Arun District Council Strategic Housing Market Assessment Validation

Transcription

1 Arun District Council Strategic Housing Market Assessment Validation Report of Findings for Opinion Research Services October 2013 Opinion Research Services The Strand Swansea SA1 1AF

2 Contents Executive Summary... 3 SHMA Validation for Arun District Council Strategic Housing Market Assessment Validation... 4 Independent Report of Findings for Arun District Council Project Overview... 4 What is a Strategic Housing Market Assessment?... 4 The Project Brief... 4 National Government Policy and Guidance... 5 National Planning Policy Framework (March 2012)... 5 CLG Practice Guidance... 6 Reviewing the Data used in the SHMA... 7 ONS Mid-Year Estimates... 8 CLG Household Headship Rates... 9 Reviewing the SHMA Population and Household Projections Considering the SHMA Population and Household Projections in Context Consultation, Listening and Engagement Stakeholders Members Key Outcomes from the Consultation Validating an Appropriate Housing Target Drivers of Population Change Projecting the requirements of the local population Impact of migration Impact of Recent Dwelling Delivery Relationship between Households and Labour Force Population Projections Household Projections Conclusions and Recommendations Community concerns Data sources Validation findings Establishing the housing target Appendix A Collated views from presentation to Members Appendix B Collated views from presentation to Stakeholders 2

3 Executive Summary SHMA Validation for Arun District Council 1. The Coastal West Sussex sub-regional Strategic Housing Market Assessment (SHMA) was prepared by GL Hearn and published in November The SHMA forms an important part of the Local Plan evidence base for the overall housing requirement for Arun District; but the study created considerable concerns amongst local communities and stakeholders. 2. Opinion Research Services (ORS) was commissioned by Arun District Council to undertake a validation of the SHMA. Whilst the validation study considered the data used in the SHMA and the methodology applied, an important part of the work was to consider the concerns raised and to afford local people an opportunity to express their views, and to query the data and assumptions used to generate the housing requirements. Key conclusions 3. The validation study concluded that the SHMA methodology and the data sources used were in line with Government recommendations, although the data sources were updated by ONS and CLG after the SHMA was published. The Coastal West Sussex Partnership authorities commissioned further analysis to consider the new data, and it has only had a moderate impact on the SHMA conclusions for Arun district. 4. The SHMA reports have taken into account key changes in both population (including migration as well as natural change) and household projections in a relatively robust manner, and the assessed level of housing required for Arun District does not appear excessive. Further, the requirements produced would appear to be compliant with the National Planning Policy Framework (2012). 5. Nevertheless, ORS recognises that the SHMA as is stands perhaps lacks transparency in certain places and possibly this has contributed to the concerns of local people. Establishing the housing target 6. On the basis of the SHMA and subsequent analysis, the Council has been advised that the range dwellings each year is the assessment of objectively assessed need for Arun district. This is consistent with the additional evidence from the validation study, which indicated a range around 620 homes each year. 7. Arun District Council has adopted a target of 580 (average) dwellings each year over the plan period, with a target of 455 (average) dwellings each year for the period to 2018/19. The overall target was consistent with the original projections based on employment growth (with migration focused towards working age), and the lower target during the initial period reflected the current economic circumstances. 8. The validation exercise has identified that there is no single approach which will provide a definitive assessment of development need ; but the Council may wish to consider if they believe that their adopted target remains appropriate in the context of the full range of evidence now available, and whether their position may be more robust if a target towards the middle or upper part of the range was adopted. 3

4 Strategic Housing Market Assessment Validation Independent Report of Findings for Arun District Council Project Overview 1. Opinion Research Services (ORS) was commissioned by Arun District Council in August 2013 to undertake a validation of the Coastal West Sussex sub-regional Strategic Housing Market Assessment (SHMA) that was prepared by GL Hearn and published in November Whilst the SHMA provides an analysis of a range of housing issues across the whole of the sub-region, the scope of the validation was focussed on the evidence-base for the overall housing requirement for Arun District. Given this scope, the validation also considered further analysis prepared by GL Hearn in the Housing Study (Duty to Cooperate) report published in May 2013 and the Updated Demographic Projections report finalised in August 2013, but yet to be published. What is a Strategic Housing Market Assessment? 3. SHMAs assess the full range of housing requirements for an area to both deepen understanding of subregional housing markets and develop future strategy, all based on the best available evidence. Critically, their evidence base shapes and assists with the production of the Local Plan which sets out the spatial policy for a local area. 4. The SHMA also provides evidence-based forecasts of need and demand for housing of different types, sizes and tenures (including market housing, intermediate tenures and social rent/affordable rent) to inform local, sub-regional and regional spatial planning and housing strategies. The Project Brief 5. The Brief was to:» Review and validate the data used in the SHMA and advise if this has been appropriately applied» Review the main issues on the SHMA raised by Members and public groups regarding: Population growth, headship rates / household size Actual house building rates in Arun Trends in migration» Review the various related documents produced locally» Advise the Council on any implications arising from the Review including a figure to set as Arun Districts housing target 4

5 6. The reports that were reviewed as part of the SHMA validation are summarised in the table below: Report SHMA update report (November 2012) Housing Study (Duty to Cooperate) (May 2013) Updated demographic projections (August 2013) What it does Sub-regional SHMA report for the Coastal West Sussex Partnership with a Local report for Arun area The Partnership includes Adur, Chichester and Worthing, plus the South Downs National Park Authority. Considers housing delivery Considers housing delivery and updates key analysis. Study based on more recently published data from ONS and includes revised household projections National Government Policy and Guidance 7. Importantly, SHMA production is informed by a series of Government practice and policy Guides, the requirements of which need to be met in order to have a credible assessment of housing requirement: objectively assessed housing need. 8. It is important to recognise that the SHMA was undertaken at a time of significant planning policy and strategy change in housing and related areas in England as new Policy and Guidance were introduced; and these changes remain on-going at the time of this SHMA validation:» Planning: a new National Planning Policy Framework (March 2012)» Guidance on the Assessment of housing and economic development needs (August 2013) 9. Significantly, there is no direction on matters such as housing targets within Policy and guidance; it is the local planning authority that is responsible for establishing the level of future housing provision required. However, any requirement needs to have been produced in compliance with Policy and Guidance. National Planning Policy Framework (March 2012) 10. This Validation study has sought to test Arun Districts SHMA housing requirements against the National Planning Policy Framework (NPPF) and latest Guidance in order to ensure outputs are compliant. NPPF has a presumption in favour of sustainable development and Local Plans should meet objectively assessed needs: At the heart of the National Planning Policy Framework is a presumption in favour of sustainable development, which should be seen as a golden thread running through both plan-making and decision-taking. Local planning authorities should positively seek opportunities to meet the development needs of their area. Local Plans should meet objectively assessed needs, with sufficient flexibility to adapt to rapid change, unless any adverse impacts of doing so would significantly and demonstrably outweigh the benefits, when assessed against the policies in this Framework taken as a whole. National Planning Policy Framework (NPPF), paragraph 14 5

6 11. NPPF also has an emphasis on the use of evidence of derive objectively assessed housing needs: and delivering a wide choice of high quality homes: To boost significantly the supply of housing, local planning authorities should use their evidence base to ensure that their Local Plan meets the full, objectively assessed needs for market and affordable housing in the housing market area. National Planning Policy Framework (NPPF), paragraph NPPF places requirements on what should be produced by SHMAs: Local planning authorities should prepare a Strategic Housing Market Assessment to assess their full housing needs, working with neighbouring authorities where housing market areas cross administrative boundaries. The Strategic Housing Market Assessment should identify the scale and mix of housing and the range of tenures that the local population is likely to need over the plan period which meets household and population projections, taking account of migration and demographic change. National Planning Policy Framework (NPPF), paragraph Further, NPPF says that the assessment of housing requirements should take into account migration and demographic change and should be unconstrained by assumptions that, for example, migration will no longer take place: Local planning authorities should prepare a Strategic Housing Market Assessment to assess their full housing needs, working with neighbouring authorities where housing market areas cross administrative boundaries. The Strategic Housing Market Assessment should identify the scale and mix of housing and the range of tenures that the local population is likely to need over the plan period which meets household and population projections, taking account of migration and demographic change. National Planning Policy Framework (NPPF), paragraph 159 CLG Practice Guidance 14. SHMA Practice Guidance recognises that a variety of approaches to deliver a robust SHMA are possible and so is not prescriptive as to the methodology to be followed and the data to be used: No one methodological approach or use of a particular dataset(s) will result in a definitive assessment of housing need and demand. CLG Practice Guidance 2007, p11 There is no one methodological approach or use of a particular dataset(s) that will provide a definitive assessment of development need. CLG Practice Guidance (Draft) 2013, section 1 6

7 15. Guidance also recognises that SHMAs need to be robust and credible assessments, but their analysis should not be seen as definitive: Housing markets are dynamic and complex. Because of this, strategic housing market assessments will not provide definitive estimates of housing need, demand and market conditions. However, they can provide valuable insights into how housing markets operate both now and in the future. CLG Practice Guidance 2007, page The 2013 SHMA Guidance also places emphasis on the role of CLG Household Projections: Household projections published by the Department for Communities and Local Government should provide the starting point estimate of overall housing need. The household projections produced by the Department for Communities and Local Government are statistically robust and are based on nationally consistent assumptions. CLG Practice Guidance (Draft) 2013, section However, SHMA Guidance does allow for the use of sensitivity testing of CLG Household projection to test where this is deemed appropriate, allowing for alternative assumptions to be used: Plan makers may consider sensitivity testing, specific to their local circumstances, based on alternative assumptions in relation to the underlying demographic projections and household formation rates. Any local changes would need to be clearly explained and justified on the basis of established sources of robust evidence. CLG Practice Guidance (Draft) 2013, section 3 Reviewing the Data used in the SHMA 18. Concerns had been raised that the SHMA analysis was based on out-of-date data, as a range of new information had become available since the original study was published in November Much of this new information is based on data that has now been published from the 2011 Census that was not available when the original SHMA was being prepared. Nevertheless, whilst it is important to recognise that this data wasn t available for the SHMA to consider, it remains appropriate to consider the impact of revised and updated data on the analysis and conclusions. 20. There are two datasets of particular importance that were both updated in April 2013, namely:» ONS Mid-Year Estimates of population: which provides information to inform population projections; and» CLG Headship Rates: this provides information that helps to translate population projections into household projections. 7

8 Number of people Opinion Research Services Arun DC: Strategic Housing Market Assessment Validation Report of Findings October 2013 ONS Mid-Year Estimates 21. ONS Mid-Year Estimates of population are of particular importance in considering the future population of Arun District. 22. The Mid-Year Estimates are published annually by the ONS for each local authority area, based on recorded births and deaths and include estimates of UK and international migration. This data was a critical input for the SHMA 2012, as it forms the basis for population projections informing the:» Fertility rates, used to project births;» Mortality rates, used to project deaths; and» Rates for in and out migration to and from elsewhere in the UK and internationally. 23. The ONS Mid-Year Estimates are retrospectively adjusted every 10-years in order to ensure consistency with the Census. This process is important as it takes account of any errors that may have arisen due to inaccuracies in the estimates. In April 2013, ONS published revised annual Mid-Year Estimates for the period ; data that was not available when the SHMA was produced. Of course, the SHMA adopted the most up-to-date data that was available at the time; but it is right to consider the possible impact of the adjusted ONS figures. 24. Figure 1 shows the revised ONS Mid-Year Estimates for Arun District alongside their original figures (on which the original SHMA analysis was based) and the Census estimates for the period from 1981 onwards. In reviewing the data, it is apparent that whilst the ONS revised figures did moderate the population estimates to an extent, the overall impact of their adjustments were relatively marginal. For example, the original estimate for Mid-2010 was 150,561 persons, which was adjusted to 149,467 persons in the revised figures: a reduction of 1,094 persons, equivalent to 0.7% of the overall population. 25. Therefore, whilst it is appropriate to challenge the data that was used in the SHMA 2012 (given that it has been subsequently updated), the changes the 2013 update has made are unlikely to significantly alter the original SHMA projections. Figure 1: Population Trends: ONS Mid-Year Estimates for Arun District (Source: ONS Mid-Year Estimates, original and revised. Census of Population 1981, 1991, 2001 and 2011) ONS MYE (revised) ONS MYE (original) Census 155, , , , , , , , , ,000 8

9 % of population that are household representatives Opinion Research Services Arun DC: Strategic Housing Market Assessment Validation Report of Findings October 2013 CLG Household Headship Rates 26. Another important data source is the CLG Household Headship Rates. The Headship Rates are published as part of the household projections produced by CLG and these are referenced in new Practice Guidance of August Headship Rates are a demographic tool used to convert population into households and are based on those members of the population who can be classed as household representatives or heads of household. 28. The proportion of people in any age cohort who will be Heads of Household (the Headship Rate ) vary between people of different ages; for example, fewer teenagers will be Heads of Household than say those aged 85 or over. However, the Headship Rates used are key to the establishment of the number of households and, further, the number of households are key to the number of homes that will be required in Arun. Getting Headship Rates accurate, therefore, is important. 29. The CLG published the Interim 2011-based Household Projections in April 2013, and this included updated Headship Rates which took account of initial data from the 2011 Census. Further data will be published by CLG which will take account of the full data from the 2011 Census, but this is not scheduled for release until October The SHMA 2012 used Headship Rates from the previous CLG publication: the 2008-based Household Projections, as it was the most up-to-date data that was available at the time. Once again, it is right to consider the possible impact of the updated 2013 figures. Figure 2 shows the Interim 2011-based Headship Rates alongside the 2008-based figures that were used by the SHMA. 31. In reviewing the data, it is apparent that changes for Arun have again been only marginal the most notable change being for the cohort aged 25-34, where the Headship Rate has reduced from 49% to 45%. What this means is that whilst the 2008-based figures estimated that there were 49 households for every 100 people in Arun, the Interim 2011-based figures reduce this estimate to 45 therefore 4 fewer households for every 100 people in this age group; and the rates for other age groups have changed too. Figure 2: 2011 Headship Rates for Arun District (Source: CLG Household Projections, 2008-based and Interim 2011-based) 100% 2008-based Interim 2011-based 90% % 70% 60% 50% % 30% 20% 10% 8 8 0% Aged Aged Aged Aged Aged Aged Aged Aged Aged 85+ 9

10 Reviewing the SHMA Population and Household Projections 32. The SHMA 2012 considered a range of population projections and produced estimates of housing numbers and employment growth for each. Figure 3 shows the summary outputs for Arun district from the original SHMA for the 20-year period The projection scenarios PROJ 1 and PROJ 2 are based on migration trends, and therefore these could be particularly affected by changes to the underlying ONS Mid-Year Estimates. Furthermore, all of the scenarios will be affected by underlying fertility and mortality rates (also derived from the ONS data) and also the Headship Rates derived from the CLG data previously discussed. Figure 3: Original SHMA outputs (November 2012) for Arun District 34. The Demographic Projection Update report was prepared by the sub-region largely to take account of the updated data previously discussed. Figure 4 shows the updated outputs for Arun from this report. Figure 4: Demographic Projection Update outputs (August 2013) for Arun District 10

11 35. Given the large number of scenarios presented, it is perhaps unfortunate that the labelling was changed between the two reports; but Figure 5 and Figure 6 provide a comparison between the two sources on the basis of our understanding of the scenarios that were considered in both reports. 36. In this context, we note that the update (paragraph 2.26) sets out that: PROJ B is comparable with PROJ 5 of the earlier Coastal West Sussex SHMA (2012). This projection is based on modelling employment growth on a 1:1 basis with changes in the resident population who are in employment. PROJ A takes account of commuting patterns to work out the likely growth in the resident workforce if commuting patterns remain the same. 37. Nevertheless, the original PROJ 5 scenario was described as employment growth with growth in out-commuting (emphasis added) whilst it was PROJ 5A that was described as employment growth with no growth in out-commuting (emphasis added). Therefore, it would seem that the update report should in fact state that PROJ A from the update is comparable with PROJ 5 of the earlier SHMA, and that PROJ B is comparable with PROJ 5A. Figure 5 and Figure 6 are based on this assumption. Figure 5: Comparing the population projections for Arun from the original SHMA and the Demographic Projection Update Scenario Original SHMA Name Population growth Demographic Update Name Population growth Net difference 10-year migration trends PROJ 1 27,652 PROJ 3 21,940-5,712 5-year migration trends PROJ 2 20,382 PROJ 4 12,529-7,853 Zero net migration PROJ 3-16,191 PROJ X -14,445 +1,746 Zero employment growth PROJ 4 12,154 PROJ Y 13, Employment growth with growth in out-commuting Employment growth with no growth in out-commuting PROJ 5 33,710 PROJ A 34, PROJ 5A 27,850 PROJ B 28, Figure 6: Comparing the housing numbers for Arun from the original SHMA and the Demographic Projection Update Scenario Original SHMA Name Housing number Demographic Update Name Housing number Net difference 10-year migration trends PROJ 1 15,075 PROJ 3 14, year migration trends PROJ 2 12,042 PROJ 4 10,106-1,936 Zero net migration PROJ 3-3,214 PROJ X -1,416 +1,798 Zero employment growth PROJ 4 8,717 PROJ Y 10,359 +1,642 Employment growth with growth in out-commuting Employment growth with no growth in out-commuting PROJ 5 17,724 PROJ A 19,483 +1,759 PROJ 5A 15,267 PROJ B 16,998 +1,731 11

12 yr trend 5-yr trend Jobs-led Internal jobs Age adjusted 10-yr trend 5-yr trend Jobs-led Internal jobs Average annual growth Opinion Research Services Arun DC: Strategic Housing Market Assessment Validation Report of Findings October 2013 Considering the SHMA Population and Household Projections in Context 38. As noted above, latest Guidance on the Assessment of housing and economic development needs (August 2013) says that: Household projections published by the Department for Communities and Local Government should provide the starting point estimate of overall housing need. 39. It is therefore important to consider the outputs from the CLG projections as a context for the outputs from the SHMA and demographic update. Figure 7 sets out the figures for Arun district from the most recent four rounds of CLG household projections. Whilst it is apparent that the specific figures have changed, the range has been relatively consistent at around additional households each year. Figure 7: CLG Household Projections for Arun District: 10-year and 25-year projections (Source: CLG Household Projections: 2004-based, 2006-based, 2008-based and Interim 2011-based) Projection Period 10-year projection Net total Annual rate Period 25-year projection Net total Annual rate 2004-based , , based , , based , , Interim 2011-based , Figure 8 compares the CLG household projections with the housing numbers from the SHMA and Demographic Update reports. The SHMA outputs tend to be lower than all previous CLG estimates. Figure 8: Comparing the CLG Household Projections with the housing numbers from the SHMA and Demographic Update 1,100 1, CLG 10-yr projections (households) SHMA (dwellings) Demographic update (dwellings) 12

13 Consultation, Listening and Engagement 41. As part of the Validation study, two separate consultation sessions were held on 19 September 2013 with:» Local stakeholders: people and representatives who had expressed concerns with the SHMA» Local Members of Arun District Council 42. The sessions were informal and participative and questions were asked and answered throughout the session. A presentation of initial findings was made using a series of PowerPoint slides. A summary of questions and answers is given in Appendix A and Appendix B. Stakeholders 43. The stakeholder sessions were held first and, in summary, the key issues raised were concerning:» How Population projections are used and work: attendees were keen to understand how projections were derived and tested.» How CLG Household projections are used and work: attendees were keen to understand how projections were derived and tested.» Projections basis to SHMA is complex and uncertain: there were concerns at the seeming lack of transparency in the 3 studies considered for the Validation exercise.» Ensuring Arun population projections are compared with neighbouring authorities: attendees were keen to understand further how the population assessment and its implications for Arun compared to neighbouring authorities.» Lack of confidence in the how data is used and the outcomes from it: attendees experience with the use of data in the previous studies was generally low.» Migration Implications: attendees were very keen to understand the implications of migration for both in-uk migrants and international migrants.» Timetable for the Evaluation project: there was also strong interest in the forward timetable for the Validation exercise and a request that stakeholders be kept informed. Members 44. The session for Members was similar in style and content to that for stakeholders. In summary, the key issues raised concerned:» History and Timeline of previous studies: Members were keen to understand how the current Validation fitted with previous studies» Issues on Household Formation and Headship rates and Head of Households: members probed the methodology on how Headship rates were used to derive Households in some detail.» Population/Household Projections Used in Previous Studies: Members also focussed on how population and household data had been used in previous studies.» Migration & Commuting Implications for Future Requirements: Members were keen to understand in detail the implications of migration for housing requirements.» Emerging Future Housing Requirement Levels: Members wanted to know the implications of the Validation study, so far, for the housing requirement of Arun District. 13

14 Key Outcomes from the Consultation 45. Both consultation sessions had positive outcomes, insofar as the participants considered that their views were being heard and that their questions and concerns were being answered as far as possible. However, there was still scepticism about the SHMA outputs in terms of the number of homes required for Arun district in future years. 46. Most of the concerns raised by both Stakeholders and Members derived from a lack of confidence in the evidence base, as the SHMA failed to provide sufficient transparency about the way in which projections were derived. 47. Both Stakeholders and Members were keen that the validation should adopt a transparent approach to the core issues within the validation exercise, namely, how population and household formation were calculated and how these work to produce a housing requirement. Further, they were keen to understand how migration acted as a driver of housing requirement. Overall, there was strong interest in the outcomes and outputs from the Validation study. Validating an Appropriate Housing Target 48. Whilst the outputs from the SHMA and the subsequent Demographic Projection Update are marginally lower than CLG projections, concerns raised during the consultation highlighted the lack of transparency in these calculations as being a key concern. 49. Given this context and as part of the Validation process, Arun District has asked for additional projections to be prepared that are presented as transparently as possible in order for everyone to understand the process that has been employed. 50. This should provide a basis for validating an appropriate housing target for the district which is generally accepted. Drivers of Population Change 51. The Population is analysed to consider what is driving any increase or decrease, in particular:» natural change: the balance of births and deaths; and» net migration: the increase or decrease brought about by people moving in and out of the area. Net migration is the balance remaining (either negative or positive) by deducting the number of out-migrants from the number of in-migrants. 52. Figure 9 shows Arun s natural change is driving a reduction in the local population over time which has been a relatively consistent trend from 1991 to 2006, although the rate of decrease has slowed in subsequent years. However, migration is driving a net increase in the population over time. Trends are consistently positive in that in-migrants are higher in number than out-migrants for the period from 1991 to However, the trend is not fixed and net in-migration has changed over time; for example, it increased more in some years (e.g. an increase of more than 3,000 people in ) than in others (e.g. an increase of less than 1,000 people in ). 53. Figure 10 provides a detailed breakdown of components of annual population change over the last 10-years, and also provides 5-year averages that show how the trends are changing. 14

15 Number of People Opinion Research Services Arun DC: Strategic Housing Market Assessment Validation Report of Findings October 2013 Figure 9: Population Trends: Components of Annual Net Change (Source: ONS Mid-Year Estimates, revised) Natural change Migration & other changes Census (annual average) +4,000 +3,000 +2,000 +1, ,000-2,000 Figure 10: Components of population change, revised in the light of the 2011 Census (Source: ONS Mid-Year Estimates, revised. Note: Other Changes includes adjustments for asylum seekers, prisoners, armed forces and other unattributable changes. All figures presented unrounded for transparency, but should only be treated accurate to the nearest 100) Year Births Deaths Natural Change UK Migration International Migration In Out In Out Other Changes Migration and Other Changes Total Change ,261 2,311-1,050 8,874 6, ,437 +1, ,222 2,242-1,020 8,176 5,996 1, ,093 +1, ,278 2, ,536 6, ,541 +1, ,357 2, ,429 6,123 1, , ,364 2, ,660 6,073 1, , ,408 2, ,907 6,506 1, , ,434 2, ,950 6,049 1, , ,538 2, ,419 5,508 1, , ,419 2, ,886 5, , ,504 2, ,608 5, year Average 1,379 2, ,545 6,006 1, , Year Averages ,296 2, ,135 6, ,079 +1, ,326 2, ,942 6,151 1, ,930 +1, ,368 2, ,696 6,162 1, ,776 +1, ,420 2, ,273 6,052 1, , ,433 2, ,164 6,013 1, , ,461 2, ,954 5,935 1, , Lowest Net Migration Highest Net Migration 1,461 2, ,954 5,935 1, , ,296 2, ,135 6, ,079 +1,150 15

16 Projecting the requirements of the local population 54. As a baseline position, we have analysed the requirements of the local population on the basis of a scenario without any migration. 55. This is different to the net nil migration scenario that was considered by the original SHMA, as that assumed that there would be migration to and from Arun, but the number of persons moving to the area will balance against the number of persons moving away: that is, migration does not change the number of people in the area, although it could impact on the local age structure. Changes to the number of people in different age cohorts will influence the number of births and deaths over the projection period. 56. Figure 11 shows a summary of the modelled projections using a zero migration scenario. Figure 11: Summary of 20-year Projection Zero Migration (Note: All figures presented unrounded for transparency, but should only be treated accurate to the nearest 100) Components of Population Change 20-year Total Zero migration Annual Average Number of births 29,410 1,470 Number of deaths 37,268 1,863 Natural change -7, Net migration 0 0 Net population change -7, The scenario shows a projected reduction from 149,800 to 142,000 persons over the 20-year period Figure 12 shows the impact on age structure. Figure 12: Population projections by gender and 5-year age cohort based on Zero Migration scenario (Note: Figures rounded to nearest 100. All calculations based on unrounded data) Age 2011 Mid-year Estimate 2031 Zero Migration Net M F Total M F Total change Aged 0-4 3,900 3,600 7,500 3,500 3,300 6, Aged 5-9 3,500 3,400 7,000 3,700 3,500 7, Aged ,800 3,600 7,400 3,800 3,600 7, Aged ,100 3,900 7,900 4,000 3,800 7, Aged ,700 3,700 7,500 3,900 3,600 7,500 0 Aged ,600 3,600 7,300 3,500 3,400 6, Aged ,500 3,700 7,200 3,800 3,600 7, Aged ,800 4,200 8,000 4,000 3,800 7, Aged ,900 5,200 10,100 3,700 3,700 7,400-2,700 Aged ,100 5,400 10,500 3,500 3,600 7,100-3,400 Aged ,500 4,900 9,500 3,400 3,700 7,000-2,500 Aged ,400 4,800 9,100 3,700 4,000 7,700-1,400 Aged ,300 6,000 11,300 4,600 4,900 9,600-1,700 Aged ,000 5,500 10,500 4,500 5,100 9, Aged ,100 4,800 8,900 3,800 4,500 8, Aged ,500 4,200 7,600 3,400 4,000 7, Aged ,600 3,500 6,000 3,500 4,500 8,000 +2,000 Aged ,500 2,700 4,100 2,500 3,400 5,900 +1,800 Aged ,800 2,500 2,000 3,200 5,100 +2,600 Total 71,500 78, ,800 68,800 73, ,000-7,800 16

17 58. Whilst this provides an important baseline for comparison, we would note that in itself it is unlikely be compliant with the requirements of the National Planning Policy Framework (NPPF). The NPPF states that local planning authorities must: identify the scale and mix of housing and the range of tenures that the local population is likely to need over the plan period which meets household and population projections, taking account of migration and demographic change National Planning Policy Framework (NPPF), paragraph 159 (emphasis added) Impact of migration 59. The detailed data from the ONS Mid-Year Estimates (Figure 10) provided data about migration flows to and from Arun district. Figure 13 details summarises this information by presenting the 10-year average (for the period ) together with the 5-year averages for the periods that had the overall net change that were highest and lowest during this time: Figure 13: Migration and other changes based on 10-year and 5-year average flows (Source: ONS Mid-Year Estimates, revised. Note: Other Changes includes adjustments for asylum seekers, prisoners, armed forces and other unattributable changes. All figures presented unrounded for transparency, but should only be treated accurate to the nearest 100) Components of Population Change 10-year average Period HIGHEST overall net 5-year averages Period LOWEST overall net Moves to Arun from elsewhere in the UK 7,545 8,135 6,954 Moves from Arun to elsewhere in the UK 6,006 6,076 5,935 Net UK migration +1,539 +2,059 +1,019 International moves to Arun 1, ,142 International moves from Arun Net International migration Other changes Migration and other changes +1,652 +2,079 +1, These figures clearly illustrate that annual migration flows can vary, and it is therefore important to understand the impact of higher or lower figures on the overall population and household projections. It is also important to consider the impact of other changes that are not factored into the individual migration flows. 61. Other changes includes ONS adjustments for asylum seekers, prisoners and armed forces, as well as other unattributable changes (including any residual errors in the migration flow data). In Arun District, it is apparent that this adjustment reduces the number of people estimated each year. Therefore, we have adjusted the gross migration flows to take account of this through reducing inward migration and increasing outward migration each by 2.6%. This ensures that the net total for migration flows is consistent with the total for migration and other changes set out in the ONS data. The adjusted flows are detailed in Figure

18 Figure 14: Migration flows used to inform the population projections Components of Population Change Mid-trend migration (10-year average) High-trend migration (highest 5-yr avg) Low-trend migration (lowest 5-yr avg) Moves to Arun from elsewhere in the UK 7,346 7,932 6,761 Moves from Arun to elsewhere in the UK 6,164 6,228 6,100 Net UK migration +1,182 +1, International moves to Arun 1, ,110 International moves from Arun Net International migration Migration and other changes +1,652 +2,079 +1, A key issue for the Council and local stakeholders concerns the characteristics of the migrant population and the areas that people moving to Arun come from, and the destination of those leaving the district. The ONS migration unit publishes detailed information about movers within the UK each year. Over the 5-year period , the data highlight (Figure 15):» Age: overall; the key age groups in-migrating are aged 45+. This suggests that people are retiring to the District and not moving again.» Source: Arun is a net exporter of migrants to all UK regions except London, South East and, marginally, Eastern. Essentially, UK migrants are coming from London and the South East. Figure 15: 5-year UK migration (Source: ONS migration unit. Note: this data source forms part of the information used to inform the Mid-Year Estimates Components of Change, but the two sources are not identical) In-migration Out-migration Net Migration BY AGE 0-15 years 4,640 3, years 6,530 7, years 9,940 9, years 7,960 4,890 +3, years 5,290 3,540 +1,750 BY REGION North East North West Yorkshire and Humberside East Midlands West Midlands Eastern 1,610 1, London 5,450 2,650 +2,800 South East 21,590 18,630 +2,960 South West 2,220 2, Wales TOTAL 34,400 29,210 +5,190 18

19 63. When considering migration within the sub-region (Figure 16), it is apparent that population trends show that Arun gains population from Worthing: 720 people over the period , which represents 14% of the overall net UK migration for the same period. There are also notable gains from Adur (350 people, 7% of the net UK total). 64. There are large flows of population both to and from Chichester (over 4,000 moves in both directions over the 5-year period, equivalent to more than 10% of moves to and from Arun), however net migration shows a balance, with only a small gain of 80 people. Similarly there is a balance in moves to and from Brighton and Hove (50 person net gain) and Lewes (20 person net loss). 65. Overall, migration trends suggest that Arun district gained around 240 persons each year from the rest of the sub-region over the period , which represents almost a quarter (23%) of the overall net UK migration for the same period. When we consider the age structure of this population, it is apparent that there is a strong bias towards older people: 100 aged over 65 and a further 110 aged The small gain of young adults aged and children aged under 16 is largely offset against a loss of population aged moving to Chichester, Brighton and Hove and Worthing. Figure 16: 5-year migration within the Coastal West Sussex sub-region (Source: ONS migration unit) COASTAL WEST SUSSEX SUB-REGION Inmigration Outmigration Net Migration Net Migration by Age Worthing 4,630 3, Adur Chichester 4,340 4, Brighton & Hove 1,100 1, Lewes TOTAL 11,110 9,930 +1, Annual Average 2,220 1, Figure 17 considers the other local authorities of England and Wales and shows the ten areas with the highest net population gain over the 5-year period and the highest net population loss over the period. Figure 17: 5-year migration : highest gains and losses of population by Local Authority (Source: ONS migration unit) In-migrants moving TO Arun Out-migrants moving FROM Arun Croydon +490 Havant -120 Sutton +470 Portsmouth -110 Horsham +330 Bournemouth -80 Hillingdon +290 Bath and North East Somerset -70 Reigate and Banstead +250 Hounslow and Kingston-upon-Thames Crawley for each area Merton +180 Mid Sussex +150 South Somerset, Southampton and West Berkshire Canterbury, Cardiff, Oxford, Poole, South Holland and Winchester -60 for each area -50 for each area 19

20 Impact of Recent Dwelling Delivery 67. Concerns were raised by some stakeholders about the inherent circularity in establishing the number of homes to be built in future on the basis of past migration patterns, as the number of recent migrants will inevitably have been influenced by the availability of local housing and the number of homes that have been built in the past. 68. Figure 18 shows the relative in increases in dwelling stock across South East England. This suggests that the rate of increase in Arun was actually marginally lower than the regional average, although higher than all other authorities in the Coastal West Sussex sub-region other than Chichester. Figure 18: Rate of dwelling stock increase per 1,000 dwellings for South East LAs between 2001 and 2011 (Source: UK Census of Population 2001 and Note: this does not take account of any local growth constraints) England South East Milton Keynes Ashford Dartford Tonbridge and Malling West Oxfordshire Swale Slough Basingstoke and Deane Maidstone Gosport Shepway Tunbridge Wells Eastleigh Dover Eastbourne Winchester Chichester Canterbury Horsham Reading West Berkshire Thanet Mid Sussex Isle of Wight Hart East Hampshire Epsom and Ewell Aylesbury Vale Test Valley Vale of White Horse Portsmouth Oxford Reigate and Banstead Southampton Hastings Fareham Windsor and Maidenhead Wycombe Cherwell Rother Wealden New Forest Medway Tandridge Woking Arun Mole Valley Wokingham South Bucks Lewes Gravesham Worthing Crawley Waverley Sevenoaks Elmbridge Surrey Heath Bracknell Forest Brighton and Hove Chiltern Havant South Oxfordshire Runnymede Guildford Adur Spelthorne Rushmoor

21 Relationship between Households and Labour Force 69. Stakeholders also recognised the importance of the relationship between housing and labour supply both within individual local authorities and across the sub-region. Between 2001 and 2011 across England, the number of economically active people has increased from an average of 1.16 to 1.23 per household, with an equivalent increase from 1.23 to 1.28 across the South East. 70. Figure 19 shows the relative change for all local authorities across South East England. This suggests that the rate of increase in Arun was actually marginally higher than the regional average (from 1.00 to 1.08), although this represents workers living in the area and many may work outside the local authority. Figure 19: Change in average number of economically active persons per household for South East LAs between 2001 and 2011 (Source: UK Census of Population 2001 and Note: this does not take account of communal establishments) England South East Brighton and Hove Oxford Eastbourne Southampton Worthing Canterbury Epsom and Ewell Woking Elmbridge Slough Spelthorne Dartford Arun Reigate and Banstead Hastings Lewes Crawley Portsmouth Adur Gravesham Rother Thanet Ashford South Bucks Dover Wealden Chichester Guildford Tunbridge Wells Reading Tandridge Runnymede Sevenoaks Mole Valley Isle of Wight Milton Keynes Windsor and Maidenhead South Oxfordshire Mid Sussex Rushmoor Shepway Tonbridge and Malling Medway Surrey Heath Waverley New Forest Maidstone Swale Cherwell Havant West Oxfordshire Winchester Horsham Chiltern Wycombe Eastleigh Bracknell Forest Aylesbury Vale East Hampshire Vale of White Horse Basingstoke and Deane Fareham West Berkshire Test Valley Hart Gosport Wokingham

22 Number of People Opinion Research Services Arun DC: Strategic Housing Market Assessment Validation Report of Findings October 2013 Population Projections 71. Given the above context, we produced detailed 20-year projections based on the average migration for the last 10-year period, and sensitivity tested this based on the highest and lowest 5-year figures. These scenarios are described as mid-trend migration, high-trend migration and low-trend migration in the following section. Figure 20 and Figure 21 show the overall population projections for the four scenarios over the period and the projected 5-year age cohorts by gender. Figure 20: Population projections comparing Zero, High-, Mid- and Low-trend Migration scenarios Trend Zero Migration Low-trend Migration Mid-trend Migration High-trend Migration 185, , , , , , , , , , , , ,000 Figure 21: Population projections by gender and 5-year age cohort based on Zero, High-, Mid- and Low-Trend Migration scenarios (Note: Figures rounded to nearest 100. All calculations based on unrounded data) Age High-trend Migration 2031 Mid-trend Migration 2031 Low-trend Migration 2031 Zero Migration M F Total M F Total M F Total M F Total M F Total Aged 0-4 3,900 3,600 7,500 3,900 3,700 7,700 3,600 3,400 7,100 3,300 3,100 6,400 3,500 3,300 6,900 Aged 5-9 3,500 3,400 7,000 4,300 4,100 8,400 4,000 3,800 7,800 3,600 3,500 7,100 3,700 3,500 7,200 Aged ,800 3,600 7,400 4,600 4,400 9,000 4,300 4,100 8,500 4,000 3,900 7,900 3,800 3,600 7,500 Aged ,100 3,900 7,900 4,800 4,600 9,400 4,600 4,300 9,000 4,400 4,100 8,500 4,000 3,800 7,800 Aged ,700 3,700 7,500 4,200 3,700 8,000 4,000 3,500 7,500 3,800 3,200 7,100 3,900 3,600 7,500 Aged ,600 3,600 7,300 3,800 3,400 7,200 3,600 3,200 6,700 3,400 2,900 6,300 3,500 3,400 6,900 Aged ,500 3,700 7,200 4,100 3,900 8,100 3,900 3,600 7,500 3,700 3,200 6,900 3,800 3,600 7,300 Aged ,800 4,200 8,000 4,700 4,600 9,300 4,400 4,200 8,600 4,100 3,800 7,900 4,000 3,800 7,800 Aged ,900 5,200 10,100 4,900 5,100 10,000 4,600 4,800 9,400 4,300 4,500 8,800 3,700 3,700 7,400 Aged ,100 5,400 10,500 4,700 5,000 9,700 4,500 4,800 9,300 4,300 4,500 8,800 3,500 3,600 7,100 Aged ,500 4,900 9,500 4,600 4,800 9,400 4,300 4,600 9,000 4,100 4,500 8,600 3,400 3,700 7,000 Aged ,400 4,800 9,100 4,900 5,300 10,200 4,700 5,100 9,800 4,500 5,000 9,400 3,700 4,000 7,700 Aged ,300 6,000 11,300 6,300 7,000 13,300 6,100 6,800 12,800 5,800 6,500 12,300 4,600 4,900 9,600 Aged ,000 5,500 10,500 6,900 7,700 14,500 6,600 7,300 13,900 6,300 7,000 13,300 4,500 5,100 9,600 Aged ,100 4,800 8,900 6,100 6,800 12,800 5,800 6,500 12,300 5,600 6,200 11,800 3,800 4,500 8,300 Aged ,500 4,200 7,600 5,100 5,700 10,900 5,000 5,500 10,400 4,700 5,300 10,000 3,400 4,000 7,400 Aged ,600 3,500 6,000 4,800 5,700 10,500 4,700 5,500 10,200 4,500 5,300 9,800 3,500 4,500 8,000 Aged ,500 2,700 4,100 3,100 3,900 7,000 3,000 3,700 6,800 2,900 3,600 6,600 2,500 3,400 5,900 Aged ,800 2,500 2,200 3,500 5,700 2,200 3,400 5,500 2,100 3,300 5,400 2,000 3,200 5,100 Total 71,500 78, ,800 88,100 93, ,200 83,800 88, ,000 79,500 83, ,900 68,800 73, ,000 22

23 72. The trend-based projections range from 162,900 based on the Low-trend Migration scenario up to 181,200 based on the High-trend Migration scenario, which represent 20-year increases of 13,100 persons and 31,400 persons respectively (a range of 18,300 persons between the two scenarios). This clearly illustrates the importance of migration assumptions to the overall projections. Household Projections 73. Prior to considering household projections, it is necessary to identify the household population and separate out the population assumed to be living in communal establishments. The Census identified that Arun had a total population of 149,500 persons on 27 March 2011, of which 3,800 lived in communal establishments and the remaining 145,700 formed the household population. There were a total of 66,700 households with at least one usual resident, therefore an average household size of 2.18 persons. 74. To establish the future communal establishment population, we have assumed that the number of people aged under 75 living in communal establishments will remain constant over the projection period; however, for those aged 75 or over, a proportion of the total population is assumed to live in communal establishments. This proportion is based on data from the 2011 Census for each age-gender cohort; however the projections also take account of changing trends over time. This is consistent with the approach that is used for the CLG household projections. 75. As previously discussed, headship rates are used to convert population projections to household projections. The ONS Mid-2011 Population Estimate identified that Arun had a total population of 149,800 persons at the end of June 2011, which suggests a household population of around 146,000 persons. By applying the headship rates (by age and gender) from the CLG 2008-based household projections, the ONS Mid-2011 Population Estimate translates to 67,700 households with an average household size of 2.16 persons. The headship rates from the CLG 2011-based household projections translate to 66,800 households in 2011, with an average household size of 2.19 persons. 76. It would appear that the CLG 2011-based headship rates provide a more realistic basis in the context of the current population. Nevertheless, it could be argued that the current economic circumstances are unrealistically suppressing household formation and if new households were more readily able to form and live independently, then headship rates might return to the levels assumed in the 2008-based projections. 77. Given this context, the analysis has therefore considered the impact of both 2008-based and 2011-based headship rates on the projected number of households based on the High-, Mid- and Low-trend Migration scenarios, after taking account of the population assumed to be living in communal establishments. 78. An example of the detailed calculation is set out in Figure 21. In summary:» The top of the table shows the calculations for the baseline population in 2011, and the bottom shows the projected population for 2031 based on the Mid-trend Migration scenario;» Columns A to C show the total male population, the proportion assumed to live in communal housing and the resulting communal establishment male population; and this analysis is repeated for the female population in Columns D to F;» Column G shows the resulting household population: the total male population plus the total female population, less those living in communal establishments;» Columns H to L show the 2008-based and 2011-based headship rates (Columns H and K respectively), and the resulting number of households for each. 23