay would not make up for benefits lost, but not for medium and higher earning jobs. The Coalition Government is planning to introduce a Universal Credit system where graduated payments mean that households will always be better off in employment. The needs of particular groups 14. There is a substantial analysis of the various particular groups identified in Guidance as requiring examination. There are an estimated 41,465 households in Wiltshire with one or more members in an identified support needs group - this represents 21.5% of all households. Just under half (43.2%) of all social rented dwellings contain a person with a support need. Support needs households report a lower than average income per head and are less likely to be able to afford market housing. x

Figure ES5 Proportion of households within tenure containing a support needs member 50.0% 43.2% 40.0% 30.0% 20.8% 20.0% 11.8% 15.0% 10.0% 0.0% Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented 15. Some 54,484 households in Wiltshire contain only older 1 people (28.2%) and a further 14,622 (7.6%) contain both older and non-older people. Less than a quarter (22.7%) of older person households are unable to afford market housing in Wiltshire. 16. Black and Minority Ethnic (BME) households tend to be more likely to live in rented accommodation. About two thirds of households where the household head is non-white are renters, compared to 36.6% among White British households. 17. Non-heterosexual households were considerably more likely to be private renters, and less likely to own their house without a mortgage 18. There are 54,582 households which contain at least one child in Wiltshire. Of these 17.8% are lone parent families, 53.2% are two parent families with at least one young child (aged 9 or under) and 28.9% are two parent families with only older children (aged 10-18). 78.4% of single parent families live in social/private rented accommodation.. 19. About 8,995 households in Wiltshire (4.7%) are estimated to be composed entirely of people under 30 and do not contain children. Half of young households (51.4%) live in the private rented sector. 1 Older people are defined as those over the state pension eligibility age (currently 65 for men, 60 for women). xi

Urban/rural comparisons 20. The Defra definition of rural allows settlements under 10,000 to exist within rural areas. By this measure about 20% of England s population live in rural areas. In Wiltshire the figure is 69%. We created an urban/rural typology for Wiltshire, which was used to analyse and understand the differences across the County using the household survey data. 21. There a high concentration of pensioners in rural areas. Lone parents and single non-pensioners decline as rurality increases. Income and capital (savings and equity) are higher in rural areas than in urban Wiltshire. 22. Car ownership is higher in rural areas among both high and low income families. Rural households are both most likely to work from home and to travel outside of Wiltshire for work. Area Not in employment Table ES2 Commuting by typology Work from home / mobile worker Work in local area* Work elsewhere in Wiltshire Work outside Wiltshire Urban 1 33.2% 7.2% 22.2% 21.8% 15.6% 51,462 Urban 2 35.8% 7.2% 20.5% 22.9% 13.5% 46,893 Urban 3 24.8% 5.7% 24.5% 24.6% 20.5% 12,620 Rural 1 36.7% 8.8% 12.7% 15.4% 26.5% 16,576 Rural 2 31.5% 15.3% 9.5% 22.3% 21.4% 65,448 Countywide Total 33.0% 10.0% 16.8% 21.9% 18.3% 193,000 * working in an area which is both in the same Community Area and sub-area Affordable Rent 23. Flexible Tenancies are being introduced as a new tenure. They will not give the tenant security of tenure for life and will allow Affordable Rent to be charged. Affordable Rent will be based on the open market value of each property and the intention is that it should be set at 80% of this value. We modelled the potential cost of Affordable Rent at 80% of median of the middle market of private rents. We also worked out equivalent costs for Affordable Rent at 70% and 60%. 24. The ability of these products to be afforded by suitable households could then be tested. The following table shows how many households on the Register (only some of whom will be in housing need as defined by CLG) can afford Affordable Rent at 80% of the median, and down to 60% of the median: xii

Table ES3 Affordability of households on housing register Households on Register % of households on Register Market housing 4,571 25.3% Shared Ownership (75%) 913 5.0% Shared Ownership (50%) 1,435 7.9% Affordable Rent at 80% 137 0.8% Affordable Rent at 70% 1,341 7.4% Affordable Rent at 60% 2,601 14.4% Social rent 216 1.2% Need LHA 6,884 38.0% Total number of households 18,096 100.0% 25. This table shows that very few households on the Register could afford Affordable Rent at 80%, that could not also afford a shared ownership property. If the Affordable Rent were set at 60% the potential demand for it (excluding those able to afford shared ownership) from households on the Register is 3,130. If the Affordable Rent level were set at 70%, it would generate a potential demand from 1,150 households. 26. The housing needs model suggests that the future annual supply of affordable dwellings will consist of 97 shared ownership homes, 2,282 social rented properties and 9 Affordable rent units. If all social rented re-lets were converted to Affordable Rent the supply of this tenure could be 2,291 units per year. The registered housing providers in Wiltshire are currently considering converting 50% of all social rented relets to an Affordable Rent tenure. This would lead to a potential supply of 1,150 homes per year. This is exactly the same as the potential demand for Affordable Rent at 70%. 27. Using these figures, the most sensible rate at which to set Affordable Rent is: 70%. If this fails to generate sufficient demand from the Register a rate of 60% would be more appropriate. 28. We examined the costs of building various kinds of affordable housing. The following diagram illustrates the level of profit or subsidy involved in them. As can be seen, it is not possible to build Affordable Rent at below 80% without subsidy. The level of subsidy required for 70% and 60% Affordable Rent is not great compared with social rent. Moreover if existing social relets are converted to Affordable Rent there is a net profit (when considering future income streams). If Affordable Rent were set at 70%, one conversion of a social relet to Affordable Rent would pay the net subsidy required to build 5 dwellings of at the same price. This is quite an efficient trade off. xiii

Figure ES6 Development Surplus by Tenure (3 Bedroom) 50,000 44,624 40,000 32,577 30,000 20,000 19,984 10,000 7,392-41,376 2,947-3,976-7,853-16,253-29,553-26,289 0-10,000-20,000-30,000 Freehold Purchase Shared Ownership 60% Shared Ownership 50% Shared Ownership 40% Open Market Rent Affordable Rent 80% Affordable Rent 70% Affordable Rent 65% Affordable Rent 60% Affordable Rent 50% Social Rent - 40,000-50,000 NB this table takes into account the varying land price paid by developers for different types of housing land. Balancing the housing market 29. In order to provide a long term view of the changes required to the housing stock, the Long Term Balancing Housing Markets model (LTBHM) has been developed. This takes the population and household projections and, adjusting occupation patterns to ensure all households are adequately housed, determines an ideal stock profile for Wiltshire in 2026 30. The overall ONS forecast for households 2011-2026 is 34,000 in Wiltshire. The LTBHM suggests that about 55% of the additional housing should be market housing and 45% affordable. However in terms of subsidy cost the affordable housing would be much cheaper to produce than the traditional social housing. About 40% of it would be shared ownership and 60% Affordable Rent. There is hardly any net requirement for social rent over the period. Tenure Table ES4 Tenure of new accommodation required in Wiltshire over the next 15 years Current tenure profile Tenure profile 2026 Change required % of change required Market 148,965 167,701 18,736 55.1% Shared Ownership 825 6,642* 5,817 17.1% Affordable Rent 0 8,907 8,907 26.2% Social rented 16,757 LHA 26,454 } 43,750 540 1.6% Total 193,000 227,000 34,000 100.0% *Indicates the ability of households to afford Shared ownership rather than the demand for this tenure, which is particularly low in Wiltshire. xiv

31. In terms of size of new homes required over the next 15 years, most of the market and shared ownership housing should be of 2- and 3-bed sizes. The other major element, Affordable Rent, has a requirement spread mainly over the 1- to 3- bed range. Overall level of newbuild 32. Whilst the LTBHM model has been based on the ONS population and household projections which indicate household growth of 34,000 in Wiltshire over the next 15 years, the Council have now adopted a target of 37,000 new homes across the County between 2006 and 2026. This equates to 27,188 new dwelling between 2011 and 2026, once dwellings that have been completed in the last five years have been taken into account. This does not affect the proportions of tenure and size mix discussed in the previous subsection; the proportionate results of the LTBHM can be adjusted to the lower target being pursued by the Council. Policy implications and issues 33. The final chapter of the SHMA reviews policy implications. It is of course for the Council to set policy, studies such as this provide only part of the evidence base. However some implications are drawn out in this chapter. They include: i) Housing need is quite high, and although relatively lower in rural areas, it is there that the policy issue is strongest. That is because of the very low supply in rural areas. The two main rural areas (West and South) have affordable relet supply of (well) below 20%. Of urban areas only Trowbridge shows this pattern. It is common in England to find low affordable supply in rural areas due to the effects of Right to Buy on the former affordable stock in villages. Clearly this is an important policy issue. ii) In relation to particular groups the outstanding policy issue is the ageing population. One response to that which is available in Wiltshire concerns equity release for the many older owner-occupiers. This is likely to become an ever more important policy issues. This is especially true given the high older population in rural parts of Wiltshire. iii) Affordable Rent could be a valuable tenure in Wiltshire. We suggest that a level of 70% of median market rent, rather than the standard 80%, would be the best level. However this is just a policy suggestion: there are several possible choices. iv) In terms of recommended tenure split, the existing 40% target could be maintained or increased slightly. In relation to the types of affordable housing the evidence suggests 40% shared ownership and 60% Affordable Rent, but there is not historically been a high level of take-up for shared ownership products within Wiltshire, at least on the types of sites on which it has been made available. It may be that this is an issue for policy consideration. v) The analysis throughout the SHMA has found that there are three housing markets areas in Wiltshire. The map below depicts how they correspond to the Community Areas and should be defined in terms of housing policy. xv

Figure ES 7 Comprehensive Housing Market Areas in Wiltshire xvi

1. Introduction Summary i) This Strategic Housing Market Assessment (SHMA) report is a comprehensive analysis of the Wiltshire Housing Market. It includes a review of the current local situation, a discussion of the housing market dynamics and suggested policy responses. This Strategic Housing Market Assessment (SHMA) is a key element of the evidence base for developing the Wiltshire Local Development Framework (LDF). ii) The report follows the general structure of the CLG Strategic Housing Market Assessment Practice Guidance of August 2007. The research has been carried out to the requirements of Planning Policy Statement 3 (PPS3) and Planning Policy Statement 12 (PPS12). The document also takes account of changes since the publication of the main CLG Guidance; in particular as regards Affordable Rents, the new social tenure and reforms to Local Housing Allowance. iii) A range of evidence was used to inform the outputs of this study. This includes data from a survey of local households (primary data), secondary sources a consultation with key stakeholders in the County and face-to face-interviews with estate/letting agents. Background 1.1 Wiltshire Council is a Unitary Authority. It is a substantially rural county, though containing a number of towns and a city within it. The following map from the 2009 Draft South West Spatial Strategy shows the four districts of the former two tier County (Swindon being excluded from the new authority and also the scope of the present study). 1.2 The recent change in administrative status inevitably revealed that the existing district studies of the housing market and housing needs in Wiltshire were not consistent across the County. After some consideration it was decided by the Council to carry out a new household survey to enable analysis across the whole area of the new authority and provide a robust up-to-date information source on the local population. 1

Figure 1.1 Market areas in Wiltshire Source: Draft South West Spatial Strategy 2009 1.3 This study a key element in the developing Wiltshire Local Development Framework (LDF), and is designed to be a robust part of the information base for that framework. This is a substantial piece of work that will inform the Council s wider planning and decision making particularly concerning housing. 1.4 Early in 2011 Wiltshire Council commissioned Fordham Research to undertake this Strategic Housing Market Assessment (SHMA). Fordham Research ceased trading on 20th July 2011 and the business was closed. Wiltshire Council then commissioned RS Drummond-Hay MRICS ACIH to complete the report. The Brief 1.5 The Brief requires a robust study, covering the whole of the County: The aim of this study is to deliver robust evidence to inform and justify policy proposals relating to housing, capable of withstanding scrutiny at an Examination in Public, and which is fully compliant with appropriate planning guidance, in particular, PPS3 and the SHMA PGv2 (page 8 of Request for Quotation) 2

1.6 The most specific output required by PPS 3 (and echoed at greater length in the Practice Guidance (Figure 1.1 of the Practice Guidance Version 2)) is: Based on the findings of the Strategic Housing Market Assessment and other local evidence, Local Planning Authorities should set out in Local Development Documents: i) The likely overall proportions of households that require market or affordable housing, for example, x% market housing and y% affordable housing ii) iii) The likely profile of household types requiring market housing e.g. multi-person, including families and children (x %), single persons (y %), couples (z %) The size and type of affordable housing required. [Planning Policy Statement 3: Housing (June 2010) paragraph 22] 1.7 The following table lists the detailed requirements of the Guidance and where they are found in the report: 3

Table 1.1 Practice Guidance Steps in Wiltshire HMA Report Chapter of the SHMA report Step identified in the Practice Guidance Version 2 (August 2007) Page in Practice Guidance Ch 1 Step 3.1.2 National and Regional Economic Policy 20 Ch 3 Step 3.1.1 Demography and Household Types Step 3.1.3 Employment Levels and Structure Step 3.1.4 Incomes & Earnings Step 3.2.1 Dwelling Profile Step 3.2.2 Stock Condition Step 3.2.3 Shared Housing and Communal Establishments 19 21 22 23 24 25 Ch 4 Step 3.3.1 The cost of buying or renting a property 26 Ch 3, 4 Ch 5,8 Step 3.3.2 Affordability of housing Step 3.3.3 Overcrowding and Under-Occupation Step 3.3.4 Vacancies, Available Supply and Turnover by Tenure Step 5.1.1 Homeless households and those in temporary accommodation Step 5.1.2 Over-crowding and concealed households Step 5.1.3 Other groups Step 5.1.4 Total current housing need Step 5.2.1 New household formation Step 5.2.2 Proportion unable to afford entry-level market housing Step 5.2.3 Existing households falling into need Step 5.2.4 Total newly arising need Step 5.3.1 Affordable dwellings occupied by households in need Step 5.3.2 Surplus stock Step 5.3.3 Committed supply of new affordable units Step 5.3.4 Units to be taken out of management Step 5.3.5 Total affordable housing stock available Step 5.3.6 Future annual supply of social re-lets (net) Step 5.3.7 Future annual supply of intermediate affordable housing Step 5.3.8 Future annual supply of affordable housing units Step 5.4.1 Choices with the existing affordable housing stock Step 5.4.2 Requirement for affordable housing of different sizes Step 5.4.3 The private-rented sector Step 5.5.1 Estimate of net annual housing need 29 30 31 43 44 44 45 45 46 46 46 47 47 48 48 48 48 49 50 50 50 51 52 Ch 7, App 4 Step 3.4.2 Trends and drivers 33 Ch 9 Step 4.1 Projecting changes in the future numbers of households Step 4.2 Future Economic Performance Step 4.3 Future affordability 35 36 37 Ch 6 Specific Groups (no formal steps) 54-60 Ch 10 Step 3.4.3 Issues for future policy/strategy Step 4.4 Bringing the evidence together Step 5.5.2 Key issues for future policy/strategy Step 5.5.3 Joining across the assessment 33 38 53 53 Throughout Step 3.4.1 Mapping market characteristics and future growth 32 Source: CLG Practice Guidance 2009 4

1.8 To this are added various requirements of the Localism Act. The most notable for the purposes of this study is Affordable Rent. This new social tenure is the main object of planned future housing investment by the HCA (Homes and Communities Agency) so we have included a detailed analysis of it s potential scope within chapter 8. Range of evidence used in this study 1.9 In accordance with Practice Guidance a range of data sources were used to create a robust and diverse evidence base. In addition to the use of data, information was gathered from a number of relevant publications and through discussions with local stakeholders. Below we provide a brief summary of the main sources of information used in this report: 1.10 Secondary data: There are a range of data sources already available at the local, regional and national level which provide a good background to the housing market and how it is changing. Data sources considered in analysis include Land Registry data, the 2001 Census, the Annual Survey of Hours and Earnings (ASHE), Housing Strategy Statistical Appendix (HSSA) data and published household/population projections. In addition, a number of relevant documents were reviewed to inform the policy context in which the Wiltshire housing market exists. 1.11 Primary data: In addition to studying secondary data sources it was necessary to conduct a local household survey for Wiltshire. The survey data allowed many of the gaps in secondary data to be filled. Most notably the survey data established how households act within the housing market and an understanding of why they do so. It should be noted that the survey dataset is wide-ranging, and; the analysis presented in this report is mostly restricted to those issues directly related to the functioning of the housing market. However, the dataset is available to the Council for use in further research. The methodology used to collect and validate the data that forms the household survey is described in chapter 2. 1.12 Estate agent information: A key part of the project was a series of face-to-face interviews with estate/letting agents in the local area. The dynamics of the housing market (e.g. what types of households are seeking properties) were discussed at these interviews, which provided a good overview of what has actually been happening locally. The interviews were supplemented by an internet property/rent price search to establish the costs of housing in the local area. 1.13 Stakeholder consultation: Stakeholder views have been sought through the process and incorporated into the study. Stakeholders were consulted at two events, one prior to the survey analysis, in order to inform and direct the work, and one afterwards to obtain feedback on the results. The notes from these meetings are contained at Appendix 6. 1.14 The production of the SHMA has been an iterative process; the views of stakeholders have been necessary to help the research evolve. The resultant document addresses the particular situation in Wiltshire in addition to presenting the outputs required by both PPS3 and the Practice Guidance. 5

1.15 An SHMA is designed to be an ongoing process, and the Steering Group will build on the relationships developed through discussions both before and during the present SHMA process to ensure the sustained involvement of a wide range of agencies in developing further understanding of the Wiltshire housing market. Analysis contained in this report 1.16 This SHMA report presents a comprehensive analysis of the Wiltshire Housing Market. It contains two housing models, each with a different objective. Chapter 5 presents the CLG housing need model, calculated in accordance with (chapter five of) the Practice Guidance. This assesses the absolute need for affordable housing in Wiltshire using established definitions. The outputs produced enable comparison with previous estimates and estimates made in other areas of the country. The model is an analysis of the situation now. 1.17 A second model, the Long Term Balancing Housing Markets model (LTBHM) is detailed in Chapter 9 of the report. This model considers the nature of housing required in Wiltshire over the longer plan period (the majority of outputs address changes required over fifteen years). This model addresses the full housing market rather than focusing principally on the affordable sector. Whilst the two models address different aspects of the housing market, the results from both of them are used together to inform the suggested policy responses that are presented in chapter 10 of this document. 1.18 The full structure of the report is: Chapter 2 sets out the approach to data collection and sample validation of the household survey, which forms a crucial component of the evidence base used in this report and therefore needs to be transparently described. This chapter also introduces the sub-areas and Community Areas that the County has been divided into to provide lower-level analysis. The results of the lower-level analysis are presented in Appendix 2. Chapter 3 of the report provides socioeconomic context to the study area based primarily on secondary data. The chapter presents the latest background information available on the local population, the labour market in the County and the nature of the housing stock. 1.19 Chapters 4 examines the housing markets in Wiltshire in considerable detail. The chapter establishes the cost of different tenures of housing and how they vary across the County. It considers the evidence for different housing markets that exist in Wiltshire before concluding on a typology that best fits the County. This chapter is supported by Appendix 3 which provides more detail on the evidence of the housing market boundaries concluded. Appendix 4 further investigates the search patterns used by households to determine what time of home they move to, which provides considerable insight into the dynamics of the Wiltshire housing market. As indicated above Chapter 5 contains the results of the CLG housing need model in Wiltshire. Chapter 5 also considers the changes to the Local Housing Allowance cap, identifying the number of households in Wiltshire likely to be affected. 6

1.20 The household survey is a vast resource and contains robust information on subsets of the population as well as the County as a whole. Chapter 6 therefore provides detailed analysis of the housing market situation of various sub-groups of households that the Practice Guidance indicates as appropriate for specific consideration. These include ethnic minorities, support needs groups, pensioners, younger people and families. 1.21 Wiltshire is a large and diverse County and there are a range of ways that it can be logically divided to present data at a more geographically detailed level. This includes the sub-area and Community Area analysis noted above, but the dataset can also be disaggregated by the rurality of the location. This enables a comparison of the rural and urban areas of Wiltshire. This is presented in chapter 7. 1.22 Chapter 8 introduces the new tenure of Affordable Rent and assesses it s ability to meet housing need. The chapter goes on to consider the relative viability of different Affordable Rent options before concluding at what level that it should usefully be set at. As described above chapter 9 presents the Long Term Balancing Housing Markets model. Finally chapter 10 summarises the key conclusions of the report and the potential policy actions that these imply are presented for consideration. 7

2. Primary data collection Summary i) Primary data analysed within the report is based on a postal questionnaire completed by Wiltshire residents in February 2011. The survey was distributed to a stratified random sample of 26,500 households in the County. A total of 5,948 households completed the survey. The dataset is weighted using ONS 2008 household projections alongside other robust secondary data. ii) The study was designed to provide outputs for two sets of areas; comprehensive outputs for eleven sub-areas, and descriptive outputs for each of twenty Community Areas. Introduction 2.1 A household survey was undertaken in Wiltshire in February 2011. This was a substantial piece of new research and provides an up-to-date information source on the Wiltshire housing market. The household survey provides primary data that can be used to undertake statistical analysis of the area. It forms a major part of the evidence base used to produce this SHMA, therefore it is important to explain how this household survey was conducted and the data collected used. Gathering the data 2.2 The postal questionnaire survey was sent to 26,500 households across Wiltshire. These households (the sample) were drawn at random from the Council Tax Register, covering all areas and tenure groups in Wiltshire. Although random, the sample was stratified in order to achieve a sufficient response rate for each of the sub-areas and Community Areas the Council wanted to provide individual analysis for (shown below). The total sample of 26,500 was calculated as the minimum required for this level of analysis. 2.3 The survey included questions intended to examine current housing circumstances, the detail of past and future moves, and the general situation of the household (e.g. number of children, income, education levels). A copy of the questionnaire is provided in Appendix 7. In total 5,948 completed postal surveys were returned. This is significantly in excess of the 1,500 per local authority recommended by the Practice Guidance and allows reliable analysis at a County-wide and sub-area level. The overall response rate to the survey was good at 22.4%. 9

Sub-area analysis 2.4 The study was designed to provide outputs for two sets of areas; comprehensive outputs for eleven sub-areas, and descriptive outputs for each of twenty Community Areas. The eleven sub-areas used are shown in the table and map below. The sub-areas are intended to separate out urban and rural areas, and provide insights into the housing circumstances at this scale. Many of the smaller towns have been grouped together in order to allow statistically robust analysis. Table 2.1 Sub-areas in Wiltshire Sub-area name Areas contained Bradford-on-Avon Chippenham Devizes Rural North and Central North Small Towns North Towns Salisbury Salisbury Plain Rural South Trowbridge West Towns Bradford-on-Avon Chippenham Devizes Rural areas in the former districts of Kennet, North Wiltshire and West Wiltshire Cricklade, Malmesbury, Marlborough and Pewsey Calne, Corsham and Wootton Bassett Salisbury and Wilton Amesbury, Bulford, Durrington, Larkhall, Ludgershall and Tidworth Rural areas in the former district of South Wiltshire, including Mere, Tisbury and Downton Trowbridge Melksham, Warminster and Westbury Source: Wiltshire Council (2011) 10

Figure 2.1 Sub-areas in Wiltshire Source: Wiltshire Council (2011) 2.5 Descriptive analysis was also possible for the Community Areas of Wiltshire, shown on the map below. These areas are used by the Council for various administrative and statistical purposes. Survey outputs for these both the sub-areas and the Community Areas are provided in Appendix 2 11

Figure 2.2 Community Areas in Wiltshire Source: Wiltshire Council (2011) 2.6 The response rate to the household survey varied from 27.3% of surveys returned in Bradford-on- Avon to 18.0% in the military part of Salisbury Plain. Overall, the number of responses to the survey in each area was sufficient to provide accurate analysis at the level originally envisaged. 12

Weighting the dataset 2.7 The dataset is weighted so that the survey response best reflect all households in the area studied. Weighting a dataset is a process whereby every response to the survey is assigned a number, equivalent to the number of real households in the County it is taken to represent. Where a type of household is thought to be over-represented in the response to the survey, the number attached to this group of respondents (the weight ) is lowered. On the other hand if a type of household is thought to be under-represented in the survey response, they are assigned a higher weight. 2.8 This process uses secondary data to determine the real numbers of households in particular categories. For example, the number of households living in social rented housing is known by the Council, and so the data can be weighted to reflect this, compensating for this group having a lower or higher response rate than average. 2.9 It is particularly important to determine the total number of households in the County. To determine the number of households in 2011, we used the ONS 2008-based household projections (published in 2010). The estimated total number of households in 2011 in Wiltshire used by the report is therefore 193,000. 2.10 The other variables used to weight the data are listed below, alongside an indication of the secondary data source the appropriate profile was derived from. Secondary data sources are projected forward where possible to produce estimated figures for 2010. tenure: Wiltshire Housing Strategy Statistical Appendix (HSSA) (2011), Census (2001), English Housing Survey (EHS) (2009) sub-areas: Wiltshire Council Tax Registers (2011) Council Tax Band: Wiltshire Council Tax Registers (2011) household type: ONS 2008-based household projections (2010), Census (2001) population age groups: ONS 2008-based population projections (2010), Census (2001) accommodation type: Census (2001), Survey of English Housing (2008), English Housing Survey (EHS) (2009) car ownership: Census (2001), National Travel Survey (2009). 2.11 As an example, the table below shows an estimate of the current tenure split in Wiltshire and also the sample achieved for each tenure. The data shows that in 2011 around 62.7% of households are owner-occupiers with 19.2% living in the social rented sector and 18.1% resident in private rented accommodation. 13

Table 2.1 Number of households in each tenure group: changes since the 2005 HNA Tenure Weighted SHMA data Total number of households % of households Unweighted SHMA data Total number of households % of households Owner-occupied (no mortgage) 65,357 33.9% 2,714 45.6% Owner-occupied (with mortgage) 55,512 28.8% 2,018 33.9% Social rented 37,115 19.2% 603 10.1% Private rented 35,016 18.1% 613 10.3% Total 193,000 100.0% 5,948 100.0% Source: Wiltshire SHMA, (2011) Quality, accuracy and data protection 2.12 Quality is paramount to work carried of this type. The methodology used is established and has always found to be robust. The work has been undertaken by people with considerable knowledge and experience in their field. 2.13 The data entry system (used for inputting the data from postal questionnaires) has built-in logic checks that come into effect if illogical data are inputted. This is important to ensure the accuracy of the data as it cannot be assumed to be precise without careful checking and, where necessary, correction. On receipt of inputted data we apply various checks. By carrying out such checks we ensure that all data sets are logical and correct. 2.14 All the information that respondents provide is treated as confidential and used only for the purpose of this SHMA. This report contains grouped responses only (i.e. does not provide results for any individual household). The household-level dataset which is supplied to the Council is anonymous; addresses of properties surveyed are not supplied. 2.15 The postal questionnaires forms contain a unique reference number for each household sampled, from which the Project Manager can identify the respondent s postcode. This enables the number of responses in each ward or sub-area to be ascertained. 2.16 The files containing ID numbers and address details are saved in separate locations in order to safeguard against computer fraud. Names of respondents are never requested (and are not provided to us in the first instance), to comply with the Data Protection Act 1998. RS Drummond-Hay MRICS ACIH is data protection registered. 14

3. Demographic and economic context Summary i) The household population in Wiltshire in 2011 is 462,000 and the number of households is 193,000. The population as a whole and the number of resident households in Wiltshire appears to have been increasing at a similar rate since 2001. The average household size in the County has remained stable at 2.4 persons per household. ii) Levels of owner-occupation are generally high in all areas except Salisbury and Chippenham. The far south west of the County has the highest proportion of private rented accommodation. iii) There were 179,000 employee jobs in Wiltshire in 2008. The number of jobs has increased by over a tenth (12.6%) since 1998. The number of VAT registered businesses in the County at the end of 2007 was 18,370; this is an increase of 3,670 since 1997 (25.0%). The median earned income for full-time employed Wiltshire residents in 2010 was 25,392. iv) HSSA data indicates that the overall increase in the housing stock in Wiltshire over the period 2000 to 2010 was lower than experienced regionally but higher than the national rate. 3.1 Two of the main drivers of the housing market are the resident population and the local labour market. They affect the nature of housing demand including household formation rates and households investment in housing. This chapter documents the current socio-economic profile in Wiltshire and recent trends that have occurred. The information presented compares the circumstances in the County to the regional and national situation. This provide useful background knowledge before the local housing market is examined in more detail in subsequent chapters. Population 3.2 Between 1981 and 1991, the population of Wiltshire increased slightly slower than was the case regionally. However since 1991, the population increase in Wiltshire has been greater than the increase in the population of the South West. Since 1981 the rate of population increase in Wiltshire and across the South West has been consistently greater than that of England as a whole. Between 1981 and 2009, the increase in population in Wiltshire was 21.6%. Regionally, the increase was 19.3% and nationally 10.7%. 15

Figure 3.1 Population change in Wiltshire (1981-2009) Change in population (1981 Index=100) 120% 110% 100% Wiltshire South West England 90% 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 Source: ONS mid-year population estimates 1981-2009 (from Nomis) 3.3 The 2001 Census indicates that 96.2% of the population are White-British, which is slightly more than the regional figure of 95.4% and much higher than the national average (87.0%). There is a correspondingly smaller non-white ethnic population. 3.4 ONS have produced some estimates of the changes in population by ethnicity to 2007, although these are classed as experimental statistics and should be treated with caution. They suggest that the BME population 2 of Wiltshire increased from 3.8% to 6.4% of the total population since the Census (this includes white non-british groups). Figure 3.3 presents the ethnicity of the population in the County according to the latest (2007) estimates. 2 Defined as all ethnic groups other than White - British 16

Figure 3.2 Ethnicity of the Wiltshire population, 2007 0.7% 2.2% 93.6% 6.5% 1.0% 1.1% 0.7% 0.8% White - British White - Irish White - Other Mixed Asian Black Chinese or Other Source: ONS Resident Population Estimates by Ethnic Group, June 2007 3.5 The population, as a whole and the number of resident households, in Wiltshire appears to have been increasing at a similar rate since 2001. The average household size in the County has therefore remained stable at 2.4 persons per household. Table 3.1 Household and population change, 2001 to 2008 and 2011 2001 2008 2011*** % change Households* 177,000 187,000 193,000 9.04% Population** 432,973 454,000 462,000 6.70% Average household size 2.45 2.43 2.39 - Sources: *CLG household estimates **ONS mid-year population estimates *** Data for 2011 are projections, not estimates Tenure profile 3.6 Analysis of 2001 Census data reveals that 72.0% of households in the County were owner-occupiers, slightly lower than the regional equivalent but higher than the national figure. Around a sixth of households resided in the social rented sector, higher than the South West and lower than nationally. The private rented sector makes up a smaller proportion of households than both the regional and national level. 17

Figure 3.3 Tenure (2001) Wiltshire 31.1% 40.9% 14.3% 7.9% 5.8% South West England 34.1% 29.2% 39.5% 39.0% 13.5% 19.3% 9.6% 8.8% 3.8% 3.2% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Owner-occupation (no mortgage) Owner-occupation (with mortgage) Social rented Private rented Other Source: Census 2001 3.7 The maps below show the proportion of households living in each of the three broad tenure groups (owner-occupation, social rent and private rent) by ward. There are notable variations within the County; levels of owner-occupation are generally high in all areas except Salisbury and Chippenham, the far south west of the County has the highest proportion of private rented accommodation and there are small proportions of social rented housing across the Salisbury plains and higher densities of this tenure in some urban areas. 18

Figure 3.4 Tenure by ward Owner-occupied Social rented Private rented Source: Census 2001 19

Employment 3.8 Measured by the most recent Annual Business Inquiry (ABI), there were 179,000 employee jobs in Wiltshire in 2008. Despite periodical falls in jobs over the period, the number of jobs in Wiltshire has increased by over a tenth (12.6%) since 1998. The South West as a whole experience greater growth of 14.8% whilst for Great Britain the figure was 9.5%. Figure 3.5 Indexed employment growth in Wiltshire (1997-2007) 120% Wiltshire Change in employee jobs (1997 Index=100) 115% South West Gr eat Britain 110% 105% 100% 1998 2000 2002 2004 2006 2008 95% Year Source: ONS Annual Business Inquiry Employee Analysis 2008 (from Nomis) 3.9 In terms of employment, the Public admin, education & health sector is by far the largest sector of employment, representing 28.5% of all jobs in the County in 2008; similar to the levels in the South West and Great Britain as a whole (data is from ONS s Annual Business Inquiry). Distribution, hotels and restaurants account for 25.0% of employee jobs in the County, the same as the regional figure. The Finance, IT and other business sectors are proportionally smaller in the County and region at around 19% of employee jobs compared to 22% in Great Britain as a whole. 3.10 BERR (the Department for Business, Enterprise and Regulatory Reform) provides information about the number of VAT registered businesses in an area and how this has changed over time. This can provide a good indication of the state of the economy as an increase in VAT registered business would suggest either new companies moving to the area or an increase in local entrepreneurship. The number of VAT registered businesses in the County at the end of 2007 was 18,370; this is an increase of 3,670 since 1997 (25.0%). This proportional increase is higher than that recorded for the region (19.5%), and for Great Britain (21.3%). 20

3.11 Job density is a measure of the number of jobs per person of working age in an area. According to ONS data, in 2008 there were 0.82 jobs per working-age person in the County. This is the same as the ratio for the South West region whilst the ratio for Great Britain is 0.79. 3.12 ONS publish the number of people claiming Jobseeker s allowance (JSA) on a monthly basis. This provides a very up-to-date measure of the level of unemployment of residents in an area. The figure below shows the change in the proportion of the working-age population claiming JSA in Wiltshire since January 2006. The figure indicates that although Wiltshire historically has had a relatively low unemployment level in recent years, in the 12 months from summer 2008 the rate of unemployment in the County has shown a notable increase, mirroring the national and regional trends. This levelled off and somewhat and reduced in 2010. However, the current trend suggests that unemployment is rising again. Overall the number of people claiming job seekers allowance in Wiltshire in February 2011 was more than double the number recorded in the same month in 2008. Figure 3.6 Level of unemployment in Wiltshire (2006-2011) Proportion of working age population claiming job s eek ers allowance 5% 4% 3% 2% 1% 0% Wiltshire South West England Jan-06 Jul-06 Jan-07 Jul-07 Jan-08 Jul-08 Jan-09 Jul-09 Jan-10 Jul-10 Jan-11 Year Source: ONS Claimant count 3.13 The Annual Population Survey provides information on employment by occupation, categorising all working people resident within an area into one of nine groups depending on the nature of the skills that they use. These nine groups are graded from managerial jobs (Groups 1-3) to unskilled jobs (Groups 8-9). As illustrated below, half (50.9%) of employed residents in Wiltshire are classified as being in groups 1 to 3, which is higher than the regional and national levels. 21

Figure 3.7 Occupation structure 2009/10 Wiltshire 50.9% 21.6% 12.4% 15.1% South West 45.0% 21.2% 16.2% 17.7% England 43.9% 23.0% 15.8% 17.4% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Group 1-3: Senior, Professional or Technical Group 6-7:Personal service, Customer service and Sales Group 4-5: Administrative, skilled trades Group 8-9: Machine operatives, Elmentary occupations Source: ONS Annual Population Survey 2009/10 3.14 There is some data available to offer an insight into the trends of migrant workers in Wiltshire. The National Insurance Recording System collates the number of National Insurance Number registrations by non-uk nationals. This does not offer an indication of the current number of migrant workers in the area as it only records new migrant workers registering for a National Insurance Number, nor does it paint an accurate picture of the trends of non-uk nationals arriving in the area as it can often take several years for arrivals to register for a National Insurance Number. 3.15 The figures suggest that during 2009/10, 1,300 non-uk nationals registered for National Insurance Numbers in Wiltshire. Those registered originated from a range of different countries, the largest proportion were from Poland (23%), followed by Philippines (6%), India (5%) and Romania (5%). Income 3.16 Income has a crucial effect on the range of choice a household has in their future accommodation. The median earned income for full-time employed Wiltshire residents in 2010 was 25,392, according to the ONS Annual Survey of Hours and Earnings (ASHE), higher than the South West median (at 24,466) but lower than the median for England as a whole ( 26,268). The figure below shows that this pattern is repeated when lower and upper quartile incomes are considered. 22

Figure 3.8 Annual gross income of full-time employees (2010) 40,000 35,000 36,332 34,489 37,247 Annual Gross Income 30,000 25,000 20,000 15,000 10,000 5,000 25,392 24,466 18,096 17,831 18,577 26,268 0 Wiltshire South West England Lower Quartile Median Upper Quartile Source: ONS Annual Survey of Hours and Earnings (2010) 3.17 ASHE data suggests that the median income in Wiltshire has barely changed (-0.2%) over the last three years. Within the same time period the regional increase was 2.1% and the national increase was 2.8%. 3.18 It is possible to compare the incomes of full-time employed Wiltshire residents with those of nonresidents working in jobs located in the County. The median income of the latter group is 23,999, which is about 1,400 lower than the median income of (employed) County residents. This implies that many Wiltshire residents travel to work to areas outside of the County were salaries are higher. 3.19 ASHE data enables a useful comparison between areas, but the data covers employees only and so cannot be used to provide a full profile of the area. Furthermore the data is presented at an individual level, rather than at a the household level, on which a housing market assessment must be based. It is necessary to supplement this analysis of ASHE data with a profile of household income in Wiltshire recorded by the household survey. The primary survey obtained information from all types of household in the County and collected income information at the household rather than individual level. 3.20 Survey results for household income estimate the mean gross household income level to be 38,106 per annum, including households without any members in employment. The median income is noticeably lower than the mean, at 28,100 per annum. 3.21 Figure 3.9 shows the distribution of incomes in Wiltshire. It is clear that there is a wide range of incomes in the County as the largest income groups are 10k- 20k and 20k- 30k followed by 60k+. 23

Figure 3.9 Distribution of annual gross household income in Wiltshire 25% 20% 23.0% 21.7% 20.5% % of households 15% 10% 8.6% 12.5% 10.2% 5% 3.5% 0% 0-10k 10-20k 20-30k 30-40k 40-50k 50-60k+ 60k+ Source: Wiltshire household survey 2011 3.22 Almost half (46.2%) of households with incomes of less than 20,000 have a retired household reference person (HRP). Households where the HRP is in employment have a mean income of around 45,629 (median of 36,858). Dwelling stock 3.23 Analysis of the current stock of housing allows a broad assessment of the range and condition of properties currently within the County. The Council s 2009/2010 Housing Strategy Statistical Appendix (HSSA) records that there are 200,488 dwellings in the County. The table below presents a summary of key information on the Wiltshire housing stock compared with regional and national equivalents. Table 3.2 Dwelling stock profile in Wiltshire Wiltshire South West England 10 year change in total stock (2000 2010)* 8.5% 9.6% 7.8% Proportion of dwellings that are flats** 10.0% 16.1% 19.3% Proportion of dwellings which contain 5 or more rooms** 74.2% 70.0% 67.4% Proportion of dwellings in lowest Council Tax Band (A)*** 12.1% 17.5% 25.2% Vacancy rate in the private sector* 2.7% 2.5% 3.2% Proportion of household spaces in shared dwellings** 0.2% 0.3% 0.4% Sources: *HSSA/HIP data (2000 & 2010), ** 2001 Census, *** CLG (2009) 3.24 HSSA data indicates that the overall increase in the housing stock in Wiltshire over the period 2000 to 2010 was lower than experienced regionally but higher than the national rate. 24

3.25 According to the 2001 Census, only a tenth of the County s dwellings are flats, lower than the regional average and about half the national level. Over a third of dwellings (35.8%) were detached and almost another third were semi-detached dwellings. 3.26 The 2001 Census contains information about the size of properties (in terms of the number of rooms excluding bathrooms, toilets and hallways). The data shows that dwellings in the in the County tend to have more rooms than was the case regionally: almost three quarters (74.2%) of dwellings had five or more rooms. 3.27 A good indication of the quality and price structure of the housing stock is the distribution of dwellings by Council Tax Band. Only around a tenth (12.1%) of properties in the County fall into the lowest Council Tax Band; this is almost half of the national average and lower than the regional average, illustrating the relatively high property values in the County. 3.28 A certain level of vacant dwellings is necessary to ensure the housing market is dynamic. HSSA data indicates that the proportion of market dwellings vacant in Wiltshire is the same as the regional average. 3.29 Shared dwellings are a household space within part of a converted or shared house' in which a household does not have exclusive use of a part of the dwelling (including bathroom and toilet). Households may reside in shared dwellings as a result of an inadequate supply of affordable housing in a local area. Only 308 households in Wiltshire were sharing a dwelling at the time of the 2001 Census, amounting to 0.2% of the total. 25

4. The current housing market Summary i) According to data from the Land Registry, the mean house price in Wiltshire in Q4 of 2010 was 241,810. Over the last ten years, the average price in Wiltshire more than doubled, rising by 102%. ii) Entry-level purchase prices and entry level private rents are lower in urban than rural areas. The cost of social rented lettings in Wiltshire is significantly below private rented housing, indicating a gap between these sectors. This gap widens with size of property. iii) Wiltshire has a reasonable level of self-containment - 62.8% of Wiltshire households that had moved in the last two years moved within the County and 74.2% of Households Reference Persons work within the County. iv) 28 interviews were conducted with estate agents to understand in detail the housing dynamics operating. Based on these findings along with survey data, the following conclusions are made: v) There are three market areas across Wiltshire (the east, the west and the south) and a further five distinct sub-markets. vi) The east is influenced by the affluent commuter areas of Reading, Newbury, Oxford and London as well as Swindon more locally. The west has a diverse number of smaller urban centres set in rural hinterland and is the most balanced area. The south is the most self-contained. Introduction 4.1 An effective SHMA is founded on a thorough understanding of local housing what it costs and how this varies. This chapter will use a range of sources to present this information for Wiltshire, including Land Registry data, an online survey of house prices and interviews with estate and letting agents. 4.2 This chapter begins by using Land Registry to put current property prices in a regional and historical context. The chapter then moves on to identify the cost of different tenures in Wiltshire and the gaps that exist between the tenures. This is followed by a discussion of the level of self-containment in Wiltshire and the existence of a Wiltshire Housing Market. 4.3 It is important that in a large geographical area such as Wiltshire, housing and planning policies are made with a degree of flexibility in order to meet the local circumstances. To this end the second part of the chapter tries to identify boundaries within the County that reflect how the market is practically 27

experienced and will form a suitable scale for localised policy to be set. Information from interviews with estate and letting agents on the local housing market is presented before an analysis of potential housing markets is undertaken. 4.4 The investigation and subsequent definition of the housing market areas and sub-markets has been approached through a combined quantitative and qualitative methodology. We have: reviewed the available secondary data analysed past and anticipated move data collected through the household survey analysed the travel to work patterns collected through the household survey consulted with stakeholders (particularly estate agents) Relative prices 4.5 According to data from the Land Registry, the mean house price in Wiltshire in Q4 of 2010 was 241,810. This compares to the South West region average of 228,672 and the national average of 239,760. 4.6 The figure below shows the change in mean prices over the last ten years in the County, region and across England as a whole. The data shows that significant price increases were experienced in all three cases. Over the last ten years, the average price in Wiltshire more than doubled, rising by 102%. This is slightly lower than the regional and national increases over the same time period (131% and 122% respectively). The graph also illustrates the drop in house prices in 2008 and 2009, and the partial recovery. Figure 4.1 Land Registry price changes 2000-2010 (mean) Average price 300,000 250,000 200,000 150,000 100,000 Wiltshire South West England 50,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: Land Registry 2000-2010 28

4.7 The figure below shows the change in lower quartile prices over the same time period. The trend in terms of change over time is similar to that shown above. Lower quartile prices in Wiltshire appear to more closely match those of the South West, whereas mean prices in the County are closer to national figures. Figure 4.2 Land Registry price changes 2000-2010 (lower quartile) 180,000 160,000 140,000 Wiltshire South West England Average price 120,000 100,000 80,000 60,000 40,000 20,000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Source: Land Registry 2000-2010 The cost of housing in Wiltshire 4.8 This section of the report considers the cost of different tenures in Wiltshire. It will focus on comparing rural areas with urban areas which shows distinct differences in prices, although at total of 17 price areas have been identified in the County (the equivalent data for these 17 price markets is presented in Appendix 1). 4.9 To fully understand the affordability of housing in an area it is necessary to collect data on the cost of housing by bedroom size. This allows the ability of households to be able to afford market housing of the appropriate size (determined by the statutory bedroom standard) to be assessed. 4.10 Entry-level prices 3 were obtained by bedroom size for each of the main settlements in the County via an online search of properties advertised for sale and to let during February 2011. These costs were qualified by the qualitative research with estate and letting agents in the area. The entry-level price for owner-occupied property in the rural and urban areas of Wiltshire is presented in the figure below. 3 In accordance with the Practice Guidance entry-level costs are based on lower quartile prices. 29

Figure 4.3 Entry-level purchase price by size of dwelling 300,000 280,931 Entry-level purchase price 250,000 200,000 150,000 100,000 93,195 81,193 139,771 111,793 198,686 148,664 214,315 50,000 0 1 bed 2 bed 3 bed 4 bed Rural Urban Source: Survey of entry-level house prices, February 2011 4.11 The weekly cost of entry-level private rented accommodation is presented in the figure below. Figure 4.4 Entry-level private rents by size of dwelling (per week) 250 230 Entry-level weekly rental cost 200 150 100 103 89 133 122 171 154 199 50 0 1 bed 2 bed 3 bed 4 bed Rural Urban Source: Survey of entry-level rents, February 2011 4.12 The cost of social rented accommodation by dwelling size in Wiltshire can be obtained from Continuous Recording (CORE) which is a national information source on social rented lettings. The table below illustrates the cost of recent social rented lettings in Wiltshire. As can be seen the costs are significantly below those for private rented housing, particularly for larger houses, indicating a significant potential gap between the social rented and market sectors. 30

Table 4.1 Social rented costs in Wiltshire Bedrooms Rent (per week) 1 bed 72 2 bed 82 3 bed 96 4 bed 96 Source: CORE 2009/2010 4.13 The Coalition Government have introduced a new tenure to affordable housing: Affordable Rent. Under this scheme rents will be set at the level of up to 80% of the open market rent of each property available. The specific details of this policy and its implications are discussed more fully in Chapter 8. As Affordable Rent is not available at the time of writing the costs of the new tenure have been modelled (this process is explained in more detail in Chapter 8). The prices shown below are based at 80% of the median private rented property. Table 4.2 Potential costs of Affordable Rent housing (per week) Bedrooms Rural Urban 1 bed 91 84 2 bed 122 107 3 bed 154 132 4 bed 242 175 Source: Survey of entry-level rents February 2011 4.14 The weekly costs of Shared Ownership housing in Wiltshire is displayed in the table below. This is based on households buying a 75% share in the house (typical for Wiltshire) with a mortgage length of 25 years and supplying a 20%. Table 4.3 Entry-level cost of Shared Ownership housing (per week) Bedrooms Rural Urban 1 bed 92 80 2 bed 138 110 3 bed 196 147 4 bed 277 212 Source: Survey of entry-level rents, February 2011 31

Analysis of housing market gaps 4.15 Housing market gaps analysis has been developed to allow easy comparisons of the income required to access different tenures, in order to facilitate the testing of newbuild proposals, and to show generally the nature of the housing ladder in a particular locality. 4.16 The tables below show a comparison of the weekly costs of housing for the different tenures described in this chapter. Measurement of the size of the gaps between these rungs of the ladder helps assess the feasibility of households moving between the tenures - the smaller the gaps, the easier it is for a household to ascend the ladder. Table 4.4 Weekly costs by tenure Bedrooms Social rent Affordable Rent Shared Ownership Entry-level private rent Entry-level purchase* Urban 1 bed 72 84 80 89 92 2 bed 82 107 110 122 127 3 bed 96 132 147 154 169 4 bed 96 175 212 199 244 Rural 1 bed 72 91 92 103 106 2 bed 82 122 138 133 159 3 bed 96 154 196 171 226 4 bed 96 242 277 230 320 *weekly costs are based on 25 year repayment mortgage at an interest rate of 3.95% (Nationwide 2011) Source: Survey of entry-level house prices February 2011; CORE 2009/2010 Table 4.5 Weekly costs by tenure gaps Bedrooms Social rent Affordable Rent Social rent Private rent Affordable Rent Private rent Private rent Private purchase Urban 1 bed 16.7% 23.6% 7.1% 3.4% 2 bed 30.5% 48.8% 14.8% 4.1% 3 bed 37.5% 60.4% 16.8% 9.7% 4 bed 82.3% 107.3% 13.7% 22.6% Rural 1 bed 26.4% 43.1% 12.9% 2.9% 2 bed 48.8% 62.2% 9.3% 19.5% 3 bed 60.4% 78.1% 10.7% 32.2% 4 bed 152.1% 139.6% -5.0% 39.1% Source: Survey of entry-level house prices February 2011; CORE 2009/2010 32

4.17 The gap between the cost of cost of entry-level market accommodation and social rent increases with the size of the property, rising to 150% of social rent levels in four bedroom properties. This suggests that an Affordable Rent product could be beneficial to a number of households. However, Affordable Rent (taken as 80% of the median private rented value) is not necessarily much cheaper than entry level private rents (and is more expensive for four bedroom properties in the rural area). The issue of the affordability of different Affordable Rent levels is discussed in more detail in chapters 5 and 8. Defining the Housing Market Area 4.18 PPS3 defines housing market areas as geographical areas defined by household demand and preference for housing. They reflect key functional linkages between places where people live and work. 4.19 As recognised by the CLG advice note Identifying sub-regional housing market areas (March 2007) there is no single preferred methodology to defining housing market areas and a range of information sources should be considered. The most frequently adopted methodologies have been based upon household migration patterns and Travel to Work areas. The advice note suggests the adoption of a pragmatic approach to identifying housing market areas, following local authority boundaries. 4.20 According to 2001 Census data, Wiltshire has a reasonable level of self-containment 4 in terms of migration: 51.8% of households moving into a dwelling in the County moved from another dwelling in the County. The 2011 household survey suggests a higher level of self-containment, with 62.8% of Wiltshire households that had moved in the last two years moving within the County. 4.21 Census migration data suggests that Wiltshire has particular links with the neighbouring authorities of Swindon, Bath & North East Somerset and Test Valley. However, the data also indicates that people move in to and out of Wiltshire from a wide variety of locations. This is supported by the household survey which found that 25.4% of households that had moved to the County in the last two years came from further a field i.e. not neighbouring or nearby local authorities. 4.22 In terms of commuting flows, data from the 2001 Census indicates that Wiltshire has a high level of self-containment with 75.9% of (working) Wiltshire residents working within the County. The household survey similarly suggests that 74.2% of (working) Household Reference Persons (HRP) work within the County. Other key areas of employment are Swindon and Bath & North East Somerset (according to both Census and household survey data), although local residents work in a range of areas outside of the County. 4.23 Census data also shows commuting in-flows to Wiltshire. Overall, 84.3% of people employed in Wiltshire in 2001 also lived in the County. The most place of residence common place of residence of 4 Defined as the proportion of household moves taking place within the County 33

workers from outside of Wiltshire was Swindon; although this applied to only 2.1% of employed people in Wiltshire. This indicates that commuting into the County for work occurs in small flows from a wide range of neighbouring areas. 4.24 For some purposes Wiltshire can be seen as part of a wider housing market, but equally there exist many local sub-markets which are perceived by households actually moving in the County. (These are considered further in Appendix 4 which looks at the way households make their housing choices and move within Wiltshire.) In terms of this report, we use the Council boundary to note the practical boundary of the whole housing market studied, as that follows the recommendation of the CLG Advice Note on housing market areas. The Wiltshire Housing Market however contains some smaller housing markets which are discussed later in this chapter. Interviews with Estate and Letting Agents 4.25 During January 2011, 28 interviews were conducted with estate agents dealing with sales, and 12 with agents who worked on the rental market. Agents representing all parts of the County were interviewed. The interviews were conducted in order to better understand Wiltshire s housing market and to ensure report findings accurately reflect to the local context. The agents were interviewed in their offices. 4.26 The survey was conducted through semi structured interviews. The agents were approached without prior appointment, in their offices/shops and asked if they would take part in the research. They were told that the research was being undertaken for the Council so that new housing and planning policies could be created to ensure that the correct housing was built in the correct places across the County. The purpose of the survey was fourfold: i) To check entry level prices for home ownership and market rent ii) To gather information on housing markets and how areas relate to each other iii) To ascertain agents expectations on the future of the market iv) To identify the types of homes the agents perceived to be most in demand or in surplus. Agents views on Wiltshire s Housing Markets 4.27 This section provides a summary of the agents interviews, a fuller account is presented in Appendix 5. 4.28 Wiltshire is considered by many of the agents to be an in between place. Whilst it has areas of distinct identity it was universally felt that the neighbouring large towns, outside the County, have a very strong influence over the individual towns within the County, even Salisbury. 4.29 Surprisingly few agents expressed a strong link between Salisbury and other Wiltshire towns, suggesting Salisbury may be quite a self-contained and distinct market; however agents in Salisbury did not consider the city to be self-contained. 34

4.30 Having said this, agents felt that both Salisbury and Salisbury Plain have their own distinct identities. Salisbury is a self-contained city that people move and work within, while Salisbury Plain is strongly influenced by the various military camps and the army s garrison. 4.31 Across the County the general perception is that that at least half of people looking to move house were either moving from outside the locality (not necessarily the County), or worked outside the County. 4.32 The agents indicated that Wiltshire s towns are generally perceived to be affordable relative to those in surrounding counties, making the County very attractive to commuters. The towns in the northeast of the County such as Wootton Basset and Cricklade are more desirable than the suburbs of Swindon. 4.33 The eastern half of the County looks towards Reading, Newbury and London, the southern part of the County looks towards Bournemouth, Southampton and Portsmouth and the western half looks towards Bath and Bristol. In the north of the County, the mainline train and the M4 motorway are very strong influences. The villages and some of the towns to the north (for instance Marlborough and Malmesbury) are seen as highly attractive, and expensive. Wiltshire s housing markets 4.34 The household survey dataset was analysed to try and identify the existence of smaller housing markets within the Wiltshire Housing Market. The results of this analysis are presented in Appendix A3. The conclusion of this analysis echoed the agents comments; It is clear that smaller housing markets exist in the County and there is considerable overlap between the various market areas. 4.35 Having carefully considered the evidence, we have taken a pragmatic view: i) The County is divided into three main housing markets; the south, the west and the east. ii) Whilst these cover the whole County, two towns (Calne and Devizes) are felt to sit along the boundary between two market areas, and do not belong fully to either. These form distinct sub-markets. iii) Further distinct sub-markets have been identified within the broader main housing market areas. One (Salisbury Plain) lies across two main areas, whilst each of the other two sits wholly within one market area. 4.36 The market and sub-market areas identified are described in summary form below, and are depicted in map form in Figure 4.5. Main market areas 4.37 The South the area to the south of Salisbury Plain. This area is centred on Salisbury and looks south. Salisbury Plain is a relatively sparsely populated area that forms a natural geographic 35

boundary with areas to the north. Salisbury itself is sufficiently distinct from the rural areas and the towns to the west of it to justify being identified as a sub-market within this wider market. 4.38 The West the area to the west of Devizes and Calne. The division of northern part of the County is broadly into two. Devizes and Calne can sit quite comfortably in either so have been defined as a single sub-market. The Northwest has strong links with the City of Bath and the major regional centre of Bristol. The M4 and links to the M5 are strong influences on this market. 4.39 The East the area to the east of Devizes and Calne. This part of the County looks towards the prosperous towns of Reading and Newbury and even further away to Oxford and London. Swindon in the northeast is a very strong influence on this part of the County. The areas close to Swindon have also been recognised as a separate sub-market. Distinct sub-markets 4.40 Salisbury Plain This a distinct place overlapping the bottom portion of the Northeast and Northwest market areas. It is sparsely populated and those towns within the area tend to be lower priced areas than the remainder of the County. The area is not completely self-contained however. The presence of the numerous military camps and towns and the emerging Super Garrison set the area a little apart from the rest of the County. 4.41 Salisbury City is a relatively small city that it is in the heart of the southern market area that we have identified. Since much of the area is rural we have designated Salisbury a separate sub-market. 4.42 Outer Swindon Swindon is a large and growing town. Swindon s boundaries are drawn closely to the edge of the built up area and so its influence on neighbouring villages outside its administrative area is strong. 4.43 Devizes and Calne are similar in terms of housing market areas as they are both major towns in central Wiltshire. Devizes and Calne can sit quite comfortably in either the Northwest or the Northeast so have each been defined as a single sub-market. 4.44 Appendix 3 contains further detail, analysis and comment on these market areas. 36

Figure 4.5 Wiltshire s Housing Markets Source: Fordham Research 2011 37

5. Housing need Introduction 5.1 Housing need is a term created in the 1990s to help provide a means-tested estimate of the requirement for affordable housing in an area. It is defined in PPS3 as the number of households who lack their own housing or who live in unsuitable housing and who cannot afford to meet their housing needs in the market. In addition to calculating the size of the housing need in Wiltshire, this chapter will examine the type of affordable accommodation required to meet this need and identify the locations in the County where housing need is greatest. Finally the chapter goes onto consider the changes to the Local Housing Allowance cap, identifying the number of households in Wiltshire likely to be affected. 39

5.2 The calculation of housing need follows the 16 steps outlined in the Strategic Housing Assessments Practice Guidance (August 2007) separated into three broad stages. Each of these stages will be calculated separately before the overall annual need for affordable housing is derived. Current need 5.3 The first stage of the model assesses current need. This begins with an assessment of housing suitability and affordability and also considers homeless households before arriving at a total current need estimate (gross). Unsuitable housing 5.4 A key element of housing need is an assessment of the suitability of a household s current housing. The Practice Guidance sets out a series of nine criteria for unsuitable housing which has been followed in this report. In Wiltshire it is estimated that a total of 11,373 households are living in unsuitable housing (this represents 5.9% of all households in the county). 5.5 The most common reason for unsuitable housing is households living in an overcrowded property, followed by the housing being unsuitable for a household with support needs or mobility problems, with accommodation being too expensive the third most common reason for unsuitable housing. In-situ solutions 5.6 The survey has highlighted that 11,373 households are in unsuitable housing. However, it is most probable that some of the unsuitability can be resolved in the households current accommodation. Households living in housing deemed unsuitable for the following reasons were not considered to have an in-situ solution: end of tenancy, accommodation too expensive, overcrowding, sharing facilities, harassment. 5.7 The survey data therefore estimates that of the 11,373 households in unsuitable housing, 8,281 (or 72.8%) do not have an in-situ solution and therefore require a move to alternative accommodation. Affordability 5.8 The ability of each of these 8,281 households to afford entry-level market housing of an appropriate size is tested using the affordability criteria described in the glossary. The dataset indicates that of the 8,281 households, 5,821 cannot afford market housing and are living in unsuitable housing (and require a move to alternative accommodation). This represents 3.0% of all existing households in the county. These 5,821 households are considered to be in current housing need. 5.9 For the purposes of the housing needs assessment, households considered to be in current housing need have been split into two categories: current occupiers of affordable housing (this includes occupiers of social rented and shared ownership accommodation), and households from other tenures. It is estimated that some 3,129 households in need currently live in affordable housing (35 of which are in a shared ownership property) and 2,692 reside in other tenures. 40

Homeless households 5.10 The housing needs assessment is a snapshot that assesses housing need at a particular point in time. There will, in addition to the existing households in need, be some homeless households who were in need at the time of the assessment and should also be included within any calculation of backlog need. 5.11 To assess the number of homeless households we have used CLG data on statutory homelessness which is drawn from the information contained in the Council s P1(E) Homeless returns. The important point about this information is the note underneath. This should be a snapshot of the numbers in accommodation on the last day of the quarter, not the numbers taking up accommodation during the quarter. This is important given the snapshot nature of the survey. 5.12 Not all of the categories of homelessness are added to our assessment of current need. This is because, in theory, they will be part of our sample for the Housing Needs Assessment. For example, households housed in Council or RSL accommodation should already be included as part of the housing need such household addresses should appear on the Council Tax file from which the dataset sample was derived. After considering the various categories, we have decided there are two which should be included as part of the homeless element: Households in Bed and Breakfast Accommodation and Households living in Hostels. Therefore of the homeless households identified by the Council, 42 households shall be added to the estimate of current housing need. Total current need (gross) 5.13 Table 5.1 below summarises the first stage of the overall assessment of housing need as set out by the Strategic Housing Assessments Practice Guidance (August 2007). The data shows that there are an estimated 5,863 households currently in need in Wiltshire. Table 5.1 Backlog of housing need Step Notes Number 1.1 Homeless households and those in temporary accommodation 42 1.2 Overcrowding and concealed households 1.3 Other groups Two steps taken together 5,821 1.4 equals Total current housing need (gross) 1.1+1.2+1.3 5,863 Source: Wiltshire household survey 2011 Future need 5.14 In addition to Current Need, there will also be Future Need. This forms the second stage of the housing needs assessment model. This is split, as per CLG guidance, into two main categories. These are as follows: 41

new household formation ( proportion unable to buy or rent in market) existing households falling into need. 5.15 The estimate of the number of newly forming households in need of affordable housing is based on an assessment of households that have formed over the past two years. Such an approach is preferred to studying households stating likely future intentions as it provides more detailed information on the characteristics of these households contributing to this element of newly arising need. 5.16 Table 5.2 shows details of the derivation of the need arising from new household formation. The table begins by establishing the number of newly forming households over the past two years an affordability test is then applied. Table 5.2 Derivation of newly arising need from new household formation Aspect of calculation Number Sub-total Number of households moving in past two years 38,458 Minus households NOT forming in previous move -31,571 6,887 Times proportion unable to buy or rent in the market 44.8% Estimate of newly arising need 3,083 Annual estimate of newly arising need 1,542 Source: Wiltshire household survey 2011 5.17 The table shows that an estimated 6,887,households are newly formed within the county over the past two years (3,444 per annum). The survey estimates that 44.8% of newly forming households are unable to afford market housing without some form of subsidy in Wiltshire. This results in an annual estimate of the number of newly forming households falling into need of 1,542. Existing households falling into need 5.18 This is an estimate of the number of existing households who will fall into housing need over the next two years (and then annualised). The basic information for this is households who have moved home within the last two years and affordability. A household will fall into need if it has to move home and is unable to afford to do this within the private sector (examples of such a move will be because of the end of a tenancy agreement). A household unable to afford market rent prices but moving to private rented accommodation may have to either claim Local Housing Allowance (formerly Housing Benefit) or spend more than a quarter of their gross income on housing, which under the Strategic Housing Assessment Practice Guidance (August 2007) is considered unaffordable (or indeed a combination of both). 42

5.19 Households previously living with parents, relatives or friends are excluded as these will double-count with the newly forming households already considered in the previous table. The data also excludes moves between social rented properties. Households falling into need in the social rented sector have their needs met through a transfer to another social rented property, hence releasing a social rented property for someone else in need. The number of households falling into need in the social rented sector should therefore, over a period of time, roughly equal the supply of transfers and so the additional needs arising from within the social rented stock will be net zero. Table 5.3 shows the derivation of existing households falling into need. Table 5.3 Derivation of Newly Arising Need from existing households Aspect of calculation Number Sub-total Number of households moving in past two years 38,458 Minus households forming in previous move -6,887 31,571 Minus households transferring within affordable housing -2,076 29,495 Times proportion unable to buy or rent in the market 24.0% Estimate of newly arising need 7,070 Annual estimate of newly arising need 3,535 Source: Wiltshire household survey 2011 5.20 The table shows that a total of 29,495 existing households are considered as potentially in need (14,748 per annum). Using the standard affordability test for existing households it is estimated that 24.0% of these households cannot afford market housing. Therefore our estimate of the number of households falling into need excluding transfers is 7,070 households over the two-year period. Annualised this is 3,535 households per annum. Total future need (gross) 5.21 The data from the two steps described above can now be put into the needs assessment table as illustrated in Table 5.4. It indicates that future need will arise from a total of 5,077 households per annum. Table 5.4 Future need (per annum) Step Notes Number 2.1 New household formation (gross per year) 3,444 2.2 Proportion of new households unable to buy or rent in the market leaves 1,542 44.8% 2.3 Existing households falling into need 3,535 2.4 Total newly arising housing need (gross per year) 2.1 2.2 2.3 5,077 Source: Wiltshire household survey 2011 43

Affordable housing supply 5.22 The supply of affordable housing to meet housing need comprises the third stage of the housing needs assessment model. The affordable housing supply stage is split between existing stock that is available to offset the current need and the likely future level of supply. Available stock to offset current need 5.23 The stock available to offset the current need includes stock from current occupiers of affordable housing in need, surplus stock from vacant properties and committed supply of new affordable units. Units to be taken out of management are removed from the calculation. Affordable dwellings occupied by households in need 5.24 Firstly, it is important when considering net need levels to discount households already living in affordable housing. This is because the movement of such households within affordable housing will have an overall nil effect in terms of housing need. As established when calculating current need (stage 1), there are currently 3,129 households currently in need already living in affordable housing. Surplus stock 5.25 A certain level of vacant dwellings is normal as this allows for transfers and for work on properties to be carried out. The Practice Guidance suggests that if the vacancy rate in the affordable stock is in excess of 3% then some of the vacant units should be considered as surplus stock which can be included within the supply to offset housing need. Wiltshire records a vacancy rate in the social rented sector of 2.2%. As the vacancy rate in Wiltshire is lower than the 3% benchmark, no vacant dwellings are considered available to be brought back into use to increase the supply of affordable housing. Committed supply of new affordable units 5.26 The Practice Guidance recommends that this part of the assessment includes new social rented and intermediate housing which are committed to be built over the period of the assessment. For the purposes of analysis we have taken Wiltshire HSSA (Housing Strategy Statistical Appendix) data showing the number of planned and proposed affordable units for the period 2010-2012 as a guide to new provision. 5.27 Overall the 2010 HSSA data suggests that there are 1,317 affordable dwellings planned or proposed for 2010/11 and 2011/2012. 5.28 Proposals exist for a Super Garrison which would combine some of the military establishments within Wiltshire. This process is expected to generate a supply of affordable housing as former military housing is placed outside the wire. It is too early to make definite assessments of the additional supply. Given the level of housing need it is not likely that this additional supply will substantially alter the conclusions of this SHMA. 44

Units to be taken out of management 5.29 The Practice Guidance states that this stage involves estimating the numbers of social rented or intermediate units that will be taken out of management. The main component of this step will be properties which are expected to be demolished (or replacement schemes that lead to net losses of stock). At the time of reporting the proposed number of affordable dwellings expected to be taken out of management in the future was unknown and hence a figure of zero has been used in this step of the model. Total available stock to offset current need 5.30 Having been through a number of detailed stages in order to assess the total stock available to offset current need in the County we shall now bring together all pieces of data to complete this part of the needs assessment. This is presented in the table below. The data shows that there are an estimated 4,446 properties available to offset the current need in Wiltshire. Table 5.5 Current supply of affordable housing Step Notes Number 3.1 Affordable dwellings occupied by households in need 3,129 3.2 Surplus stock 0 3.3 Committed supply of affordable housing 1,317 3.4 Units to be taken out of management 0 3.5 Total affordable housing stock available 3.1+3.2+3.3-3.4 4,446 Source: Wiltshire household survey 2011 Future supply of affordable housing 5.31 The future supply of affordable housing is the flow of affordable housing arising from the existing stock that is available to meet future need. It is split between the annual supply of social relets and the annual supply of relets within the intermediate sector. The future supply of social rented housing 5.32 The Practice Guidance suggests that the estimate of likely future relets from the social rented stock should be based on past trend data which can be taken as a prediction for the future. 5.33 CORE data provides an indication of the number of lettings in the RSL sector whilst HSSA data provides an indication of the number of lettings in Council owned housing. It is estimated that in 2008-09 there were 1,391 dwellings let to new social rent tenants in Wiltshire and 1,512 in 2009-10. The average for the two-year period is 1,452 per annum. 45

Intermediate supply 5.34 In most local authorities the amount of intermediate housing (mostly Shared Ownership) available in the stock is fairly limited. However, it is still important to consider to what extent the current supply may be able to help those in need of affordable housing. 5.35 Therefore we include an estimate of the number of Shared Ownership units that become available each year. Applying the relet rate for social rented housing to the estimated stock of Shared Ownership housing, it is estimated that around 46 units of Shared Ownership housing will become available to meet housing needs from the existing stock of such housing. Total future supply 5.36 This step is the sum of the previous two. The total future supply is estimated to be 1,498 comprised of 1,452 units of social rented accommodation and 46 units of Shared Ownership: Table 5.6 Future supply of affordable housing (per annum) Step Notes Number 3.6 Annual supply of social relets (net) 1,452 3.7 Annual supply of intermediate housing available for relet or resale at sub-market levels 46 3.8 Annual supply of affordable housing 3.6+3.7 1,498 Source: Wiltshire household survey 2011 5.37 The 16 steps detailed above (set across the three broad stages) are brought together in the housing needs assessment model: 46

Table 5.7 Housing needs assessment model for Wiltshire Stage and step in calculation Notes Number STAGE 1: CURRENT NEED (Gross) 1.1 Homeless households and those in temporary accommodation 42 1.2 Overcrowding and concealed households Two steps 1.3 Other groups taken together 5,821 1.4 Total current housing need (gross) 1.1+1.2+1.3 5,863 STAGE 2: FUTURE NEED 2.1 New household formation (gross per year) 3,444 2.2 Proportion of new households unable to buy or rent in the market leaves 1,542 44.8% 2.3 Existing households falling into need 3,535 2.4 Total newly arising housing need (gross per year) 2.1 2.2 2.3 5,077 STAGE 3: AFFORDABLE HOUSING SUPPLY Current supply 3.1 Affordable dwellings occupied by households in need 3.2 Surplus stock 0 3,129 3.3 Committed supply of affordable housing 1,317 3.4 Units to be taken out of management 0 3.5 Total affordable housing stock available 3.1+3.2+3.3-3.4 4,446 Future supply 3.6 Annual supply of social relets (net) 3.7 Annual supply of intermediate housing available for relet or resale at sub-market levels 1,452 3.8 Annual supply of affordable housing 3.6+3.7 1,498 Source: Wiltshire household survey, 2011 46 5.38 The Practice Guidance states that these figures need to be annualised to establish an overall estimate of net housing need. The first step in this process is to calculate the net current need. This is derived by subtracting the estimated total stock of affordable housing available (step 3.5) from the gross current need (step 1.4). This produces a net current need figure of 1,417 (5,863-4,446). 5.39 The second step is to convert this net current need figure into an annual flow. The Practice Guidance acknowledges that this current need can be addressed over any length of time although a period of less than five years should be avoided. For the purposes of this study the quota of five years proposed in the Practice Guidance will be used. Therefore to annualise the net current need figure, it will be divided by five. This calculation results in a net annual quota of 283 (1,417/5) households who should have their needs addressed. 47

5.40 The final step is to sum the net annual quota of households who should have their needs addressed with the total newly arising housing need (step 2.4) and subtract the future annual supply of affordable housing (step 3.8). This leads to an annual need estimate of 3,862 (283+5,077-1,498). These figures are summarised in the table below. Table 5.8 Summary of housing needs assessment model Element Number Current need (Step 1.4)/5 1,173 Current supply (Step 3.5)/5 889 Net current need 283 Future need (Step 2.4) 5,077 Future supply (Step 3.8) 1,498 Net future need 3,579 Total net annual need 3,862 Total gross annual need 6,250 Total gross annual supply 2,388 Total net annual need 3,862 Source: Wiltshire household survey 2011 Location of housing need 5.41 The table below shows gross housing need by sub-area within Wiltshire. In terms of the proportion of all households in gross need, Salisbury Plain has the highest figure. It is also interesting to note that West Towns account for nearly one-fifth of all the gross need in the County. 48

Table 5.9 Gross housing need and sub-area Housing need Sub-area In need Number of h holds in sub-area % of total h holds in need % of those in need Bradford-on-Avon 63 4,226 1.5% 1.0% Chippenham 329 14,326 2.3% 5.3% Devizes 359 7,303 4.9% 5.7% North and Central Rural 841 49,176 1.7% 13.4% North Small Towns 433 9,090 4.8% 6.9% North Towns 565 17,161 3.3% 9.0% Salisbury 699 21,183 3.3% 11.2% Salisbury Plain 689 12,620 5.5% 11.0% South Rural 582 19,484 3.0% 9.3% Trowbridge 537 15,912 3.4% 8.6% West Towns 1,152 22,326 5.2% 18.4% Total 6,250 192,807 3.2% 100.0% Source: Wiltshire household survey 2011 5.42 However, knowing the variation in tenure profile across the County we would expect the supply of affordable housing to differ considerably across Wiltshire. Using information from the survey data we are able to estimate how the gross supply in the County is apportioned between its different subareas. From this information we can ascertain a net housing needs figure for each of the eleven subareas in Wiltshire, as shown in the table below. 49

Table 5.10 Net housing need and sub-area Housing need Sub-area Gross annual need Gross annual supply Net annual housing need % of net shortfall Supply as % of need Bradford-on-Avon 63 48 15 0.4% 75.8% Chippenham 329 235 94 2.4% 71.4% Devizes 359 197 162 4.2% 54.8% North and Central Rural 841 151 689 17.8% 18.0% North Small Towns 433 259 174 4.5% 59.8% North Towns 565 292 273 7.1% 51.7% Salisbury 699 379 320 8.3% 54.2% Salisbury Plain 689 130 559 14.5% 18.9% South Rural 582 92 490 12.7% 15.9% Trowbridge 537 114 422 10.9% 21.3% West Towns 1,152 490 662 17.1% 42.5% Total 6,250 2,388 3,862 100.0% 38.2% NB: The table shows where the need arises, not where it should be met. Households in need in an area may state a preference for being housed elsewhere in Wiltshire. These figures exclude households are already in the affordable sector in need Source: Wiltshire household survey, 2011 5.43 The table shows that the largest net need is found in North and Central Rural followed by West Towns. The table also indicates that households in need in South Rural are likely to have the longest wait to access affordable housing as the annual supply represents only 15.9% of the annual need. Type of affordable accommodation required 5.44 Households in need were tested for their ability to afford a range of affordable products. These are: Shared Ownership with a 75% equity share Shared Ownership with a 50% equity share Affordable Rent at 80% market rents Affordable Rent at 70% market rents Affordable Rent at 60% market rents Current social rent. 5.45 Any household unable to afford any of these products is assumed to require assistance with their accommodation costs in the form of Local Housing Allowance (formerly Housing Benefit). The table below shows the suitability of these various products for households in need in Wiltshire. 50

5.46 The table indicates that 9.2% of households in gross need would be able to afford Shared Ownership with a 75% equity share and a further 15.1% would be able to afford Shared Ownership with a 50% equity share. Whilst Affordable Rent set at 80% of market-entry rents would be suitable for just 4.1% of households in need, Affordable Rent set at 70% would be affordable to a further 11.3% of households. Table 5.11 Affordability of households in housing need (gross) Product In need Housing need % of those in need Shared Ownership 75% 575 9.2% Shared Ownership 50% 943 15.1% Affordable Rent 80% 257 4.1% Affordable Rent 70% 704 11.3% Affordable Rent 60% 924 14.8% Social rent 115 1.8% Need Local Housing Allowance 2,732 43.7% Total 6,250 100.0% Source: Wiltshire household survey, 2011 5.47 The results can be simplified by grouping the two Shared Ownership products into a single category and by grouping the three Affordable Rent options into a single classification. The results of this process are presented in the table below. The table indicates that almost a quarter of households in gross need could afford a shared equity product and a further 30% would be suitable for Affordable Rent. Table 5.12 Affordability of households in housing need (gross) simplified Product In need Housing need % of those in need Shared Ownership 1,518 24.3% Affordable Rent 1,885 30.2% Social rent 115 1.8% Need Local Housing Allowance 2,732 43.7% Total 6,250 100.0% Source: Wiltshire household survey, 2011 51

5.48 This tenure distribution can be compared to the likely supply of different forms of affordable housing according to the housing needs assessment model. This is presented in the table below. The table indicates that the majority of the supply (95.6%) will be social rented properties. The likely supply is deducted from the gross need to determine the net position. This shows that intermediate housing forms the majority of the new affordable provision required in Wiltshire. The data suggests that almost half should be Affordable Rent, 36.8% Shared Ownership and 14.6% social rented properties. The results indicate an annual requirement for 1,876 Affordable Rent homes and 1,421 shared ownership properties Product Table 5.13 Type of affordable housing required (net) Gross annual need Gross annual supply Net annual housing need % of net shortfall Shared Ownership 1,518 97 1,421 36.8%* Affordable Rent 1,885 9 1,876 48.6% Social rent 115 Need Local Housing Allowance 2,732 } 2,282 565 14.6% Total 6,250 2,388 3,862 100.0% *Indicates the ability of households to afford Shared ownership rather than demand, which is particularly low in Wiltshire Source: Wiltshire household survey, 2011 5.49 At the time of writing, registered housing providers in Wiltshire were considering converting 50% of all relets to an Affordable Rent tenure. (This possibility is considered further in chapter 8.) The impact of this has been modelled in the table below. The table shows that this practice would halve the level of need for Affordable Rent accommodation however, the need for housing suitable for households dependent on LHA triples. Table 5.14 Type of affordable housing required (net) with 50% all relets Affordable rent Product Gross annual need Gross annual supply Net annual housing need % of net shortfall Shared Ownership 1,518 97 1,421 36.8%* Affordable Rent 1,885 1150 735 19.0% Social rent 115 Need Local Housing Allowance 2,732 } 1,141 1,706 44.2% Total 6,250 2,388 3,862 100.0% *Indicates the ability of households to afford Shared ownership rather than demand, which is particularly low in Wiltshire Source: Wiltshire household survey, 2011 52

Size of affordable housing required 5.50 To establish the size of affordable accommodation required to meet housing need, data from the household survey has been used alongside data from the HSSA provided by the Council. The table below shows the estimated need for different sizes of accommodation from three different groups. These are: households currently or projected to be in need (from survey data) households currently or projected to be in need (from survey data) who are also in a group likely to be considered as having a priority (pensioner households, households with children or households where someone has a special/support need) households on the housing waiting list in April 2010 as submitted on the HSSA. 5.51 The table below shows some variation between the data sources and groups studied although in all cases the majority of the requirement is for one and two bedroom units. Taking a crude average of all of these figures suggests that around 80% of the requirement is for one and two bedroom dwellings. All of these figures are gross, however, and therefore do not take account of the supply of different sizes of accommodation. Table 5.15 Estimated size requirement for additional affordable housing (gross) Households in need* Households in need* (priority) Housing Register Crude average 1 bedroom 42.3% 26.0% 55.0% 41.1% 2 bedrooms 36.9% 48.1% 30.0% 38.3% 3 bedrooms 16.3% 20.9% 12.9% 16.7% 4+ bedrooms 4.5% 4.9% 2.1% 3.8% Total 100.0% 100.0% 100.0% 100.0% *The size requirement resulting from the analysis of survey data is based on a strict bedroom standard (defined in the glossary) Source: Wiltshire household survey, 2011 and Wiltshire Council HSSA 2010 5.52 In order to obtain a net need figure by bedroom size, the likely size of the affordable supply needs to be taken into account. This information is shown in the table below. The net need figures by bedroom size have been derived by applying the crude averages shown in the table above to the gross need figure of 6,250. Data from the household survey on the number of bedrooms in affordable dwellings that have recently been let has informed the profile of the size of the affordable dwellings likely to become available to meet the housing need. 53

Table 5.16 Estimated size requirement for additional affordable housing (net) Gross annual need Gross annual supply Net annual need % of net annual need Supply as a % of need 1 bedroom 2,569 790 1,779 46.1% 30.8% 2 bedrooms 2,396 1,305 1,091 28.3% 54.5% 3 bedrooms 1,046 266 779 20.2% 25.5% 4+ bedrooms 239 26 213 5.5% 10.9% Total 6,250 2,388 3,862 100.0% 38.2% Source: Wiltshire household survey, 2011 and Wiltshire Council HSSA 2010 5.53 The table suggests that 46% of the net need is for one bedroom accommodation, 28% for two bedroom dwellings and a quarter for homes with three or more bedrooms. The final column shows that the need relative to supply is greatest for larger accommodation just over 10% of the (gross) need for four bedroom accommodation is expected to be met compared to over half of the (gross) need for two bedroom accommodation. Sensitivity analysis 5.54 The housing needs assessment model assesses the number of households in need based on the affordability assumptions required by the CLG Practice Guidance. It is possible, however, to examine how the model would be affected if the affordability assumptions were altered. The table below presents the model results where households were considered able to afford market rented housing in cases where the rent payable would constitute no more than 30%, 35% and 40% of gross household income, rather than 25% used in the standard model. The affordability assumptions for home ownership remain unchanged. 54

Table 5.17 Impact of different affordability assumptions on affordable housing requirement in Wiltshire Rent payable constitutes no more than: 30% of gross household income 35% of gross household income 40% of gross household income Backlog need (annual) 1,049 918 683 Backlog supply (annual) 871 837 708 Net backlog need (annual) 177 81-24 Future need (annual) 4,041 3,432 2,499 Future supply (annual) 1,498 1,498 1,498 Net future need (annual) 2,543 1,934 1,001 Total net annual need 2,721 2,015 976 Total gross annual need 5,090 4,350 3,182 Total gross annual supply 2,369 2,335 2,206 Total net annual need 2,721 2,015 976 Source: Wiltshire household survey, 2011; various secondary sources 5.55 The table indicates that the number of households in need would decrease from 3,862 to 2,721 if 30% of gross household income could be spent on rent. This would decrease further to 2,015 if 35% of income could be spent on rent and to 976 if the affordability assumption was changed to 40%. Local Housing Allowance & the impact of household benefit cap 5.56 Local Housing Allowance (LHA) is the replacement for the former Housing Benefit. It is designed to make up the shortfall in people s ability to pay for the housing they need. LHA may represent 100% or some lower percentage of the overall rent paid. A major difference between LHA and the former Housing Benefit is that it is paid directly to the tenant and not to the (private or public) landlord as was the case with Housing Benefit. 5.57 The Practice Guidance requires the extent of the private rented sector (through the Local Housing Allowance system) to meet the needs of households in need to be estimated. We have therefore used survey data to look at the number of new Local Housing Allowance (LHA) supported private rented housing lets over the past two years. In Wiltshire, it is estimated that over the past two years, 2,652 LHA supported lettings have been made (1,326 per annum). 5.58 Since the election of the Coalition Government in May 2010 two polices have been announced that could limit the capacity of LHA to support households in need: In April 2011 the overall LHA cap was reduced from the median value of properties on the market to the 30th percentile. This is assessed locally. These local caps must fall within the national upper LHA limits ( 250 per week for a one bedroom home, 290 55

per week for two bedrooms, 340 per week for three bedrooms and 400 per week for home with four or more bedrooms). A cap will be introduced in 2012 which limits the total benefit income that a household can receive at 18,200 for one person households and 26,000 for all other households. 5.59 These policies are also discussed in Chapter 8 in reference to wider changes to housing policy, primarily the introduction of Affordable Rent and fixed term tenancies. This section considers to what extent the LHA and total household benefit caps will impact households in Wiltshire. Local housing allowance cap 5.60 Prior to April 2011, LHA caps were set at the median value of a private rented sector property for each dwelling size (one to five bedrooms). These levels were determined within local Broad Rental Market Areas (BRMAs) by the Valuation Office Agency. Since April 2011 the LHA cap has been reduced to 30% of private rented properties in an area. In addition, the provision of a LHA level for five bedrooms has been removed (with the four bedroom cap applying to all properties with four or more bedrooms). 5.61 There are eight BRMA in Wiltshire. However, the vast majority of households in the County live in one of three BRMAs: Salisbury, Swindon and West Wiltshire. The LHA caps for these three BRMAs in July 2011 were as follows: Table 5.18 Maximum LHA entitlement per month by BRMA and number of bedrooms BRMA 1 bedroom 2 bedrooms 3 bedrooms 4 bedrooms Salisbury 500 625 750 950 Swindon 430 525 650 850 West Wiltshire 425 525 625 850 Source: Direct gov (July 2011) 5.62 As can be seen in the table above, Salisbury is generally the most expensive BRMA in Wiltshire, whilst the housing costs in Swindon and West Wiltshire are similar, according to the Valuation Office Agency. 5.63 To assess the impact of the reduction of the LHA cap on households in Wiltshire, each household has to be allocated a BRMA to determine which LHA cap they are subject to. For the purposes of the analysis two assumptions have been made. Firstly, all households live in one of the three main BRMAs. In reality, a very small minority of households live in a different BRMA (Newbury, Basingstoke, Yeovil or Cheltenham). Secondly, households have been allocated a BRMA by Community Area however, the actual BRMA boundaries slightly differ from Community Area boundaries. Wootton Bassett & Cricklade and Malborough have been placed in the Swindon BRMA; Pewsey, Tidworth and the former district of Salisbury have been placed in the Salisbury BRMA; all the remaining Community Areas have been placed in the West Wiltshire BRMA. 56

5.64 The two tables below show the proportion of households living in the private rented sector and claiming LHA whose housing costs exceed the LHA cap. It is possible that not all of these households have been affected by the cap: they may not have received LHA covering their full housing costs and therefore the LHA cap has remained above what they receive. The tables do however indicate which households are most likely to have experienced a fall in their LHA entitlement since April 2011. 5.65 The table below shows that private rented households in receipt of LHA living in one and four bedroom properties are the most likely to be affected by the cap. For both property sizes, over half of households claiming LHA, have housing costs that exceed the cap. For two and three bedroom homes this reduces to a quarter and a third respectively. Table 5.19 % of households claiming LHA affected by the LHA cap by number of bedrooms Property size % LHA claimants affected 1 bedroom 50.8% 2 bedrooms 25.7% 3 bedrooms 33.1% 4 bedrooms 55.8% Source: Wiltshire household survey, 2011 5.66 The table below shows that the housing costs of LHA claimants in the Salisbury BMRA are the least likely to exceed the LHA cap. This is likely to be due to the higher LHA caps in the Salisbury BRMA (Table 5.18). The housing costs of around a third of LHA claimants in the Swindon BRMA exceed the cap, however there are far fewer households in this sample so the data is less reliable. LHA claiming households in the West Wiltshire BRMA are the most likely to have housing costs higher than the LHA cap (44.2% of these households). Table 5.20 % of households claiming LHA affected by the LHA cap by BRMA BRMA % LHA claimants affected Salisbury 18.0% Swindon 32.6% West Wiltshire 44.2% Source: Wiltshire household survey, 2011 57

Total household benefit cap 5.67 The cap for the total amount in benefits a household can receive may also impact the amount of LHA that a household is entitled to. As part of recent reforms to the benefits system, it is proposed to place a 26,000 cap on the total benefits that can be received by any multi-person households not in employment, or a 18,200 cap on single person households not in employment. This is in line with the median income of a household in employment and is intended to include Local Housing Allowance and Housing Benefit. When benefits are capped under this scheme, there will be an order of precedence, in which housing-related benefits will come last. Therefore all other benefits will be paid before housing-related benefits, which will be reduced in order to fit the cap. This section analyses the potential impact of this in Wiltshire. 5.68 A total of 26,593 households in Wiltshire state that they receive LHA or Housing Benefit in order to help with the payment of their rent. Of these, 16,910 do not have a household member in employment, which includes 7,950 pensioner households who are excluded from the cap. The remaining 8,960 households could theoretically be affected by the total cap on household benefits and are analysed below. 5.69 The survey measured the level of benefits currently received by these households, excluding their LHA or Housing Benefit payment. The table below shows these results, The table also shows how much these household receive in benefits including LHA (if LHA covered their housing costs in full). All the households considered here receive LHA but may not receive it in full so in some cases the total amount may be an over estimate. Income received in benefits (per year) Table 5.21 Potential LHA payments to households affected by the cap Single person households ( 18,200 cap) Multi-person households ( 26,000 cap) Without LHA With LHA* Without LHA With LHA* Under 10k 3,544 2,573 3,862 1,265 10k - 15k 282 1,048 923 2,304 15-18,200k 38 173 197 790 18,200-26k 0 70 114 675 More than 26k 0 0 0 62 Total over cap 0 70 0 62 * Assumes that LHA covers the full cost of housing which may not be the case Source: Wiltshire household survey 2011 5.70 It shows that the benefit cap is unlikely to impact many households in Wiltshire (a maximum of 132, 0.1%) if their current situation remained the same. However, the introduction of the Affordable Rent tenure (which will charge higher rents than social rented property), will increase the housing costs of some of these households, and therefore their requirement for LHA,which could push them above the cap. 58

5.71 It is therefore helpful to reverse this analysis and consider how many households would potentially be unable to claim LHA in full (because their current income from other benefits does not leave room under the cap for the full LHA). The table below summarises this. It considers all non-pensioner households in Wiltshire that do not contain anyone in employment (14,636 households in total) regardless of weather they receive LHA currently. The final row totals the number of households that could find a LHA claim constrained under the household benefit cap. This is based on the LHA caps in Salisbury BRMA where the caps are relatively high. Table 5.22 Potential LHA payments to households affected by the cap Potential money left under cap for LHA (per week) Minimum bedrooms required under bedroom standard 1 2 3 4 More than 300 5,098 3,539 943 142 250-300 1,909 291 214 46 200-250 849 142 124 0 150-200 392 3 120 59 Less than 150 378 0 0 114 Total vulnerable to cap 378 0 120 173 Source: Wiltshire household survey, 2011 5.72 Overall the number of households that could be affected by the benefit cap if they needed to claim LHA is relatively small, potentially around 650 (although this estimate is approximate due to the small sample). However, it should be noted that unemployed households requiring four bedrooms face a much higher risk of reaching the cap: around half of these households are likely to have an LHA claim constrained by the household cap. 5.73 The number of households affected by the overall benefit cap is likely to be small, primarily because of the low numbers of unemployed households in Wiltshire, the vast majority of which require only one or two bedroom accommodation (and therefore would have lower LHA requirements). However, if the number of unemployed larger households were to increase this could result in a rise in affected households. 5.74 It is worth noting that the amount of benefit received above is based on survey responses. Some households will receive tax refunds (principally Council Tax rebates) which they may or may not have included in their figure for benefits received but which would be counted under the total benefits cap. This could push a small number of additional households into the bands facing difficulty. 59

5.75 Finally, at the time of writing the household benefit cap was intended to be introduced in 2012. However, there has been extensive criticism of the policy for not taking into account the range of situations that households experience that would entitle them to a large amount in state benefit. It is anticipated that the final policy, when it is implemented, will be slightly more flexible than it is at present. For example there may be more leniency on the cap for households containing someone with a disability or large families in expensive areas. The existence of a benefit trap 5.76 The survey data can be used to try to identify to what extent households are caught in a benefit trap and do not seek employment because their income would fall. The table below shows the median annual benefit income received (excluding LHA/Housing benefit) according to household composition and number of household members working. The table contains categories which have small sample sizes; hence detailed analysis should be viewed with caution. Pensioner households are excluded from this analysis. 5.77 The table shows that generally total benefits received falls as the number of employed people in the household increases. However, once two or more people are employed benefit income does not appear to change. 5.78 Multi-adult households containing children appear to experience the greatest fall (60%) in benefit income when one person in the household is employed compared to where nobody is employed. However, there are relatively few of these households (only 3.9% of multi-adult households with children contain no employed household members) and the sample the information is based on is small, so the data here is not reliable. 5.79 Single (non-pensioner) adult households also experience a fall in over 50% in benefit income when a household member is employed. Lone parent households experience a notable decline in benefit when one person is in employment (over 2,000 per year), however lone parent households receive a greater amount in benefits than any other household type with one person employed. 60

Table 5.23 Median benefit income (excluding housing related benefits) according to household composition and number of household members working Household type Number of household members working 0 1 2 3 or more Single non pensioner 3,357 1,443 - - 2 adults, no children 2,307 1,673 1,407 1,580 Lone parent 7,477 5,184 - - 2 + adults, one child 7,776* 2,605 1,594 1,542* 2 + adults, two children 7,158* 2,186 1,783 2,502 Total 4,232 1,838 1,545 1,647 * small sample size Source: Wiltshire household survey, 2011 5.80 The table below shows the same data as the table above but includes earned income as well as benefit income (excluding LHA/Housing benefit). It highlights that while there is a notable reduction in benefit income for households in employment, the average earned income for all household types is higher than any forgone benefit income. The table shows that as a result of one member of the household working, average household income almost doubles. This impact is felt mostly by households which contain two adults and two children their income increases by 154%. Lone parents income increases the least as a result of employment by 27%. 5.81 While overall household income increases with the number of household members working, the impact of this diminishes as the number of workers increases. This reflects the greater number of parttime workers in households with more than one person in employment. Table 5.24 Median household income (including non housing related benefits) according to household composition and number of household members working Household composition Number of household members working 0 1 2 3 or more Single non pensioner 10,358 24,609 - - 2 adults, no children 22,016 29,054 42,872 47,647 Lone parent 15,593 19,755 - - 2 + adults, one child 13,590* 25,650 43,339 39,586* 2 + adults, two children 15,879* 40,382 49,606 52,200 Total 13,926 26,516 44,965 47,575 * small sample size Source: Wiltshire household survey, 2011 61

5.82 The table below shows the tenure of households containing nobody in employment (excluding pensioner households). Just under a quarter (23.4%) of non-working households own their own home without a mortgage, which equates to 11.6% of all owner-occupiers without a mortgage. This could indicate a significant number of retired owner-occupiers that have not yet reached the official pension age. Social rented households are most likely not to contain anyone in employment. Almost a third (31.0%) of social rented households do not contain anyone in employment. Over half of the unemployed households in Wiltshire reside in this tenure. Table 5.25 Household tenure by work status excluding pensioner households H holds not working All h holds % of tenure group not working % of h holds not working Owner-occupied (nm) 3,354 28,876 11.6% 23.4% Owner-occupied (wm) 735 52,447 1.4% 5.1% Social rented 7,976 25,706 31.0% 55.5% Private rented 2,299 31,488 7.3% 16.0% Total 14,363 138,516 10.3% 100.0% Source: Wiltshire household survey, 2011 5.83 The analysis presented in this section suggests that whilst some unemployed households may fear that finding employment would result in a reduction of overall income, the likelihood of this is quite low. Whilst some unemployed households will see a notable drop in their income from benefits once in employment, the additional earned income is likely to be greater than this loss. Lone parent families appear to gain the least from moving into employment. 5.84 A major limitation of the analysis thus far is that the benefit income has excluded LHA/Housing Benefit payments. For the vast majority receiving LHA/Housing Benefit this will be the largest benefit that they receive. In addition, the benefit is means tested and therefore households could reasonably fear losing their LHA/Housing Benefit entitlement if they were to find employment. The remainder of this section attempts to determine the extent to which LHA/Housing Benefit exacerbates the existence of a benefit trap. 5.85 The table below shows the number of households in Wiltshire receiving LHA by the number of people employed. It shows that whilst unemployed households are much more likely to receive LHA/Housing Benefit (62.4% compared to an average of 13.3%), over half of households in receipt of LHA/Housing Benefit contain an employed person. 62

Table 5.26 Household LHA by working status excluding pensioner households No employed people in h hold H holds with LHA All h holds % of group with LHA % of h holds with LHA None 8,960 14,363 62.4% 48.5% One 7,207 52,445 13.7% 39.0% More than one 2,299 71,708 3.2% 12.5% Total 18,466 138,516 13.3% 100.0% Source: Wiltshire household survey, 2011 5.86 Unemployed households may be particularly concerned about the losing their entitlement to LHA/Housing Benefit if they were to move into employment. However, the table above suggests that it would not necessarily be withdrawn if they found employment. 5.87 In order to gauge fully the extent to which LHA/Housing Benefit exacerbates the existence of a benefit trap it would be useful to consider how much LHA/Housing Benefit households receive. However, this was not specifically asked in the survey. For the purposes of this analysis we have estimated the amount of LHA/Housing benefit received by each household. This has been done for illustrative purposes only: the figures are based on assumptions and should be treated with caution. The survey asked households to indicate their housing costs and whether they receive LHA/Housing benefit to cover some or part of these housing costs. Therefore we have assumed that total annual benefit income of households is as follows: Households not receiving LHA/Housing benefit: total benefit income equals their benefit income excluding housing related benefits. Households receiving LHA/Housing benefit to cover all of their housing cost: total benefit income equals their benefit income excluding housing related benefits plus their housing costs. Households receiving LHA/Housing benefit to cover part of their housing cost: total benefit income equals their benefit income excluding housing related benefits plus half of their housing costs. 5.88 The following two tables replicate tables 5.23 and 5.24 by considering how benefit income and total income (earned income plus benefit income) varies by the employment status of the household for a range of household types. However, the data in the tables below also includes the estimated amount of benefit income from LHA/Housing benefit. 5.89 The table below, which looks at total annual benefit income (including housing related benefits), shows that the benefit income of unemployed households increases significantly when LHA/Housing benefit is included (as compared to the data in Table 5.23), whilst the total benefit income of all other households changes only marginally. This suggests that LHA/Housing benefit has a significant impact on the profitability of working for unemployed households. 63

Table 5.27 Median benefit income (including estimated housing related benefits) according to household composition and number of household members working Household type Number of household members working 0 1 2 3 or more Single non pensioner 6,230 1,443 - - 2 adults, no children 2,566 1,673 1,419 1,580 Lone parent 10,714 6,021 - - 2 + adults, one child 10,968* 2,685 1,594 1,542* 2 + adults, two children 12,095* 2,186 1,783 2,502 Total 6,522 1,910 1,545 1,647 * small sample size Source: Wiltshire household survey, 2011 5.90 The table below shows total annual earned income plus benefit income (including housing related benefits) by the number of employment people in the household. It shows that when LHA/Housing benefit is taken into account, the additional income received from being in employment lessens. However, the averages suggest that all households groups are notably wealthier when members are employed. Table 5.28 Median household income (including all benefits) according to household composition and number of household members working Household composition Number of household members working 0 1 2 3 or more Single non pensioner 12,729 24,788 - - 2 adults, no children 22,181 29,054 42,872 47,647 Lone parent 18,085 22,512 - - 2 + adults, one child 18,341* 26,137 43,339 39,586* 2 + adults, two children 19,406* 40,382 49,606 52,200 Total 16,568 26,516 44,965 47,575 * small sample size Source: Wiltshire household survey, 2011 5.91 Whilst it appears that households in employment are better off than unemployed households, unemployed households are likely to experience a substantial drop in benefit income if they find employment. Therefore it seems reasonable that it would only be worthwhile for an unemployed household to accept a job with a salary that was sufficient to offset the lost benefit income. 64

5.92 Finally, the Universal Credit, which gradually reduces benefit entitlement as employment increases to ensure that households are always better off if they are employed, is planned to be introduced during this Government. Therefore, once this is established, the above discussion on a potential benefit trap would in principle become irrelevant. 65

6. Particular household groups 6.1 This chapter uses survey data to profile particular groups of households that the Practice Guidance indicates as suitable for specific focus. Some groups may be disadvantaged, and some may not, but the additional detail of their circumstances should be of value in considering policy options for them. This chapter looks separately at support needs households, older person households, BME households, non-heterosexual households, families with children and young people 67

Households with support needs 6.2 The household survey collected information about households who indicated that one or more household members needed support for one or more of the reasons listed below: frail elderly a medical condition a physical disability a learning disability a mental health issue a sensory disability other. 6.3 Overall there are an estimated 41,465 households in Wiltshire with one or more members in an identified support needs group this represents 21.5% of all households in the County. It should be noted that a household found to have a support need will not necessarily need to move to alternative accommodation. In many cases the support need can be catered for within the household s current home whilst for others the issue may be the need for support rather than a specific type of accommodation. 6.4 The table below shows the number of households with each type of support need. Households with a medical condition are the predominant support needs group. The next largest group is those with a physical disability. These two categories represent the majority of all support needs households. The numbers of households in each category exceed the total number of support needs households because people can have more than one category of support need. Table 6.1 Support needs categories Category Number of households % of all households % of support needs households Medical condition 22,880 11.9% 55.2% Physical disability 13,464 7.0% 32.5% Frail elderly 7,908 4.1% 19.1% Mental health problem 6,465 3.3% 15.6% Learning difficulty 5,777 3.0% 13.9% Severe sensory disability 2,790 1.4% 6.7% Other 3,265 1.7% 7.9% Source: Wiltshire household survey, 2011 68

Overview 6.5 For the larger support needs groups, information about their location within Wiltshire is shown below. Towns in western Wiltshire have a particularly high proportion of frail elderly households (4.8%), with lower proportions in the north and south of the County, particularly in Salisbury Plain (1.2%). This is likely to be due to the high proportion of tied military accommodation in this area. 6.6 Those with a medical condition show a slightly different distribution, concentrated particularly in the medium sized towns such as Devizes, with fewer in the northern rural areas and small towns. People with a physical disability are most likely to be found in Salisbury Plain (8.9%) and Bradford-on-Avon (9.0%), and least likely to be found in the small towns in the north of the County (5.3%). Table 6.2 Location of support needs households Sub-area Frail elderly Medical condition Physical disability All support needs Bradford-on-Avon 5.1% 13.4% 9.0% 23.9% Chippenham 3.7% 10.1% 6.4% 19.4% Devizes 3.8% 15.2% 6.2% 25.8% North and Central Rural 4.8% 10.0% 6.4% 20.2% North Small Towns 4.6% 8.8% 5.3% 18.1% North Towns 4.2% 15.7% 7.0% 23.6% Salisbury 3.3% 11.7% 6.6% 22.7% Salisbury Plain 1.2% 12.0% 8.9% 19.3% South Rural 3.2% 11.1% 7.6% 17.7% Trowbridge 5.2% 12.6% 7.7% 23.8% West Towns 4.8% 14.2% 7.3% 25.2% Total 4.1% 11.9% 7.0% 21.5% Source: Wiltshire household survey, 2011 6.7 Support needs households most commonly reside in the social rented sector, as shown in the figure below, which indicates that 43.2% of all social rented dwellings contain a person with a support need. 69

Figure 6.1 Proportion of households within tenure containing a support needs member 50.0% 43.2% 40.0% 30.0% 20.0% 20.8% 11.8% 15.0% 10.0% 0.0% Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented Source: Wiltshire household survey, 2011 6.8 The table below shows the tenure distribution for each support needs group compared with all support needs households, all households without a support need and all households in the County as a whole. All support needs groups were more likely than average to reside in the social rented sector, particularly households containing someone with a learning difficulty or a mental health problem. Table 6.3 Tenure and support needs group Support need Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented Total Frail elderly 54.2% 5.0% 35.0% 5.8% 100.0% Medical condition 31.7% 15.8% 40.4% 12.2% 100.0% Physical disability 36.5% 10.5% 43.2% 9.8% 100.0% Learning difficulty 7.3% 18.7% 56.9% 17.0% 100.0% Mental health problem 22.1% 14.4% 47.2% 16.3% 100.0% Severe sensory disability 37.8% 29.0% 28.0% 5.2% 100.0% Other 21.4% 15.8% 49.4% 13.5% 100.0% All support needs households 32.9% 15.7% 38.7% 12.7% 100.0% All non-support needs households 34.1% 32.3% 13.9% 19.6% 100.0% All households 33.9% 28.8% 19.2% 18.1% 100.0% Source: Wiltshire household survey, 2011 6.9 The table below shows the household size of the various support needs groups. Households containing people with support needs were on average about the same size as non-support needs households, but this conceals wide variation. Frail elderly households were particularly small, with 63.9% containing just one person, while nearly half of households containing a person with a learning difficulty contain four or more people. 70

Table 6.4 Household size and support needs group Support need 1 person 2 people 3 people 4 or more people Total Average h hold size Frail elderly 63.9% 32.5% 1.6% 2.0% 100.0% 1.4 Medical condition 33.4% 40.5% 11.8% 14.3% 100.0% 2.2 Physical disability 39.9% 41.7% 9.8% 8.6% 100.0% 1.9 Learning difficulty 10.4% 22.3% 21.3% 46.0% 100.0% 3.4 Mental health problem 23.5% 41.3% 18.6% 16.7% 100.0% 2.4 Severe sensory disability 37.2% 35.5% 9.9% 17.3% 100.0% 2.2 Other 23.4% 30.9% 19.4% 26.3% 100.0% 2.6 All support needs households 34.7% 39.3% 11.1% 14.9% 100.0% 2.2 All non-support needs households 28.4% 38.5% 12.8% 20.3% 100.0% 2.3 All households 29.8% 38.7% 12.4% 19.1% 100.0% 2.3 Source: Wiltshire household survey, 2011 6.10 The figure below shows the broad household types of support needs households. Overall, they tend to be more likely to be older person households, but this is not the case for all support needs groups. Households containing a person with a learning difficulty are mostly families. Mental health problems are distributed almost evenly across all household types, although they particularly affect lone parent households. Figure 6.2 Household type and support need group Frail elderly 63.9% 27.2% 1.5% 6.8% Medical condition 21.4% 18.3% 12.0% 18.3% 8.5% 5.7% 8.2% 7.6% Physical disability 29.8% 23.0% 10.1% 14.4% 8.7% 4.1% 6.0% 3.9% Learning difficulty Mental health problem 4.0% 1.5% 6.1% 8.9% 13.1% 5.1% 9.1% 17.4% 26.1% 19.8% 13.8% 12.2% 15.2% 31.6% 8.1% 8.2% Severe sensory disability 31.6% 12.3% 5.6% 10.3% 13.2% 6.6% 9.6% 10.8% Other 11.6% 7.4% 11.8% 7.4% 11.8% 7.1% 20.0% 22.9% All support needs households 24.1% 17.7% 10.6% 14.5% 9.7% 6.5% 7.3% 9.6% All non-support needs households All households 12.0% 14.6% 12.5% 13.6% 16.4% 15.1% 21.6% 20.1% 7.9% 8.3% 4.7% 5.1% 9.4% 9.0% 15.5% 14.2% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Single pensioner 2+ pensioners Single non-pensioner Couple, no children Multi-adult Lone parent 2+ adults, 1 child 2+ adults, 2+ children Source: Wiltshire household survey, 2011 71

Financial situation 6.11 All types of support needs household report a lower than average income per head, even when benefits are taken into account. The lowest incomes per head were found among those households with mental health problems and learning difficulties, as well as the group who identified themselves as having other support needs. Figure 6.3 Household income and benefits by support needs group, per head Frail elderly Medical condition Physical disability 8,000 8,900 8,500 1,260 1,440 1,490 Learning disability Mental health problem 5,900 5,900 1,490 1,700 Severe sensory disability 9,000 1,200 Other 5,900 1,280 All support needs h'holds 8,500 1,280 All non-support needs h'holds 14,800 680 All h'holds 12,700 700 0 5,000 10,000 15,000 Annual gross household income (including non-housing benefits) Source: Wiltshire household survey, 2011 6.12 Support needs households are less likely to be able to afford market housing than other households some 41.9% are unable to afford, compared to 19.5% of other households. A total of 26.1% of support needs households would require LHA even if resident in social rented housing. 6.13 Support needs households are more likely to be in unsuitable housing than other households 11.4% of support needs households are unsuitably housed compared to 4.4% of non-support needs households. The table below shows the proportion of each support needs group estimated to be living in unsuitable housing. As can be seen, those with learning difficulties, mental health problems or other support needs are particularly likely to face housing problems. 72

Table 6.5 Proportion of support needs groups living in unsuitable housing Support need % of households Frail elderly 7.1% Medical condition 11.5% Physical disability 12.6% Learning difficulty 25.6% Mental health problem 15.8% Severe sensory disability 6.5% Other 18.6% All support needs households 11.4% All non-support needs households 4.4% All households 5.9% Source: Wiltshire household survey, 2011 Improvements required 6.14 Support needs households were asked to indicate if there was a need for improvements to their current accommodation and/or services they required. The responses are detailed in the figure below. The most common requirement was for a handyperson service to help maintaining the home, and the most common physical adaptation was for a low level shower unit. Figure 6.4 Support needs households - improvements to accommodation & services Handyperson service 6,310 Low level shower unit 4,467 Extra handrails Other bathroom/kitchen alterations More support services to home Emergency alarm Support managing finances / with forms Downstairs toilet Lift or stair lift Need to move to alternative housing with specialist adaptations or care/support Car parking space near to door Other alterations to improve accessibility Alterations to the kitchen 3,572 3,249 2,936 2,526 2,423 2,417 2,317 2,187 1,865 1,397 1,254 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 Source: Wiltshire household survey, 2011 Households 73

Older persons 6.15 In this study older people are defined as those over the state pension eligibility age (currently 65 for men, 60 for women). The household survey suggests that 54,484 households in Wiltshire contain only older people (28.2%) and a further 14,622 (7.6%) contain both older and non-older people. 6.16 Using the ONS 2008-based population projections in combination with the survey dataset, we can estimate that the number of older person only households is likely to increase by 52% in the next 15 years (2011-2026), to an estimated 82,889. The number of households containing both older and nonolder people is likely to remain about the same, increasing by just 1% to an estimated 14,769. Overview 6.17 The locations with the highest proportion of older person only households are Bradford-on-Avon (37.9%) and the South Rural area (36.8%). The smallest percentages of older person only households are found in Salisbury Plain (15.1%) and Chippenham (22.1%). Table 6.6 Location of older person households Sub-area No older people Some older people All older people Total Bradford-on-Avon 54.1% 8.1% 37.9% 100.0% Chippenham 71.7% 6.2% 22.1% 100.0% Devizes 66.4% 6.6% 27.0% 100.0% North and Central Rural 60.6% 10.4% 29.0% 100.0% North Small Towns 63.7% 8.6% 27.7% 100.0% North Towns 63.8% 5.5% 30.6% 100.0% Salisbury 66.2% 7.0% 26.8% 100.0% Salisbury Plain 78.2% 6.7% 15.1% 100.0% South Rural 55.1% 8.1% 36.8% 100.0% Trowbridge 69.1% 5.4% 25.5% 100.0% West Towns 63.5% 5.9% 30.7% 100.0% Total 64.2% 7.6% 28.2% 100.0% Source: Wiltshire household survey, 2011 6.18 The figure below shows the tenure older person households are resident in. The data shows that 72.6% of older person only households are owner-occupiers, nearly all of these without a mortgage. There are very few in the private rented sector; just 6.5% live in this tenure compared to 24.4% of households that contain no older people. About a fifth (20.9%) live in the social rented sector. 6.19 Further analysis shows that more than half (55.8%) of all owner-occupied properties without a mortgage in Wiltshire are occupied by older person only households. This type of household also occupy nearly a third (30.7%) of all social rented properties. 74

Figure 6.5 Tenure of older person households All older people 67.0% 5.6% 20.9% 6.5% Some older people 55.5% 19.5% 16.4% 8.5% No older people 16.8% 40.0% 18.8% 24.4% All households 33.9% 28.8% 19.2% 18.1% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented Source: Wiltshire household survey, 2011 Under-occupation and household mobility 6.20 The survey suggested that almost all (99.9%) of older person only households are comprised of one or two people yet more than half (59.4%) of older person households reside in accommodation with three or more bedrooms. This suggests that there could be potential scope to free up larger units for younger families if older person households should choose to move into suitable smaller units. 6.21 The vast majority of older person households with two or more spare bedrooms live in the owneroccupied sector (89.2%), with only 7.1% (2,279 households) in the social rented sector. This would limit the scope to reduce under-occupation through Council initiatives. A total of 76.3% of underoccupying older person households in Wiltshire indicated that they were very satisfied with their existing accommodation, with only 1.6% either fairly or very dissatisfied. 6.22 A relatively small proportion of older person households were planning to move, or had moved recently. Only 7.3% of households containing only older people moved in the last two years (prior to the survey) compared with 26.8% of households containing no older people. Only 9.7% intended to move in the next two years, compared with 29.5% of households containing no older people. Accommodation preferences 6.23 If Wiltshire are seeking to reduce under-occupation among older person households in the owneroccupied sector, it is crucial to know what these households are looking for. The reasons older person households listed for choosing a property are analysed in the figure below. Issues particularly important for older people include suitability for retirement, the perceived safeness of an area, and issues relating to accessibility. This included both within the home (e.g. steps/stairs) and access to services, transport and in particular to friends and family. Older person households tended to be slightly less concerned than average about the size of a house, its value for money, or its condition on purchase. Older person households also showed a slightly greater preference than average for newlybuilt over older homes. 75

Figure 6.6 Older person households: reasons for choice of dwelling (past moves) Price/value for money Number of bedrooms 39.7% 42.8% 50.1% 51.4% Close to friends/family 27.9% 46.7% Close to work 7.5% 34.2% Close to school 0.2% 15.7% Close to services 20.3% 35.6% Transport connections 15.8% 27.8% Good neighbourhood Safe area 32.6% 41.7% 43.6% 43.4% Accessibility needs 8.9% 25.1% General house condition 34.4% 39.5% Suitable for retirement 1.5% 48.0% Newly-built home 5.3% 8.8% Older home 7.6% 10.8% Design of housing 14.1% 16.7% Garden 41.7% 41.6% Lack of alternatives 17.3% 20.6% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% Older person only households Households containing no older people Source: Wiltshire household survey, 2011 6.24 In total, 37.1% of older person only households moving into properties in Wiltshire in the last two years came from outside the County, accounting for 9.9% of all moves into Wiltshire in that time period. 6.25 Looking at future moves, 55.9% of moving older person only households wish to move to owneroccupied accommodation and 29.5% to a social rented home. In terms of dwelling type, some 42.2% wish to move to a detached house, 30.1% to a bungalow and 16.5% to a flat. The remainder (11.2%) want to move to a semi-detached or terraced house. 6.26 Moving older person households show almost identical area preferences to other households, with 26.5% seeking to move out of Wiltshire, compared to 26.3% of moving households as a whole. Almost exactly half of older person households (51.2%) want to remain in the same Community Area (CA) as they currently live, compared to 49.5% of all moving households. Looking at the urban/rural dimension, 76

about 27.4% of older moving households wish to live in rural parts of the County, similar to the figure of 27.0% for all moving households. Financial situation 6.27 As shown in the table below, only 2.7% of older person only households were found to be in unsuitable housing, compared to 5.9% of households across the County as a whole. While older person households have lower incomes than other households, they have higher levels of capital (calculated as savings and equity combined). However, it should not be assumed that all older people have access to capital. Table 6.7 Median income, benefits and savings, and unsuitable housing: older people Older people Income* Income per head* Average benefits per head** Capital % in unsuitable housing All older people 16,100 11,000 1,200 211,500 2.7% Some older people 25,400 12,000 640 235,800 3.3% No older people 32,600 14,600 680 44,300 7.6% Overall 25,500 12,700 700 109,600 5.9% * per annum, excluding benefits ** per annum, excluding LHA/HB Source: Wiltshire household survey, 2011 6.28 In terms of affordability, some 22.7% of older person households are unable to afford market housing in Wiltshire. This compares to 24.3% of all households in the County. It is worth noting however that only 6.7% can afford to rent privately compared to 18.6% overall, while 70.1% can afford to buy outright compared with 48.4% overall. This reflects their financial situation, having low average incomes and high average savings. Black and Minority Ethnic (BME) households 6.29 Due to the very small BME population in the County, it was not possible to weight the data to correct for lower or higher response rates among these groups. The figures provided here for the overall group sizes should therefore be treated with caution. However, the size of the estimated BME popualtion according to the housheold survey (after weighting by all other variables) is presented in the table below. 77

Table 6.8 Ethnicity in Wiltshire (not adjusted for potential response bias) Ethnicity Households % of households Population % of population White British 184,644 95.7% 418,182 94.9% White Irish 1,511 0.8% 2,843 0.6% White Other 4,065 2.1% 11,328 2.6% Mixed 938 0.5% 2,678 0.6% Black / Black British 717 0.4% 1,998 0.5% Asian / Asian British 674 0.3% 2,052 0.5% Other 449 0.2% 1,538 0.3% Total 193,000 100.0% 440,619 100.0% Source: Wiltshire household survey, 2011 6.30 The small sample seriously restricts the amount of analysis that can be done of these groups. To enable some basic analysis to be done, we have aggregated all non-white groups, and combined White Irish and White Other. 6.31 The figure below shows how tenure varies by housheold ethnicity. It is clear that BME households tend to be more likely to live in rented accommodation. About two thirds of households where the household head is non-white are renters, compared to just 36.6% among White British households. Figure 6.7 Tenure of BME households White British 34.6% 28.7% 19.4% 17.2% White Other 20.2% 31.9% 12.5% 35.3% Non-White 10.1% 25.0% 21.7% 43.2% All households 33.9% 28.8% 19.2% 18.1% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented Source: Wiltshire household survey, 2011 6.32 Analysis of the income, benefits and housing situation of BME households is shown in the table below. Although the figures are approximate due to the small sample, it is clear that non-white households have a lower income per head than White British households. BME households generally have a much lower level of capital, with the non-white households recording a median figure of just 700. This is because fewer than half of such households are owner-occupiers, with equity being the main source of capital for most households. The level of benefits (excluding LHA or Housing Benefit) does not vary significantly by ethnic group. 78

6.33 The proportion of non-white households in unsuitable housing is significantly higher than average, at 16.5% compared to 5.7% among White British households. However, there is no signficant difference from the average for White Other households. A slightly higher proportion of non-white households stated that they suffer from harassment or threats; however it cannot be confirmed why this is or whether it could be accounted for by other factors (e.g. location of residence). Table 6.9 Median income, benefits and savings, and unsuitable housing: ethnic groups Ethnicity Income* Income per head* Average benefits per head** Capital % in unsuitable housing % suffering harassment or threats White British 25,500 12,800 700 113,100 5.7% 2.4% White Other 30,800 12,600 680 26,500 6.1% 0.5% Non-White 20,300 8,700 690 700 16.5% 4.6% Total 25,500 12,700 700 109,600 5.9% 2.4% * per annum, excluding benefits ** per annum, excluding LHA/HB Source: Wiltshire household survey, 2011 Sexual orientation 6.34 The survey also asked about sexual orientation. It was not possible to weight the data to correct for lower or higher response rates among these groups. The figures provided here should therefore be treated with caution. However, the proportions according to the housheold survey (after weighting by all other variables) are presented in the table below. Table 6.10 Sexual orientation in Wiltshire (not corrected for response rates) Sexual orientation Households % of households Heterosexual / Straight 168,636 87.4% Gay Man 1,034 0.5% Lesbian / Gay Woman 923 0.5% Bi-sexual 743 0.4% Other 444 0.2% Prefer not to state 21,221 11.0% Total 193,000 100.0% Source: Wiltshire household survey, 2011 6.35 For the remainder of the analysis, all non-heterosexual respondents (other than prefer not to state who are listed separately) have been grouped together due to the small sample size. 6.36 Non-heterosexual respondents were slightly more likely than average to live in unsuitable housing (8.3% compared to 5.8%). They were not significantly more or less likely to experience harassment or threats than other households. 79

Table 6.11 Unsuitable housing by sexual orientation Sexual orientation % in unsuitable housing % experiencing harassment / threats Heterosexual / Straight 5.8% 2.5% Non-heterosexual 8.3% 2.0% Prefer not to state 6.6% 1.9% Total 5.9% 2.4% Source: Wiltshire household survey, 2011 6.37 In terms of tenure, non-heterosexual households were considerably more likely to be private renters, and less likely to own their house without a mortgage. Figure 6.8 Sexuality: tenure Heterosexual 33.0% 30.0% 18.4% 18.6% Non-heterosexual 18.4% 33.3% 15.0% 33.3% Prefer not to state 43.0% 18.5% 26.4% 12.2% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented Source: Wiltshire household survey, 2011 Families with children 6.38 The Practice Guidance acknowledges the importance of providing suitable housing for families, especially those with children, to ensure that communities are appropriately mixed. For the purposes of analysis children are defined in the same way as the 2001 Census those under 16, and those aged 16-18 if in full time education. 6.39 The household survey estimates that there are 54,582 households which contain at least one child in Wiltshire. Of these 17.8% are lone parent families, 53.2% are two parent families with at least one young child (aged 9 or under) and 28.9% are two parent families with only older children (aged 10-18). 6.40 Overall, the largest proportion of households with children is found in the Salisbury Plain sub-area. This area has a highly unusual distribution of family types, with almost all being two parent households with younger children. This is likely to be due to the high proportion of tied military accommodation in the area. There are slightly fewer families than average in Bradford-on-Avon (21.3%) and Trowbridge (24.7%). The highest proportions of lone parent families are found in towns, in particular in Salisbury. 80

Table 6.12 Location of households with children Sub-area Lone parent Older children Younger children All families Bradford-on-Avon 2.9% 6.1% 12.3% 21.3% Chippenham 6.1% 9.2% 15.0% 30.3% Devizes 6.7% 5.5% 16.5% 28.7% North and Central Rural 3.2% 8.6% 15.8% 27.5% North Small Towns 6.1% 8.8% 16.0% 30.8% North Towns 5.2% 7.8% 16.8% 29.9% Salisbury 7.0% 6.4% 14.2% 27.6% Salisbury Plain 6.1% 7.7% 24.3% 38.1% South Rural 3.4% 10.5% 12.2% 26.1% Trowbridge 6.0% 7.5% 11.1% 24.7% West Towns 6.1% 8.4% 12.5% 27.1% Total 5.1% 8.2% 15.0% 28.3% Source: Wiltshire household survey, 2011 6.41 The figure below shows the tenure of the three groups of family households. Two parent households are particularly likely to be owner-occupiers, although a significant proportion of those with younger children are renters. About 17% in each case live in social rented housing. Tied accommodation, most of it military, is not a major tenure in the County as a whole, making up only 3.3% of all housing. However, 8.6% of two parent families with young children live in this tenure, occupying 39.6% of all tied accommodation in the County. 6.42 Single parent families show a radically different tenure distribution to the other groups, with 78.4% living in rented accommodation. Further analysis shows that 67.5% of lone parent households in Wiltshire claim Local Housing Allowance or Housing Benefit, compared to a county average of 38.8%. They account for a total of 19.4% of all the claimants in the County. 81

Figure 6.10 Tenure of family households All families Single parent Older children 10.8% 7.3% 14.4% 18.8% 45.5% 47.0% 54.5% 22.2% 15.6% 31.4% 16.9% 6.1% 5.9% 1.2% 3.7% Younger children 7.7% 51.0% 16.8% 15.8% 8.6% All households 33.9% 28.8% 19.2% 14.9% 3.3% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented Tied / Military Source: Wiltshire household survey, 2011 6.43 Looking at household mobility, the survey indicates a total of 24.8% of families with children have moved in the last two years, compared to 18.0% of other households. Lone parent households were particularly likely to have moved recently (32.6%), compared to 30.0% of two parent households with younger children, and 10.4% of two parent households with older children. 6.44 As is shown in the chart below, families had specific reasons for choosing a particular property. Most families considered the size of the property to be an important concern. Notably, nearly a third (32.5%) considered the local schools when moving to their current property. They were also more likely than average to consider the design of the housing (e.g. semi-detached) and the garden. 82

Figure 6.11 Families: reasons for choice of dwelling (past moves) Price/value for money Number of bedrooms Close to friends/family Close to work Close to school Close to services Transport connections Good neighbourhood Safe area Accessibility needs General house condition Suitable for retirement Newly-built home Older home Design of housing Garden Lack of alternatives 26.2% 31.3% 28.7% 31.6% 32.5% 3.3% 22.7% 21.9% 10.5% 21.5% 33.3% 33.7% 7.2% 12.9% 32.8% 0.2% 12.1% 4.0% 6.6% 8.7% 11.3% 20.3% 13.8% 24.4% 17.7% 41.8% 46.1% 41.7% 44.6% 42.2% 45.0% 39.9% 52.6% 58.1% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% Households containing children Other households Source: Wiltshire household survey, 2011 6.45 Survey data, shown in the table below, indicates that lone parent families in Wiltshire are far more likely than average to live in unsuitable housing 20.0% live in unsuitable housing compared to just 5.9% of households overall. They account for 17.2% of all unsuitable housed households in the County. Both groups of two parent families are more likely than average to live in unsuitable housing, with 9.7% of those with older children and 9.5% of those with younger children in this situation. 6.46 Two parent families generally have a high income as a household, with a median income of over 40,000. However, when the number of people in the household is considered, the income per head is lower than for non-family households. Lone parent households show exceptionally low earned incomes both as households and per head, although this income is typically supplimented by an above average sum in benefits. 83

Table 6.13 Median income, benefits and savings, and unsuitable housing: families Income* Income per head* Average benefits** Capital % in unsuitable housing Lone parent 10,400 3,700 5,900 200 20.0% Older children 43,600 11,900 1,800 138,100 9.7% Younger children 40,400 9,900 1,900 39,200 9.5% No children 23,200 14,700 1,800 140,600 3.7% County average 25,500 12,700 1,900 109,600 5.9% * per annum, excluding benefits ** per annum, excluding LHA/HB Source: Wiltshire household survey, 2011 Young people 6.47 About 8,995 households in Wiltshire (4.7%) are estimated to be composed entirely of people under 30 and do not contain children. This group are located disproportionately in the largest towns. The subarea with the highest proportion of younger households is Chippenham (8.9%), closely followed by Salisbury Plain (8.7%) and Trowbridge (7.8%). However, although it is the largest urban area, younger households only comprise 5.1% of the household population in Salisbury. The rural areas generally contain a very low proportion of young households, just 3.2% in the North and Central Rural sub-area and 1.6% in the South Rural sub-area. 6.48 As shown in the chart below, more than half of young households live in the private rented sector (54.1%), with over a quarter being owner-occupiers (29.6%). Very few are social renters, perhaps unsurprisingly given the relatively limited supply of this tenure in recent years. Figure 6.12 Tenure of young households Young households 2.2% 27.4% 9.4% 54.1% 6.9% Other households All households 35.4% 33.9% 28.8% 28.8% 19.7% 19.2% 14.1% 16.0% 1.9% 2.1% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private rented Tied / Military Source: Wiltshire household survey, 2011 84

6.49 Young people are particularly mobile, with 69.3% reported to have moved in the last two years. The figure below shows what they considered when looking for their new home. Young households are particularly conscious of value for money, and the condition of a property. They are also particularly likely to choose a property for its closeness to their workplace. Figure 6.13 Young people: reasons for choice of dwelling (past moves) Price/value for money Number of bedrooms Close to friends/family 28.1% 37.0% 46.9% 49.6% 53.7% 58.8% Close to work Close to school Close to services Transport connections 3.3% 15.6% 14.7% 17.3% 17.7% 23.7% 27.6% 46.1% Good neighbourhood Safe area Accessibility needs 8.0% 11.5% 32.7% 38.4% 42.5% 43.8% General house condition Suitable for retirement Newly-built home Older home Design of housing Garden Lack of alternatives 0.0% 4.7% 9.4% 10.5% 11.9% 10.1% 14.3% 16.5% 16.4% 20.8% 37.7% 36.9% 42.7% 45.1% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% Young household Other households Source: Wiltshire household survey, 2011 6.50 The proportion of younger households in unsuitable housing is significantly lower than the average, at 3.5% compared to 6.0% among other households. However, because few younger households are owner-occupiers, they have a significantly lower level of capital, restricting their ability to buy property. 85

Table 6.14 Median income, benefits and savings, and unsuitable housing: young people Income* Income per head* Average benefits** Capital % in unsuitable housing Young household 28,800 20,400 1,500 1,300 3.5% Other household 25,200 12,500 1,900 118,500 6.0% County average 25,500 12,700 1,900 109,600 5.9% * per annum, excluding benefits ** per annum, excluding LHA/HB Source: Wiltshire household survey, 2011 86

7. Rural as compared with urban housing Summary i) Wiltshire s population is about two thirds rural, compared with the national average of a fifth. ii) There a high concentration of pensioners in rural areas. Lone parents and single non-pensioners decline as rurality increases. Income and capital (savings and equity) are higher in rural areas than in urban Wiltshire. iii) Car ownership is higher in rural areas among both high and low income families. Rural households are both most likely to work from home and to travel outside of Wiltshire for work. Introduction 7.1 Wiltshire is a largely rural County, but with a notable urban population. Housing issues faced in rural areas are often different to those faced in urban areas. This chapter will create an urban/rural typology for Wiltshire, which will be used to analyse and understand the differences across the County using the household survey data. Initially the recent government publications addressing rural housing will be summarised. Rural housing policy background 7.2 A major step in addressing rural issues was the Rural White Paper ( Our Countryside: the future 2000). The White Paper identified a number of trends that have led to problems for those living in rural areas; the decline of job opportunities, an associated decline in the incomes that can be earned locally, a slow erosion of service infrastructure and in some areas a reduction in the resident population. 7.3 The White Paper also addresses the issue of deprivation in rural communities: a problem often made worse by their isolated state. One of the key issues highlighted in the White Paper is the lack of affordable housing provision in rural areas to assist those in poverty, which has been exacerbated by the Right to Buy effect being more pronounced in rural than in urban areas. 7.4 The most recent major study of rural housing issues is the The Taylor Review of rural economy and affordable housing: Living Working Countryside (Matthew Taylor, CLG 2008). Taylor summarises his findings in terms of two broad categories: villages and hamlets and market towns. His conclusions for villages and hamlets are: 87

the choice is between becoming ever more exclusive enclaves for the wealthy and retired, or building the affordable homes to enable people who work in these communities to continue to live in them. In many cases just a handful of well designed homes, kept affordable in perpetuity for local people, will make all the difference (Foreword) 7.5 In the case of market towns, an entirely different conclusion is arrived at: Market towns face a different choice. The demand to live in a rural community means many market towns are growing fast. One option is to stick with current planning practice which are too often ringing country towns with anonymous housing estates, business and retail parks...[he argues for a more imaginative approach, based on a few enlightened examples] (Foreword) 7.6 The challenges faced are quite different, but the one constant is the pressure for new housing these areas will face in the future. This is forcefully illustrated by one of the diagrams in the report, which sets out the projected population growth across the country, disaggregated by the rurality of the area: 7.7 The diagram is clear; over the next two decades all types of rural area will grow faster than urban ones, and the most rural will grow the fastest. As Taylor suggests it is crucial that the nature of the growth permitted will produce the best results for these rural communities. Wiltshire s rural area 7.8 It is worth noting that the official rural classification by Defra allows for the fact that rural is an umbrella term that includes many urban areas. Defra includes as rural settlements below 10,000 population. Defra uses a three level tiering to classify rural areas: Town and fringe Village Hamlet (including isolated dwellings) 88

7.9 Defra data indicates that nationally about 20% of the population falls into one of these three rural categories. The 2009 Defra data for Wiltshire suggests that 68% of the population of the County live in rural areas (297,223 people). The following Defra map compares the rurality of Wiltshire with other areas of England and Wales. Figure 7.1 Map of Urban /Rural classification of Counties Source: Defra 2001 89

7.10 We have used the sub-area classification identified in Chapter 2 as the basis for differentiating the urban and rural areas of Wiltshire within the household survey. This is because the boundaries closely follow the edges of large settlements, allowing a meaningful urban/rural split. The sampling scheme was also designed to ensure the outputs for each of these areas were robust and accurate. The urban and rural typology applied to the household dataset is set out in the table below. Table 7.1 Urban and rural typology used for Wiltshire analysis Typology Sub-area Towns included Population Urban 1 (Principal settlements) Chippenham Chippenham 33,100 Salisbury Salisbury 46,100 Trowbridge Trowbridge 35,600 Calne 15,900 North Towns Corsham 12,200 Urban 2 (Smaller Towns over 10,000 population) Wootton Bassett 11,100 Devizes Devizes 16,200 West Towns Melksham 19,300 Warminster 16,700 Westbury 12,000 Urban 3 (Salisbury Plain) Salisbury Plain Salisbury Plain 32,100 Rural 1: Small towns (under 10,000 population) Bradford-on-Avon Bradford-on-Avon 8,700 North Small Towns Rural North and Central Cricklade 4,400 Malmesbury 5,900 Marlborough 7,500 Pewsey 3,400 Downton 2,200 Mere 2,900 Tisbury 1,700 Rural 2: Rural villages Rural North and Central - 114,100 Rural South - 39,500 Source: Wiltshire household survey, 2011 7.11 The distinctions reflect the national policy that rural should refer only to areas with settlements less than 10,000. The two purely rural sub-areas have substantial populations, but they are spread over a wide area and do not contain any substantial towns. Salisbury Plain is in a category of its own. It could be considered to be rural, since it consists only of settlements which are technically villages, but some of these settlements are now quite large, and being primarily inhabited by military personnel are not typical rural areas. It is therefore treated as its own category. This system of three urban and two rural categories is used for the analysis of the household survey dataset which follows. 90

Household characteristics 7.12 The following table provides information on the tenure of households according to the urban/rural classification set out above. The data indicates that the level of owners without a mortgage increases with rurality, but there is not a large difference in the scale of the private rented sector between the areas. Other than a very small social rented sector in the rural villages area, this tenure is also relatively evenly spread across the County. This corresponds to the broad picture of the most rural areas being dominated by older owner-occupiers who have paid off their mortgages, while the former social rented stock has been lost to Right to Buy. Area Owner-occupied (no mortgage) Table 7.2 Tenure pattern by typology Owner-occupied (with mortgage) Tenure Social Rented Private Rented Tied / Military Urban 1 30.2% 29.7% 22.1% 16.4% 1.6% 51,462 Urban 2 31.6% 32.1% 21.1% 14.5% 0.7% 46,893 Urban 3 14.8% 23.7% 29.1% 13.0% 19.5% 12,620 Rural 1 36.3% 25.9% 22.9% 13.6% 1.2% 16,576 Rural 2 41.4% 27.4% 12.7% 14.6% 3.9% 65,448 Countywide Total 33.9% 28.8% 19.2% 14.9% 3.3% 193,000 Source: Wiltshire household survey, 2011 7.13 The table below shows the different type of households resident in each area. Both rural areas display a higher than average proportion of pensioner households. Lone parent and single non-pensioner households decline as rurality increases. Area Single pensioner 2+ pensioner Table 7.3 Household types by typology Couple, no children Tenure Multi-adult Lone parent 2+ adults, one child 2+ adults, 2+ children Urban 1 13.8% 11.2% 20.3% 17.9% 9.3% 6.4% 7.3% 13.7% 51,462 Urban 2 16.4% 13.6% 16.4% 18.5% 6.7% 5.9% 9.7% 12.8% 46,893 Urban 3 7.8% 7.4% 13.2% 22.3% 11.3% 6.1% 12.0% 20.0% 12,620 Rural 1 18.1% 15.5% 14.3% 17.8% 7.8% 5.4% 7.4% 13.6% 16,576 Rural 2 14.4% 16.1% 10.8% 23.1% 8.1% 3.1% 9.6% 14.8% 65,448 Single nonpensioner Countywide 14.6% 13.6% 15.1% 20.1% 8.3% 5.1% 9.0% 14.2% Source: Wiltshire household survey 2011 Total 193,000 91

Financial situation 7.14 The table below presents a financial profile of households across the five areas used within the urban/rural typology. Income, especially income per head, increases with rurality, and even more strikingly so does capital. There is no urban/rural distinction in the level of benefits received. Table 7.4 Financial information by typology (medians) Area Income* Benefits** Capital Income per head Total households Urban 1 24,800 1,910 75,400 12,700 51,462 Urban 2 21,600 1,980 78,200 11,300 46,893 Urban 3 28,400 1,850 5,100 12,000 12,620 Rural 1 25,200 1,930 139,300 13,100 16,576 Rural 2 29,200 1,770 195,800 14,800 65,448 25,500 1,880 109,600 12,700 193,000 * gross per household, excluding benefits ** excluding LHA / Housing Benefit Source: Wiltshire household survey, 2011 7.15 The table below shows the proportion of households in each area that can be considered as low income (gross household income excluding benefits less than 20,00 per year). The data suggests that there is no strong pattern except that Urban 3 (Salisbury Plan) and Rural 2 (Rural South) show fewer low income households than the other areas. Table 7.5 Low income households by typology Area Income less than 20,000* Income more than 20,000 Total Urban 1 39.8% 60.2% 51,462 Urban 2 46.4% 53.6% 46,893 Urban 3 32.2% 67.8% 12,620 Rural 1 40.5% 59.5% 16,576 Rural 2 30.5% 69.5% 65,448 Countywide Countywide 37.8% * gross per household, excluding benefits Source: Wiltshire household survey, 2011 62.2% 193,000 7.16 The table below shows the ability of households to afford market housing. The data indicates that households in the most rural area are most likely to be able to afford market housing and those in Salisbury Plan (Urban 3) are least able to. 92

Table 7.6 Affordability of housing by typology Area Afford to buy Afford to private rent* Cannot afford market housing Total Urban 1 51.3% 23.6% 25.1% 51,462 Urban 2 54.3% 17.8% 27.9% 46,893 Urban 3 32.8% 35.9% 31.3% 12,620 Rural 1 56.7% 16.6% 26.7% 16,576 Rural 2 65.0% 16.8% 18.1% 65,448 Countywide 56.0% 20.1% 23.9% 193,000 * but not to buy Source: Wiltshire household survey, 2011 Travel patterns 7.17 The rurality of a location will obviously have a big impact on the how households are able to access different things, as rural areas are associated with increased remoteness. The table below considers how car ownership varies across the urban/rural typology. The table indicates, as would be expected, the proportion of households with no cars decreases as rurality increases and also that that the most rural area (Rual 2) has by far the largest number of households with two or more cars. Over half (55.7%) of households in this area have at least two cars, suggesting that this is not just a consequence of greater wealth, but is practically necessary. Table 7.7 Car ownership by typology Area No cars One car Two or more cars Total Urban 1 21.6% 47.5% 30.9% 51,462 Urban 2 19.0% 47.2% 33.8% 46,893 Urban 3 13.6% 49.4% 37.0% 12,620 Rural 1 13.4% 53.3% 33.3% 16,576 Rural 2 9.7% 34.6% 55.7% 65,448 County-wide 15.7% 43.7% 40.6% 193,000 Source: Wiltshire household survey, 2011 7.18 The following table displays the workplace location of household heads by the urban/rural typology. The data shows that a much higher proportion of workers operate from home or are mobile, in Rural 2 while Rural 1 also shows a high presence of home workers. This mirrors patterns recorded nationally by Defra research. Urban households are much more likely to work locally (over 20% of urban households work in the local area compared to around 10% of rural households). The proportion of households working outside Wiltshire is greatest in the rural parts of the County showing the presence of a notable number of long-distance (and often affluent) commuters in the these areas. 93

Table 7.8 Commuting by typology Area Not in employment Work from home / mobile worker Work in local area* Work elsewhere in Wiltshire Work outside Wiltshire Total Urban 1 33.2% 7.2% 22.2% 21.8% 15.6% 51,462 Urban 2 35.8% 7.2% 20.5% 22.9% 13.5% 46,893 Urban 3 24.8% 5.7% 24.5% 24.6% 20.5% 12,620 Rural 1 36.7% 8.8% 12.7% 15.4% 26.5% 16,576 Rural 2 31.5% 15.3% 9.5% 22.3% 21.4% 65,448 County-wide 33.0% 10.0% 16.8% 21.9% 18.3% 193,000 * working in an area which is both in the same Community Area and sub-area Source: Wiltshire household survey, 2011 7.19 The table below presents the median gross annual household income (excluding benefits) for the groups featured in the previous table. This corroborates the last comment above, with the highest average earned incomes recorded for rural households that work elsewhere. The other very high earners are rural households that work from home. Overall the median earned income of households working in their local area is 29,000 compared to 40,800 for those that work further away. Table 7.9 Household income* by typology and employment location (medians) Area All employed households Work from home / mobile Employed in local area** Employed elsewhere Retired Urban 1 34,900 33,800 29,600 38,200 17,900 Urban 2 31,200 32,500 25,500 35,700 17,100 Urban 3 34,900 32,100 33,300 38,500 18,100 Rural 1 37,800 46,700 23,700 40,100 18,200 Rural 2 40,800 32,800 30,600 47,000 22,100 Countywide 35,600 33,000 29,000 40,800 19,100 * gross per household, excluding benefits ** in both same sub-area and same Community Area Source: Wiltshire household survey, 2011 7.20 The final table shows access to facilities. The Rural 2 area shows much higher levels of difficulty (combining quite and very ) in accessing shopping than any other group: nearly 14% finding it difficult. The County s other rural inhabitants show average levels of difficulty with regard to shopping. 7.21 In terms of access to town centres (important for young people in particular) the greatest level of combined difficulty is shown in Urban 3 (Salisbury Plan). The two rural areas not surprisingly show lower levels of easy access than the two more urban sets of areas. 94

Area Table 7.10 Access to shopping facilities by typology Grocery shopping Town centre Easy Quite Difficult Very Difficult Easy Quite Difficult Very Difficult Urban 1 94.0% 4.6% 1.4% 92.7% 5.5% 1.7% Urban 2 95.0% 4.1% 0.8% 92.6% 6.6% 0.8% Urban 3 92.5% 6.9% 0.6% 78.6% 17.8% 3.6% Rural 1 93.6% 5.1% 1.3% 85.2% 12.3% 2.6% Rural 2 86.3% 11.1% 2.6% 80.1% 15.5% 4.3% Countywide 91.5% 6.9% 1.6% 86.9% 10.6% 2.6% Source: Wiltshire household survey, 2011 95

8. Affordable Rent: scope in Wiltshire Introduction 8.1 The Localism Bill is introducing Flexible Tenancies as a new tenure within the affordable sector. Flexible Tenancies will be significantly different to the currently used social rented tenure. Flexible tenancies will not give the tenant security of tenure for life and will allow Affordable Rent to be charged. Their introduction is likely to have an affect on the wider housing market and a significant impact on the affordable sector. This chapter aims to provide the information necessary for the Council to ensure their introduction is as successful as possible. 8.2 The most important issue for the Council to determine is the level at which Affordable Rent should be set. This chapter will consider this from two perspectives; firstly at what level does Affordable Rent best assist households requiring affordable accommodation and secondly at what level does the product provide best value for money and facilitate the largest production of further affordable homes. 8.3 The chapter will begin by setting out the policy context surrounding Affordable Rent and it s impact on social rent. It then moves on to look at affordable rent in Wiltshire and calculate it s potential cost. This is followed by an analysis of the affordability of these various options for households likely to move to this type of housing. 8.4 The chapter then goes on to discuss the anticipated changes to the funding of affordable homes nationally, before providing an introduction to development economics. The chapter then considers the value of a range of Affordable Rent products, which helps determine the size of subsidy they would require to be built. Finally the chapter concludes at what level the Council should set Affordable Rent to best balance these conflicting pressures. 97

Affordable Rent: a new tenure 8.5 In 2011-15 Affordable Homes Programme Framework published by the Homes and Communities Agency in 2011, Grant Shapps, the Minister said: [T]he money must go further. So we are introducing new flexibilities for providers on using existing assets, and a new offer on rents. The objective of these flexibilities, including the new Affordable Rent product, is to enable providers to deliver up to 150,000 new affordable homes (Ministerial Foreword). 8.6 This statement is amplified in various ways in the Framework document. The key facts are as follows: Affordable Rents can be set at up to 80% of open market rental value It is formally defined as a social tenure (paragraph 3.20) It is intended to be made available to those on the Housing Register, though it will be attractive to some of those on the Register who are on higher incomes and may not be in formally defined housing need Flexible Tenancies must not be less than two years. 8.7 Paragraph 3.24 says that accommodation which becomes Affordable Rent must be permanently available for letting. This does not prevent staircasing up to ownership but does prevent switching to the social rent structure in the future. 8.8 The Framework also indicates that the same procedures for allocating social rented homes should be used for Affordable Rent: 3.20 Allocations and nominations processes for Affordable Rent homes are expected to mirror the existing frameworks for social rented housing. Providers will be under the same statutory and regulatory obligations when allocating Affordable Rent homes as they are when allocating properties for social rent. 8.9 The most interesting point is the up to 80% of OMV (Open Market Value). The conditions under which an Affordable Rent below 80% may be set are addressed in the following two paragraphs quoted from the Framework document. 3.10 While offers which include Affordable Rent for new supply and/or conversions at less than 80% of market rents will be considered, it is expected that providers utilise the flexibility to charge rents of up to 80% of market rents to maximise financial capacity. The HCA would need to understand how any proposal to charge lower rents would help to meet particular housing needs, deliver value for money for the taxpayer and generate the capacity required to deliver new supply aspirations. 98

3.11 There may be specific circumstances where it is appropriate to set rents at less than 80% of market rents. For example, providers may wish to charge a lower rent where a rent at 80% of market rent would exceed or be close to the relevant Local Housing Allowance (LHA) cap, or if the local rented market was considered to be particularly weak or fragile (for example on an existing estate where there may be few market rented properties). The HCA would wish to explore with providers the rationale for considering rents at less than 80% of market rents. In all cases, an Affordable Rent should be no lower than the rent calculated based on the current target rent regime. In cases where an Affordable Rent would otherwise be lower than the target rent for a property, the target rent will constitute a floor for the rent to be charged. 8.10 Paragraph 3.10 emphasises the need to maximise the amount of affordable housing created for each pound of money spent: it is expected that providers utilise the flexibility to charge rents of up to 80% of market rents to maximise financial capacity 8.11 Taken by itself, this would suggest that there is in fact very little flexibility: the HCA wants to see rents at 80% because that way there will be more money than if they are set below 80% 8.12 But paragraph 3.11 softens this message: There may be special circumstances where it is appropriate to set rents at less than 80% of market rents.the HCA would wish to explore with providers the rationale for considering rents at less than 80% of market rents 8.13 The tone of this statement is still directed to 80% as the default position, but does entertain the argument that there may be evidence to justify rents lower than 80%. This statement by the HCA can be set beside many statements by Ministers to the effect that up to is a very important part of the message. For example the statement by Grant Shapps (below): There s another couple of billion plus for the new Affordable Rent, which is the entirely new scheme of up to 80%--- not at 80%, which has often been inferred, wrongly of the market rent [Response to Q47 of Oral Evidence from Ministers to the Communities and Local Government Committee on the Comprehensive Spending Review 21 st December 2010] 8.14 The key point in this case is his underlining of the up to 80%. It seems clear that in the wider political agenda it is important that this downward flexibility should exist. At the same time, the HCA has suffered a considerable cut in the funding it had hoped for, and clearly wants to make as much affordable housing as possible from that money is available. The difficulty in striking this balance will be discussed in the second half of this chapter. 99

The impact on social rents 8.15 It is made clear in the Framework that existing social rented units can be relet as Affordable Rent when they become vacant. Indeed this will be a major source of new Affordable Rent units. 1.3 Affordable Rent will form the principal element of the new supply offer. At the same time, new flexibilities will allow a proportion of social rent properties to be made available at re-let at an Affordable Rent, with the additional capacity generated from those re-lets applied to support delivery of new supply. 2.11 In the new model there will be four broad funding streams which contribute to the development of new supply: i) the additional borrowing capacity that can be generated from the conversion of social rent properties to Affordable Rent (or other tenures) at re-let, as well as borrowing capacity generated by the net rental income stream of the new properties developed; 2.17 Providers are invited to consider offering conversion to Affordable Rent of existing committed social rent schemes begun under the 2008-11 National Affordable Housing Programme (NAHP) which will achieve practical completion in the new programme period. We encourage such proposals as a way of generating additional financial capacity for new development in a way that can offer certainty of volume, rent, location and timing. This will not lead to recovery of existing funding, but will generate additional financial capacity for the delivery of new supply. 8.16 The analysis in this chapter does not address the question of the level of conversion of relets of social rented units, since it is mainly focussed upon the financial implications of this tenure, and the degree to which it can assist in meeting housing need. The chapter is not devoted to the source of the Affordable Rented units. Local Housing Allowance 8.17 Paragraph 5.58 of this report details the changes to Local Housing Allowance (LHA) availability that have been introduced. There is an interaction between LHA availability and Affordable Rent: at the most basic level there is no point in setting an Affordable Rent when nobody could afford it either because they did not have the income or could not obtain sufficient LHA to pay the rent due to the new caps. The following extract from the Framework expands on that: 3.7 The Tennant Services Authority is therefore not proposing to restrict the maximum rent that Registered Providers can charge for Affordable Rent properties based on the Local Housing Allowance. However, landlords will wish to consider the local market context when setting rents, including the relevant Local Housing Allowance for the Broad Rental Market Area in which the property is located. They should also take into account wider benefit policy such as the proposal, subject to the passage of the Welfare Reform Bill, to cap total 100

household benefit payments so that workless families do not receive more in welfare than the median earned income after tax and National Insurance contribution earnings of working families. War widows and households with a member entitled to Disability Living Allowance, Constant Attendance Allowance or Working Tax Credit will be exempt from the cap. 8.18 Although therefore the Tennant Services Authority (now part of the Homes and Communities Agency), will not restrict the Affordable Rent to the LHA, it expects Registered Providers to take account of the wider situation in setting those rents. Understanding Affordable Rent 8.19 Affordable Rent will be based on the open market value of each property. For all affordable properties, the market price will be inferred from values in the private rented sector (PRS). Therefore the PRS needs to be understood. This section outlines how the nature of the PRS can impact how competitive Affordable Rent levels are likely to be. 8.20 The spread of rents in the PRS will have implications for the rates of Affordable Rents, as is illustrated by the figure below. It shows the spread of rents in two hypothetical housing markets for the same property size. The short black vertical lines show the key points on the distribution (minimum, lower quartile, median, upper quartile and maximum) and the green boxes show the inter-quartile range or the middle market. The scale shows the median rent level as 100% and the variation of rents can be seen as a proportional difference to the median; the typical Affordable Rent line is therefore at 80% (the red dashed line). The yellow dashed line indicates the LHA cap (the 30 th percentile of rents available in the market) Figure 8.1 The impact of the open market on Affordable Rent levels Source: Fordham Research (2011) 8.21 In a dense market the majority of the rents available will lie within a small range. In this market the Affordable Rent level is likely to be well below the majority of open market rents and offer an appealing alternative to the private sector. Although Affordable Rent rates vary by the value of each property, on the whole property value does not vary much in the dense market and Affordable Rent should be 101

consistently lower than private sector alternative. The Affordable Rent level should also be lower than the LHA cap. 8.22 In a broad market the range of rents available, even in the middle of the market, is very wide. In this case if a median or higher value property was let at an Affordable Rent there would still be a substantial proportion of properties in the open market available at a cheaper rate (albeit, lower quality properties). In addition the Affordable Rent rate could also be similar to or even above the LHA level. For Affordable Rent properties to offer a cheaper alternative to the open market homes in this broad market, lower value open market properties would be the viable option for converting to Affordable Rent. 8.23 In reality, most PRS have a dense market with a broad upper-middle market. In these markets, as long as non-luxury properties (those in the bottom 75%) are converted to Affordable Rent, the rates offered would be cheaper than the majority of the open market alternatives. However, in some cases where there is a broad lower-middle market, median or higher value properties would not be a viable Affordable Rent option. It is an important point as, on the occasion that the typical Affordable Rent level is more expensive than entry-level private rents (the lower quartile), priority can be given to ensuring that the kind of properties that are converted to Affordable Rent are appropriate for the area. Understanding the private rented sector in Wiltshire 8.24 The section considers the breadth of the private rented market for each property size in Wiltshire. The table below shows the cost at the key points of the rental distribution, as recorded during the price survey described in chapter 4. This analysis considers the rural and urban price markets in Wiltshire separately 8.25 It can be seen from the figures in the table below that each property size operates in largely distinct markets as there are big differences between the median rents for each property size and overlap within the inter-quartile ranges is minimal. The extremes of each market overlap somewhat with the next size of dwelling. For example, somebody in a high quality one-bed dwelling could live in a median priced two-bed property at the same rent but they would have to accept a noticeable drop in quality. 102

Table 8.1 Private sector rent level in Wiltshire (cost per month) Urban House size One bed Two bed Three bed Four bed Minimum 320 470 600 750 Lower Quartile 395 525 675 850 Median 450 575 725 950 Upper Quartile 500 675 795 1,175 Maximum 625 775 899 1,355 Inter-quartile range 105 150 120 325 % difference between quartiles 26.6% 28.6% 17.8% 38.2% Rural House size One bed Two bed Three bed Four bed Minimum 350 509 650 807 Lower Quartile 450 575 745 995 Median 498 663 850 1,300 Upper Quartile 544 750 1,000 1,700 Maximum 649 995 1,740 3,500 Inter-quartile range 94 175 255 705 % difference between quartiles 20.8% 30.4% 34.2% 70.9% Source: Estate Agent survey (March 2011) Estimating the Affordable Rent 8.26 For the purposes of policy analysis it is important to model the likely levels of Affordable Rent, and their impact on local households. This allows a policy judgement as to what level they should be set. But to do that requires some approximation: we cannot analyse every PRS dwelling in Wiltshire. We have considered various forms of averaging to derive a median market rent, from which the Affordable Rent at 80% could be calculated. The most effective, we believe, is to take the median from the middle range of observed rents. This avoids both the worst PRS dwellings and the luxury ones. We believe that the middle market gives the best indication of the true price of the sector. That is what has been used as the basis for the Affordable Rents set out below. (Affordable Rent levels in each price area of Wiltshire are presented in Appendix 1). 103

Table 8.2 Affordable Rent ranges for Wiltshire (cost per month) Urban House size One bed Two bed Three bed Four bed Minimum 320 420 540 680 Median 360 460 580 760 Maximum 400 520 640 880 Range 80 100 100 200 Rural House size One bed Two bed Three bed Four bed Minimum 360 460 600 800 Median 400 530 680 1,040 Maximum 440 600 800 1,200 Range 80 140 200 400 Source: Estate Agent survey (March 2011) 8.27 Two trends of Affordable Rent levels in Wiltshire can be seen in the table above: As bedroom size increases the range of possible Affordable Rents increases. In the rural areas the costs of Affordable Rent will be higher than the urban equivalent and the range of possible Affordable Rents is greater in rural areas. Affordable Rents compared to open market rents 8.28 The table below compares the observed ranges of rent in the PRS with the Affordable Rents shown in previous section for comparison. Social rent and LHA levels are also included. The tables show that social rent levels in Wiltshire are consistently below the entire range of rates for Affordable Rent products and the gap between social rent and Affordable Rent increases with property size. 8.29 Four LHA levels have been listed in the tables below, the first is the 30 th percentile of the private rented sector 5, according to the estate agent survey conducted for this report and the remaining three are the actual LHA caps for the three main BRMAs in Wiltshire 6. In terms of comparing the estate agent data with the actual caps the tables below show that the LHA caps in the Salisbury BRMA tend to be higher than the 30 th percentile rents and in Swindon BRMA and West Wiltshire BRMA the caps tend to be lower. However, this is not surprising, as the estate agent data has been split by rural/urban and not BRMA therefore it is likely to be skewed upwards due to the higher costs in Salisbury. Nonetheless, LHA caps do not take into account the higher market rents in rural areas within a BRMA and therefore households requiring LHA are more likely to reach the cap if they live in a rural area. 5 The Coalition reduced the LHA cap to this from the 50 th percentile in April 2011 (paragraph 5.58 provide further explanation). 6 These are set by the Valuation Office (paragraph 5.60 provides further explanation). 104

8.30 In most markets the LHA rates are above the median and often the maximum Affordable Rent level. However, the LHA cap is below the maximum and median rent for three and four bedroom properties in the rural area. In these markets, if the intended households for Affordable Rent homes require the entire rent to be covered by LHA, the properties available should be drawn from the lower end of the market, in which case the rent level would be below the LHA cap. 8.31 In some cases there is an overlap between the maximum Affordable Rent rate and the entry level private rent. If, in these markets, high end properties were made available as Affordable Rent products, they would offer the chance for households to move into a high quality property at below open-market rents however, there would still be suitable cheaper properties available in the open market. 8.32 In terms of providing an Affordable Rent product that is above the social rent level but suitably below the entry-level market rent, the tables suggest that the most suitable properties to be made available for Affordable Rent would be ones equivalent to those in the lower-middle section of the open market. Table 8.3 Rent levels by tenure urban (cost per month) House size One bed Two bed Three bed Four bed PRS Upper Quartile 500 675 795 1,175 Median 450 575 725 950 Lower Quartile 395 525 675 850 Affordable Rent Maximum 400 520 640 880 Median 360 460 580 760 Minimum 320 420 540 680 Social rent Typical rent* 312 355 416 416 LHA cap 30 th percentile (Fordham survey) 400 550 690 880 Salisbury BRMA** 500 625 750 950 Swindon BRMA** 430 525 650 850 West Wiltshire BRMA** 425 525 625 850 Source: Estate Agent survey (March 2011) *Source: Core 2010 **Source: Direct Gov 2011 105

Table 8.4 Rent levels by tenure rural (cost per month) House size One bed Two bed Three bed Four bed PRS Upper Quartile 544 750 1,000 1,700 Median 498 663 850 1,300 Lower Quartile 450 575 745 995 Affordable Rent Maximum 440 600 800 1,200 Median 400 530 680 1,040 Minimum 360 460 600 800 Social rent Typical rent* 312 355 416 416 LHA cap 30 th percentile (Fordham survey) 450 600 750 1,100 Salisbury BRMA** 500 625 750 950 Swindon BRMA** 430 525 650 850 West Wiltshire BRMA** 425 525 625 850 Source: Estate Agent survey (March 2011) *Source: Core 2010 **Source: Direct Gov 2011 Affordability of Affordable Rent 8.33 Affordable Rent is designed to reduce Housing Registers, and to accommodate both households in priority need, and others who may not be in priority need but who have the income to afford the new tenure. We therefore analyse both households on the Register and households in need, the latter to provide context. 8.34 In addition to testing Affordable Rent based on 80% of the market median (set out above) we have also looked at Affordable rent based on 70% and 60% of the market median (which are calculated using the same approach). In carrying out the affordability assessment we have used the standard 25% of gross income on housing test, rather than a higher one. This is because, for households on low incomes, as those in housing need mainly are, anything much higher than 25% of income on housing leaves very little to live upon. Affordability of Affordable Rent for households in housing need 8.35 The table below, a replica of table 5.11, illustrates how many households in defined housing need are able to afford different kinds of affordable housing products. It shows that of the 1,773 households in need that are able to afford Affordable Rent at 80%, 85.5% of them could also afford a Shared Ownership property. However if the Affordable Rent level were lowered to 60% of private rent values then the number of households in need able to afford Affordable Rent, but not Shared Ownership, would be 1,885 households per year. 106

8.36 The largest group of households in need are those unable to afford any accommodation without support from LHA. The analysis above shows that the LHA cap should be above the expected Affordable Rent levels in the majority of cases. Therefore households unable to afford could be housed in Affordable Rent properties at 80% with the support of LHA. Table 8.5 Affordability of households in need (annual) Households in need % of households in need Shared Ownership (75%) 574 9.2% Shared Ownership (50%) 942 15.1% Affordable Rent at 80% 257 4.1% Affordable Rent at 70% 704 11.3% Affordable Rent at 60% 924 14.8% Social rent 115 1.8% Need LHA 2,732 43.7% Total number of households 6,250 100.0% Source: Wiltshire household survey, 2011 8.37 The table below splits the figures shown in the table above by bedroom size. It shows that Affordable Rent at 80% would be most suitable for households in need of three bedroom accommodation over a fifth of these households (128 households) would be able to afford Affordable Rent at 80% but not Shared Ownership properties. Affordable Rent at 60% would be most suitable for households requiring the other property sizes. Table 8.6 Breakdown the proportion of size and type of Affordable Rent home required by those in need (households) One bed Two bed Three bed Four bed Shared Ownership (75%) 8.9% 13.5% 2.8% 0.0% Shared Ownership (50%) 21.8% 13.5% 1.8% 13.0% Affordable Rent at 80% 1.4% 0.0% 21.2% 1.8% Affordable Rent at 70% 7.3% 17.0% 10.4% 4.6% Affordable Rent at 60% 9.7% 18.8% 16.9% 21.8% Social rent 0.0% 1.7% 0.0% 26.6% Need LHA 50.9% 35.4% 46.9% 32.2% Total number of households 2,433 (100%) 1,989 (100%) 603 (100%) 52 (100%) Source: Wiltshire household survey, 2011 107

Households on the Register, including those in need 8.38 This sub-section repeats the above analysis but this time considers households on the Housing Register, which are not necessarily households in need. In Wiltshire around half of the households in need are on the Housing Register and around 15% of households on the Housing Register are in housing need. 8.39 The table below shows that around a quarter of households on the Housing Register are able to afford suitable accommodation in the open market. It also shows that if Affordable Rent is set at the level of 60% the need for social rent is minimal. In terms of Affordable Rent at 80%, it appears that if suitable Shared Ownership properties were available, this product would only be appropriate for 0.8% of households on the Register. Table 8.7 Affordability of households on Housing Register Households on Register % of households on Register Market housing 3,446 25.0% Shared Ownership (75%) 710 5.1% Shared Ownership (50%) 1,092 7.9% Affordable Rent at 80% 105 0.8% Affordable Rent at 70% 1,045 7.6% Affordable Rent at 60% 1,980 14.3% Social rent 121 0.9% Need LHA 5,301 38.4% Total number of households 13,800 100.0% Source: Wiltshire household survey, 2011 8.40 The table below splits the figures in the table above by bedroom size. In terms of households of the Housing Register, Affordable Rent at the 80% level is only able to meet need a small proportion of the need for one bedroom accommodation and none of the need for two to four bedroom dwellings. For Affordable Rent to actually to be affordable to a significant proportion of households on the Housing Register it would need to be reduced to a level of 70% for three bedroom accommodation and 60% for one, two and four bedroom accommodation. 8.41 For households on the Housing Register, as with those in need, it appears that the majority of households lie at either end of the affordability scale i.e. they are either able to afford market/shared Ownership housing or they are LHA dependent. 108

Table 8.8 Breakdown the proportion of size and type of Affordable Rent home required by those on the Housing Register (households) One bed Two bed Three bed Four bed Market housing 33.0% 13.5% 18.1% 23.8% Shared Ownership (75%) 3.8% 7.8% 5.6% 0.0% Shared Ownership (50%) 8.6% 9.4% 1.6% 8.8% Affordable Rent at 80% 1.4% 0.0% 0.0% 0.0% Affordable Rent at 70% 4.8% 8.5% 18.2% 0.0% Affordable Rent at 60% 12.4% 17.1% 14.6% 23.2% Social rent 0.0% 0.2% 0.0% 35.1% Need LHA 36.0% 43.5% 42.0% 9.1% Total number of households 7,518 (100%) 4,128 (100%) 1,832 (100%) 322 (100%) Source: Wiltshire household survey, 2011 Supply of affordable housing to meet Affordable Rent 8.42 Table 5.13 suggests that the future annual supply of affordable dwellings will consist of 97 shared ownership homes, 2,282 social rented properties and 9 Affordable rent units. If all social rented re-lets were converted to Affordable Rent the supply of this tenure could be 2,291 units per year. The registered housing providers in Wiltshire are currently considering converting 50% of all social rented relets to an Affordable Rent tenure. This would lead to a potential supply of 1,150 homes per year. What is the need for Affordable Rent? 8.43 The analysis presented above shows that very few households on the Register could afford Affordable Rent at 80%, that could not also afford a shared ownership property. If the Affordable Rent were set at 60% the potential demand for it (excluding those able to afford shared ownership) from households on the Register is 3,130. This suggests that if Affordable Rent were set as low as 60% there would be much more demand for it from those on the Register than the supply could allow to be housed. 8.44 If the Affordable Rent level were set at 70%, it would generate a potential demand from 1,150 households. This is exactly the same as the potential supply (should half of social rented re-lets be converted). Using these figures, the most sensible rate at which to set Affordable Rent is: 70%. If this fails to generate sufficient demand from the Register the Council would have to consider lowering the rate to 60%. Generating new affordable dwellings vs. meeting need 8.45 One of the key aims of the Coalition Government s policy on affordable housing is to make the Homes and Communities Agency s (HCA) much reduced budget go further (the HCA s budget has declined from over 8.4 billion to less than 2.5 billion). The hope and objective of Affordable Rents is that by 109

charging higher rents the affordable housing developers would require less grant and subsidy and thus the development of affordable housing would therefore fund its self. The theory being that if the developer could charge a higher rent then it can borrow more money to finance the construction of more affordable units. Thus Affordable Rent is expected to yield almost 1.5 times the number of new affordable units as compared with social renting. 8.46 Flexible Tenancies will be able to be granted for more than just newbuild properties. Some of the relets will able to be at Affordable Rents rather than social rents - the extra income from the relets going to fund further development. 8.47 However there is a trade-off. Although it is desirable, in order to meet housing need, that the Affordable Rent level should be set below 80%, this means that the rent flow available to the housing association or developer to secure additional borrowing or funding is shrunk. This second half of this chapter will therefore consider some of the financial issues behind development and look at how high the Affordable Rent actually has to be to make up the shortfall in grant due to the cutting of the HCA budget. 8.48 These changes in the HCA s budget have caused considerable uncertainty within the affordable housing sector. The lack of grant and the fact that there is little prospect of grant being available in significant amounts for the foreseeable future means that some Housing Associations are now considering Stock Market listings. Traditionally the development of affordable housing has been dominated by not-for-profit organisations and grant has been restricted to them. The HCA is now encouraging a greater diversity of providers including conventional for profit developers. Current Grant Funding 8.49 For many years the HCA and Local Planning Authorities (LPAs) have aspired to ensure that affordable housing is delivered without grant. When LPAs have negotiated with developers during the planning process about the number and type of affordable housing to be provided through the section 106 agreement and planning conditions, the initial basis of those discussions has been that the affordable units would be made available without any grant. 8.50 The reality has been rather different with the developer either transferring the serviced land for affordable housing to the Private Registered Provider (RP) for no cost or the RP purchasing the completed units from the developer with grant assistance from the HCA. 8.51 The amount of grant paid by the HCA has been assessed project by project depending on that site s financial characteristics. Typically the grant outside London is about 35,000 per unit in urban areas and around 42,000 in the rural areas, where the costs of development tend to be higher. 8.52 The aim of the extra income from Affordable Rents is to replace this grant. The RP will be able to service new borrowings to make up the gap in grant. The funding available to subsidise Affordable 110

Housing is going to be massively reduced and restricted to very few schemes Affordable Rents are the only show in town for now, so Councils have to work within this new framework. Development Economics of Affordable Rents 8.53 The economics of development is very simple. Either the income from a development exceeds the costs and a profit is generated or the costs exceed the income and a loss arises. The equations is: Gross Development Value (The combined value of the complete development) LESS Cost of creating the asset, including a profit margin (land + construction + fees + finance charges) = Profit or Loss 8.54 In the development of affordable housing for rent the value of the development is the amount of the income that the completed let unit will produce. This is the amount an investor or other RP would pay for the completed unit. This will depend on the amount of the rent, the cost on managing the property (letting, rent collection, repairs etc) and other uses to which it may be able to be put to at some time in the future. If, for example, the unit could be sold on the open market in the future then a buyer may be willing to pay more to take into account the long term value (know by valuers as the reversion). 8.55 As noted at the start of this chapter, the HCA has published 2011-15 Affordable Homes Programme Framework. This contains the rules and guidance around Flexible Tenancies and Affordable Rents. It is quite a brief document so some areas are not perfectly clear. It says 3.24 There will be a presumption that new Affordable Rent properties which receive funding under the new programme will be permanently available for letting. Flexible tenancies have been introduced to meet the differing needs of prospective tenants but the homes themselves are expected to be available to meet need over the long-term, and it is on that basis that funding will be made available. We recognise that circumstances may change over time and any future disposal of properties will require TSA consent in the usual way, including consultation with the relevant local authority. 8.56 Based on this we know that the reversionary period will not command a premium price as the new property can only be used for Affordable Rent (this only appears to apply to new properties and not relets) so the worth of the unit is that of the income it produces i.e. the rent. 8.57 What is the rental stream worth either to the RP or to somebody else? There are two aspects to this. i) How much additional borrowing the extra income from the Affordable Rent will support. 111

ii) What is a unit let on Affordable Rent actually worth 8.58 This figure obviously depends on the level at which Affordable Rent is set. 8.59 The way that RPs raise their money for building housing needs to be considered. RPs operate in the commercial borrowing markets. They have two main sources of finance; they can issue a bond or they borrow conventionally. 8.60 When an organisation issues a bond it borrows a large amount of money for a fixed period of time and at a fixed interest rate in the money markets. Housing Associations that are RPs are registered with the HCA and one has never been allowed to fail they are in effect guaranteed by the State (although their borrowings do not appear on the national balance sheet). That is to say, if they were to fail the Government would step in and make good the losses. The bond is normally secured on an income stream from a portfolio of Social Rented properties or similar. Traditionally RPs have been well regarded in the financial markets and have been able borrow relatively inexpensively. 8.61 Only the larger Housing Associations are able to issue bonds (due to the cost associated with the process) although there is a movement towards smaller organisations clubbing together to issuing a joint bond. 8.62 The conventional borrowing route is simpler and works in the same way as for any commercial organisation trying to borrow money. The financial institution will consider the merits and risks of the project, the income and the security on offer. It will then set the terms to reflect that. 8.63 In both bond and the conventional borrowing the actual interest rate will depend on the prevailing interest rates in the capital markets, base rates and the individual circumstances. Typically interest rates are between 5% and 7.5%. Assuming a 5.5% interest rate, an additional rent of a little under 65 per week is required to compensate for borrowing 60,000, whilst an additional rent of about 37.50 per week is needed to compensate for borrowing 35,000. To set this in context the social rent on the property would probably be about 96 per week. 8.64 Clearly not all this additional rent needs to come from the home being developed. Some could come from the additional rent generated by relets. If the additional rent from say two relets and one newly developed home were combined then only a third of the weekly increase would be required. Each developer will have to consider its own situation and the locality (i.e. the need and demand for housing and different rental levels) when deciding how to proceed with developments. Value of Affordable Rented Homes 8.65 Regardless of how much additional borrowing an Affordable Rented property is able to fund in terms of interest payments it is unlikely that any financial institution would lend more than the property in question is actually worth. But what is the worth? As no Flexible Tenancies have yet been granted 112

and no Affordable Rent charged there is no established market. It is therefore not possible to say with any certainty how the market will approach the valuation of homes subject to this new tenure. 8.66 The value can however been considered in the wider context. There are a number of factors that will affect the value. The amount of rent, the costs of managing the home and the returns sought by investors and property owners in the wider market. 8.67 The initial rent is important but so is the way that is changes over time. The HCA s 2011-15 Affordable Homes Programme Framework says: 3.9 In order to provide protection and certainty for tenants, providers and funders, the maximum annual rent increase on an Affordable Rent property will be Retail Price Index (RPI) + 0.5%. RPI will be taken as at September of the previous year. Additionally there will be a requirement to rebase the rent on each occasion that a new Affordable Rent tenancy is issued (or renewed) for that property to ensure that the rent remains at no more than 80% of market rent (inclusive of service charges) this requirement overrides the RPI + 0.5% limit. 8.68 At first sight this appears attractive to the property owner as the rent will always increase a little faster than inflation. This means that even if rents in the private rental market are going down the Affordable Rents will go up. The exception is when the home is relet when the rent will return to a maximum of 80% of the Open Market Rent. 8.69 It is likely that the market will consider Affordable Rents to be a product somewhere between social rent and Open Market Rent. When considering the worth of assets producing an income, valuers commonly use Years Purchase (YP) to capitalise the rents i.e. to work out the capital value of an income stream. The YP is the factor by which the rent is multiplied to calculate the capital value (calculated at 1/yield). The lower the yield the higher the value as the income is considered to be of higher quality (and lower risk) so an investor is willing to pay more for less income. It is likely that the closer the rent is to a social rent then the lower the yield (so higher the value) will be and the higher the rent the higher the yield (so lower the value). It will be interesting to see how this is actually treated by the market but there is certainly a logic to this approach. The following table shows how values may vary as rents vary and impact on the YP. 113

Table 8.9 Variation in the worth of different income levels in Wiltshire (3 Bedrooms) Open Market Rent Affordable Rent Affordable Rent Affordable Rent Affordable Rent Affordable Rent Social rent 80% 70% 65% 60% 50% Open Market Rent 760 875 Affordable Rent 700 613 569 525 438 Social rent 461 Annual Income 9,120 8,400 7,350 6,825 6,300 5,250 5,532 Yield 8.00% 7.00% 6.50% 6.50% 6.25% 6.00% 5.50% Years Purchase 12.50 14.29 15.38 15.38 16.00 16.67 18.18 Capitalised Rent 114,000 120,000 113,077 105,000 100,800 87,500 100,582 Source: Wiltshire household survey 2011 8.70 In calculating the above, the Open Market Rent used is the median rent across the County of 760 per month. The Affordable Rent is calculated based on an Open Marker Rent of 875 per month to reflect the higher specification and space standards required by RPs. Clearly the setting of Affordable Rents is going to be sensitive not only to meeting the need for affordable housing but also relative to its value. 8.71 The table above shows the current market and social rents sought in Wiltshire with a range of alternative Affordable Rents also presented. The cost of developing one of these Affordable Rent units is about 117,000 assuming a land cost of 10,000 per unit and a 17% developers profit (on costs). It should be noted the land cost would be substantially more for a private rented property (and higher still for an owner-occupied home). 8.72 It is possible to compare the capitalised value of a home in a particular tenure (the bottom row in the table above for rented tenures and the purchase price for ownership models) to the cost of developing that type of home. The difference is the development surplus, which is presented in the figure below. 114

Figure 8.2 Development Surplus by Tenure (3 Bedroom) 50,000 44,624 40,000 32,577 30,000 20,000 19,984 10,000 7,392-41,376 2,947-3,976-7,853-16,253-29,553-26,289 0-10,000-20,000-30,000 Freehold Purchase Shared Ownership 60% Shared Ownership 50% Shared Ownership 40% Open Market Rent Affordable Rent 80% Affordable Rent 70% Affordable Rent 65% Affordable Rent 60% Affordable Rent 50% Social Rent - 40,000-50,000 Source: Wiltshire household survey 2011 8.73 The figure shows that freehold purchase is clearly profitable, but building for private rent is not (as evidenced by the lack of Build to Rent happening in the market). Only building Affordable Rent at 80% of Open Market Rent covers the costs of development although developing it at 60% to 70% requires a substantially lower subsidy than developing social rent under the current funding models. 8.74 Using three bedroom units as the example, the levels of subsidy required for the various Affordable Rent levels considered are shown the table below. Table 8.10 Subsidy required for a new 3 bedroom affordable unit at various levels of rent Tenure status Affordable Rent at 80% OMR Level of subsidy None Affordable Rent at 70% OMR 3,976 Affordable Rent at 65% OMR 7,853 Affordable Rent at 60% OMR 16,253 Affordable Rent at 50% OMR 29,553 Social rent 26,289 Source: Wiltshire household survey 2011 The use of relets 8.75 The above figures assume that the production of a new Affordable Rented home is a stand alone project without subsidy from outside. The HCA is to permit relets of existing social rented units to be let under Flexible Tenancies and at Affordable Rents so long at the income from the relets over and above the existing social rents is used to finance new development. 115

8.76 The table below looks at how many existing social rents need to be converted to Affordable Rents to make up the development shortfall. The table indicates for example that one additional social rented property being relet at 70% Affordable Rent, would provide an additional 22,313 capitalised rent. This figure is enough to offset the development shortfall from five new 70% Affordable Rent properties being built. Alternatively one relet at 65% Affordable Rent could provide sufficient additional funding to enable two new 65% Affordable Rents homes to be developed. Table 8.11 Alternative Newbuild / Re Let Mixes New Affordable Rent Units Social rent Relets at Affordable Rent Balance Number % OMR Surplus / Deficit Number % OMR Capitalised Additional Rent 1 80% 2,947 0 80% 0 2,947 5 70% -19,879 1 70% 22,313 2,435 2 65% -15,705 1 65% 18,436 2,731 1 60% -16,253 2 60% 20,073 3,820 1 50% -29,553 0 50% 0-29,553 NA: Read tables across rows Source: Wiltshire household survey 2011 8.77 In practice this calculation will need to be carried out on a project by project basis as the rents on the newbuild and relets will vary from property to property. The above table assumes that the open market rental value of an old social rented unit is equal to that of a new Affordable Rent this is unlikely to be the case. However the above does illustrate the difficult decisions that the Council will have to make when deciding on future housing mixes and types. What level should the Affordable Rent be set at? 8.78 It is clear that Affordable Rent newbuild requires a subsidy if it is to be set below 80%. The first part of the chapter showed that it must be set at 70% or even 60% if it is to draw the requisite numbers from the Register. 8.79 The way in which this can be funded is by converting social relets into Affordable Rent relets. The question of how much of this should be done is a political one, upon which we cannot comment. What is clear, however, is that from a funding point of view an Affordable Rent level down to 60% could be afforded with suitable conversion of social relets to Affordable Rent. 116

9. Improving market balance over the longer term Introduction 9.1 The previous chapters in this report have focused on current market pressures, this chapter considers what accommodation is required to provide housing market balance over the long-term. This is an important exercise because there is a lag in the planning system, which means that it is not possible to respond immediately to imbalances between the nature of accommodation required and the stock currently available. It is therefore appropriate to consider the intervention required to the housing stock over the long-term to enable future action to be planned effectively. 9.2 Although there is not a housing market model in the Practice Guidance, there is comment on the importance of studying mix and balance. This is summarised below before carrying out the analysis. The following extract from paragraph 20 of PPS3 addresses the issue of mixed communities: Key characteristics of a mixed community are a variety of housing, particularly in terms of tenure and price and a mix of different households such as families with children, single person households and older people. 117

9.3 The SHMA Practice Guidance (August 2007) emphasises, as its second core output, the analysis of balance as can be seen from the following extract from page 10 (repeated on page 34): Analysis of past and current housing market trends, including balance between supply and demand in different housing sectors and price/affordability. 9.4 This chapter describes a model that uses secondary data in combination with the household survey dataset to compare the current housing stock against the stock of housing required in the future. The purpose of this model is to identify the new accommodation required to adequately house the future population in the County and ensure that the housing market is balanced. 9.5 This chapter initially presents the ONS 2008-based population projections for Wiltshire and describes the predicted changes in both the population size and composition. The chapter then introduces a model which identifies the adjustment required to provide a sufficient range of accommodation to adequately house each household type and balance the housing stock, culminating in suggested profiles for new housing in terms of type, tenure and dwelling size. Demographic projections 9.6 The 2008-based population projections from ONS indicate that the population within Wiltshire is projected to increase by 10.7% between 2011 and 2026 (an increase of 49,400 people). 9.7 Figure 11.1 shows the projected change within each five-year age cohort between 2011 and 2026. The data suggests that there is likely to be an increase in all age groups, with the exception of 40-54 year olds, those under five and those aged between 15 and 19. The number of people aged 60 and over is expected to experience the largest increase, with the number of people aged 90 and over expected to increase by 115.6%. These changes may have implications in terms of the proportion of economically active people living in the County. 118

Figure 9.1 Forecast population change by age group in Wiltshire, 2011-2026 %age population change 2006-2026 140% 120% 100% 80% 60% 40% 20% 0% -20% 0-4 -0.7% 6.7% 5-9 0.7% 10-14 15-19 -1.0% -8.2% 20-24 3.9% 25-29 15.0% 30-34 9.2% 35-39 -13.8% 40-44 -20.3% 45-49 -4.0% 50-54 27.1% 55-59 18.7% 60-64 24.7% 65-69 35.5% 70-74 67.9% 75-79 62.4% 80-84 60.0% 85-89 115.6% 90+ -40% Source: ONS 2008-based sub-national population projections 9.8 The ONS also published 2008-based household projections alongside the population projections. These show that the number of households within Wiltshire is projected to increase by 17.6% between 2011 and 2026 (an increase of 33,004 households). The fact that households are projected to increase at a faster rate than population suggests a proliferation of smaller households. The 2008- based household projections also provide detail on the likely composition of the future household population in Wiltshire. The projections indicate that single person households are likely to increase by 43.4% and lone parent households by 41.0% in Wiltshire over the next 15 year, whilst other household groups will increase by just 1.8%. 9.9 The ONS population and household projections have been applied to the household survey dataset to provide an estimated household profile for 15 years time. Relating the ONS projections to the Council s work 9.10 Alongside this SHMA work, Wiltshire Council has been developing its Housing Requirements figures. This is detailed in Topic Paper 17: Housing Requirements Technical Paper issued by the Council in June 2011for consultation in relation to the Core Strategy. 9.11 The Technical Paper reviews household growth projections based on different drivers: population led ; economic led and job alignment led. These different projections produce a series of estimates for the number of additional dwellings required in Wiltshire during the plan period 2006-2026. They range from just under 36,000 to nearly 58,000. The population led one projection shows a growth of 43,200 households for the period 2006-2026. 119

9.12 The ONS 2008 based household projection suggests household growth of 33,000 between 2011 and 2026. By adding the total number of dwellings built between 2006-2011 to this figure it is possible to compare the ONS household projections for Wiltshire to the population led projection presented in the Council s Technical Paper. Some 8,314 dwellings were built between 2006-2010 and the projected completions for 2010-2011 are 1,498. Added to the projected growth for 2011-2026, this leads to a total of 43,812 new dwellings over the period 2006-2026. This is very close to the figure of 43,200 set out in the Technical Note. 9.13 The Technical Note reports on a detailed consultation process which has had an important effect on the preparation of the targets for individual communities within Wiltshire (para ES13) and has led to an adoption of the target of 37,000 across the County for the period 2006-2026. Allowing for dwellings already built between 2006-2011, it leaves a total of 27,188 new dwellings required in the 15 year period 2011-2026. This total has therefore been developed in conjunction with local communities (It) will support the projected employment growth and go some way towards addressing outcommuting. 9.14 The resulting 27,188 newbuild for the period 2011-2026, which includes nearly 9,000 dwellings permitted but not yet built, is the result of a policy process that lies outside the SHMA. The SHMA has proceeded on the basis of the 2008 based ONS household projections. These ONS projections contain an assumed greater level of net in-migration than the figure of 27,188 arrived at within the Council s Technical Paper. The extent of in-migration that should be encouraged (as it can t be actually controlled) is a policy decision, which relates to the capacity of individual (often rural) communities to absorb it. Hence the value of the consultation process discussed in Topic Paper 17. 9.15 The figure of 27,188 set for the period 2011-2026 in Topic Paper 17 is therefore some 7,000 lower than the unconstrained ONS household forecast used to drive this model. This should be borne in mind when considering the LTBHM results that follow. However the proportionate results of the LTBHM can be adjusted to the lower target being pursued by the Council. Adequacy of the housing stock 9.16 For the purpose of the LTBHM model, the housing market is considered balanced if the local population is adequately accommodated. It is therefore initially appropriate to assess the adequacy of the current accommodation to house the residents of Wiltshire. This is determined through response to the household survey. 9.17 A household is considered adequately housed currently unless: The household is in unsuitable housing (as defined by CLG Guidance) and cannot resolve this unsuitability without moving to a new property - these households are allocated to a property with characteristics (in terms of type, size and tenure) more suitable for their needs. 120

The households is overcrowded (according to the bedroom standard as which corresponds to Government Guidance) these households are assumed to require a property large enough for overcrowding not to take place. 9.18 In addition, if a household states that they need to move now due to their accommodation s size, cost or the services available within the property being currently unsuitable for the resident household. The size, type and tenure of dwelling these households expect to achieve when they move are presumed to represent the nature of the accommodation that they require. 9.19 The table below shows the proportion of each household type currently requiring alternative accommodation in order to be adequately housed. The table shows that some 7.9% of households are classified as inadequately housed currently. Lone parents are the household group least likely to reside in adequate accommodation, whilst pensioner households are most likely to be adequately housed. Table 9.1 Types of households inadequately housed currently Household type Number inadequately housed All households Proportion inadequately housed Single pensioners 892 28,242 3.2% Two or more pensioners 532 26,242 2.0% Single non-pensioners 2,891 29,233 9.9% More than one adult and no children 4,120 54,700 7.5% Lone parent 1,686 9,758 17.3% Two or more adults, one child 1,965 17,334 11.3% Two or more adults, two or more children 3,108 27,490 11.3% Total 15,194 193,000 7.9% Source: Wiltshire household survey, 2011 9.20 Some further adjustments are also made to use the affordable stock and any housing subsidy paid most economically (this adjustment also allows the introduction of Affordable Rent to be assessed): Households resident in the rented sector on Local Housing Allowance (LHA) that can afford market, Shared Ownership or Affordable Rented accommodation are assumed to require this, to ensure that the stock is being most appropriately and efficiently used. Households in social rented accommodation (non LHA) that can afford market, Shared Ownership or Affordable Rented accommodation are assumed to require this to ensure that the stock is being most appropriately and efficiently used. 121

9.21 All these adjustments inform the ideal accommodation profile for the current population. However, rather than prescribing the accommodation required to address the current mismatch between the current ideal accommodation profile and the current stock, the ideal accommodation profile for each household type is applied to the household population in 15 years time. The model therefore assumes that the pattern of accommodation required by each household type remains constant. Tenure of housing required 9.22 The model is able to provide detail on the tenure of new dwellings required. The table below shows the tenure profile of households resident in Wiltshire currently. The table indicates that almost fourfifths of households are resident in market accommodation (without the aid of LHA), 0.4% live in a Shared Ownership home, 8.7% live in a social rented property (without the aid of LHA) and 13.7% live in rented accommodation with the aid of LHA. Table 9.2 Current tenure profile in Wiltshire Tenure Number of households Percentage of households Market 148,965 77.2% Shared Ownership 825 0.4% Social rented 16,757 8.7% LHA (both private and social rented) 26,454 13.7% Total 193,000 100.0% Source: Wiltshire household survey, 2011 9.23 The tenure of Affordable Rent is to be introduced in the future and the distinction in the affordable sector will be between those able to afford Affordable Rent (or Shared Ownership) and those requiring subsidy for their housing costs (those needing LHA). Taking this into account, the table below shows the ideal tenure profile for the County in 15 years time (if all households are to be adequately housed and the affordable stock is to be used most efficiently). The data shows that in 2026 the housing stock should comprise 73.9% market dwellings, 2.9% Shared Ownership properties, 3.9% Affordable Rented homes and 19.3% dwellings occupied with the support of LHA. Table 9.3 Ideal tenure profile in 2026 in Wiltshire Tenure Number of households Percentage of households Market 167,701 73.9% Shared Ownership 6,642 2.9% Affordable Rent 8,907 3.9% LHA 43,750 19.3% Total 227,000 100.0% Source: Wiltshire household survey, 2011 122

9.24 The table below shows the tenure profile required by households resident in the County in 15 years time in comparison to the tenure profile recorded currently. The difference between these two distributions is the change required to the housing stock over this period. The results show that 55.1% of new housing should be in the market sector, 17.1% should be Shared Ownership properties, 26.2% Affordable Rent and 1.6% new homes for households requiring LHA (its is assumed that the current LHA and social rented stock will also house these households). Tenure Table 9.4 Tenure of new accommodation required in Wiltshire over the next 15 years Current tenure profile Tenure profile 2026 Change required % of change required Market 148,965 167,701 18,736 55.1% Shared Ownership 825 6,642* 5,817 17.1% Affordable Rent 0 8,907 8,907 26.2% Social rented 16,757 LHA 26,454 } 43,750 540 1.6% Total 193,000 227,000 34,000 100.0% *Indicates the ability of households to afford Shared ownership rather than demand, which is particularly low in Wiltshire Source: Wiltshire household survey, 2011 9.25 The model outlined above recommends Shared Ownership and Affordable Rent properties to together equate to approximately 6.8% of overall stock in 2026. It suggests that the split of new affordable housing (excluding LHA housing) required up to 2026 should be 43% Shared Ownership and 57% Affordable Rent. However, this is based on what households can afford and not demand for these products. In particular, information form the Wiltshire Council s Housing department suggests that Shared Ownership products are unpopular and that demand would not be sufficient to meet the supply suggested by the model. Feedback from the stakeholder consultation and data from the household survey supports this. 9.26 The household survey indicates that only 1.2% of moving households would like to move to a Shared Ownership property compared with 5.2% who would like to rent a property privately and 26.1% who would like a social rented property. In addition 54.6% of households that previously lived in Shared Ownership properties and have moved in the past two years, now rent a property privately. this suggests that Shared Ownership has not been entirely effective at allowing households to get on the property ladder. Despite having a Shared Ownership property previously, only a quarter (23.6%) of these households considered this tenure as an option for their next home. 123

9.27 Current Shared Ownership households were the most likely tenure group to claim that one of the reasons for choosing their current home was due to a lack of alternatives. In addition, Shared Ownership households were also the least likely tenure group to claim to be satisfied with their current housing (75% of shared ownership households compared to 90% of all households). The survey findings therefore support the evidence that Shared Ownership is not a popular tenure. 9.28 In terms of building affordable housing in the future, it seems that it would be more suitable in Wiltshire to combine the building target for Affordable Rent properties and Shared Ownership properties together. The exact proportion of each to be built can then be decided in response to local demand. Size of housing required within each tenure 9.29 The table below presents the size of market accommodation required in Wiltshire in 15 years time in comparison to the size profile recorded in the sector currently. The implied change to the housing stock is also presented. The table shows that some 46.0% of new market dwellings should be three bedroom properties, with 40.2% containing two bedrooms 11.8% having one bedroom and 2.0% having four or more bedrooms. Dwelling size Table 9.5 Size of new market accommodation required in Wiltshire over the next 15 years Current size profile Size profile 2026 Change required % of change required One bedroom 6,232 8,441 2,209 11.8% Two bedrooms 30,767 38,306 7,539 40.2% Three bedrooms 63,136 71,752 8,617 46.0% Four or more bedrooms 48,830 49,202 372 2.0% Total 148,965 167,701 18,736 100.0% Source: Wiltshire household survey, 2011 9.30 This analysis can be repeated for Shared Ownership housing and is presented in the table below. The data indicates that of the 5,817 Shared Ownership dwellings required within the County, 49.8% should be two bedroom properties with a further 36.7% three bedroom accommodation. 124

Table 9.6 Size of new Shared Ownership accommodation required in Wiltshire over the next 15 years Dwelling size Current size profile Size profile 2026 Change required % of change required One bedroom 104 532 428 7.4% Two bedrooms 479 3,373 2,894 49.8% Three bedrooms 207 2,339 2,132 36.7% Four or more bedrooms 36 398 362 6.2% Total 825 6,642 5,817 100.0% Source: Wiltshire household survey, 2011 9.31 The table below shows the equivalent results for the Affordable Rented sector, as there is not currently any of this tenure in existence it will all be new. The table shows that of the 8,097 additional Affordable Rented units required within the County over the next 15 years, over half should be two bedroom properties, with almost a quarter one bedroom dwellings. Table 9.7 Size of new Affordable Rented homes required in Wiltshire over the next 15 years Dwelling size Size profile 2026 % of change required One bedroom 2,095 23.5% Two bedrooms 4,588 51.5% Three bedrooms 1,964 22.0% Four or more bedrooms 259 2.9% Total 8,907 100.0% Source: Wiltshire household survey, 2011 9.32 The table below presents the size of LHA accommodation required in Wiltshire in 15 years time in comparison to the size profile recorded in the social rented and LHA rented sectors currently. The implied change to the housing stock is also presented. The table shows that there will be more households using LHA requiring one and four bedroom accommodation in 2026 than there are properties of this size within the social rented and LHA rented sector currently. However the opposite is true for two and three bedroom accommodation, with fewer LHA households in 2026 requiring this size home than there are dwellings of this size in the social rented and LHA rented sector currently. 125

Table 9.8 Size of new LHA accommodation required in Wiltshire over the next 15 years Dwelling size Current size profile Size profile 2026 Change required One bedroom 10,015 12,296 2,281 Two bedrooms 18,507 16,278-2,229 Three bedrooms 13,736 13,041-695 Four or more bedrooms 951 2,134 1,183 Total 43,210 43,750 540 Source: Wiltshire household survey, 2011 9.33 These surpluses would logically be used as Affordable Rented accommodation. If this were to happen then the size of Affordable Rented and LHA accommodation required in 2026 would be as is described in the table. Table 9.9 Size of new Affordable Rent and LHA accommodation required in Wiltshire over the next 15 years accounting for surpluses Dwelling size Affordable Rent LHA Both % of both One bedroom 2,095 2,281 4,376 46.3% Two bedrooms 2,359 0 2,359 25.0% Three bedrooms 1,269 0 1,269 13.4% Four or more bedrooms 259 1,183 1,442 15.3% Total 5,983 3,464 9,447 100.0% Source: Wiltshire household survey, 2011 Future specialist housing needs 9.34 Of the estimated 227,000 households in Wiltshire in 2026 it is expected that 5.1% will have some kind of specialist housing need. They would require specialist housing or floating support services to allow them to live independently. This section outlines the characteristics of these 11,760 households in order to allow the Council to anticipate and plan the services that they are likely to require. 9.35 The table below shows where we expect these households to be located (based on trends in 2011). It appears that the West Towns subarea will have both the highest number and proportion of households requiring specialist housing support. The proportion of households in need of support in the rural areas is reasonably low. 126

Table 9.17 Location of households requiring housing support in 2026 Households requiring housing support % of all households Bradford-on-Avon 546 10.4% Chippenham 364 2.2% Devizes 729 8.3% North and Central Rural 1,150 2.0% North Small Towns 828 7.7% North Towns 1,029 5.0% Salisbury 1,460 5.8% Salisbury Plain 557 4.0% South Rural 702 3.0% Trowbridge 1,410 7.5% West Towns 2,986 10.9% Total 11,760 5.2% Source: Wiltshire household survey, 2011 9.36 It is estimated that 8,262 of these households (70.3%) will contain someone with a support need. The table below identifies the type of support needs these households will have. The most common need amongst households requiring housing support is Medical condition, followed by Frail Elderly and Physical Disability. Around a quarter of these households will contain someone with more than one support need but only 6.8% of households will contain more than one person with a support need. The numbers of households in each category exceed the total number of support needs households because people can have more than one category of support need. 127

Table 9.18 Support needs categories of households requiring housing support in 2026 Category Number of households % of households requiring support Medical condition 4,362 37.1% Frail elderly 3,193 27.2% Physical disability 3,050 25.9% Mental health problem 1,002 8.5% Learning difficulty 346 2.9% Severe sensory disability 416 3.5% Other 642 5.5 Multiple support needs 2,925 24.9% More than one person with support need 801 6.8% Any support need 8,262 70.3% Source: Wiltshire household survey, 2011 9.37 Although only 27.2% of households are identified as Frail/Elderly, it is estimated that 81.9% of households in need of specialist housing support in 2026 will be older person only households. Further analysis suggests that areas in the North of the County will have particularly high proportion of households that require housing support that are older person only households. In the more urban areas to the West of the County, it is more likely that households that require housing support will not be older person households but will have a support need. 9.38 The figure below shows the appropriate tenure for households in need of housing support in 2026 (as suggested by the LTBHM). It shows that these households are much less likely to be able to afford suitable accommodation on the open market. Around half of these households would not be able to afford market accommodation or an affordable tenure (i.e. Shared Ownership or Affordable Rent) and would require state support to meet their housing costs. 128

Figure 9.8 Appropriate tenure for households requiring housing support in 2026 60.0% 52.2% 50.0% 40.0% 30.0% 20.0% 10.0% 25.2% 22.6% 0.0% Market Affordable tenure State subsidised Source: Wiltshire household survey, 2011 9.39 The model indicates that 78.2% of the households expected to require housing support in 2026 will contain only one person, and almost all of the remainder will contain two people. 9.40 Finally, the figure below shows the appropriate dwelling type for households in need of housing support in 2026. This information can indicate the kinds of adaptations some households will require, for example non-bungalows may require ramps / a stair lift. Figure 9.9 Proportion of households requiring housing support in 2026 by dwelling type Dwelling type 7.2%5.2% 53.9% 34.8% 0% 25% 50% 75% 100% Detaced/Semi-deatched Terraced Flat Bungalow Source: Wiltshire household survey, 2011 LTBHM outputs in 2016 and 2021 9.41 Whilst the LTBHM model has been designed to provide a longer term perspective on the change in housing required, it is possible to produce outputs for the more immediate future that can be used as waymarks towards the figures for 2026. This section therefore presents results from the LTBHM for the housing required in 2016 and 2021. This is achieved using the same approach as described for the main 2026 outputs, but basing the projection on the ONS population and household estimates for the respective years. 129

Requirements for 2021 9.42 The table below shows the tenure profile required by households resident in Wiltshire in 10 years time and the change in stock this implies. Overall in 2021 some 74.6% of dwellings should be market homes, 2.5% shared ownership properties, 3.6% Affordable Rented homes and 19.2% dwellings occupied with the support of LHA. 9.43 The data suggest that over the next 10 years, 49.0% of new housing should be in the market sector, 18.7% should be Shared Ownership properties and 32.3% Affordable Rent. There will be a surplus of homes for households requiring LHA. Logically these would be used to house those able to afford Affordable Rent and the Affordable Rent requirement would fall accordingly. This would leave a requirement for 6,092 Affordable Rented homes between 2011 and 2021, which represents 27.0% of the total new housing required. The market requirement would then account for 52.8% of new housing and shared ownership 20.2%. Tenure Table 9.19 Tenure of new accommodation required in Wiltshire over the next 10 years Current tenure profile Tenure profile 2021 Change required % of change required Market 148,965 160,908 11,943 49.0% Shared Ownership 825 5,389 4,564 18.7% Affordable Rent 0 7,864 7,864 32.3% Social rented 16,757 LHA 26,454 } 41,439-1,772 0.0% Total 193,000 215,600 22,600 100.0% *Indicates the ability of households to afford Shared ownership rather than demand, which is particularly low in Wiltshire Source: Wiltshire household survey, 2011 9.44 The figure below shows the size of new accommodation required in each tenure by 2021. 130

Figure 9.10 Size of new accommodation required in Wiltshire by 2021 (by tenure) 7,000 Number of dwellings required 6,000 5,000 4,000 3,000 2,000 1,000 0-1,000-2,000-3,000-4,000 5,606 1 bedroom 2 bedroom 4,113 4,055 3 bedroom 4+ bedroom 2,094 1,758 1,758 1,771 1,261 963 875 1,195 362 350 280 market shared ownesrhip affordable rent LHA -643-3,199 Source: Wiltshire household survey, 2011 Requirements for 2016 9.45 The table below shows the tenure profile required by households resident in Wiltshire in 5 years time and the change in stock this implies. Overall in 2016 some 75.1% of dwellings should be market homes, 2.2% shared ownership properties, 3.6% Affordable Rented homes and 19.1% dwellings occupied with the support of LHA. 9.46 The data suggest that over the next 5 years, 28.3% of new housing should be in the market sector, 24.3% should be Shared Ownership properties and 47.5% Affordable Rent. There will be a surplus of homes for households requiring LHA. Deducting these from the Affordable Rent total would results in revised requirement of 39.1% market, 33.6% shared ownership and 27.4% Affordable Rent. Tenure Table 9.19 Tenure of new accommodation required in Wiltshire over the next 5 years Current tenure profile Tenure profile 2016 Change required % of change required Market 148,965 153,343 4,378 28.3% Shared Ownership 825 4,583 3,758 24.3% Affordable Rent 0 7,349 7,349 47.5% Social rented 16,757 LHA 26,454 } 38,926-4,285 0.0% Total 193,000 204,200 11,200 100.0% *Indicates the ability of households to afford Shared ownership rather than demand, which is particularly low in Wiltshire Source: Wiltshire household survey, 2011 131

9.47 The figure below shows the size of new accommodation required in each tenure by 2016. Figure 9.11 Size of new accommodation required in Wiltshire by 2016 (by tenure) Number of dwellings required 5,000 4,000 3,000 2,000 1,000 0-1,000-2,000 1 bedroom 3,775 2 bedroom 3 bedroom 2,378 4+ bedroom 1,676 1,6281,510 1,587 1,703 702 547 288 332 283 market shared ownesrhip affordable rent -43 LHA -378-1,209-3,000-4,000 Source: Wiltshire household survey, 2011-3,579 9.48 Presenting the results for these waymarks shows that the affordable requirement over the 15 year period is relatively front loaded, with the proportion of new market housing required increasing over time. However, whilst these results provide useful context, the model outputs for 2026 are the most important and the ones that the Council should use to inform their overall policy for new housing. Reconciling future estimates for affordable housing 9.49 The figures in Table 9.4 indicate the levels of newbuild required by the projected increase in population and households, adjusting occupation patterns to ensure all households are adequately housed. This suggests there is a total requirement for 15,264 new affordable dwellings over the 15 year model period, which equates to about 1,000 new affordable dwellings per annum. 9.50 Table 5.8 shows the level of housing need recorded in Wiltshire following the CLG housing needs assessment model. This indicates a need for 3,862 affordable dwellings each year. This figure does not correlate well with the LTBHM equivalent, but that is because the needs estimate assesses the size of the problem and not its solution. There are two market realities that the housing need model does not reflect but are accounted for within the LTBHM model. 9.51 Firstly the significant reduction in affordable stock that has occurred in recent history has meant that households have a reduced expectation of accessing an affordable home. in addition the polarisation of the stock of affordable homes has also reduced the number of households that aspire to reside in one of these properties. This means that even though there are a notable number of households spending more than of their income on their rent in the market sector than the guideline level (25% of 132

gross household income), households are accepting that this will continue as it is unlikely an affordable home will be available for them. Furthermore, in some cases households are happy to pay more for market housing than would be desirable (and is arguably sustainable) as they do not intend to live in an affordable home. The reality is that 30% of gross income on rent better reflects the prevailing market conditions in Wiltshire. Table 5.17 showed that if this is taken as the affordability criteria then the level of net need reduces to 2,721 households per annum. 9.52 Secondly, the housing need model does not accurately reflect affordable housing supply as it operates in the current housing market. The definition of affordable housing used in PPS3 excludes market housing supported with LHA from being considered affordable housing. In reality market housing (principally the private rented sector) is used to compensate for the shortfall of genuine affordable housing and houses households who theoretically would be more appropriately accommodated in a social rented or intermediate property. Lettings of this housing, identified as 1,362 per annum in paragraph 5.57, represents a supply to meet the housing need. Deducting this figure from the revised housing need estimate of 2,722 described above results in a requirement for 1,359 new affordable dwellings each year. 9.53 This revised figure of 1,359 more closely corresponds with the results of the LTBHM model, particularly those showing the housing required over the next five years (the same period that the housing need model examines), which are presented in table 9.19. This indicates that there is a requirement for 6,821 new affordable homes between 2006 and 2011, which equates to 1,364 dwellings per year. 9.54 The tenure of affordable housing required recorded in the housing need model (table 5.13) and in the LTBHM model (table 9.4) broadly correspond. In both cases Affordable Rent forms around half of the new affordable accommodation required and Shared Ownership the majority of the remainder. Tenure required if a 30% affordability threshold is used 9.55 The table below shows the tenure profile required by households resident in the County in 15 years time if it is presumed that a household can afford 30% of its gross income on the rental cost rather than the 25% set out by Practice Guidance. This is compared to the tenure profile recorded currently. The results show that using the affordability threshold of 30% does increase notably the number of households able to afford Shared Ownership and Affordable Rented products to the extent that there would be a surplus of over 30,000 social rented and LHA private rented dwellings in Wiltshire. This surplus is greater than the requirement for Affordable Rented dwellings meaning theoretically that no new dwellings of this tenure would be required (just the re-labelling of the current social rented stock) and that all new provision could be Shared Ownership and market homes (56.6% of which would be market dwellings). 133

9.56 It should however be noted that raising the proportion of income that households spend on their rent for sub-market housing is likely to leave the household in financial difficulties as the incomes are already low. Tenure Table 9.9 Tenure of new accommodation required in Wiltshire over the next 15 years based on 30% affordability Current tenure profile Tenure profile 2028 Change required % of change required Market 148,965 168,220 19,255 28.4% Shared Ownership 825 32,563 31,738 46.8% Affordable Rent 0 16,852 16,852 24.8% Social rented 16,757 LHA 26,454 } 9,364-33,846 0.0% Total 193,000 227,000 34,000 100.0% Source: Wiltshire household survey, 2011 134

10: Policy Implications & Monitoring Introduction 10.1 This final chapter reviews the policy implications of the SHMA. The Executive Summary at the front of the report provides a summary of the work covered: this section simply collates the potential policy implications of the report s findings. They are policy suggestion and it is of course for the Council to decide what it does with the findings of such work, in conjunction with its stakeholders. Meetings with stakeholders were held both at the project inception and after the draft results were produced. Given its summary nature there is no separate summary for this chapter. 10.2 It is important to note that whilst the outputs of this SHMA form an important part of the evidence base for the Council s development of planning and housing polices, this study does not, on its own, determine policy. The purposes of this study and other sources of evidence is to inform rather than prescribe policy. In drawing up policy the Council will draw on a wide range of evidence. Political priority will have a strong influence on the final direction of policy. Final housing market areas for policy 10.3 In order for Wiltshire Council to produce housing targets for the County, Housing Market Areas need to be established. This SHMA contains a thorough analysis of the potential boundaries of the housing markets areas operating in Wiltshire (Chapter 4 and Appendices 3, 4 and 5). This section attempts to conclude this analysis; the identification of the comprehensive Housing Market Areas on Wiltshire which can form the basis of future Housing Policy in the County. 10.4 The findings from the estate agents survey and the search patterns analysis have enabled us to identify three final Housing Market Areas, contiguous with the boundaries of the Community Areas already used by the Council. These three Housing Market Areas are displayed on the map below, whilst the following table lists the Community Areas contained within each. Table 10.1 Community Areas and households within each Housing Market Area HMA Community Areas H holds % h holds East Devizes, Marlborough, Pewsey, Tidworth 32,673 16.9% South Amesbury Downton, Mere, Salisbury, Tisbury, Wilton 48,635 25.2% West Bradford-on-Avon, Calne, Chippenham, Corsham, Malmesbury, Melksham, Trowbridge, Warminster, Westbury, Wootton Bassett & Cricklade Source: Wiltshire household survey, 2011 111,692 57.9% 135

Figure 10.1 Comprehensive Housing Market Areas in Wiltshire Source: Wiltshire household survey, 2011 10.5 The largest Housing Market Area is the West containing 111,692 households. This is followed by the South (48,635 households) and then the East (32,673 households). The West is more than twice the size of the other HMAs (it accounts for 57.9% of households in the County). Whilst it does cover a broad area spatially it also denser containing a number of towns along the western edge. 10.6 There are many similarities between the HMAs identified in this SHMA and those identified in the Regional Spatial Strategy in 2009. The South HMA is identical to the Salisbury HMA proposed in the RSS and is based on the former Salisbury district. The East HMA is the same as the former district of Kennet, however, it is smaller than the Swindon HMA identified in the Regional Spatial Strategy. Finally, the West HMA is much larger and combines the former districts of North Wiltshire and West Wiltshire. This seems sensible as there were a number of close links between the towns along the west of the County which were previously divided into separate HMAs. 136

Housing Need 10.7 The estimated housing need in Wiltshire is 3,862 new households per annum. This tallies quite well with the Housing Register, although the two numbers are quite differently derived. The implication of this level of housing need is to support the present 40% target for affordable housing. 10.8 The geographical desegregation of the housing need estimate shows that the locations where affordable supply is lowest in relation to need are, not surprisingly the rural areas: North and Central rural and South rural, where supply is below 20% of the annual need. Salisbury Plain, a mixed urban and rural area has an equally low supply. Among towns, the lowest supply is found in Trowbridge (21%). The County average is that 38% of the need is met by annual supply. These findings may have more specific policy implications. Particular groups 10.9 The level of support needs in Wiltshire is substantial, but fairly typical: 22% of all households contain a member with a support need. These households are concentrated in the social rented sector (nearly half of households in that tenure contain a household member with a support need), but it is not likely that this finding raises new policy issues. 10.10 Another important statistic is that 36% of all households in the County are either wholly or partly made up of older people. Most of these households (three quarters) can afford market housing and many will have substantial equity in their home. This is important as it implies that there is considerable scope for equity release. This is, for many people, a major route to sustainable living in later years, and would benefit from some policy emphasis. Rural issues 10.11 Wiltshire has highly rural population: about two thirds of all households live in rural areas, compared with a fifth as the national average. Rural areas show the strongest polarity of incomes: relatively low income households and high income ones typically involving long distance commuters. This is combined with a larger older population. Rural households register much greater difficulty in accessing services, a common pattern throughout the country. 10.12 The main policy issue is in rural areas is greater provision of affordable housing. It was noted above, under housing need, that there is a very low supply of rural affordable housing to meet the rural need. Affordable Rent 10.13 This is the new social tenure, set at 80% of median market rent, or lower if there is evidence to justify it. The HCA is only likely to fund Affordable Rented housing over the next four year period. However Wiltshire s affordable housebuilding programme is mainly funded by S106 payments. 137

10.14 Nevertheless, it is likely that the Registered Providers will want to build some Affordable Rented housing. If the Affordable Rent were set at 80% of median market rent it would be affordable for relatively few households on the Housing Register. But if it were set at 70% of market rent, it would be affordable for nearly 1,150 households on the Register and a similar number of households in defined housing need. The likely annual supply of this tenure would be around 1,050 if half of social re-lets were converted to Affordable Rent, as is currently being considered by local register providers. 10.15 Affordable Rent can be achieved either by newbuild or by conversion of social relets. The latter is obviously cheaper and can be used to cross-subsidise newbuild. According to our calculations the net subsidy required for 5 new Affordable Rent dwellings (at 70%) can be met by one conversion of a social relet to Affordable Rent. That is because the level of net subsidy required for Affordable Rent at 70% of market rent is not very large. It is up to the Council to choose the level that Affordable rent is finally set: there is a considerable range of evidence in the report. New affordable housing required 10.16 Paragraphs 9.63 through to 9.68 reconcile the affordable housing requirements of the housing need and LTBHM models. The conclusion is that in current market conditions, there is a requirement for around 1,360 new affordable dwellings in Wiltshire each year. The actual production of new affordable housing in Wiltshire is currently 682 units per annum. If production could reach an average of 1,000 new affordable homes over the next 15 years then the market would become more balanced and there would be very little housing need in the County at the end of the period. Type and tenure of new housing required over the period 10.17 The LTBHM suggests that the 34,000 new homes required over the next 15 years should be split about 55% market housing and 45% affordable housing. A split of 55:45 market: affordable might appear difficult in viability terms. It would have had serious implications in the old days when affordable housing was largely social rented. But nearly all the affordable housing required is either shared ownership (40%) or Affordable Rent (60%) with hardly any new social rented housing being necessary in net terms over the period. This means that the subsidy required, if it is all achieved through S106 or, in the future Community Infrastructure Charge (CIL) should be achievable and that these developments could well be viable. 10.18 In terms of size mix, the LTBHM model would suggest that the new market and shared ownership housing should mostly be 2 and 3 bedroom homes. The Affordable Rent component, and the small LHA element require a size mix spread fairly evenly across the 1 to 3 bed range. These figures are presented in full in table 9.5 to 9.9. 138

Effect of total numbers changes 10.19 Whilst the LTBHM model has been based on the ONS population and household projections which indicate household growth of 34,000 in Wiltshire over the next 15 years, the Council have now adopted a target of 37,000 new homes across the County between 2006 and 2026. This equates to 27,188 new dwelling between 2011 and 2026, once dwellings that have been completed in the last five years have been taken into account. This does not affect the proportions of tenure and size mix discussed in the previous subsection; the proportionate results of the LTBHM can be adjusted to the lower target being pursued by the Council. Monitoring 10.20 At the start of this report we stated that an SHMA is a process although this report is, by its very nature, a snapshot in time. It is, however, important that it is kept up to date and that the policies and strategies that emerge from this and other evidence are monitored to assess whether or not they are achieving their desired ends. 10.21 The principle recommendation from this report is the affordable housing target. There are two main sources of information for this from this report. The first being the Housing Needs Figure as derived in accordance with the Guidance (calculated in Chapter 5), the second is the Long Term Balancing Housing Markets Model (as set out in Chapter 9). In addition the actual affordable housing target will be need to be viable, that is to say (and as required by paragraph 29 of PPS3) that it is deliverable by the development industry. 10.22 There will also be a political element to setting both the number of houses to be built over the plan period and the amount of affordable housing target. Of particular importance is how the private rented sector (PRS) is going to be used. Presently the PRS is used to house many people in housing need that receive help with the cost of housing through Local Housing Allowance. There is no suggestion that all households in housing need should be accommodated in affordable housing, a decision will need to be taken to strike the balance between the use of affordable housing and the PRS. 10.23 To assist the council in monitoring their progress towards achieving the market balance we have provided the council with a simple monitoring tool in the form of an Excel spread sheet. Initially this has been set up to monitor the progress toward the market balance identified in the LTBHM, shown in table 9.4 of this report (presuming 27,188 new dwellings in period to 2026 and 45% of these will be affordable). The working sheet is shown in the table below. 139

Long-term BHM - achieving market balance more quickly Market Affordable Start year 2011 Starting stock 148,965 17,582 End year 2026 Target % affordable for tenure balance 16.9% Total build (all housing) 27,188 Years taken to reach balance 37 Total build % affordable 45.0% Tenure balance achieved at 2048 Year Market stock Affordable stock % of build affordable Build rate (override) Build rate % of stock affordable Balance Achieved 2011 148,965 17,582 45.0% 1,813 10.6% NO 2012 149,962 18,398 45.0% 1,813 10.9% NO 2013 150,959 19,213 45.0% 1,813 11.3% NO 2014 151,956 20,029 45.0% 1,813 11.6% NO 2015 152,953 20,845 45.0% 1,813 12.0% NO 2016 153,949 21,660 45.0% 1,813 12.3% NO 2017 154,946 22,476 45.0% 1,813 12.7% NO 2018 155,943 23,291 45.0% 1,813 13.0% NO 2019 156,940 24,107 45.0% 1,813 13.3% NO 2020 157,937 24,923 45.0% 1,813 13.6% NO 2021 158,934 25,738 45.0% 1,813 13.9% NO 2022 159,931 26,554 45.0% 1,813 14.2% NO 2023 160,928 27,370 45.0% 1,813 14.5% NO 2024 161,925 28,185 45.0% 1,813 14.8% NO 2025 162,922 29,001 45.0% 1,813 15.1% NO 2026 163,918 29,817 45.0% 1,813 15.4% NO 2027 164,915 30,632 45.0% 1,813 15.7% NO 2028 165,912 31,448 45.0% 1,813 15.9% NO 2029 166,909 32,264 45.0% 1,813 16.2% NO 2030 167,906 33,079 45.0% 1,813 16.5% NO 2031 168,903 33,895 45.0% 1,813 16.7% NO 2032 169,900 34,710 45.0% 1,813 17.0% YES 2033 170,897 35,526 45.0% 1,813 17.2% YES 2034 171,894 36,342 45.0% 1,813 17.5% YES 2035 172,890 37,157 45.0% 1,813 17.7% YES 10.24 This tool allows the council to add in the number of houses built each year and the proportion of these that were affordable. The spreadsheet the automatically updates a revised target in terms of total new build and the proportion to be affordable, to allow progress towards the long-term aim of a balanced market to be monitored. 140

Glossary Affordability A measure of whether households can access and sustain the cost of private sector housing. There are two main types of affordability measure: mortgage and rental. Mortgage affordability assesses whether households would be eligible for a mortgage; rental affordability measures whether a household can afford private rental. Mortgage affordability is based on conditions set by mortgage lenders using standard lending multipliers (3.5 times household income). Rental affordability is defined as the rent being less than a proportion of a household s gross income (in this case 25% of gross income). Affordable housing Affordable housing includes social rented and intermediate housing, provided to specified eligible households whose needs are not met by the market. Affordable housing should be at a cost which is below the costs of housing typically available in the open market and be available at a sub-market price in perpetuity (although there are some exceptions to this such as the Right-to-Acquire). [There is an ambiguity in PPS3: Housing, where intermediate housing is defined as being below market entry to rent, while affordable housing is defined to be below the threshold to buy (normally much higher than the private rental one). But in principle the Guidance defines affordable housing as below the market threshold, and rationally speaking, that includes the private rented as well as purchase sectors]. Affordable Rent A new social tenure brought in through the current Localism Bill by the Coalition Government. It is defined as having a rent that is 80% of the median market rent. It is to be assessed on a per dwelling basis. But for the purpose of analysis in this report it is assumed to coincide with 80% of the median rent for the middle range of the market in each dwelling size group. The rent, if set at 80% is much higher than a social rent, and is designed partly provide housing for the higher end of the Housing Register but also to generate extra affordable housing via the capitalised value of the higher rents involved. At the time of finalising this report it is an untried tenure. Annual need The combination of the net future need plus an allowance to deal progressively with part of the net current need. 141

Bedroom standard The bedroom standard is that used by the General Household Survey, and is calculated as follows: a separate bedroom is allocated to each co-habiting couple, any other person aged 21 or over, each pair of young persons aged 10-20 of the same sex, and each pair of children under 10 (regardless of sex). Unpaired young persons aged 10-20 are paired with a child under 10 of the same sex or, if not possible, allocated a separate bedroom. Any remaining unpaired children under 10 are also allocated a separate bedroom. The calculated standard for the household is then compared with the actual number of bedrooms available for its sole use to indicate deficiencies or excesses. Bedrooms include bed-sitters, box rooms and bedrooms which are identified as such by respondents even though they may not be in use as such. Current need Households whose current housing circumstances at a point in time fall below accepted minimum standards. This would include households living in overcrowded conditions, in unfit or seriously defective housing, families sharing, and homeless people living in temporary accommodation or sharing with others. Disaggregation Breaking a numerical assessment of housing need and supply down, either in terms of size and/or type of housing unit, or in terms of geographical sub-areas within the District. Entry-level market housing The survey of prices and rents is focussed on entry-level prices/rents. That is to say the price/rent at which there is a reasonable supply of dwellings in reasonable condition. The purpose of this approach is to ensure that when assessments are made of say first-time buyers, that the prices are the appropriate ones for the typical members of this group. Financial capacity This is defined as household (income 3) savings equity (the value of the property owned by owner-occupiers, typically the family home, net of mortgage). This provides an indication, when put on a capital basis, of the amount which the household could afford to pay for housing. Since equity is now a substantial part of the overall financial capacity of the large fraction of owner-occupiers it is essential to use this measure rather than the old price/income ratio to measure the activity of a housing market. Forecast Either of housing needs or requirements is a prediction of numbers which would arise in future years based on a model of the determinants of those numbers and assumptions about (a) the behaviour of households and the market and (b) how the key determinants are likely to change. It involves understanding relationships and predicting behaviour in response to preferences and economic conditions. 142

Headship rates Measures the proportion of individuals in the population, in a particular age/sex/marital status group, who head a household. Projected headship rates are applied to projected populations to produce projected numbers of households. Household One person living alone or a group of people who have the address as their only or main residence and who either share one meal a day or share a living room. Household formation The process whereby individuals in the population form separate households. Gross or new household formation refers to households which form over a period of time, conventionally one year. This is equal to the number of households existing at the end of the year which did not exist as separate households at the beginning of the year (not counting successor households, when the former head of household dies or departs). Household reference person For the purposes of our study the survey respondent is taken to represent the household reference person (HRP). Housing Market Area The geographical area in which a substantial majority of the employed population both live and work, and where most of those changing home without changing employment choose to stay. Housing need Housing need is defined as the number of households who lack their own housing or who live in unsuitable housing and who cannot afford to meet their housing needs in the market. Housing Register A database of all individuals or households who have applied to a local authority or RSL for a social tenancy or access to some other form of affordable housing. Housing Registers, often called Waiting Lists, may include not only people with general needs but people with support needs or requiring access because of special circumstances, including homelessness. Intermediate Housing PPS3 defines intermediate housing as housing at prices and rents above those of social rent but below market prices or rents and which meet the criteria set out above. These can include shared equity products (e.g. HomeBuy), other low cost homes for sale and intermediate rent. Lower quartile The value below which one quarter of the cases falls. In relation to house prices, it means the price of the house that is one-quarter of the way up the ranking from the cheapest to the most expensive. 143

Mean The mean is the most common form of average used. It is calculated by dividing the sum of a distribution by the number of incidents in the distribution. Median The median is an alternative way of calculating the average. It is the middle value of the distribution when the distribution is sorted in ascending or descending order. Migration The movement of people between geographical areas primarily defined in this context as local authority Districts. The rate of migration is usually measured as an annual number of households, living in the District at a point in time, who are not resident in that District one year earlier. MoD (Ministry of Defence) Wiltshire contains a number of military bases mostly located within the Salisbury Plain. More military staff live in the county than any other area in the South West (Audit Commission, Wiltshire Comprehensive Area Assessment 2009). Most military staff based in Wiltshire are provided with housing owned by the MoD. Net need The difference between need and the expected supply of available affordable housing units (e.g. from the re-letting of existing social rented dwellings). Newly arising need New households which are expected to form over a period of time and are likely to require some form of assistance to gain suitable housing together with other existing households whose circumstances change over the period so as to place them in a situation of need (e.g. households losing accommodation because of loss of income, relationship breakdown, eviction, or some other emergency). Newly forming households Adult individuals, couples or lone parent families living as part of other households of which they are neither the head nor the partner of the head and who need to live in their own separate accommodation, and/or are intending to move to separate accommodation rather than continuing to live with their host household. Occupancy rating This is used in the 2001 Census as a measure of under occupancy and overcrowding. It relates the actual number of rooms in the dwelling to the number of rooms required by the members of the household (based on relationships between them and their ages). An occupancy rating of -1, for example, implies that there is one room too few and that there is overcrowding in the household. 144

Overcrowding Definition used by Fordham Research (in analysis of household survey data) and the Survey of English Housing: a dwelling which is below the bedroom standard (see above). Definition used by the 2001 Census: a dwelling with a negative occupancy rating (see above). Primary data Information that is collected from a bespoke data collection exercise (e.g. surveys, focus groups or interviews) and analysed to produce a new set of findings. Projection Either of housing needs or requirements is a calculation of numbers expected in some future year or years based on the extrapolation of existing conditions and assumptions. For example, household projections calculate the number and composition of households expected at some future date(s) given the projected number of residents, broken down by age, sex and marital status, and an extrapolation of recent trends in the propensity of different groups to form separate households. Relets Social rented housing units which are vacated during a period and become potentially available for letting to new tenants. Sample survey Collects information from a known proportion of a population, normally selected at random, in order to estimate the characteristics of the population as a whole. Secondary data Existing information that someone else has collected. Data from administrative systems and some research projects are made available for others to summarise and analyse for their own purposes (e.g. Census, national surveys). SHMA (Strategic Housing Market Assessment) SHMA derives from government guidance suggesting that the evidence base required for the good planning of an area should be the product of a process rather than a technical exercise. Social rented housing PPS3 defines social rented housing as rented housing owned by local authorities and registered social landlords, for which guideline target rents are determined through the national rent regime, the proposals set out in the Three Year review of Rent Restructuring (July 2004) were implemented in policy in April 2006. It may also include rented housing owned or managed by other persons and provided under equivalent rental arrangements to the above, as agreed with the local authority or with the Housing Corporation as a condition of grant. 145

Support Needs Relating to people who have specific needs: such as those associated with a disability. Specialised housing Refers to specially designed housing (such as mobility or wheelchair accommodation, hostels or group homes) or housing specifically designated for particular groups (such as retirement housing). Under-occupation An under-occupied dwelling is one which exceeds the bedroom standard by two or more bedrooms. Unsuitably housed households All circumstances where households are living in housing which is in some way unsuitable, whether because of its size, type, design, location, condition or cost. Households can have more than one reason for being in unsuitable housing. 146

Appendix 1: Housing costs by sub-area A2.1 This Appendix compares the cost of the different tenures of housing in the seventeen price areas identified within the County. This information is then used to look at the size of the gaps between these different tenures. This replicates the information presented for the broader price areas presented in tables 4.4 and 4.5. Bedrooms Bradford-on-Avon Table A1.1 Weekly costs of housing in different tenure by price area Social rent Affordable Rent Shared Ownership Entry-level private rent Entry-level purchase* 1 bed 72 103 116 106 134 2 bed 82 138 147 147 170 3 bed 96 165 212 185 245 4 bed 96 221 306 240 354 Calne 1 bed 72 79 87 84 100 2 bed 82 106 105 122 122 3 bed 96 127 143 154 165 4 bed 96 170 202 199 234 Chippenham 1 bed 72 84 88 91 101 2 bed 82 113 108 126 125 3 bed 96 134 147 159 169 4 bed 96 181 214 206 248 Corsham 1 bed 72 92 88 91 102 2 bed 82 123 122 126 140 3 bed 96 146 163 158 188 4 bed 96 197 252 205 291 Cricklade 1 bed 72 90 75 101 87 2 bed 82 119 113 130 130 3 bed 96 151 160 167 185 4 bed 96 237 227 225 262 147

Bedrooms Devizes Table A1.1 Weekly costs of housing in different tenure by price area Social rent Affordable Rent Shared Ownership Entry-level private rent Entry-level purchase* 1 bed 72 76 84 85 97 2 bed 82 102 108 117 125 3 bed 96 121 155 148 179 4 bed 96 163 202 191 234 Malmesbury 1 bed 72 90 86 101 99 2 bed 82 119 129 130 149 3 bed 96 151 184 167 212 4 bed 96 237 260 225 300 Marlborough and Pewsey 1 bed 72 96 108 112 125 2 bed 82 128 162 144 187 3 bed 96 162 224 184 259 4 bed 96 254 326 248 377 Melksham 1 bed 72 82 71 91 82 2 bed 82 110 92 125 106 3 bed 96 131 132 158 152 4 bed 96 176 202 205 233 Rural North & Central 1 bed 72 91 84 104 97 2 bed 82 120 131 135 151 3 bed 96 158 184 172 212 4 bed 96 245 294 235 340 Rural South 1 bed 72 91 102 103 118 2 bed 82 121 152 132 175 3 bed 96 147 205 160 237 4 bed 96 257 283 230 326 Salisbury 1 bed 72 99 91 92 106 2 bed 82 126 141 149 163 3 bed 96 146 179 170 207 4 bed 96 210 255 236 295 148

Bedrooms Salisbury Plain Table A1.1 Weekly costs of housing in different tenure by price area Social rent Affordable Rent Shared Ownership Entry-level private rent Entry-level purchase* 1 bed 72 92 79 101 92 2 bed 82 123 109 140 126 3 bed 96 146 161 176 186 4 bed 96 197 223 228 258 Trowbridge 1 bed 72 83 73 85 85 2 bed 82 111 106 118 122 3 bed 96 133 129 149 149 4 bed 96 179 200 193 231 Warminster 1 bed 72 82 74 90 86 2 bed 82 110 112 125 129 3 bed 96 131 143 157 165 4 bed 96 176 184 203 213 Westbury 1 bed 72 75 70 83 81 2 bed 82 101 104 115 120 3 bed 96 120 130 145 150 4 bed 96 161 182 188 211 Wootton Bassett 1 bed 72 81 80 92 93 2 bed 82 108 110 127 127 3 bed 96 129 154 160 178 4 bed 96 173 215 207 248 Source: Wiltshire household survey, 2011 149

Bedrooms Bradford-on-Avon Table A1.2 Housing market gaps by price area Social rent Affordable Rent Social rent Private rent Affordable Rent Private rent Private rent Private purchase 1 bed 43.2% 47.4% 3.0% 25.9% 2 bed 68.4% 79.5% 6.6% 15.3% 3 bed 71.4% 93.0% 12.6% 32.4% 4 bed 130.5% 149.9% 8.4% 47.3% Calne 1 bed 10.2% 17.3% 6.5% 18.3% 2 bed 29.6% 48.9% 14.9% -0.5% 3 bed 31.8% 60.1% 21.4% 7.1% 4 bed 77.3% 107.4% 16.9% 17.4% Chippenham 1 bed 17.0% 26.5% 8.1% 11.1% 2 bed 37.7% 54.0% 11.8% -1.3% 3 bed 40.1% 65.5% 18.2% 6.6% 4 bed 88.4% 114.4% 13.8% 20.3% Corsham 1 bed 27.2% 26.1% -0.8% 12.1% 2 bed 49.6% 53.5% 2.6% 11.5% 3 bed 52.2% 65.0% 8.4% 18.9% 4 bed 104.7% 113.7% 4.4% 41.9% Cricklade 1 bed 24.5% 40.7% 13.0% -14.5% 2 bed 45.2% 59.0% 9.5% -0.3% 3 bed 57.5% 74.4% 10.7% 10.5% 4 bed 146.7% 134.6% -4.9% 16.2% Devizes 1 bed 5.4% 17.4% 11.4% 14.2% 2 bed 24.0% 42.9% 15.2% 6.4% 3 bed 26.2% 53.6% 21.8% 21.3% 4 bed 69.7% 99.0% 17.3% 22.3% Malmesbury 1 bed 24.5% 40.7% 13.0% -2.1% 2 bed 45.2% 59.0% 9.5% 14.2% 3 bed 57.5% 74.4% 10.7% 26.5% 4 bed 146.7% 134.6% -4.9% 33.0% 150

Bedrooms Marlborough and Pewsey Table A1.2 Housing market gaps by price area Social rent Affordable Rent Social rent Private rent Affordable Rent Private rent Private rent Private purchase 1 bed 33.4% 54.9% 16.2% 11.9% 2 bed 55.5% 75.1% 12.6% 30.5% 3 bed 68.7% 92.0% 13.9% 40.5% 4 bed 164.2% 158.3% -2.3% 52.0% Melksham 1 bed 14.1% 25.7% 10.2% -9.5% 2 bed 34.2% 53.0% 14.0% -15.4% 3 bed 36.5% 64.5% 20.5% -3.8% 4 bed 83.6% 113.1% 16.0% 14.0% Rural North & Central 1 bed 26.8% 44.4% 13.9% -7.0% 2 bed 45.8% 64.6% 12.9% 12.1% 3 bed 64.4% 78.7% 8.7% 23.7% 4 bed 155.5% 144.6% -4.3% 44.6% Rural South 1 bed 26.4% 42.8% 13.0% 15.1% 2 bed 47.4% 61.4% 9.5% 32.4% 3 bed 53.1% 67.0% 9.0% 47.8% 4 bed 167.9% 139.6% -10.6% 41.9% Salisbury 1 bed 37.5% 27.8% -7.1% 14.8% 2 bed 53.9% 81.7% 18.1% 9.2% 3 bed 52.4% 76.8% 16.0% 21.7% 4 bed 119.1% 145.6% 12.1% 25.0% Salisbury Plain 1 bed 27.2% 40.1% 10.2% -9.0% 2 bed 49.6% 70.6% 14.0% -10.1% 3 bed 52.2% 83.4% 20.5% 5.5% 4 bed 104.7% 137.5% 16.0% 12.9% Trowbridge 1 bed 15.6% 18.6% 2.6% -0.8% 2 bed 35.9% 44.4% 6.2% 3.4% 3 bed 38.3% 55.2% 12.2% -0.2% 4 bed 86.0% 101.0% 8.1% 19.7% 151

Bedrooms Warminster Table A1.2 Housing market gaps by price area Social rent Affordable Rent Social rent Private rent Affordable Rent Private rent Private rent Private purchase 1 bed 13.6% 24.9% 10.0% -4.8% 2 bed 33.6% 52.1% 13.8% 3.4% 3 bed 36.0% 63.5% 20.2% 4.9% 4 bed 82.9% 111.7% 15.8% 4.6% Westbury 1 bed 4.3% 15.6% 10.9% -2.5% 2 bed 22.7% 40.7% 14.7% 4.3% 3 bed 24.8% 51.3% 21.2% 3.0% 4 bed 67.9% 96.0% 16.8% 11.9% Wootton Bassett 1 bed 12.0% 27.3% 13.7% 1.2% 2 bed 31.7% 54.9% 17.6% 0.2% 3 bed 34.0% 66.6% 24.3% 11.3% 4 bed 80.2% 115.8% 19.7% 19.6% Source: Wiltshire household survey, 2011 152

Appendix 2: Sub-area and Community Area outputs Introduction A2.2 This appendix initially provides a profile of the eleven sub-areas listed in table 2.1 (and illustrated in figure 2.1). The Appendix then provides equivalent analysis for the twenty Community Areas of Wiltshire set out in figure 2.2. This provides additional geographical detail to the County-level data shown in the main report. Sub-Areas A2.3 Wiltshire was divided into eleven sub-areas, designed to reflect the urban/rural divisions in the County. This section provides some background data on these areas from the survey. Many of these statistics are unavailable from any other source at a sub-county level. Table A2.1 Tenure pattern by sub-area Tenure Sub-Area Owner-occupied (no mortgage) Owner-occupied (with mortgage) Social rented Private Rented Tied / Military Total Bradford-on-Avon 41.1% 27.1% 16.8% 14.4% 0.5% 4,226 Chippenham 29.6% 35.6% 19.8% 14.9% 0.0% 14,366 Devizes 26.2% 23.4% 34.2% 15.6% 0.6% 7,303 Rural N. & Central 41.8% 27.3% 11.8% 15.0% 4.0% 49,176 N. Small Towns 34.3% 25.9% 26.7% 11.5% 1.6% 9,090 North Towns 33.3% 36.1% 17.6% 13.0% 0.0% 17,265 Salisbury 28.6% 24.8% 25.1% 18.0% 3.5% 21,183 Salisbury Plain 14.8% 23.7% 29.1% 13.0% 19.5% 12,620 Rural South 39.5% 27.0% 16.3% 14.2% 3.0% 19,532 Trowbridge 32.7% 30.9% 20.3% 15.8% 0.3% 15,912 West Towns 32.1% 31.8% 19.6% 15.2% 1.3% 22,326 Total 33.9% 28.8% 19.2% 14.9% 3.3% 193,000 Source: Wiltshire SHMA, 2011 A2.4 Wiltshire Council has provided the exact numbers of military housing by Community Area as displayed in Table A2.9. This data is not available by sub-area level. There will be some differences between the actual proportion of Military housing and our estimates in the table above. The reasons for the differences between the data are partly due to the inclusion of non-military households who live in accommodation that is tied to their job and also due to the difficulties of surveying MoD barracks in a household survey. 153

Table A2.2 Household types by sub-area Tenure Sub-Area Single pensioner 2+ pensioner Couple, no children Single nonpensioner Multiadult Lone parent 2+ adults, one child 2+ adults, 2+ children Total Bradford-on-Avon 21.8% 16.1% 15.5% 18.6% 6.7% 2.9% 6.7% 11.7% 4,226 Chippenham 12.2% 9.9% 19.0% 19.7% 8.9% 6.1% 8.4% 15.7% 14,366 Devizes 15.8% 11.2% 18.3% 17.4% 8.6% 6.7% 8.7% 13.3% 7,303 Rural N. & Central 14.3% 14.7% 11.5% 24.6% 7.4% 3.2% 10.3% 14.0% 49,176 N. Small Towns 16.2% 11.5% 14.8% 17.7% 9.0% 6.1% 8.6% 16.2% 9,090 North Towns 15.7% 15.0% 14.5% 19.1% 5.9% 5.2% 10.3% 14.3% 17,265 Salisbury 14.5% 12.3% 21.6% 14.1% 9.9% 7.0% 8.0% 12.6% 21,183 Salisbury Plain 7.8% 7.4% 13.2% 22.3% 11.3% 6.1% 12.0% 20.0% 12,620 Rural South 15.4% 21.4% 9.2% 18.2% 9.7% 3.4% 6.9% 15.8% 19,532 Trowbridge 14.4% 11.1% 19.7% 21.5% 8.7% 6.0% 5.4% 13.2% 15,912 West Towns 17.3% 13.4% 17.2% 18.4% 6.6% 6.1% 9.6% 11.4% 22,326 Total 14.6% 13.6% 15.1% 20.1% 8.3% 5.1% 9.0% 14.2% 193,000 Source: Wiltshire SHMA, 2011 Table A2.3 Dwelling types by sub-area Sub-Area Dwelling Type Detached Semi-detached Terraced Bungalow Flat Total Bradford-on-Avon 21.6% 21.5% 29.0% 14.0% 13.9% 4,226 Chippenham 26.4% 25.5% 22.1% 7.0% 19.0% 14,366 Devizes 17.9% 22.7% 19.0% 10.4% 30.0% 7,303 Rural N. & Central 43.6% 24.6% 9.2% 20.2% 2.4% 49,176 N. Small Towns 29.5% 23.8% 22.3% 7.5% 16.8% 9,090 North Towns 26.1% 25.7% 20.1% 12.3% 15.8% 17,265 Salisbury 21.9% 20.7% 24.5% 8.5% 24.4% 21,183 Salisbury Plain 17.1% 22.2% 26.8% 14.6% 19.2% 12,620 Rural South 40.3% 23.5% 9.4% 22.3% 4.4% 19,532 Trowbridge 26.8% 22.5% 26.1% 9.5% 15.0% 15,912 West Towns 25.2% 26.2% 17.8% 16.7% 14.1% 22,326 Total 30.7% 23.9% 17.8% 14.7% 12.9% 193,000 Source: Wiltshire SHMA, 2011 154

Table A2.4 Housing issues by sub-area Sub-Area Over-crowding Housing Issue (note one household may have more than one problem) Underoccupation Temporary / Eviction risk Receiving LHA/HB* Unsuitable housing Total Bradford-on-Avon 1.3% 46.1% 3.0% 13.4% 4.7% 4,226 Chippenham 1.3% 41.0% 4.3% 14.5% 4.2% 14,366 Devizes 2.7% 33.3% 5.4% 28.5% 7.8% 7,303 Rural N. & Central 1.0% 57.5% 4.3% 8.3% 4.8% 49,176 N. Small Towns 1.3% 39.0% 2.2% 15.5% 3.3% 9,090 North Towns 1.2% 40.6% 4.0% 13.2% 5.9% 17,265 Salisbury 2.5% 35.4% 2.9% 16.7% 6.9% 21,183 Salisbury Plain 4.8% 26.9% 5.8% 15.1% 11.3% 12,620 Rural South 1.7% 56.3% 2.9% 8.0% 5.3% 19,532 Trowbridge 4.0% 41.9% 4.0% 17.7% 7.0% 15,912 West Towns 1.4% 40.1% 3.8% 19.1% 5.9% 22,326 Total 1.9% 44.9% 3.9% 13.8% 5.9% 193,000 * Reported by respondent to survey Source: Wiltshire SHMA, 2011 Table A2.5 Financial information by sub-area (medians) Sub-Area Income* Benefits** Savings Equity*** Capital Income* per head Bradford-on-Avon 25,100 1,830 8,710 215,000 158,700 14,100 Chippenham 27,700 1,870 3,980 154,400 80,200 13,400 Devizes 17,500 2,240 1,320 163,700 10,800 9,600 Rural N. & Central 28,900 1,750 7,400 239,000 184,400 14,800 N. Small Towns 25,900 1,880 7,510 208,200 120,300 12,600 North Towns 23,100 1,930 5,240 145,200 92,100 11,800 Salisbury 24,500 1,910 2,860 193,300 46,400 12,500 Salisbury Plain 28,400 1,850 1,020 135,500 5,100 12,000 Rural South 29,200 1,890 14,380 277,400 227,500 14,500 Trowbridge 23,000 1,970 3,290 145,000 81,700 12,400 West Towns 21,600 1,950 2,140 144,100 78,300 11,000 Total 25,500 1,880 4,530 188,500 109,600 12,700 * gross per household, excluding benefits ** excluding LHA / Housing Benefit *** owner-occupiers only Source: Wiltshire SHMA, 2011 155

Table A2.6 Income* per head by sub-area and employment status (medians) Sub-Area All employed households Employed in local area** Employed elsewhere % employed in local area** Bradford-on-Avon 17,100 10,300 20,200 27.9% Chippenham 17,600 12,800 20,800 44.6% Devizes 12,900 11,300 16,000 52.3% Rural N. & Central 16,800 14,600 19,400 36.5% N. Small Towns 13,600 12,800 14,800 39.3% North Towns 13,800 12,300 15,500 41.6% Salisbury 14,800 12,000 16,900 50.8% Salisbury Plain 13,300 11,600 15,600 42.1% Rural South 15,600 13,100 18,000 40.6% Trowbridge 15,400 13,800 16,300 39.8% West Towns 12,600 10,400 15,800 44.2% Total 15,500 12,500 17,300 41.6% * gross per household, excluding benefits ** in both same sub-area and same Community Area, includes work from home and mobile workers Source: Wiltshire SHMA, 2011 Table A2.7 Satisfaction levels Life as a whole Current housing Sub-Area Very satisfied Fairly satisfied Not satisfied* Very satisfied Fairly satisfied Not satisfied* Total Bradford-on-Avon 40.4% 44.9% 14.7% 60.2% 33.7% 6.0% 4,226 Chippenham 28.5% 52.1% 19.5% 50.0% 37.8% 12.2% 14,366 Devizes 27.8% 48.9% 23.3% 51.9% 31.3% 16.8% 7,303 Rural N. & Central 39.4% 41.0% 19.6% 63.7% 30.9% 5.4% 49,176 N. Small Towns 37.8% 44.3% 17.9% 54.0% 34.2% 11.9% 9,090 North Towns 32.9% 45.6% 21.5% 55.0% 36.1% 8.9% 17,265 Salisbury 33.8% 41.8% 24.4% 55.0% 32.5% 12.5% 21,183 Salisbury Plain 25.6% 51.3% 23.1% 43.4% 42.1% 14.5% 12,620 Rural South 38.7% 44.5% 16.8% 65.1% 29.2% 5.7% 19,532 Trowbridge 30.7% 48.0% 21.3% 53.8% 34.3% 11.9% 15,912 West Towns 27.2% 52.2% 20.6% 53.4% 32.6% 14.0% 22,326 Total 33.8% 45.8% 20.5% 56.8% 33.3% 9.9% 193,000 * including neither satisfied nor dissatisfied Source: Wiltshire SHMA, 2011 156

Community Areas A2.5 There are 20 Community Areas in Wiltshire, each covering an urban centre or small town and the immediately surrounding rural area. Although some of these areas have the same name as sub-areas, it is important to remember that they are not the same. The sample for these areas did not allow complex analysis of the market, but was designed to allow descriptive data to be produced from the survey. The following tables present a range of data on the community Areas. Table A2.8 Tenure pattern by Community Area Tenure Community Area Owneroccupied (no mortgage) Owneroccupied (with mortgage) Social rented Private Rented Tied / Military Total Amesbury 25.1% 25.9% 23.4% 14.3% 11.3% 12,503 Bradford-on-Avon 42.7% 32.8% 13.3% 10.9% 0.3% 7,776 Calne 27.9% 37.8% 17.9% 14.2% 2.2% 9,469 Chippenham 30.9% 36.5% 17.5% 13.4% 1.6% 18,652 Corsham 33.4% 31.6% 13.6% 19.4% 2.1% 8,618 Devizes 33.9% 23.8% 26.7% 13.9% 1.8% 13,256 Downton 39.8% 35.8% 7.7% 12.5% 4.2% 8,676 Malmesbury 41.0% 29.6% 10.6% 18.2% 0.7% 8,272 Marlborough 35.4% 16.9% 24.3% 19.8% 3.7% 7,189 Melksham 36.1% 35.4% 19.8% 8.7% 0.0% 11,606 Mere 44.2% 18.0% 18.5% 14.4% 4.8% 2,486 Pewsey 34.3% 24.2% 24.1% 9.5% 7.9% 5,939 Salisbury 26.7% 24.4% 27.1% 19.0% 2.7% 17,748 Tidworth 15.9% 21.4% 30.1% 13.5% 19.1% 6,289 Tisbury 41.8% 22.8% 20.6% 13.6% 1.2% 3,274 Trowbridge 37.2% 28.4% 18.1% 16.0% 0.3% 16,927 Warminster 39.5% 21.7% 18.8% 18.1% 1.8% 10,514 Westbury 31.6% 38.1% 13.2% 15.8% 1.3% 8,038 Wilton 32.1% 18.9% 28.9% 15.8% 4.2% 3,948 Wootton Bassett & Cricklade 42.7% 28.4% 11.8% 13.1% 4.0% 11,820 Total 33.9% 28.8% 19.2% 14.9% 3.3% 193,000 Source: Wiltshire SHMA, 2011 157

Table A2.9 Service family accommodation (SFA) by Community Area SFA stock % housing SFA stock Amesbury 1,739 31.2% Bradford on Avon 0 0.0% Calne 0 0.0% Chippenham 215 3.9% Corsham 370 6.7% Devizes 59 1.1% Downton 20 0.4% Malmesbury 0 0.0% Marlborough 0 0.0% Melksham 0 0.0% Mere 0 0.0% Pewsey 160 2.9% Salisbury 113 2.0% Tidworth 1,352 24.3% Tisbury 5 0.1% Trowbridge 36 0.7% Warminster 591 10.6% Westbury 11 0.2% Wilton 133 2.4% Wootton Bassett & Cricklade 763 13.7% Total 5,567 2.9% Source: Wiltshire Council, Secretariat Officer for DIO Operations Accommodation (May, 2011) 158

Table A2.10 Household types by Community Area Tenure Community Area Single pensioner 2+ pensioner Single nonpensioner Couple, no children Multi-adult Lone parent 2+ adults, one child 2+ adults, 2+ children Total Amesbury 9.9% 12.7% 9.3% 21.0% 10.7% 4.5% 12.8% 18.9% 12,503 Bradford-on-Avon 19.0% 16.1% 14.3% 22.1% 4.1% 2.5% 6.9% 15.1% 7,776 Calne 15.0% 13.7% 16.0% 17.1% 7.0% 3.6% 10.7% 16.8% 9,469 Chippenham 11.7% 11.5% 16.0% 20.2% 7.7% 5.2% 10.2% 17.5% 18,652 Corsham 15.7% 13.6% 17.9% 19.9% 5.2% 7.5% 10.9% 9.3% 8,618 Devizes 17.0% 11.7% 18.3% 20.2% 6.7% 6.0% 9.3% 10.7% 13,256 Downton 14.9% 20.9% 11.7% 14.5% 11.4% 3.8% 7.5% 15.3% 8,676 Malmesbury 9.7% 12.2% 10.4% 32.4% 4.7% 2.3% 8.2% 20.1% 8,272 Marlborough 11.4% 12.6% 11.6% 20.8% 7.5% 3.6% 15.3% 17.2% 7,189 Melksham 18.4% 13.9% 14.2% 19.2% 9.2% 4.1% 6.3% 14.8% 11,606 Mere 22.4% 22.8% 8.3% 19.0% 8.8% 3.7% 3.4% 11.6% 2,486 Pewsey 18.3% 10.5% 10.4% 28.2% 5.6% 4.1% 8.7% 14.2% 5,939 Salisbury 13.9% 10.1% 24.0% 14.5% 9.9% 7.7% 7.6% 12.3% 17,748 Tidworth 3.4% 7.0% 16.8% 24.5% 10.9% 8.2% 12.1% 17.1% 6,289 Tisbury 17.1% 21.1% 10.5% 18.8% 6.4% 1.5% 7.1% 17.5% 3,274 Trowbridge 13.8% 13.1% 19.8% 20.6% 10.7% 5.2% 5.0% 11.8% 16,927 Warminster 16.4% 22.7% 10.6% 16.9% 7.3% 6.0% 8.8% 11.3% 10,514 Westbury 18.5% 11.3% 19.7% 23.8% 5.6% 3.1% 8.3% 9.7% 8,038 Wilton 18.4% 23.2% 8.4% 14.4% 12.3% 4.1% 5.5% 13.7% 3,948 Wootton Bassett & Cricklade 17.8% 11.3% 10.7% 19.8% 9.7% 6.7% 11.5% 12.5% 11,820 Total 14.6% 13.6% 15.1% 20.1% 8.3% 5.1% 9.0% 14.2% 193,000 Source: Wiltshire household survey, 2011 159

Table A2.11 Dwelling type by Community Area Dwelling Type Community Area Detached Semidetached Terraced Bungalow Flat Total Amesbury 27.8% 23.6% 20.1% 21.1% 7.5% 12,503 Bradford-on-Avon 37.7% 16.5% 20.3% 16.4% 9.1% 7,776 Calne 30.1% 22.1% 19.8% 11.4% 16.7% 9,469 Chippenham 33.5% 24.2% 18.8% 8.9% 14.6% 18,652 Corsham 25.5% 30.3% 17.9% 13.3% 12.9% 8,618 Devizes 26.7% 25.9% 13.6% 16.3% 17.5% 13,256 Downton 41.4% 20.7% 7.0% 24.7% 6.2% 8,676 Malmesbury 46.9% 23.0% 11.8% 14.9% 3.4% 8,272 Marlborough 33.7% 24.6% 13.5% 18.8% 9.4% 7,189 Melksham 32.3% 28.6% 12.5% 13.3% 13.4% 11,606 Mere 30.8% 26.9% 12.9% 20.7% 8.7% 2,486 Pewsey 36.8% 28.2% 14.1% 16.0% 4.9% 5,939 Salisbury 20.7% 20.6% 26.3% 6.1% 26.3% 17,748 Tidworth 17.9% 19.9% 23.9% 13.3% 24.9% 6,289 Tisbury 42.0% 24.7% 15.0% 14.9% 3.5% 3,274 Trowbridge 27.0% 22.1% 24.8% 12.7% 13.4% 16,927 Warminster 28.4% 23.8% 17.3% 20.3% 10.2% 10,514 Westbury 30.9% 20.8% 16.5% 18.9% 12.8% 8,038 Wilton 32.8% 20.7% 15.2% 18.7% 12.6% 3,948 Wootton Bassett & Cricklade 32.8% 30.7% 15.2% 14.7% 6.7% 11,820 Total 30.7% 23.9% 17.8% 14.7% 12.9% 193,000 Source: Wiltshire SHMA, 2011 160

Table A2.12 Housing issues by Community Area Community Area Housing Issue (note one household may have more than one problem) Overcrowding Underoccupation Temporary / Eviction risk Receiving LHA/HB* Unsuitable housing Total Amesbury 3.8% 38.3% 4.6% 7.1% 10.3% 12,503 Bradford-on-Avon 0.7% 52.6% 1.7% 7.1% 2.9% 7,776 Calne 1.8% 48.5% 4.9% 7.6% 8.4% 9,469 Chippenham 1.0% 42.8% 3.6% 6.5% 3.2% 18,652 Corsham 0.0% 49.5% 3.4% 4.6% 6.5% 8,618 Devizes 2.3% 40.8% 6.8% 9.4% 6.0% 13,256 Downton 1.5% 60.9% 2.7% 1.4% 2.8% 8,676 Malmesbury 0.0% 55.8% 5.8% 3.5% 1.4% 8,272 Marlborough 4.3% 45.2% 8.1% 1.6% 6.7% 7,189 Melksham 0.3% 46.7% 2.8% 5.7% 6.4% 11,606 Mere 0.9% 53.5% 1.7% 4.7% 3.3% 2,486 Pewsey 1.9% 50.4% 0.3% 9.0% 6.7% 5,939 Salisbury 2.8% 32.9% 3.3% 8.9% 7.2% 17,748 Tidworth 3.8% 30.7% 3.1% 5.3% 9.2% 6,289 Tisbury 0.0% 60.5% 5.9% 8.6% 6.1% 3,274 Trowbridge 3.8% 43.2% 5.1% 8.5% 6.4% 16,927 Warminster 2.6% 42.7% 4.2% 9.2% 5.4% 10,514 Westbury 0.0% 43.5% 1.7% 10.9% 2.6% 8,038 Wilton 2.5% 42.0% 1.3% 3.4% 6.1% 3,948 Wootton Bassett & Cricklade 0.8% 49.5% 3.2% 6.9% 7.5% 11,820 Total 1.9% 44.9% 3.9% 6.9% 10.3% 193,000 * Reported by respondent to survey Source: Wiltshire household survey, 2011 161

Table A2.13 Financial information by sub-area (medians) Community Area Income* Benefits** Savings Equity*** Capital Income* per head Amesbury 30,000 1,810 1,540 189,200 18,700 12,600 Bradford-on-Avon 29,100 1,630 10,200 223,400 181,100 15,500 Calne 23,700 1,850 2,580 140,800 74,400 12,000 Chippenham 29,600 1,810 4,500 166,800 100,500 14,800 Corsham 25,900 1,930 8,610 190,500 131,100 13,300 Devizes 19,200 2,060 2,130 182,100 77,600 10,000 Downton 33,000 1,720 22,260 267,300 247,900 15,500 Malmesbury 31,700 1,740 12,010 241,800 193,300 15,700 Marlborough 26,700 1,630 5,500 287,400 61,200 12,900 Melksham 22,900 1,900 3,050 171,600 123,800 11,100 Mere 23,900 1,910 10,860 257,400 181,900 13,300 Pewsey 31,100 1,790 3,830 270,100 149,700 14,900 Salisbury 24,100 1,950 2,530 186,700 37,100 12,500 Tidworth 25,200 1,820 950 152,600 5,200 12,000 Tisbury 26,700 2,100 12,710 289,400 203,000 15,700 Trowbridge 23,000 1,970 4,060 151,800 100,500 12,300 Warminster 21,600 2,080 2,350 197,600 98,000 11,500 Westbury 26,700 1,760 2,900 145,600 102,500 14,400 Wilton 24,400 2,170 3,040 270,600 76,200 10,500 Wootton Bassett & Cricklade 25,400 1,990 7,750 179,000 144,400 12,200 Total 25,500 1,880 4,530 188,500 109,600 12,700 * gross per household, excluding benefits ** excluding LHA / Housing Benefit *** owner-occupiers only Source: Wiltshire household survey, 2011 162

Table A2.14 Satisfaction levels Life as a whole Current housing Sub-Area Very satisfied Fairly satisfied Not satisfied* Very satisfied Fairly satisfied Not satisfied* Total Amesbury 31.6% 47.4% 21.1% 49.3% 38.1% 12.7% 12,503 Bradford-on-Avon 53.7% 33.5% 12.8% 67.1% 29.0% 3.8% 7,776 Calne 31.2% 46.3% 22.4% 52.7% 37.9% 9.4% 9,469 Chippenham 28.8% 53.3% 17.9% 53.8% 36.3% 9.8% 18,652 Corsham 41.0% 35.3% 23.7% 59.4% 34.5% 6.2% 8,618 Devizes 32.1% 43.4% 24.5% 53.0% 33.6% 13.4% 13,256 Downton 35.1% 47.2% 17.6% 72.0% 26.0% 2.0% 8,676 Malmesbury 40.6% 38.5% 20.9% 68.4% 22.2% 9.4% 8,272 Marlborough 30.1% 51.0% 19.0% 50.8% 37.8% 11.4% 7,189 Melksham 28.3% 55.2% 16.5% 50.1% 37.4% 12.5% 11,606 Mere 39.7% 49.2% 11.1% 59.5% 33.8% 6.7% 2,486 Pewsey 42.9% 40.4% 16.7% 60.3% 35.8% 4.0% 5,939 Salisbury 34.6% 39.2% 26.2% 53.1% 33.6% 13.3% 17,748 Tidworth 25.0% 46.2% 28.7% 44.5% 45.6% 10.0% 6,289 Tisbury 39.2% 42.7% 18.1% 64.7% 25.4% 9.9% 3,274 Trowbridge 30.2% 49.3% 20.5% 55.4% 33.8% 10.9% 16,927 Warminster 32.5% 45.3% 22.2% 61.4% 24.3% 14.3% 10,514 Westbury 31.1% 49.6% 19.3% 60.2% 31.6% 8.2% 8,038 Wilton 37.9% 47.8% 14.3% 63.2% 27.5% 9.3% 3,948 Wootton Bassett & Cricklade 34.5% 45.9% 19.7% 59.8% 32.8% 7.4% 11,820 Total 33.8% 45.8% 20.5% 56.8% 33.3% 9.9% 193,000 Source: Wiltshire SHMA, 2011 * including neither satisfied nor dissatisfied 163

Appendix 3: HMA characteristics Introduction A3.1 This appendix contains the analysis of the household survey dataset that was used to help inform the location of the housing market boundaries within Wiltshire. These market boundaries are summarised at the end of Chapter 4. A range of housing market aspects were considered and the differences and similarities between the three principal housing market areas (HMAs) East, West and South are displayed within this Appendix. Data is also presented for three of the distinct sub-markets contained within these larger markets the Salisbury Plain, Outer Swindon and Salisbury. Patterns in accommodation type A3.2 The table below shows dwelling types in Wiltshire by HMA. It shows that the West has the lowest level of flats and the highest level of detached homes. Meanwhile, the spread of dwellings in the South and East are very similar. The East contains slightly more bungalows than the other two market areas. Table A3.1 Dwelling type by HMA Dwelling type East South West Total Count % Count % Count % Count % Detached 9,272 28.4% 14,171 29.1% 35,768 32.0% 59,211 30.7% Semi-detached 8,129 24.9% 10,710 22.0% 27,263 24.4% 46,102 23.9% Terraced 5,105 15.6% 9,190 18.9% 20,065 18.0% 34,360 17.8% Bungalow 5,308 16.2% 7,599 15.6% 15,472 13.9% 28,379 14.7% Flat 4,858 14.9% 6,966 14.3% 13,124 11.8% 24,948 12.9% Total 32,674 100% 48,636 100% 111,692 100% 193,002 100% Source: Wiltshire household survey, 2011 A3.3 Among the sub-markets, Swindon Outer contains a markedly higher proportion of detached properties, and fewer flats/maisonettes. The Salisbury Plain sub-market has fewer detached, and more terraced units. Salisbury contains fewer detached and semi-detached dwellings, and more terraced houses and flats. Of the sub-areas, Salisbury has fewer bungalows. A3.4 Differences in house size (in terms of bedroom numbers) are comparatively minor, for the main market areas at least. The East has fewer one bedroom dwellings, and more two beds. A3.5 Of the sub-markets, Salisbury Plain has fewer one and four bed dwellings but more two bed homes. Outer Swindon has fewer one and two-bedroom dwellings, and more properties with at least four beds. Salisbury has more units with one and two bedrooms, and fewer with four or more bedrooms. 165

A3.6 The table below shows tenure in Wiltshire by HMA. Differences on tenure are again comparatively slight. The East has fewer owner-occupiers, and more rented accommodation, particularly social rented, than elsewhere. The West has more owner-occupiers, and lower proportions of social and private renting. The South has the greatest proportion of private rented households Table A3.2 Tenure by HMA Tenure Owner-occupied (no mortgage) Owner-occupied (with mortgage) East South West Total Count % Count % Count % Count % 10,070 30.8% 15,070 31.0% 40,216 36.0% 65,356 33.9% 7,157 21.9% 12,616 25.9% 35,739 32.0% 55,512 28.8% Social rented 8,604 26.3% 10,672 21.9% 17,840 16.0% 37,116 19.2% Private rented 6,842 20.9% 10,277 21.1% 17,898 16.0% 35,017 18.1% Total 32,674 100.0% 48,636 100.0% 111,692 100.0% 193,002 100.0% Source: Wiltshire household survey, 2011 A3.7 Among the sub-markets, the MoD dominated Salisbury Plain has fewer owners and more households renting. Outer Swindon has significantly more owner-occupiers than average, especially those with a mortgage, and less social rented. In Salisbury there are higher proportions renting, and fewer owneroccupiers. A3.8 The West has a higher proportion of second homes than the other two areas, though still only 1.4%. The Salisbury Plain sub-market also has a higher proportion of second homes. A3.9 The table below show the average property value in each HMA whilst the figure below shows the entry level price. On the whole the average price in the West is lower than the same property size in the North and South. In terms of entry level price, the South and East are very similar with the East becoming more expensive for four bedroom properties. Table A3.3 Average house price by HMA Property size East South West Median Mean Median Mean Median Mean 1 bed 125,000 125,000 117,000 130,000 100,000 105,000 2 bed 165,000 185,000 170,000 180,000 140,000 155,000 3 bed 250,000 280,000 230,000 250,000 190,000 220,000 4 bed 360,000 410,000 330,000 360,000 300,000 350,000 Source: Wiltshire estate agent survey March 2011, 2011 166

Figure A3.1 Entry level purchase price by HMA Entry level purchase price 350,000 300,000 250,000 200,000 150,000 100,000 93,000 140,000 195,000 290,000 93,000 150,000 190,000 275,000 85,000 120,000 160,000 240,000 50,000 0 East South West 1 bed 2 bed 3 bed 4 bed Source: Wiltshire estate agent survey March 2011, 2011 Patterns in household circumstance A3.10 The table below outlines the composition of households within each HMA. Of the main market areas, the East has fewer pensioners, fewer single non-pensioners, and more families with children. The South has more pensioners, fewer childless couples, and a slightly lower proportion of households with children. The Salisbury Plain sub-market has fewer pensioners, fewer single persons, and more households with children. In Outer Swindon there are more single pensioners. Salisbury contains more single non-pensioners, fewer childless couples, and more lone parents but slightly fewer families with children. Table A3.4 Household types by HMA Household type East South West Total Count % Count % Count % Count % Single pensioner 4,384 13.4% 6,843 14.1% 17,015 15.2% 28,242 14.6% 2+ pensioners 3,525 10.8% 7,377 15.2% 15,341 13.7% 26,243 13.6% Single non-pensioner 4,933 15.1% 7,314 15.0% 16,986 15.2% 29,233 15.1% 2+ adults, no children 9,844 30.1% 13,117 27.0% 31,739 28.4% 54,700 28.3% Lone parent 1,812 5.5% 2,566 5.3% 5,381 4.8% 9,759 5.1% 2+adults, one child 3,612 11.1% 4,131 8.5% 9,591 8.6% 17,334 9.0% 2+adults, 2+children 4,564 14.0% 7,288 15.0% 15,639 14.0% 27,491 14.2% Total 32,674 100% 48,636 100% 111,692 100% 193,002 100% Source: Wiltshire household survey 2011 167

A3.11 Unsuitable housing is slightly less prevalent in the West (5.2% of housheolds) compared with the East and South (both 6.9%). The South is the HMA with the greatest propotion of support needs housheolds. The table below shows levels of overcrowding by HMA. The East has the highest levels of overcrowding (at 2.9%) whilst the West has the highest levels of under-occupancy (47.6%). Table A3.5 Overcrowding type by HMA Occupation level East South West Total Count % Count % Count % Count % Overcrowding 958 2.9% 1,223 2.5% 1,461 1.3% 3,642 1.9% Ok 18,127 55.5% 26,522 54.5% 58,105 52.0% 10,2754 53.2% Under occupancy 13,588 41.6% 20,889 43.0% 52,126 47.6% 86,603 44.9% Total 32,674 100% 48,636 100% 111,692 100% 193,002 100% Source: Wiltshire household survey, 2011 A3.12 As is shown in the table below, the average income level across the HMAs is broadly similar. However, the difference between the mean and the median income is greatest in East suggesting a larger degree of income inequality (see the table below). Households in the East also tend to have a lower level of savings (looking at the median). The West records the lowest level of equity, reflecting the lower price of housing here. Table A3.6 Financial information by HMA HMA Income Savings Equity Median Mean Median Mean Median Mean East 26,669 38,201 2,437 36,097 207,761 209,644 South 29,657 38,659 5,420 41,054 220,583 216,645 West 28,030 37,837 5,057 33,737 173,927 182,751 Total 28,099 38,106 4,530 35,980 188,481 194,348 Source: Wiltshire household survey, 2011 Patterns of work and travel A3.13 In the East HMA the proportion of households with no car is lower, whilst more households have two or more cars. In the South more households have one car, and fewer have two cars or more. In the Salisbury Plain sub-market fewer households have no cars, and more have two. Salisbury has more households with no car, or with one car, and fewer with two or more cars. A3.14 The proportion of households not in work ranges from 34.9% in the East to 42.3% in the South. If these households are excluded, we can see more clearly how patterns of travel to work for the main respondent vary between the various HMAs and sub-markets. These are set out in the table below: 168

Table A3.7 Location of work for household main respondent (those in work only) by HMA / HMA sub-area Location Main HMA East West South Salisbury Plain Sub-HMAs Wiltshire average Swindon Salisbury Overall Within County 50.4% 59.1% 62.2% 62.2% 36.0% 71.0% 57.5% Neighbouring County 26.4% 17.6% 13.3% 13.3% 42.0% 11.0% 19.0% Work from home 10.6% 10.6% 10.6% 10.6% 8.5% 7.0% 10.6% Mobile worker 5.7% 4.7% 4.2% 4.2% 6.9% 4.1% 4.8% Not on map 6.9% 8.1% 9.6% 9.6% 6.6% 6.8% 8.1% Total (in work) 100.0% 100.0% 100.0% 100.0% 100.0% 0.0% 100.0% Source: Wiltshire household survey 2011 A3.15 The table shows that the East is the market area with the lowest proportion of employed household heads working within Wiltshire; and since the Salisbury Plain sub-area has a higher proportion (as might be anticipated) the difference would be even more marked if these were excluded. The East also has the highest numbers working in an adjacent District. A3.16 The South is the most self-contained at 62.2% (the figure is even higher, at 71.0%, for Salisbury alone). Household heads in the South are least likely to work in an adjoining District, but are most likely to work in a more distant location (presumably the larger South Coast employment centres, or London). As might be expected more households in the Outer Swindon area travelled to work in a neighbouring District more in fact than worked within the County. A3.17 In the East, consistent with the above, fewer household heads travelled less than 10 miles to work; but more travelled 30 or more. A much higher proportion in the South (87.7%) travelled less than ten miles. The proportion travelling 30 miles or more was also high in the Swindon sub-area. Patterns in previous homes evidence A3.18 There was little variation in the proportions of recent movers between HMAs, although there were slightly fewer movers in the South. Those in the Salisbury Plain sub-market were slightly more likely to have moved in the last two years. The table below sets out the previous tenure for those households who had recently moved. 169

Table A3.8 Tenure of previous home (recent movers only) by HMA / HMA sub-area Previous tenure Main HMA East West South Salisbury Plain Sub-HMAs Wiltshire average Swindon Salisbury Overall Living with parents/relative/friends 9.0% 16.7% 5.1% 9.3% 4.9% 5.3% 12.1% Owner-occupied 24.8% 35.6% 26.9% 16.1% 45.0% 23.5% 31.1% Social rented 8.3% 6.9% 8.6% 10.2% 6.8% 7.4% 7.7% PRS 26.7% 30.8% 38.6% 21.5% 35.8% 38.4% 31.2% Shared house/flat 1.7% 1.9% 3.4% 2.2% 0.0% 5.5% 2.2% Armed services accommodation 24.9% 3.7% 9.2% 34.1% 6.5% 11.2% 9.9% Tied to job 0.3% 1.5% 4.5% 0.0% 0.0% 3.1% 1.7% Shared Ownership 0.5% 0.1% 0.3% 1.0% 0.0% 0.0% 0.4% Lodger 0.7% 0.3% 0.0% 0.0% 1.1% 0.0% 0.7% Previously homeless/ta 3.1% 2.5% 3.5% 5.6% 0.0% 5.7% 2.8% Total recent movers 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Source: Wiltshire household survey 2011 A3.19 It can be seen that in the West, households were more likely to have lived with parents, or to have been owner-occupiers. Fewer came from the private rented categories. In the East fewer had lived with parents, or been owner-occupiers. Many more had been in the Armed Service. In the South fewer had lived with parents, and there were slightly fewer owner-occupiers, however the proportion from the wider privately rented sector was much higher. The proportions previously in social rented homes varied very little. A3.20 Among the sub-markets, Salisbury Plain had fewer previous owner-occupiers, and as suggested above more from Armed Services accommodation. In Outer Swindon fewer households that had lived with parents, and more came from owner-occupation. Salisbury contained less households that had previously lived with parents, or in owner-occupation, and more from private rent. Salisbury recorded the highest proportion of households that had been previously homeless. A3.21 Turning to the location of the previous home; in the East a much lower proportion of households (46.4%) had previously lived within the County than in the other two areas (overall 61.3%). More came from a neighbouring area, or from further afield. Many households in Salisbury Plain, unsurprisingly, had come from a distant location and fewer from the County or nearby. Over three quarters of households in Salisbury had moved from within the County area. A3.22 Reasons for households moving are set out in the figure below for each HMA area: 170

Figure A3.2 Reasons for move (past moves) 10.9% Move to cheaper accommodation 12.3% 13.7% 16.3% Previous home too small 16.6% 22.9% 7.5% Previous home too big 11.4% 10.9% 4.6% Access problems 4.9% 1.5% 8.3% Problem with condition of home 7.4% 12.6% 14.2% Relationship breakdown 12.2% 12.6% 8.2% Move to live with partner 11.0% 4.0% 1.1% Move closer to transport 4.2% 3.1% 12.7% Move closer to family 14.7% 14.0% 27.5% Move closer to employment/other facilities 17.5% 15.4% 4.2% Move closer to shops and services 4.2% 4.8% 14.6% Move to better environment 21.1% 24.7% 7.4% Move to safer area 6.6% 7.8% 4.2% Move to school catchment 2.0% 4.7% 6.3% Move to live independently 13.7% 10.7% 1.0% Move to receive care 2.9% 1.7% 0.5% Unable to manage home 5.4% 1.9% 11.7% Eviction/end of tenancy 8.9% 9.5% 0% 5% 10% 15% 20% 25% 30% East West South Source: Wiltshire household survey 2011 171

A3.23 In the East, a move to be close to employment is more common than average; cheaper accommodation, home too big, unable to manage home, and a move to better environment were cited less often than in the other HMAs. Households in the West were more likely to have moved to live with a partner, live independently, or because they were unable to manage. In the South the reasons home too small, house condition, and better environment were higher than average. A3.24 In the Salisbury Plain sub-market a move to join a partner, and for job reasons, were particularly prevalent. In Outer Swindon moves due to relationship breakdown were more common than average. Households in Salisbury were more likely to move because of home condition, or to live independently. Type of home sought A3.25 The survey collected a range of data from households that had recently moved home, not only on the nature of their previous home, but what types of property they considered before acquiring their current home. Analysis of these results is presented in considerable detail in Appendix A4, this section briefly summarises the most important distinctions in these results between the HMAs. A3.26 Looking at the tenures considered prior to moving to their current home, households in the East were less likely than average to consider owner-occupation, and more likely to consider Other (i.e. in the main, Armed Forces accommodation). In the South fewer had considered the social rented sector, and more private rent. Salisbury followed the pattern for the South HMA. Households in Outer Swindon were particularly likely to consider home ownership (especially with a mortgage). A3.27 Turning to the number of bedrooms, households in the East were most likely to consider larger properties (four or more bedrooms ) and least likely to consider a one bedroom dwelling. This pattern was particularly marked in Outer Swindon. Households in the South were most interested in one bedroom accommodation, with relatively few considering three bedrooms. A3.28 The maximum purchase budget considered by moving households varies notably between market areas. Households in the West were the most likely to look at prices up to 150k, and the least likely to consider prices above 250k. Conversely households in the East were much less likely to consider prices under 150k, and much more likely to look at prices above 400k. Households in the South were less likely to look at prices under 200k, and much more likely to look at prices above 250k. Households in Outer Swindon were particularly likely to look at prices above 400k. Turning to the rental budget, the geographical differences are less great between the HMAs and sub-markets. A3.29 Across the County, the actual price households paid for their current home was generally lower than their maximum purchase budget; for instance, whilst 15.9% of households had a maximum budget of 150k, 23.3% of households paid under 150k. The scale of the reduction varied between the HMAs, with the South showing the biggest reductions and the West the smallest. 172

The choice of home A3.30 As well as establishing the reasons households moved from their previous homes (figure A3.2) the survey looked at the reasons households choose their current home. This is presented in the figure below. The figure indicates differences between the HMAs in the relative importance of the reasons for the choice of the present home. In the East, the most important factors were correct number of bedrooms, garden and then good neighbourhood. Number of bedrooms came first in the West, with price/value for money and good neighbourhood close behind, though garden and general condition were also very frequently mentioned. In the South, price/value for money was most important the highest figure in the chart -followed by number of bedrooms and, some way behind, good neighbourhood. A3.31 Looking at the figures, those for the East seem to reflect a narrower range of criteria used by these households, as the total of all reasons (452%) is lower than in the South and, particularly, the West (590%). The figures for the Salisbury Plain sub-area suggest that this may reflect in part the differences in the nature of the dwelling choice between Armed Forces and civilian households. 173

Figure A3.3 Reasons for choice of dwelling (past moves) Price/value for money 34.0% 52.9% 58.6% Correct number of bedrooms 42.2% 53.2% 53.5% Close to family & friends Close to work 22.3% 25.4% 29.3% 32.8% 31.9% 37.4% Close to school Close to services 9.2% 14.9% 16.5% 17.4% 24.4% 21.2% Good transport connections 11.8% 19.7% 22.7% In good neighbourhood 35.3% 41.1% 49.7% In safe area 28.7% 31.5% 37.5% Home met accessibility needs 7.8% 7.4% 12.0% General condition of home 26.7% 35.5% 45.0% Suitable for retirement New build Older home Types of housing 6.5% 8.2% 7.6% 1.8% 6.9% 5.5% 8.1% 12.7% 9.1% 18.6% 16.7% 13.7% Garden 35.0% 37.7% 47.9% Lack of alternatives 15.6% 18.7% 23.6% 0% 10% 20% 30% 40% 50% 60% 70% East West South Source: Wiltshire household survey, 2011 174

Overview A3.32 The survey data confirm that there are differences in the nature of the three market areas. All contain a wide range of household types, and the differences in house type and tenure profile are relatively slight. A3.33 However, the East appears to be influenced to a significant extent by mature households families with children moving into Wiltshire from areas to the east, like London, and often travelling back there to work. At the same time it contains significant numbers of Armed Forces households, whose move and dwelling choice are driven by a different set of influences, and who may move more frequently. Much of the area is rural, with some fine landscape: Marlborough is the most significant urban centre. The northernmost margin of the area looks very much to Swindon, and perhaps also to the travel links nearby. In choosing a house, value for money and the number of bedrooms were most important, perhaps reflecting a comparison with prices elsewhere rather than within Wiltshire (as the housing in the East is relatively expensive for the County). A3.34 In the West there is a considerable diversity of smaller urban centres set in a rural hinterland. In some respects it is the most balanced area, catering for a range of both emerging and mature households, though there is less renting than elsewhere in the County. Property is cheaper than in the other two areas and households moving there had smaller budgets. Households value a range of factors in choosing a home, including bread-and-butter issues like property condition, garden and neighbourhood. A3.35 The South is the most self-contained of the three market areas. More households work within the County and travel a short distance to work then elsewhere. Much of the South is very rural, and it contains more older households, and greater car ownership, Salisbury s role as a significant urban centre shows in its market characteristics. There are more flats, and terraced properties. Salisbury has a considerable population that work locally but also an appreciable commuter outflow from the City. 175

Appendix 4. Search patterns A4.1 This Appendix uses household survey data to look at the search patterns through which households arrive at their eventual home moves. A range of information was collected on the survey form from households that had recently moved home, on the other homes that they had considered moving to. This data shows how households adapt to market conditions in Wiltshire and provides a better understanding of the housing market. Whilst the majority of the information presented in this chapter is based on households moving to their current house, an overview of households planned future moves is also shown. Past moves: size and type of dwelling considered A4.2 The table below compares a households property size (what it achieved) with the property sizes it considered when looking to move. The table shows the proportion of households in each current bedroom size that considered different sized properties when searching for their current home (e.g. 63.7% of recently moving households in Wiltshire that currently live in a one bedroom property, considered two bedroom properties during their search). The percentages in the columns do not total to 100% because many households considered more than one property size in their search. Bedrooms considered Table A4.1 Property size considered (by recent movers) by current property size Bedrooms in current property One Two Three Four Five plus Total One bedroom 96.8% 23.6% 4.7% 0.6% 0.0% 23.7% Two bedrooms 63.7% 92.7% 42.0% 11.1% 0.0% 56.9% Three bedrooms 4.2% 44.0% 85.6% 65.5% 24.5% 55.0% Four bedrooms 2.1% 2.9% 32.5% 90.6% 70.0% 27.5% Five plus bedrooms 1.9% 0.6% 9.1% 20.6% 88.2% 8.9% Total number of households 5,531 13,264 12,946 5,699 1,018 38,458 % movers by current number of bedrooms 14.4% 34.5% 33.7% 14.8% 2.6% 100.0% NB households could give more than one answer Source: Wiltshire household survey 2011 A4.3 The table above shows that around two thirds of recent movers moved to two or three bedroom accommodation. The table also shows that two and three bedroom properties are the most commonly considered by moving households (56.9% and 55.0% respectively). 177

A4.4 It is clear is that many households did not have an exact idea of the number of bedrooms they required when they started looking; with over half of all moving households considering a different size to what they finally achieved. This could perhaps be explained by the wide range of attractiveness of locations: so people would look at smaller dwellings in expensive locations and similarly at larger dwellings in cheaper locations. A4.5 The table shows that households that currently living in two bedroom dwellings were more likely to consider a property with more than two bedrooms than a property with one bedroom (e.g. 44.0% looked at a three bedroom property and 23.6% a one bedroom property). This trend reverses for households living in three or more bedroom properties; these households were more likely to have considered a smaller property to the one they finally achieved rather than a larger one. A4.6 The table below mirrors the data above but shows how the dwelling types considered compare with they type achieved. The table shows the proportion of households in each current dwelling type that considered a range of dwelling types when searching for their current home. The percentages in the columns do not total 100 because many households considered more than one dwelling type in their search. Dwelling types considered Table A4.2 Dwelling types considered (by recent movers) by current dwelling types Detached Semidetached Current dwelling type Terraced Flat Bungalow Total Detached house 94.9% 58.5% 65.1% 27.4% 51.9% 59.4% Semi-detached house 56.4% 93.0% 90.7% 40.7% 34.6% 65.2% Terraced 26.6% 49.3% 91.8% 43.0% 17.1% 48.0% Purpose-built flat 10.5% 15.8% 17.4% 76.6% 9.2% 28.7% Converted flat 11.2% 12.1% 13.7% 57.4% 13.6% 23.2% Maisonette 6.2% 5.5% 9.9% 39.7% 8.8% 15.1% Bungalow 22.1% 27.9% 31.7% 27.9% 89.9% 34.9% Mobile home 3.1% 1.5% 2.9% 6.1% 2.3% 3.3% Total number of households 8,160 8,408 8,089 9,169 4,632 38,458 % movers by current dwelling type 21.2% 21.9% 21.0% 23.8% 12.0% 100.0% NB households could give more than one answer Source: Wiltshire household survey 2011 A4.7 The table above shows that, although only 21.2% of recent movers live in detached houses, nearly 60% of all recent movers had looked at detached dwellings as an option. The table also shows that the semi-detached house was the dwelling type most likely to be considered by a household even though only 21.9% of households ended up living in this type of dwelling. 178

A4.8 Looking at the trends across the table it appears that households consider a range of dwelling types, presumably based on what they perceive they can afford i.e. a household is likely to have considered the next more/less expensive dwelling type to the one that they ended up choosing. For example, a household currently living in a flat was more likely to consider a terraced housing during their previous move than a semi-detached house or detached house. A4.9 Looking at the pattern for terraced and semi-detached houses there appears to be an aspirational trend in the types of housing that household considered. For example, 90.7% of households that currently live in a terraced house considered semi-detached properties (usually more expensive) compared to 17.4% that considered flats (usually less expensive). This suggests that households are more likely to look at properties that would stretch their budget than fall significantly below it. That is consistent with an aspirational approach. Past moves: tenure choice and outcome A4.10 The table below continues in the format of Tables A4.1 and A4.2 but considers tenure. The table shows the proportion of households in each current tenure that considered the various tenures available when searching for their current home. Again the percentages in the columns do not total 100% because many households considered more than one tenure in their search. Table A4.3 Current tenure by tenures considered (by recent movers) Tenures considered Current tenure OO nm OO wm SR PRS Tied Total Owner-occupier (no mortgage) 90.1% 6.1% 0.5% 1.0% 3.3% 11.9% Owner-occupier (with mortgage) 12.1% 96.3% 3.4% 4.5% 10.8% 24.3% Social rented 1.8% 0.0% 88.7% 6.0% 3.8% 18.9% Private rented 2.4% 0.6% 7.3% 94.4% 22.8% 43.1% Shared Ownership 0.3% 2.9% 1.2% 1.3% 1.4% 1.5% Other 0.3% 0.0% 6.1% 0.3% 66.6% 7.5% Total number of households 4,199 7,778 6,862 16,008 3,611 38,458 % movers by current tenure 10.9% 20.2% 17.8% 41.6% 9.4% 100.0% NB households could give more than one answer so that percentages should not be added Source: Wiltshire household survey 2011 A4.11 The private rented sector is typically the most dynamic tenure in any council area, and this is also the case in Wiltshire where over 40% of moves are within or into the private rented sector. A4.12 The overall pattern is more polarised than with other searches: relatively few people looked outside the tenure group in which they ended up this suggests that tenure is a major factor in a household s property choice and one few will compromise on. The exceptions are: 179

Owners with no mortgage mostly looked at the same tenure (90.1%) but 12.1% considered using a mortgage to buy the home. Owners with a mortgage mainly looked at that tenure (96.3%) but 6.1% looked at no mortgage solutions, presumably involving moving to a cheaper home than they finally did. Social renters mainly looked at that tenure (88.7%) but a significant minority (7.3%) considered private renting. Private renters focussed on private renting (94.4%) but 4.5% considered purchasing with a mortgage, and 6% considered social renting. Finally households living in tied accommodation (accommodation linked to the job of a household member, almost all of this being military accommodation in Wiltshire) were the most likely to consider other tenures. Around a fifth of these households considered the PRS and a tenth considered owner-occupation before moving to their current home. Past moves: maximum price and actual purchase price A4.13 The figures in the table below show the maximum purchase price households were prepared to pay when they were looking for their current home, and the price they finally paid. These are distributions and not individual cross-tabulations: they do not show the individual maximum purchase price and the actual outcome. However the data does provide a useful comparison. A4.14 Whilst there is a broad correspondence in the profiles, it appears households generally bought housing for less than the budget they had initially in mind. For example, significantly more households (23.3%) ended up paying less than 150,000, than had budgeted to spend within this figure (15.9%). But broadly the two distributions are similar: expectations coincided with outcomes. Table A4.4 Maximum budget compared to actual price for current home (recent buyers only) Price (000) Maximum purchase budget No. of households % of households No. of households Actual purchase price % of households 50-100 324 2.7% 456 3.8% 100-150 1,601 13.2% 2,333 19.5% 150-200 2,682 22.0% 2,836 23.7% 200-250 2,099 17.3% 2,021 16.9% 250-300 1,627 13.4% 1,009 8.4% 300-350 619 5.1% 748 6.2% 350-400 434 3.6% 587 4.9% 400+ 2,784 22.9% 1,985 16.6% Total 12,170 100.0% 11,977 100.0% Source: Wiltshire household survey, 2011 180

Past moves: maximum rent and actual rent A4.15 The table below considers the same information, but for market rents. The figures show the maximum rental budget for households when they were looking for their current home, and the rent level they finally paid. Again the figures suggest that households acquired cheaper housing than they had originally expected; only 4.4% of households had budgeted to spend less than 350, but some 15.7% of households ended up paying less than this. The difference between the maximum budget and price paid is generally greater in the rental market than that found for home ownership, which suggests that planning for rent expenditure is more difficult than planning for buying. It should be noted that a substantial number of households budgeted for and pay rents at the upper end of the scale (over 600 per month). It is possible that the pattern at the upper end of the market may be different, but the data collected does not permit this to be tested. Table A4.5 Maximum budget compared to actual cost for current home (recent renters only) Monthly rent Maximum rent budget Actual rent price Count % Count % less than 300 482 1.8% 2,174 8.2% 300-350 688 2.6% 1,993 7.5% 350-400 1,451 5.5% 2,428 9.1% 400-450 4,024 15.3% 2,311 8.7% 450-500 1,774 6.7% 1,675 6.3% 500-550 2,843 10.8% 2,016 7.6% 550-600 2,388 9.1% 3,409 12.8% 600+ 12,712 48.2% 10,547 39.7% Total 26,362 100.0% 26,554 100.0% Source: Wiltshire household survey 2011 Search pattern areas A4.16 The household survey asked residents who had moved within the last two years which areas they had considered in their property search. They were able to record up to six different areas. In instances where more than one area was considered, further analysis was undertaken to establish which areas were closely correlated. These links are displayed in Figure 5.1. A4.17 The red arrows indicate two strongly linked areas. These are areas where 40% of more of respondents that chose one area also considered the other. A black border has been used to separate areas which have hardly any links between them. This map aims to identify areas that recent movers in Wiltshire considered similar areas that residents considered suitable in terms of meeting their housing requirement. 181

A4.18 In terms of the patterns that the map shows, the larger western towns are closely linked to their neighbours, indicating it was common for households to consider more than one of these towns when looking for a home. Likewise, Salisbury has a number of strong links with the neighbouring areas which, although are very different in degrees of urban-ness, appear to be considered equally as desirable. A4.19 On the whole the trend is for towns to be connected to their rural surroundings; such as Malborough, Malmesbury, Calne and Devizes. However, this was not true in Warminster which appeared to be more associated with the western towns to the north and not the rural areas in its immediate surroundings. 182

Figure A4.1 Search pattern areas groups in Wiltshire Source: Fordham Research (2011) Reasons for moving and reasons for choosing A4.20 The table below provides detail on the reasons cited by households for moving from their previous move. It shows that the most commonly cited reason for moving from the last home was to move to a better environment. (19.9% of households). 183

Table A4.6 Reasons for moving from previous home (recent movers only) Reasons for move All moving households Count % % pensioner By household type % nonpensioner/ adult only % containing children Move to cheaper accommodation 4,933 12.8% 12.6% 14.3% 10.6% Previous home too small 6,759 17.6% 2.8% 13.9% 27.6% Previous home too big 3,857 10.0% 28.7% 11.5% 2.2% Access problems 1,715 4.5% 13.7% 3.9% 2.6% Problem with condition of home 3,310 8.6% 2.4% 10.1% 8.1% Relationship breakdown 4,898 12.7% 11.4% 12.7% 13.1% Move to live with partner 3,434 8.9% 1.8% 11.2% 7.5% Move closer to transport 1,173 3.1% 5.9% 3.3% 1.9% Move closer to family/friends 5,398 14.0% 30.3% 10.0% 15.5% Move closer to employment/other facilities 7,285 18.9% 0.8% 18.7% 24.7% Move closer to shops and services 1,862 4.8% 12.0% 4.2% 3.7% Move to better environment 7,635 19.9% 13.9% 20.2% 21.1% Move to safer area 2,642 6.9% 1.5% 4.3% 12.4% Move to school catchment 1,195 3.1% 0.0% 0.6% 8.0% To live independently 4,374 11.4% 9.1% 14.5% 7.2% To receive care 796 2.1% 4.9% 2.4% 0.8% Unable to manage home 1,426 3.7% 9.9% 3.4% 2.4% Eviction/end of tenancy 3,752 9.8% 2.9% 9.4% 12.3% Total 38,458 100.0% 100.0% 100.0% 100.0% NB households could give more than one answer Source: Wiltshire household survey 2011 A4.21 The table below provides detail on the reasons cited by households for choosing their current home. The most common reason households choose their home was the correct number of bedrooms (50.3% of households) followed by price/value for money (48.8% of households). 184

Table A4.7 Reasons for choosing current home over others viewed (recent movers only) Reason for choosing home Count % Price/value for money 18,776 48.8% Correct number of bedrooms 19,342 50.3% Close to family & friends 11,350 29.5% Close to work 11,756 30.6% Close to school 5,226 13.6% Close to services 8,536 22.2% Good transport connections 6,779 17.6% In good neighbourhood 16,767 43.6% In safe area 12,916 33.6% Home met accessibility needs 4,196 10.9% General condition of home 14,953 38.9% Suitable for retirement 3,042 7.9% Newbuild 2,179 5.7% Older home 3,984 10.4% Types of housing 6,193 16.1% Garden 16,045 41.7% Lack of alternatives 7,709 20.0% Total 38,458 100.0% NB households could give more than one answer Source: Wiltshire household survey 2011 A4.22 The table below compares the most commonly cited reasons households moved from their previous home (table A4.6) with the most commonly cited reasons that they choose their current home (table A4.7). Table A4.8 Comparing reasons for move with reasons for choosing new home Important reasons for moving from previous home: Move to better environment (20% Move close to employment/other (19%) Previous home too small (18%) Closer to family (14%) Important reasons for choosing home: Bedroom size fits (50%) Value for money (49%) Good neighbourhood (44%) Garden (42%) Condition of home (39%) In safe area (34%) Relatively unimportant: Relatively unimportant: access problems; move closer to transport; move suitable for retirement; newbuild; older home to school catchment; to receive care; or unable to manage home Source: Wiltshire household survey 2011 185

A4.23 In considering these figures it is important to bear in mind that they cover all responses. Thus the fact that moving to a home suitable for retirement is relatively unimportant does not measure importance to the households in question: it may be very important to them, but they are a minority of all moves surveyed. A4.24 The first column focuses upon the push factors driving the household towards considering a move. The second column concerns the pull factors which condition the choice of a new home. If one looks at the six most popular reasons for choosing a new home (second column), they appear to correlate with the four most commonly mentioned reasons for leaving the previous home. The fact that the bedroom size fits in the new home addresses the previous home being too small, while the good neighbourhood and in safe area factors meet the desire to move to a better environment. The two items in the first column that do not appear in the second one are to do with location in relation to work and family. These factors do appear in table A4.7 but are ranked rather lower (seventh and eighth respectively), so there is some evidence that these aims were fulfilled. Satisfaction with housing and life A4.25 Comparing a households reason for moving from their previous home, with the reason for choosing their current home it is possible to identify households that did not meet all their objectives. It can be viewed that the final choice that these households made was limited by the nature of housing available at the time they moved. Overall a quarter (25.1%) of households that moved failed to achieve all their objectives due to a lack of alternative properties available. A4.26 It is useful to consider how these households satisfaction with their housing compares to the satisfaction level amongst other households that have recently moved. This is presented in the table below. It is clear from the table that those that have achieved a sub-optimal dwelling were less satisfied with their home than average - some 17% of these households indicated they were unsatisfied with their home compared to 5% of households whose housing met their objectives. It is worth noting however, that 62% of households that did not achieve their objectives were still satisfied with their current home. Table A4.9 Moving households satisfaction with their housing Level of satisfaction with housing Chose their home due to lack of alternatives: it did not meet all their criteria All other movers within past two years Satisfied 62% 92% Neutral 20% 4% Unsatisfied 17% 5% Total 7,709 30,750 Source: Wiltshire household survey 2011 186

A4.27 The table below repeats this analysis, but focusing on households satisfaction with life in general. The differences are more marked, less than half of those that achieved a sub-optimal dwelling were not satisfied with life in general, while this applied to only 20% of other moving households. In the general. The questions dealing with satisfaction are extremely subjective however, so the results of this analysis should be treated with caution. Table A4.10 Moving households satisfaction with life in general Level of satisfaction with life in general Chose their home due to lack of alternatives: it did not meet all their criteria All other movers within past two years Satisfied 47% 80% Neutral 25% 9% Unsatisfied 27% 11% Total 7,709 30,750 Source: Wiltshire household survey 2011 Planned future moves A4.28 The information in the table below and the succeeding ones is all focussed on planned future moves, which are of course aspirational. The information therefore should be regarded as rather more indicative than the actual evidence of past moves discussed above. A4.29 The table below shows the reasons for moving cited by households planning a future move. The data suggests that planned future moves show consistency with the reasons for past moves: the main strategic factors (environment, work and friends/family) arise as well as the locally specific one of home too small. So this finding reinforces the importance of these overall reasons. Issues that are viewed as important in public discussion, such as move to school catchment attract relatively low numbers in both past (especially) and future planned moves. Again that does not mean they are unimportant, but rather that they affect a minority of households. 187

Table A4.11 Reasons for planned future move (moving households only) Reason for planned future move All moving households Count % % pensioner By household type % nonpensioner/ adult only % containing children Move to cheaper accommodation 8,615 19.6% 22.2% 21.2% 16.8% Previous home too small 13,677 31.0% 7.5% 27.7% 42.1% Previous home too big 5,473 12.4% 37.4% 12.4% 4.9% Access problems 2,511 5.7% 19.4% 4.0% 3.6% Problem with condition of home 3,365 7.6% 7.9% 7.1% 8.2% relationship breakdown 2,049 4.6% 1.5% 5.3% 4.8% Move to live with partner 2,314 5.3% 1.6% 8.8% 2.0% Move closer to transport 2,860 6.5% 10.4% 7.1% 4.5% To move closer to family 6,980 15.8% 25.7% 14.8% 14.1% Move closer to employment/other facilities 9,283 21.1% 6.4% 27.9% 17.2% To move closer to shops and services 3,791 8.6% 18.4% 7.8% 6.6% Move to better environment 6,639 15.1% 12.8% 15.9% 14.8% Move to safer area 3,235 7.3% 5.0% 7.3% 8.0% Move to school catchment 2,099 4.8% 1.3% 1.0% 10.3% To live independently 1,250 2.8% 10.9% 1.2% 2.4% To receive care 898 2.0% 13.1% 0.5% 0.6% Unable to manage home 2,607 5.9% 19.6% 5.1% 2.8% Eviction/end of tenancy 3,615 8.2% 1.3% 7.3% 11.4% Total 44,064 100.0% 100.0% 100.0% 100.0% NB households could give more than one answer Source: Wiltshire household survey 2011 A4.30 The table below shows the size of accommodation moving households would like and expect. As is normal, the data shows that people would like a larger home than they expect to get, or be able to afford. This is a general finding, and it is likely to be a true reflection of the way in which the market operates. 188

Table A4.12 Number of bedrooms preferred and expect in future home (moving households only) Like Expect Count % Count % One 2,279 5.2% 4,212 9.6% Two 11,422 25.9% 16,150 36.7% Three 18,452 41.9% 16,721 37.9% Four 10,240 23.2% 6,003 13.6% Five 1,352 3.1% 953 2.2% Six 320 0.7% 25 0.1% Total 44,064 100.0% 44,064 100.0% Source: Wiltshire household survey 2011 A4.31 The table below shows the tenure moving households would like and expect. The difference between the two is typical - more households would like to own than would expect to with the reverse true for private renting. Similar proportions would like to social rent (from Council or HA) as expect to do so, which is usual in an area where the social rented sector is in good condition. Table A4.13 Preferred and expect tenure of future home(moving households only) Like Expect Count % Count % Buy own home 27,300 62.0% 18,141 41.2% Rent from Council 8,111 18.4% 8,117 18.4% Rent from HA/RSl 3,404 7.7% 5,202 11.8% Rent in PRS 2,311 5.2% 8,399 19.1% Rent form friend/relative 56 0.1% 299 0.7% Armed service accommodation 1,236 2.8% 1,960 4.4% Tied/linked to job 418 0.9% 523 1.2% Shared Ownership 507 1.2% 290 0.7% House/flat share 143 0.3% 263 0.6% Other 577 1.3% 871 2.0% Total 44,064 100.0% 44,064 100.0% Source: Wiltshire household survey 2011 A4.32 This information provides useful information for the Council to consider when planning the mix and type of housing to build in the future the current mix is clearly not the perfect mix. A4.33 The table below shows the type of property moving households would like and expect. As would be anticipated more households would like a detached property than expect one, with semi-detached and 189

terraced houses and also flats expected by more households than would prefer these accommodation types. Table A4.14 Preferred and expect dwelling type of future home (moving households only) Like Expect Count % Count % Detached 27,525 62.5% 13,292 30.2% Semi-detached 8,079 18.3% 14,400 32.7% Terraced 1,934 4.4% 6,348 14.4% Flat/apartment 2,724 6.2% 6,673 15.1% Maisonette 0 0.0% 237 0.5% Bungalow 3,802 8.6% 3,026 6.9% Mobile home 0 0.0% 88 0.2% Total 44,064 100.0% 44,064 100.0% Source: Wiltshire household survey 2011 Conclusions A4.34 Comparing the push reasons for deciding on a move with the pull reasons that resulted in the choice of a particular home indicates that most of the reasons for moving, at the global level, were indeed met. A4.35 When comparing the nature of housing considered with the nature of housing achieved it is clear that households were most flexible over the type of accommodation they would live in and least flexible on the property tenure. There is a notable degree of flexibility over the accommodation size households often looked at property sizes both bigger and smaller than what they achieved, but those that moved into larger houses searched more widely across the size range, presumably because they could afford to. A4.36 In relation to price/rent paid, there was a broad agreement between intended expenditure and the level that finally occurred, but there was a noticeable minority which paid less than they had originally budgeted for. 190

Appendix 5: Estate Agents Observations The following is a summary of the Agents comments and observations, firstly generally and then on a town by town basis: Wiltshire wide sales Market i) Three bed homes with parking and garden are in greatest demand. ii) Market is picking up after stagnation, and discounts from asking prices are declining. iii) Prices not rising but pressure is coming from vendors to increase asking prices iv) Mortgages are available for people with equity, but not others. v) First time buyers need parents help. vi) Very mixed buyers: 50% local 50% from elsewhere Wiltshire wide lettings Market i) Tenants are staying longer and renewing their tenancies some have now been in the same house for three years. For a good tenant, landlords are often willing to agree for the rent to remain unchanged at renewal to avoid a void ii) There is a real mix of tenants, including retired people. iii) Buy-To-Let (BTL) investors are coming back into the market, though slowly. If LHA were direct and paid monthly (not 4 week period) more properties would be available. iv) No interest was shown in a letting agents or landlord s forum v) Some BTL investors are coming back; flats are favoured due to less maintenance. vi) Rents felt to be steady and market steady few landlords buying or selling. Chippenham i) Supply of rental property rose last year but is now declining. ii) There is a wide mix of tenants, including retired people. iii) Good schools, M4 and relatively low prices important factors bringing buyers to Chippenham. iv) Prospective village buyers are coming from the outer Home Counties, and from London 191

v) Persimmon site (Brooklands) is having impact on the town, and very large discounts have reportedly been offered (up to 40%). vi) Scope may exist for more retirement housing in the town. vii) Interest in more smaller, less expensive three bed detached dwellings with garden and parking. viii) More three and four bed larger homes are needed in the villages Wootton Bassett i) Good new homes are being purchased as BTL. New homes rent very fast, and can be gone in a day ii) Area is linked very closely to Swindon iii) Demand from LHA backed tenants has increased; a preference to have rent paid direct to landlords was underlined iv) Need is for more two and three bed town houses and proper family houses with garden v) Whilst Persimmon was reportedly discounting 285k down to 210k at Station Rd, David Wilson on the same site has more consistent pricing. vi) Swindon and the Swindon market are key in the town, which is essentially a suburb of Swindon Marlborough i) Crest Nicholson are currently developing St Johns; prices range from 155k above garages to nearly 600k for five bed homes. Though no big discounts are available, homes are selling. ii) M4 is the crucial factor, more important than employment in Swindon or Newbury iii) A quite distinct market is associated with Newbury; Pewsey is not part of it. There are lots of commuters to Reading iv) Large numbers of downsizer retirees seeking homes, though no need for more retirement schemes, at present anyway. 192

Warminster i) Market includes many retirees. Warminster has a reputation as a retirement location, which may reflect the lower prices here. There is scope for some retirement schemes for older buyers. Considerable demand is believed to exist for bungalows. ii) MoD families are a strong feature of the market ; all ranks and at all levels iii) Rents are rising slowly. Even so there is a need for more 3 bed rent units iv) Persimmon scheme is holding prices down at the moment, though that is temporary v) Town is perceived to have the lowest prices in Wiltshire vi) The development at Victoria Fields is commanding a price premium Salisbury i) London is a strong influence through commuting, and others are travelling to Southampton ii) Little interest is coming from investors at the moment iii) Strong interest is being shown from downsizers and the retired iv) Currently there are too few furnished lettings v) Rents in the City are at similar levels to villages and rural area vi) There is a good base of long term renters, some five years or more vii) Strong demand exists for three to four bed houses (not flats) in the City viii) Most people work locally, and Porton Down is an important employer ix) A real shortage currently of three bed units with a decent garden and two parking places x) Super Garrison making no impact on sales market (what is it?) Tidworth no difference v Tidworth and Super Garrison big influence on lettings within the 10 mile catchment xi) Around a third of new build is going to BTL xii) Strong market links with Dorchester Andover MOD and Oxford/Reading Amesbury i) The Salisbury Plain market is felt to be quite different, with close links to the military 193

ii) Prices are still dropping, reflecting completions from Bloor and Persimmon. Persimmon properties are hugely discounted with houses marketed at 475k selling at around 300k. iii) Probably around 65% of buyers are local iv) The rental market is steady Devizes i) Village prices are high reflecting high demand. ii) The requirement for deposits is seen as the only factor holding people back iii) Significant demand is coming from cash buyers downsizing iv) Strong market activity is currently centred on Persimmon, Crest and Redrow. All three are felt to be over-pricing and then discounting, and a realistic price at the outset was suggested. v) Significant interest was shown from military buyers vi) Call to pay rent direct and attract more properties was echoed Westbury i) Although prices slipped at the end of 2010 they were now holding ii) There is little new build at present but new family housing would attract interest. iii) Super Garrison has had no visible effect, and has not so far attracted more military buyers Corsham i) The town displays strong demand from retirement buyers, reflecting lower prices than other towns, and a charming location. There may be a need for a McCarthy & Stone type of scheme ii) Cotswold Homes scheme went very well, confirming that there would be demand for new build iii) The Corsham market is closely linked to Bath and Chippenham; many buyers want to live in Bath but can t afford to. It is effectively a suburb of Bath, but 20% less expensive 194

Malmesbury i) Three bed homes with parking, within walking distance of shops, is the category in highest demand ii) The town is closely linked with Gloucestershire markets, and also M4 and M5 travel links, giving good access to Bristol, London, Reading, Swindon, Bath iii) Short term contracts at Dyson are important for the local letting market, but can sometimes be given too much weight iv) Buyers and renters want easy living clean property with parking, small garden and so on v) Many older buyers are moving in from the villages vi) Prices now felt to be back to the market peak level Melksham i) There is a considerable over-supply of older people s houses. Many are on the market ; they are nice units but unsellable due to high service charges ii) Further retirement provision is not required with so many currently unsold. Properties built in 1988 and sold for 65k peaked at 120 k but are now back down to 70k reflecting service charge level. One scheme has 12 out of total of 50 units on the market at the moment. iii) There is probably a need for retirement schemes in Trowbridge and Bradford, but not in Melksham. About half of prospective buyers are downsizers or pre-retirement iv) Lots of forces personnel are relocating currently v) There is strong demand from older buyers for Bungalows vi) Persimmon / Bloor and Charles Church are active in the town. Most purchases involve part exchange meaning values are somewhat unclear vii) MOD in Corsham is an important employer, even with ongoing changes 195

Appendix 6: Stakeholder discussion of the report s findings A6.1 Stakeholder discussions are an integral part of an SHMA. An initial meeting was held in early 2011 where the aims of the study were set out and the proposed methodology discussed. The main outcome was that some pressure for higher housing targets was expressed. By the time of the second consultation event the Council had published (June 2011) its Topic Paper 17: Housing Requirement Technical Paper. This details the discussions with communities that led to the lower target of about 27,000 for the period 2011-26 (in relation to the ONS household projection of 34,000 for the same period). A6.2 The second stakeholder event contained a presentation outlining the results of the study and then an open-floor discussion of them. The meeting seemed to accept the new total housing requirement set out in the Council s Topic Paper and the analysis in the draft report. As a consequence the discussion was quite brief. Minutes of Wiltshire and Swindon Housing Market Partnership meeting: item on Wiltshire SHMA A6.3 The meeting was held at 10.00 am on Monday 25th July 2011 at Wiltshire s Monckton Park offices in Chippenham. The speakers are identified in the list of attendees which follows this summary. A6.4 Dr Richard Fordham from Fordham Research gave a presentation and answered questions on the results from the work that has been done to produce the Wiltshire SHMA. The presentation overheads, based on the study, were circulated to the participants. A6.5 Mark Fox asked if the report would have a separate section covering the South Wiltshire Core Strategy area. Dr Fordham explained that the areas in report are based on where people search for houses not on the Core Strategy areas. The data for this came from the 6000 survey responses; this data is broken down by Community Areas. A6.6 Amanda Castellino asked if the numbers in Table 9.7 in the presentation (8.7 in the current report) included everyone on the Housing Register. Dr Fordham confirmed that the table showed the total number of people currently on the housing register; and that a question in the survey had asked if people were on the housing register to provide the data that was used. A6.7 Amanda Castellino also asked for clarification of the term 50% relets in Table 6.14 (5.14 in the current report). Dr Fordham explained that it meant that 50% of social housing was relet as affordable rent. 197

A6.8 Neil Tiley made everyone aware that the boundaries of the proposed Housing Market Areas are tied to the Community Area boundaries. A6.9 Mark Fox queried the extent of the Swindon sub market area given that the whole of north Wiltshire had been included in the previous Swindon Housing Market Area. Dr Fordham explained that the survey data did not show that there was such a large influence for Swindon on the rest of Wiltshire. Amanda Castellino confirmed that the work that Swindon Borough Council has carried also supports this finding; and that Swindon is relatively self contained with 80% of households in Swindon looking to stay in Swindon. Dr Fordham commented that in the original HNS for all then 5 districts of Wiltshire in 1993, it had been surprising to find the small relationship between Swindon and the surrounding Wiltshire districts: it is clearly a long established pattern A6.10 John Owen asked if affordability has been calculated on just income or if savings have also been included. Dr Fordham confirmed that a full assessment of affordability has been carried out. List of attendees at the Partnership Meeting James Stevens John Owen Alexander Nell Jamie Lewis Paul Davis Mark Fox Dan Washington Hannah Watkins Neil Hall Maggie Wood Paul Walsh Matt Richardson Suzy Birdseye Andy Birch Bob Hillman Phil Smith Amanda Castellino Janet O Brien Georgina Clampitt-Dix Chris Minors Neil Tiley Jane Macey Mary Noyce House Builders Federation Green Square Group Carter Jonas Hunter Page Planning Persimmon Pegasus Planning Group GL Hearn Aster Group AMEC DPDS Consulting Selwood Housing Gleeson Land Hallam Land Management Hallam Land Management Swindon Borough Council Swindon Borough Council Swindon Borough Council Wiltshire Council New Housing Wiltshire Council Spatial Planning Wiltshire Council Spatial Planning Wiltshire Council Spatial Planning Wiltshire Council Spatial Planning Wiltshire Council Spatial Planning 198

Appendix 7: Questionnaire 199

Dear Resident, Wiltshire Council is carrying out an important study of housing in the local area and would be grateful for your help. Households across Wiltshire are being asked to take part in a survey that will give the Council up to date information about the housing circumstances and aspirations of local people. This information will be used to help the Council develop its housing and planning policies to best meet these needs in the coming years, including how much housing is actually needed, and of what sizes and types. Your household has been chosen at random to be included in this survey and I would be very grateful if you would take a few minutes to complete the questionnaire and return it in the prepaid envelope. Even if you are not planning to move, the information is important in order to help us plan for the future provision. We are interested in getting a full range of views, from home owners as well as those who rent from a private landlord or the Council. The data is being collected and analysed on behalf of Wiltshire Council by Fordham Research. All of the information you give on this form will be treated as strictly confidential and will not be linked with your name or address, or to any other database. The Council will not see any of your individual replies, and nor will they be passed on to any other agency or market research organisation. To thank you for taking part, every household that responds will be entered into a 100 prize cash draw. The ID number at the top of the form will identify your address only to Fordham Research who will contact you if you win. Please return the completed questionnaire in the pre-paid envelope or send to FREEPOST RSHL- CCKZ-JSTX, Fordham Research, 64 Ship Street, Brighton, BN1 1AE by 5 th March 2011. Finally, I would like to take the opportunity to draw your attention to the other services we offer for housing support. If you would like to receive information about joining the Housing Register or about adaptations to your home, please ring Wiltshire Council on 0300 456 0100. Yours faithfully, Alistair Cunningham Director of Economy and Enterprise Notes for completing the questionnaire Most questions are answered by ticking the appropriate box: 01 For some questions you need to write a number in a larger box: people Please follow the questions in the numbered order unless the box you tick has a Go to question XX next to it, where you will need to go to the question stated. Most of the questions are about your household. A household is made up of either: One person living alone, or A group of people (not necessarily related) living at the same address with common housekeeping - so sharing at least one meal a day, or sharing a living/sitting room. If you would like help with any of the questions or want to discuss the questionnaire, please telephone Sharon Eusebe at Fordham Research on the free phone number: 0800 163 231. Questionnaires in large print are also available on request. 1

A YOUR HOUSEHOLD AND HOME A1. What type of property do you and your household live in? Detached... 1 Converted flat / apartment... 5 Semi-detached... 2 Maisonette... 6 Terraced (including end-terraced)... 3 Mobile home, caravan or other mobile structure... 7 Purpose built flat / apartment... 4 Other... 8 A2. Is your property a bungalow? Yes... 1 No... 2 A3. Is your property sheltered or supported accommodation? E.g. for older people or people with a support need No... 1 Yes (sheltered)... 2 Yes (supported)... 3 A4. Do you own or rent the home? Own outright... 1 Rent from a relative or friend of a household Own with a mortgage or loan... 2 member... 6 Rent from the Council / other social rented... 3 Tied or linked to a job... 7 Rent from a private landlord or letting agency... 4 Shared Ownership (part rent/part buy, HomeBuy). 8 Armed Services Accommodation... 5 Other... 9 A5. Is this property a second home (i.e. this is not your main residence)? Yes, it is a second home... 1 No, it is my main residence... 2 A6. How many bedrooms does your home have? bedroom(s) A7. a) How many people (including you) currently live in your household? b) How many couples are there? a) people b) couple(s) A8. Please enter information for each person currently living in your household (including you) in the table below. (If there are more than six household members, please give details for the six oldest members). Person Relationship to you (See list below) Age Sex (M/F) Working Status (See list below) Qualification Level (See list below) Support needs (See list below) Example 1 37 F 2 5 3, 6 1 You 2 3 4 5 6 Relationship Working Status Qualification level (select highest level achieved) Support need (select all that apply) 1) Living together as 1) Retired 1) No qualifications 1) No support needs a couple/married 2) Full-time employment 2) NVQ level 1 2) Frail elderly 2) Son or daughter 3) Part-time employment 3) GCSEs / NVQ level 2 3) A medical condition 3) Other relative 4) Self-employed 4) A levels / NVQ level 3 4) A physical disability 4) Friend 5) Unemployed 5) Higher education 5) A learning difficulty 5) Lodger 6) Full-time student (16+ years) below degree 6) A mental health problem 6) Other unrelated 7) Looking after home/family 6) Degree or equivalent 7) A sensory disability 8) Permanently sick/disabled 7) Postgraduate 8) Other 9) School/preschool/nursery professional qualification 10) Other A9. Do you share a kitchen and/or bathroom with anyone not listed in the table above? Yes... 1 No... 2 2

A10. Please tick whether the following are no problem, a problem, or a serious problem. Not a problem/ A A serious not applicable problem problem a) The health of someone in your household is suffering because of condition of your home 1... 2... 3 b) Your home is in major disrepair... 1... 2... 3 c) You have difficulty maintaining your home... 1... 2... 3 d) You have to share a bathroom/toilet/kitchen with another household... 1... 2... 3 e) You are lacking basic facilities (such as bathroom/toilet/kitchen)... 1... 2... 3 f) Your accommodation is too expensive... 1... 2... 3 g) You are under notice of eviction/re-possession, real threat of notice, or your lease is coming to an end... 1... 2... 3 h) Someone in your household is suffering harassment or threats of harassment from neighbours or others living in the vicinity... 1... 2... 3 i) Someone in your household has difficulty using stairs and/or lifts to or within your home.. 1... 2... 3 j) You need to be close to a relative/friend to give and/or receive care... 1... 2... 3 k) A household member is unable to get employment locally... 1... 2... 3 l) A household member is unable to access other important facilities... 1... 2... 3 m) Public transport provision near home does not provide for your household s needs... 1... 2... 3 n) Your accommodation is temporary, or on a tenancy which you feel is too insecure... 1... 2... 3 A11. If you or someone in your household has a support need, how could the accommodation or services for your (or their) need best be improved? Please also say if you already have these improvements. Need No Already Need have a) Need to move to alternative housing with specialist adaptations or care/support... 1... 2 b) Lift/Stair lift... 1... 2... 3 c) Extra handrails... 1... 2... 3 d) Other alterations to improve accessibility (e.g. ramps to front door)... 1... 2... 3 e) Downstairs toilet... 1... 2... 3 f) Low level shower unit... 1... 2... 3 g) Other alterations to the bathroom/toilet... 1... 2... 3 h) Alterations to the kitchen (e.g. low level units)... 1... 2... 3 i) Emergency alarm... 1... 2... 3 j) Car parking space near to front door of home... 1... 2... 3 k) More support services to your present home (such as a home carer, meals on wheels etc)... 1... 2... 3 l) Need support managing finances, completing forms etc... 1... 2... 3 m) Need help maintaining home e.g. handyperson service... 1... 2... 3 A12.How many cars or vans are owned or can be used by one or more members of your household? None... 1 Two... 3 One... 2 Three or more... 4 A13. How easy is it for you to get to each of the following? Easy Quite difficult Very difficult a) Grocery shops... 1... 2... 3 b) Post office... 1... 2... 3 c) Bank/building society... 1... 2... 3 d) Leisure facilities... 1... 2... 3 e) Doctor... 1... 2... 3 f) Hospital... 1... 2... 3 g) Schools/educational or training facilities... 1... 2... 3 h) Town (or City) Centre... 1... 2... 3 A14. Where do you and your partner (if applicable) work? Please refer to the map enclosed and fill in the box below with the number of the area in which you / your partner work, or tick the appropriate box. a) Self b) Partner a) Self b) Partner In an area marked on the map... 1. 1 Elsewhere (not on map)... 4... 4 Work from home... 2... 2 Not in work... 5... 5 Mobile worker... 3... 3 No partner... 6 A15. How far in miles do you and your partner (if applicable) travel to work? a) Self miles b) Partner miles 3

B PREVIOUS MOVES & ACCOMMODATION B1. When did you move to your current home? Within the last year... 1... Go to question B2 1 to 2 years ago... 2... Go to question B2 2 to 5 years ago... 3... Go to question C1 Over 5 years ago/always lived here... 4... Go to question C1 B2. What was your last home? Living with parents, relatives or friends... 01 Armed Services Accommodation... 06 Owner occupied (with / without mortgage)... 02 Tied or linked to a job... 07 Rented from a Council / Housing Association... 03 Shared Ownership... 08 Rented from a private landlord... 04 Lodger... 09 Rented part of a shared house or flat... 05 Previously homeless or in temporary accommodation. 10 B3. Where was your last home? Please refer to the map enclosed and enter the appropriate number or tick the appropriate box. In an area marked on the map... 1 Elsewhere in the UK / Abroad... 2 The following questions refer to how you chose your current home. B4. Why did you move from your previous home? Please tick as many as apply. To move to cheaper accommodation... 01 To live closer to employment or other facilities... 10 Previous home was too small... 02 To move closer to shops and services... 11 Previous home was too big... 03 To move to a better environment... 12 Access problems (e.g. steps, stairs)... 04 To move to a safer area... 13 You had a problem with the condition of your home... 05 To move into a school catchment area... 14 Relationship breakdown... 06 To live independently... 15 To move to live with partner... 07 To receive higher levels of care... 16 To move closer to transport links... 08 Unable to manage in previous home... 17 To move closer to friends/family... 09 Eviction, end of tenancy or re-possession... 18 B5. What types of property did you consider when you were thinking of moving? Please tick as many as apply. Detached house... 1 Converted flat / apartment... 5 Semi-detached house... 2 Maisonette... 6 Terraced (including end-terraced)... 3 Mobile home, caravan or other mobile structure... 7 Purpose built flat / apartment... 4 Bungalow... 8 B6. Did you plan to buy or rent this home? Please tick as many as apply. Own with a mortgage or loan... 1 Private rent... 4 Own outright... 2 Shared Ownership... 5 Social rent... 3 Other... 6 B7. What sizes of accommodation did you consider? Please tick as many as apply. One bedroom... 1 Four bedrooms... 4 Two bedrooms... 2 Five or more bedrooms... 5 Three bedrooms... 3 B8. Which areas did you consider? Please refer to the map enclosed and fill the boxes below with the number of up to six areas that you considered. 1 2 3 4 5 6 Elsewhere in the UK / Abroad (not on map)... 7 4

B9. a) When looking for your current home, what was your initial maximum budget? sale price OR rent per month b) How much did your current property actually cost? sale price OR rent per month B10. Why did you choose your current home over others that you considered? Please tick as many as apply. Price / value for money... 01 Home met accessibility needs... 10 Correct number of bedrooms... 02 General condition of home... 11 Close to friends/family... 03 Suitable for retirement... 12 Close to work... 04 Newly-built home... 13 Close to school... 05 Older home... 14 Close to services... 06 Type of housing (e.g. semi-detached)... 15 Good transport connections... 07 Garden... 16 In a good neighbourhood... 08 Lack of alternatives... 17 In a safe area... 09 Other (please state)... 18 C FUTURE HOUSING INTENTIONS C1. Does your household need and/or is it likely to move to a different home... Now / as soon as possible... 1... Go to question C2 Within a year... 2... Go to question C2 In 1 to 2 years... 3... Go to question C2 In 2 to 5 years... 4... Go to question D1 No need / not likely to move... 5... Go to question D1 C2. How many bedrooms would you a) LIKE and b) EXPECT in your next home? a) Like bedroom(s) b) Expect bedroom(s) C3. What are the main reasons for moving to a different home? Please tick as many as apply To move to cheaper accommodation... 01 To live closer to employment or other facilities... 10 Your home is too small... 02 To move closer to shops and services... 11 Your home is too big... 03 To move to a better environment... 12 Access problems (e.g. steps, stairs)... 04 To move to a safer area... 13 You have a problem with the condition of your home. 05 To move into a school catchment area... 14 Relationship breakdown... 06 To live independently... 15 To move to live with partner... 07 To receive higher levels of care... 16 To move closer to transport links... 08 To move closer to friends/family... 09 Unable to manage in current home... 17 Eviction, end of tenancy or repossession... 18 Other (please state)... 19 C4. Please state where you would a) most LIKE to move to, please also state where you b) EXPECT to move to? Please enter the appropriate area number from the enclosed map or tick another option as appropriate. a) Like b) Expect a) Like b) Expect In an area marked on map... 1... 1 Elsewhere in the UK / Abroad... 2... 2 5

C5. What type of housing: a) would you LIKE to move to and b) would you EXPECT to move to? Please tick one response for a) Like, and one response for b) Expect. a) Like b) Expect a) Like b) Expect Buy own home... 01... 01 Armed Services Accommodation... 06... 06 Rent from the Council... 02... 02 Tied-linked to a job... 07... 07 Rent from a Housing Association... 03... 03 Shared Ownership... 08... 08 Rent from a private landlord or letting agency. 04... 04 House/ flat share in the private rented sector.. 09... 09 Rent from a relative or friend of a Other... 10... 10 household member... 05... 05 C6. What type of property: a) would you LIKE to move to and b) would you EXPECT to move to? Please tick one response for a) Like, and one response for b) Expect. a) Like b) Expect a) Like b) Expect Detached... 1... 1 Maisonette... 5... 5 Semi-detached... 2... 2 Bungalow... 6... 6 Terraced (including end-terrace)... 3... 3 Mobile home, caravan or other Flat / apartment... 4... 4 mobile or temporary structure... 7... 7 C7.Would you a) like and b) need to move to sheltered or supported accommodation? E.g. for older people or people with support needs a) Like b) Need a) Like b) Need Yes (sheltered)... 1... 1 No (unspecialised housing)... 3... 3 Yes (supported)... 2... 2 D REQUIREMENTS OF FUTURE HOUSEHOLDS D1. Will any of the other people in your household need and/or are they likely to move to their own separate accommodation within the next five years? For example, a son or daughter, a parent, etc. Yes... 1... Go to question D2 No... 2... Go to question E1 D2. For each person that needs and/or is likely to move in the next five years, please give an indication of when they will need separate accommodation? Person 1 Person 2 Person 3 Now / as soon as possible... 1... 1... 1 Within a year... 2... 2... 2 In 1 to 2 years... 3... 3... 3 In 2 to 5 years... 4... 4... 4 Questions D3-D7: please answer for the person who is most likely to move home first. D3. What is their relationship to the head of household? Son / daughter... 1 Friend... 4 Parent / grandparent... 2 Other... 5 Other relative... 3 D4. Are they: Single adult(s) without children... 1 Couple with, or expecting child(ren)... 4 Single adult(s) with, or expecting child(ren)... 2 Other... 5 Couple without children... 3 D5. How many bedrooms would they a) LIKE and b) EXPECT in their next home? a) Like bedroom(s) b) Expect bedroom(s) D6. Please state where they would a) most LIKE to move to, and also state where they b) EXPECT to move to? Please enter the appropriate area number from the enclosed map or tick another option as appropriate. a) Like b) Expect a) Like b) Expect In an area marked on map... 1... 1 Elsewhere in the UK / Abroad... 2... 2 6

D7. Would they a) LIKE and b) NEED to move to sheltered or supported accommodation? e.g. for older people or people with support needs. a) Like b) Need a) Like b) Need Yes... 1... 1 No... 2... 2 E FURTHER HOUSEHOLD INFORMATION Any financial or personal information you give in this section will be treated in the strictest confidence, and will not be linked to your name or address, or linked to other databases. This information is valuable to us it is used to predict housing market demand, as well as measure the need for affordable housing. E1. Please tick below to show your weekly or monthly mortgage costs or the total rent for your dwelling. If bedrooms are rented separately, please give the rent for your room only. Please include any service charges. Week Month Week Month None... None... 01 120-129... 522-564... 08 Under 70... Under 304... 02 130-139... 565-608... 09 70-79... 304-347... 03 140-149... 609-651... 10 80-89... 348-390... 04 150-174... 652-760... 11 90-99... 391-434... 05 175-199... 761-869... 12 100-109... 435-477... 06 200-229... 870-999... 13 110-119... 478-521... 07 230 or more... 1,000 or more... 14 E2. If you are renting, is the rent you have shown above for the whole dwelling, or just for your bedroom? Whole dwelling... 1 Your bedroom (and use of shared facilities)... 2 E3. If you are renting, does the rent you have shown above include utility bills? (e.g. electricity, gas, water) Yes... 1 No... 2 E4. Is Housing Benefit or Local Housing Allowance used to cover some / all of the cost of your rent? None... 1 Some... 2 All... 3 E5. Please tick the total amount you receive in state benefits, allowances and tax credits (e.g. Child Benefit, Child Tax Credit, Council Tax Benefit). Please do not include Housing Benefit or Local Housing Allowance. Week Month Week Month None / under 50... Under 217... 01 300-319... 1,304-1,390... 08 50-74... 217-325... 02 320-339... 1,391-1,477... 09 75-99... 326-434... 03 340-359... 1,478-1,564... 10 100-149... 435-651... 04 360-379... 1,565-1,651... 11 150-199... 652-869... 05 380-399... 1,652-1,738... 12 200-249... 870-1,086... 06 400-499... 1,739-2,173... 13 250-299... 1,087-1,303... 07 More than 500... 2,174 or more... 14 E6. Please tick the total gross income (before tax) of your household, including income from pensions, employment or investments (but not benefits). Please combine income for all members of the household. Month Year Month Year Under 208... Under 2,500... 01 2,167-2,291... 26,000-27,499... 17 208-416... 2,500-4,999... 02 2,292-2,416... 27,500-28,999... 18 417-541... 5,000-6,499... 03 2,417-2,541... 29,000-30,499... 19 542-666... 6,500-7,999... 04 2,542-2,666... 30,500-31,999... 20 667-791... 8,000-9,499... 05 2,667-2,791... 32,000-33,499... 21 792-916... 9,500-10,999... 06 2,792-2,916... 33,500-34,999... 22 917-1,041... 11,000-12,499... 07 2,917-3,124... 35,000-37,499... 23 1,042-1,166... 12,500-13,999... 08 3,125-3,332... 37,500-39,999... 24 1,167-1,291... 14,000-15,499... 09 3,333-3,541... 40,000-42,499... 25 1,292-1,416... 15,500-16,999... 10 3,542-3,749... 42,500-44,999... 26 1,417-1,541... 17,000-18,499... 11 3,750-3,957... 45,000-47,499... 27 1,542-1,666... 18,500-19,999... 12 3,958-4,166... 47,500-49,999... 28 1,667-1,791... 20,000-21,499... 13 4,167-6,249... 50,000-74,999... 29 1,792-1,916... 21,500-22,999... 14 6,250-8,332... 75,000-99,999... 30 1,917-2,041... 23,000-24,499... 15 8,333-12,499... 100,000-149,999.. 31 2,042-2,166... 24,500-25,999... 16 Over 12,500...Over 150,000... 32 7

E7. Please tick your total net household savings or debt (excluding mortgage debt). More than 20,000 in debt... 01 2,500-4,999... 07 30,000-39,999... 13 10,000 20,000 in debt... 02 5,000-9,999... 08 40,000-49,999... 14 5,000-9,999 in debt... 03 10,000-14,999... 09 50,000-99,999... 15 Up to 5,000 in debt... 04 15,000-19,999... 10 100,000-149,999... 16 No savings... 05 20,000-24,999... 11 150,000-199,999... 17 1-2,499... 06 25,000-29,999... 12 200,000 or more... 18 E8. If you are an owner-occupier please estimate a) the value of your home and b), please also tick how much money (equity) you estimate you would get if you sold your home now, after paying off any remaining mortgages and other related debts. Please tick one response for a) Value, and one response for b) Equity. a) Value b) Equity a) Value b) Equity In negative equity... 01 40,000-49,999... 09... 09 No equity (no mortgage remaining)... 02 50,000-99,999... 10... 10 Less than 10,000... 03... 03 100,000-124,999... 11... 11 10,000-14,999... 04... 04 125,000-149,999... 12... 12 15,000-19,999... 05... 05 150,000-199,999... 13... 13 20,000-24,999... 06... 06 200,000-249,999... 14... 14 25,000-29,999... 07... 07 250,000-299,999... 15... 15 30,000-39,999... 08... 08 300,000 or more... 16... 16 E9. Is your household on the Council s Housing Register or a Housing Association Waiting / Transfer List? Yes... 1 No... 2 E10a. Overall, how satisfied would you say you were with your current housing? Very satisfied... 1 Neither / nor... 3 Very dissatisfied... 5 Fairly satisfied... 2 Fairly dissatisfied... 4 E10b. All things considered, how satisfied are you with your life as a whole nowadays? Very satisfied... 1 Neither / nor... 3 Very dissatisfied... 5 Fairly satisfied... 2 Fairly dissatisfied... 4 E11. How would you describe you and your partner's (if applicable) ethnic origin? a) Self b) Partner a) Self b) Partner White Asian or Asian British British... 01... 01 Indian... 11... 11 Irish... 02... 02 Pakistani... 12... 12 Any other White background... 03... 03 Bangladeshi... 13... 13 Any other Asian background... 14... 14 Mixed White & Black Caribbean... 04... 04 Other ethnic group White & Black African... 05... 05 Chinese... 15... 15 White & Asian... 06... 06 Any other... 16... 16 Any other Mixed background... 07... 07 No partner... 17 Black or Black British Caribbean... 08... 08 E12. Are you a UK citizen? African... 09... 09 Yes... 1 Any other Black background... 10... 10 No (please state country of citizenship)... 2 E13. Please state your sexual orientation. Heterosexual/Straight... 1 Lesbian/Gay Woman... 3 Other... 5 Gay Man... 2 Bi-sexual... 4 Prefer not to state... 6 E14. It is possible that we may wish to carry out another survey to follow up some of the housing issues raised in this one. Would you be willing to be included in a follow-up survey? Yes... 1 Don t know, would need more information... 3 No... 2 Please return this questionnaire in the pre-paid envelope. THANK YOU FOR COMPLETING THE QUESTIONNAIRE. 8

Maps for Wiltshire SHMA Questionnaire Please use the codes shown on the maps to fill in the boxes on the attached questionnaire. Answers written or marked on the maps will not be included. Please turn over for areas outside Wiltshire A list of areas shown on both maps is also provided overleaf.

Maps for Wiltshire SHMA Questionnaire Please use the codes shown on the maps to fill in the boxes on the attached questionnaire. Answers written or marked on the maps will not be included. List of areas shown on both maps 1 Stroud 24 Corsham, Gastard & Neston 43 Studley & villages around Calne 2 Cotswold 25 Cricklade 44 Avebury, Aldbourne & villages around Marlborough 3 Swindon 26 Devizes & Roundway 45 Box, Colerne & Rudloe Area 4 Vale of White Horse 27 Malmesbury 46 Lacock & Bowden Hill Area 5 West Berkshire 28 Marlborough & Manton 47 Villages around Bradford-on-Avon 6 Test Valley 29 Melksham, Whitley & Shaw 48 Broughton Gifford & Atworth Area 7 Winchester 30 Pewsey 49 Seend, Steeple Ashton & Poulshot Area 8 Southampton & Eastleigh 31 Salisbury, Harnham, Laverstock 50 Lavington, Bromham & villages around Devizes 9 Basingstoke & Deane & Old Sarum 51 Upavon & villages west of Pewsey 10 New Forest 32 Trowbridge, Hilperton & Staverton 52 Bedwyn & villages east of Pewsey 11 East Dorset 33 Warminster & Boreham 53 North Bradley, Southwick & West Ashton Area 12 North Dorset 34 Westbury & Westbury Leigh 13 West Dorset 35 Wilton & Quidhampton 54 Bratton, Dilton Marsh, Hawkeridge & nearby villages 14 South Somerset 36 Wootton Bassett 55 Codford & villages around Warminster 15 Mendip 37 Amesbury, Bulford, Durrington & Larkhill 56 Wylye, Winterbourne & Salisbury Plain villages 16 Bath & North East Somerset 38 Tidworth & Ludgershall Area 57 Netheravon, Collingbourne, Shute & nearby villages 17 North Somerset 39 Latton & Meysey Hampton Area 58 Mere, Semley & nearby villages 18 Bristol City 40 Ashton Keynes, Sherston, Brinkworth, 59 Tisbury, Donhead, Hindon & nearby villages 19 South Gloucestershire 20 Reading 21 Bradford-on-Avon 22 Calne & Quemerford 23 Chippenham & Pewsham Minety & villages around Malmesbury 41 Lyneham, Purton, Lydiard & villages west of Swindon 42 Hullavington, Castle Combe, Kington & nearby villages 60 Netherhampton, Bishopstone, Tefford & villages west of Wilton 61 South Newton & Stoford Area 62 Downton, Winterslow, Alderbury & nearby villages