Rural Housing Affordability

Published on December 2016 | Categories: Documents | Downloads: 28 | Comments: 0 | Views: 645
of 158
Download PDF   Embed   Report

Results of research into rural housing including definitions of local in the context of housing priorityFollow Colin Buchanan on twitter http://twitter.com/cbuchanancubed Read Colin Buchanan's blog at http://colinbuchanan.wordpress.com/

Comments

Content

Defra

Research into Rural Housing Affordability
Final Report

colinbuchanan.com

Research into Rural Housing Affordability
Final Report

Project No: 157701 May 2010 20 Eastbourne Terrace, London, W2 6LG Telephone: 020 7053 1300 Fax: 020 7053 1301 Email : [email protected]

Prepared by: ____________________________________________

Approved by: ____________________________________________ John Siraut Date: 22 March 2010

Nick Gallent/Steve Robinson/Chelsea Dosad/Ryan Emmett/John Siraut Status: Final document4
(C) Crown Copyright 2010. All rights reserved.

Issue no: 0

This report has been prepared for the exclusive use of the commissioning party and unless otherwise agreed in writing by Defra no other party may copy, reproduce, distribute, make use of, or rely on the contents of the report. No liability is accepted by Defra for any use of this report, other than for the purposes for which it was originally prepared and provided. Opinions and information provided in this report are on the basis of Defra using due skill, care and diligence in the preparation of the same and no explicit warranty is provided as to their accuracy. It should be noted and is expressly stated that no independent verification of any of the documents or information supplied to Defra has been made

Research into Rural Housing Affordability Final Report

Contents
Executive Summary 1 1.1 2 2.2 2.3 2.4 2.5 2.6 3 3.1 3.2 3.3 3.4 4 4.1 4.2 4.3 4.4 4.5 4.6 5 5.1 5.2 5.3 5.4 5.5 5.6 6 6.1 6.2 6.3 6.4 6.5 7 7.1 7.2 7.3 7.4 Introduction Background Context and study approach Local affordability Housing markets Communities, networks and priority Linking the three core issues Critical questions Selection of districts and parishes Background Selection of districts Selection of villages/parishes The evidence base & methods Housing affordability and its consequences Introduction Defining affordability and understanding its outcomes The geographical immediacy of local need Mobility, employment and affordable housing Local housing and the sustainability of rural communities Housing access and social networks 1 5 5 7 7 8 9 9 10 11 11 11 19 21 22 22 22 25 28 30 32

Influences on house prices 36 Introduction 36 Where do people come from 36 Functional characteristics 37 Social characteristics and identity 44 Differences between what attract new residents and keeps existing residents 44 Quantifying house price influences 45 Local priority Introduction Localness from a local perspective Local need as a flag of convenience Local need and economic necessity Assigning local priority, delivering sustainable communities Conclusions The study Affordability Influence on houses prices Local people 61 61 61 63 65 66 68 68 68 70 72 75 76 79 86 86

Appendix Appendix A: Literature Review Part 1: Local Affordability and its Impact on Communities Gentrification and displacement Affordability Part 2: Housing Markets and Market Determinants Understanding rural housing markets

Research into Rural Housing Affordability Final Report

Part 3: Communities, Networks and Local Priority Networks and community cohesion Localness and local rights Appendix B: Shortlisting areas for consultation Appendix C: Household survey Appendix D: Estate agent survey Appendix E: Hedonic pricing analysis Defining the dependent and explanatory variables Defining an appropriate functional form Specification of the hedonic price model: Selection of variables

96 96 98 121 123 127 129 136 136

Tables
Table 1.1: Table 3.1: Table 3.2: Table 4.1: Table 4.2: Table 4.3: Table 4.4: Table 4.5: Table 4.6: Table 4.7: Table 4.8: Table 4.9: Table 5.1: Table 5.2: Table 5.3: Table 5.4: Table 5.5: Table 5.6: Summary approach Selected case study districts Evidence base 6 14 21

Can local people working locally afford to buy homes in the village? 23 Proportion of respondents who think the supply of housing in the village is insufficient by housing type 23 If housing for local people is not available in the village, it is acceptable to expect them to move.. 27 If additional housing is built for local people, where should it go?28 How long to you expect to stay in this village? Compared to 5 years ago, do you feel you socialise with: 32 33

Over the past 5 years, have people you know well moved out of the village? 33 In the past 5 years, have people moved into the village who you have come to know well 34 Proportion stating the following factors are very or fairly important in influencing their decision to stay in the village 35 Where have you moved from? How far have you moved? Importance of transport accessibility in choice of village 37 37 38

How important is the availability of facilities in your choice of village? 39 How important is the quality of local schools in your choice of village? 41 Importance of rurality issues in choice of village 43

Research into Rural Housing Affordability Final Report

Table 5.7: Table 5.8:

Importance of community spirit in choice of village

44

Difference between what attracts and retains residents (positive number greater importance in retaining rather than attracting, negative number greater importance in attracting than retaining) 45 Results of correlation analysis: drivers of property price Hedonic model results for Mid Sussex Hedonic model results for North Norfolk Hedonic model results for South Shropshire Hedonic model results for Selby Quantifying the power of rurality Would you define the following as a “local person” 53 54 55 55 56 59 64

Table 5.9: Table 5.10: Table 5.11: Table 5.12: Table 5.13: Table 5.14: Table 6.1: Table 6.2: Table 6.3:

How long do you have to live in the village to be regarded as local 65 Are people who live in the vicinity local if… 65

Figures
Figure 3.1: Figure 3.2: Figure 3.3: Figure 3.4: Figure 3.5: Figure 3.6: Figure 3.7: Figure 3.8: Figure 4.1: Figure 4.2: Figure 4.3: Figure 5.1: Figure 5.2: Figure 5.3: Figure 5.4: Figure 5.5: Figure 5.6: Figure 5.7: Figure 6.1: Delimiting rural economies- key variables A typology of rural England and Wales Selected case study districts South Shropshire Regional Context North Norfolk regional context Selby regional context Mid Sussex regional context Overview of selected parishes Perceptions of housing affordability, Bishops Castle Local need, Cawood Time lived in the village Schools as house price drivers, Cuckfield ‘Access to solitude’ - Blakeney Mid Sussex detached property price per room (2007) North Norfolk detached property price per room (2007) Selby detached property price per room (2007) South Shropshire detached property price per room (2007) Rural/Urban aggregated definitions ‘Getting involved’, Handcross 11 12 14 15 16 17 18 19 24 26 31 40 42 48 49 50 51 52 62

Research into Rural Housing Affordability Final Report

Executive Summary
The Department for the Environment, Food and Rural Affairs (Defra) commissioned Colin Buchanan and Nick Gallent and Steve Robinson of University College London to conduct a study into rural housing centring on three inter-related issues: local affordability as it impacts on change in communities, stability and social networks; how the attributes of a rural location affect prices / affordability; the expectation of priority for housing amongst local populations, its rationale, its benefits and drawbacks and its achievability. The objective is to expand the Department’s rural evidence base, increasing its understanding of rural housing market dynamics and the part played by housing access in supporting the development of communities. The study consisted of: a literature review; case studies in four rural districts involving focus groups (mainly made up of local residents with formal and informal roles in the village e.g. shopkeeper, school teacher) and household surveys; and an hedonic price analysis to determine the influence of aspects of rurality on rural house prices. The literature review (summarised in chapter 2) led to the development of critical questions that were addressed in the focus groups and household surveys. The case study areas were selected to reflect the differing typographies of rural areas as defined by Lowe and Ward. The typologies and districts chosen were: Dynamic commuter: (relatively high population densities with young to middle-age, high-income professional classes predominating. Economically closely connected to adjoining urban areas with high levels of out-commuting) - Mid Sussex. Deep rural: (reflect traditional notions of the ‘countryside’, agriculture, particularly livestock farming and tourism form an important aspect of the local economy. Population and income levels are below rural averages) - South Shropshire. Retirement retreat: (popular retirement destinations, found particularly along the coast, retirement-related services including leisure, social care and health are significant sources of jobs) - North Norfolk. Transient rural: (situated relatively close to urban centres and, as sources of local employment are not substantial, they experience significant levels of outcommuting) – Selby. Within each of these districts, villages/parishes were selected that typified the broad characteristics of their typographies. The key findings of the study are set out under the three issues covered.

Housing affordability and its consequences
Rural housing affordability, explored in chapter 4, hinges a great deal around ‘within’ and ‘without’ affordability. Essentially, what are the consequences of the expectation that, on the one hand, housing within a village will meet local needs against the alternative expectation that access to affordable homes in nearby towns is sufficient? The case studies highlighted that the majority of respondents felt local housing was not affordable especially for young people and that supply issues, both in terms of the type of new housing built and approaches to the allocation of social/council housing were contributory factors. However, it was also recognised that while young people had had to move away from the village to find property, this move was often principally for employment reasons. There were mixed views as to whether it was acceptable for local people to have to move from the village to access affordable housing. In Mid Sussex and Selby over 60% of respondents believed it was acceptable to move to the closest town or elsewhere within the district to obtain housing while in South Shropshire and Norfolk that proportion was 47% and 30% respectively. In general in the less

1

Research into Rural Housing Affordability Final Report

transient and more dispersed case study areas it was significantly less acceptable for people to have to move away from their own village. If people should not have to move away from their village, where should new housing be provided? The focus groups tended to oppose new building believing that in the past it had not met a local housing need and as a consequence they preferred alternative solutions. The vast majority of survey respondents felt that housing should be provided within or on the outskirts of the village but to meet the needs of local people. There was also little contradiction in peoples’ views. So, for example, North Norfolk residents believed it was least acceptable for people to have to move elsewhere in the district for their housing needs and were the most receptive to having new housing built locally. Whilst the residents of Mid Sussex were more receptive to people having to move to neighbouring towns and were the most likely to prefer new housing to meet local need being built in the closest town. When asked how long they expect to remain in the village survey respondents in Mid Sussex were far more likely to move. Almost a third of respondents expected to move within ten years compared to less than a tenth in North Norfolk. Across all the districts, housing issues were the main reason for people considering moving - making up a third of reasons, suggesting that they did not expect that their future housing needs could be met locally. The next biggest reason for moving, in a fifth of cases was employment, followed by a desire to move closer to family members. Chapter 4 goes on to explore evidence for the importance of social networks in sustaining communities (from p.30), which emerged strongly from the case studies. This was in terms of the extent to which newcomers were considered active in village life and how this supported the village ‘life-cycle’ i.e. the balance between younger residents, families and older people. Life-cycle issues and their importance for the long term social sustainability of villages were raised consistently in the focus groups. It was felt that newcomers did not always support this through using local services (e.g. schools by or getting involved in the community even at the most basic level. In the surveys, residents who had lived in a village between 5-20 years typically socialised with more people than five years ago, suggesting that as time goes on newcomers’ circle of acquaintances increases within a fairly stable population as one might expect. It was clear that a village school or nursery and having children in general served as a ‘gateway’ to meeting people and forming networks. However, amongst the longer standing residents i.e. those who had lived in the villages more than 20 years, over a fifth reported that they socialised with fewer people as time went on. Older residents reported friends moving away or dying and finding few opportunities to meet newcomers, with the loss of local facilities such as pubs and less church going among younger residents. This was a factor commented upon in the focus groups where there was a concern raised about a number of older residents living in near isolation. In some cases newcomers (wealthier commuters, second home owners, younger) were seen as being very different with little in common with existing (longer-term) residents.

Influences on house prices
House prices and the factors that influence them provided a more quantitative focus in chapter 5. We approached the issue of influence from several angles: investigations in parish meetings (attempting to understand why migrants were drawn to particular locations and remained there), insights from the village surveys, a focused telephone survey of estate agents (highlighting patterns of apparent purchaser preference) and a statistical analysis of the Valuation Office residential property sales data. The initial literature review (from p.7) highlighted some of the obvious influences on house prices, linked to location, to property characteristics and to the value placed on living in a rural area (for example, the utility value derived from access to open countryside, the perception of peace and quiet and the belief that a greater community spirit can be found in rural areas). These issues were investigated in the meetings and surveys, with the aim being to tie generic attributes to the case studies and to particular locations. The key factors in attracting people to the case study villages, and

2

Research into Rural Housing Affordability Final Report

hence influences on house prices, were predominantly accessibility, local facilities, schools, rurality and the sense of community. The household surveys highlighted considerable differences between the villages in terms of the importance of transport accessibility in the choice of village of those who had recently moved in. For Mid Sussex, which attracted many out-commuters, the ability to commute conveniently to work scored highly. This was especially true in relation to rail access highlighting the importance of being able to access central London. In the other districts good links to neighbouring towns were more important than a convenient commute to work. While participants of the focus groups believed the availability of local facilities were important the household surveys stated this was less so in attracting recent incomers. With incomers having high car ownership the availability of services within the village itself was not regarded as crucial, but rather it was the ability to access such services in nearby towns that was important. The importance of the quality of local schools in attracting people to a village varied depending on the typography of the districts. In Mid Sussex, which had many professionals with young families, it was important to over 80% of recent incomers. In Selby, which attracted a more varied group of incomers it was only important to around half. However, the household surveys highlighted that the most important factors in attracting people to the villages surveyed were those related to their environment or rurality. Across the villages around 90% of respondents reported that issues such as peace and tranquillity, quality of the surrounding environment and safety and security were important in their choice of location. The next step was to take the findings of the survey and see if these could be quantified in terms of their influence on property prices (from p.45 chapter 5). Simply mapping house prices (using average price per room data) across all four districts showed prices were higher in the rural areas. However, the price of a home is some function of its structural and locational attributes which may include: Structural (number of rooms, plot size, garage, garden, age etc); Neighbourhood (e.g. crime rates, employment rate, private home ownership etc); Accessibility (e.g. distance to schools, shops, transport etc); and Surrounding environment (e.g. quality of open space, natural and built environment, noise level etc); Housing supply constraints relative to more urban areas It is not possible to obtain data for all these factors, especially those in relation to property specific issues and relative supply constraints (which were treated as a given in the rural areas surveyed) but this was possible for many of the locational attributes. For each attribute that data is available correlation analysis was undertaken to determine key explanatory variables of house prices by district. Once this had been done regressions were run to assess the degree to which the explanatory variables determined the variation in house prices in each district. The limitations of the analysis should be noted in terms of the “goodness of fit” of the results which, depending on the district, are only able to account for 50-60% of the variation in price. This reflects that some structural attributes not captured within the model are important factors influencing house prices. However, it was clear that in terms of the locational factors rurality was a key driver in all districts being responsible for around 10% of the total variation in property price. However, there are varying degrees of rurality, with prices rising as one moves further from urban areas but then they start to fall again moving into very rural areas as access to services becomes remote. Other specific factors vary by district. For example in Mid Sussex access to schools and services were most important, whereas in North Norfolk, being in an Area of Outstanding Natural Beauty (AONB) was responsible for the largest variation in price.

3

Research into Rural Housing Affordability Final Report

Local Priority
The third focus of the study, set out in chapter 6, was the potential policy response: the case for interventions that support ‘local’ needs, whether existing interventions have local support or how communities would like to see them adjusted to better reflect local realities. The household surveys looked to investigate this in much greater detail in an attempt to reveal exactly who village residents consider to be ‘local’. So how is a local defined? Someone who lives in the village is generally regarded as local, usually followed by someone who used to live in the village and still has a family connection to it. After that there are some marked differences between districts. In Mid Sussex people who are active or who work in the village but who do not live there are much more likely to be regarded as local than is the case in the other districts. A second issue within this theme is how long you have to live in the village before you are considered a local. Again Mid Sussex stands out as being more accepting, with the majority of respondents regard someone a local who has lived one year or more in the village, whereas in Selby 48% believe you have to be in the village for more than 10 years before you can be regarded as local. Across all the villages it is the case that the longer a respondent has lived in the village, the longer they believe another person should have lived in the village before they can be regarded as local. In terms of distance from the village, people who live in the closest village but not in the nearest town or the wider county are generally considered as local. Having defined a local person, the final question was concerned with whether people who are local should receive priority in terms of locally provided affordable homes. There were differences in this respect between the focus groups and the survey results. The focus groups were clearly of the view that any additional housing should principally go to local people. The survey indicated that respondents were more open to the needs of people across the district but that it was appropriate that some account should be given to local people in the allocation of affordable homes within the village. The full conclusions of the report are set out from page 68, but in summary these include: That local affordability (driven in part by lack of supply of a range of property types) is an issue that leads to young people moving away from rural areas but there is also a widely shared recognition that employment factors play an equal part. The acceptability of people having to move away to the nearest town to access housing varies between the districts, but is generally higher in more dynamic (i.e. commuting affected) areas than in more isolated areas. Networks are built around family-ties and for in-migrants, in particular, the use of local facilities especially schools. In those localities with a higher turnover of people (leading to fewer local family ties) and few facilities (and hence less interaction) it can become harder to maintain networks, especially for older people who can become isolated. Rurality is a key driver of property prices provided there is still access to key services (even if these are in neighbouring towns). Other key drivers relate to the typography of the district. So where many residents commute out of the village the ability to access employment outside the district is important. The definition of a local person again varies by typography; districts with a more transient population embrace far more people as local, while in more stable communities and in older populations people are regarded as requiring much closer links with a village and to have lived there longer to be regarded as local. Nevertheless, a willingness to become involved in the community has a big impact in attitudes, and facilities such as schools are important in assimilating newcomers. The predominant view was that the needs of local people should be taken into account in the allocation of local affordable housing and some villages had sought their own routes to achieving this. However there is also recognition among the existing population that people outside the area with housing needs should also be accommodated.

4

Research into Rural Housing Affordability Final Report

1
1.1
1.1.1

Introduction
Background
The Department for the Environment, Food and Rural Affairs (Defra) commissioned Colin Buchanan and University College London to conduct a study into a range of issues centring on the supply of housing in rural areas. Defra is interested in exploring the forces at work in rural housing markets and exploring the views of people affected by them. More specifically the objective of the study is to expand the Department’s rural evidence base, increasing its understanding of rural housing market dynamics and the part played by housing access in supporting the development of communities. The project places a strong emphasis on developing understanding from a residents’ perspective, with in-depth analyses of communities across rural England. This understandably covers a wide range of issues, but Defra are particularly interested in the following subject areas being explored further:

1.1.2

An informed discussion of the concept of ‘local affordability’.
1.1.3 Average house prices in rural places are higher than in urban places. There are fewer houses available within the price band that is described, for any particular locality, as ‘affordable’ in rural than in urban areas. However, this effect can be substantially reduced in most rural areas if the market for potential homes is expanded by a few miles. Less expensive housing, of a type suitable for first-time buyers and young families can be found in the nearest market town or urban area and often in closer proximity to the range of jobs and services that these people might need. In addition to this, to what extent is it the case that people need to stay in close proximity to ‘social networks’? Indeed to what extent do people tend to stay within the boundaries of where they originate if other things are equal, for example, the quality of the place or the opportunities available? Defra is seeking a challenging and informed discussion around these issues, drawing on the best available evidence, to enhance knowledge of this area.

1.1.4

Analysis of the power of rurality in various forms to predict rural house prices and where this sits relative to other determining factors
1.1.5 This relates to the factors which contribute to explaining rural house prices. In particular, to what extent is being rural in itself a contributor to the hedonic value of a dwelling and how much of this is universal. In other words, how much of the value of an average house is determined by (a) being rural and (b) being in this rural settlement. Of particular interest is the role of exploring amenity values and how much people are willing to pay to live in more attractive rural environments.

An informed discussion of the reasons and consequences of including ‘local ties’ in eligibility criteria for social housing or assisted home ownership
1.1.6 There is an expectation in some quarters that the descendants of rural families should have some additional rights to remain resident in their families’ neighbourhood than is afforded by their ability or willingness to do so. The final issue, therefore, is to examine the case for extending and strengthening policies that target assistance to ‘local households’, helping those having a connection with an area, through work or residence,

5

Research into Rural Housing Affordability Final Report

to access homes. 1.1.7 In response to the original project brief, the research team formulated the following approach: Table 1.1: Summary approach

Summary Approach 1 2 A B C 1.1.8

Systematic Literature Review across the three Issues 4 District Case Studies focusing on specific parishes Economic data analysis across each district Open meetings at the parish level Sample household surveys within the identified parishes

The aim was to address the three key issues in an integrated way. Firstly, to review what had already been written and the existing studies that have looked at these issues. For this reason, we proposed that the three questions be bundled together and examined through a systematic review of the existing literature. This, secondly, was to be ‘refreshed’ through four local case studies that focused (a) across the rural districts, providing an economic analysis that would help explain house price influences, on specific parishes that typified a range of local situations firstly through parish meetings with key informants (b) and then through follow-up surveys (c) which aimed to test the generality of views, opinions and explanations offered in the meetings. The approach and the data it generated, is explained more fully in the next section. A systematic review of past literature was conducted in the first phase of this project. The complete version forms an appendix to this report. From the literature review, a series of critical questions emerged: issues of continuing debate surrounding the impacts of housing affordability, influences on prices (from accessibility to the qualities of particular rural experiences) and the rights of local households. These critical questions formed a second key output and were transformed into practical questions for the parish meetings and follow-up surveys. But before the empirical work could begin, a critical stage in the project was the selection of districts and parishes able to represent the range of rural experiences across England, accepting that not all rural areas are the same but rather experience contrasting pressures, impacting on residents and communities in a variety of ways. The report is structured as follows: Chapter two provides a brief synopsis of the literature review which led to the critical questions that formed the basis for the village consultations and surveys; Chapter three explains the choice of districts and villages which were surveyed; Chapters four to six report the findings of the consultation, surveys and house price analysis; and Chapter seven presents our conclusions. The full literature review, details of surveys and hedonic house price analysis are provided in the appendices.

1.1.9

1.1.10

1.1.11

6

Research into Rural Housing Affordability Final Report

2
2.1.1

Context and study approach
This chapter sets out the approach to the study and the context for the research based on a summary of the literature review, the full version of which can be found at appendix A. Drawing on existing literature and past studies, it addresses three critical issues: local affordability as it impacts on change in communities, creates stability and helps preserve social networks; how the attributes of a rural location affect prices / affordability; the expectation of priority amongst local populations, its rationale, its benefits and drawbacks and its achievability. Under these three headings, the following key messages emerge:

2.1.2

2.2
2.2.1

Local affordability
Affordability is a key concept for planning and public policy. It is a measure of the linkage (usually expressed as a ratio) between local wages and local house prices (or the average loan credit needed to purchase a property on the open market) or between wages and rents. Successive studies have identified an affordability problem affecting some rural areas. This often relates to strong demand for local homes combined with low rural wage levels and inadequate supply of affordable housing. These three factors may come together, in certain locations, to produce difficulties for some rural households, even when the majority of households in an area are adequately housed. It is a selective problem, in most instances, rather than a general issue facing all rural households. Affordability has a spatial scale. Across entire housing markets, there may be a ‘satisfactory’ relationship between incomes and house prices. This means that the ‘ratio’ may be low and people on a typical local income may be able to afford to buy or rent a home without any apparent difficulty (being able to secure adequate loan credit). But when residential choice and need is put into the equation, the picture may become more complex. Households may not be able to exercise choice (to stay, for example, close to friends or relatives) or live close to work, because (for example), planning is restricting supply in an area (maybe in a particular village) when exogenous consumption habits are pushing up prices. This may create a localised problem that does not extend across the entire area. Seen at this local scale, low levels of affordability (expressing complex relationships between planning, consumption habits and local wages) may undermine social balance. Many issues and debates stem from this status quo: issues of basic social equity and local rights; issues of social balance and sustainability; issues of community capacity and cohesion; and questions over what constitutes rural sustainability and sustainable communities in rural areas. As access to housing is determined by income and wealth, it may underpin an observed pattern of gentrification in some locations. Those who are attracted to a rural location because of its rural nature rather than to access a local employment market can lead to changes in the socio-economic make up of the community. Whether this gentrification leads to displacement of the ‘local’ population is a disputed topic: is it driven by the consumption habits of an ex-urban population, or is change in the population profile merely symptomatic of underlying forces causing social change (economic restructuring and a consequent lack of attractive jobs). Most analyses steer a middle path, concluding that demand and supply side forces (including the supply of jobs

2.2.2

2.2.3

2.2.4

2.2.5

2.2.6

7

Research into Rural Housing Affordability Final Report

and other opportunities) conspire to produce the patterns of change seen in the modern countryside;

2.3
2.3.1

Housing markets
Rural housing markets have key functional characteristics (settlement pattern, housing mix etc), important political economy characteristics (how regulation treats them) and social characteristics (how buyers conceive rural areas and perceive the benefits that they offer). A basic understanding of rurality can provide some clues as to what drives rural housing markets. Perceptions of rural areas are important in driving and shaping patterns of demand, especially demand from non-rural households wishing to move to the countryside to gain a better quality of life or to down-shift. Physical and social perceptions are important: the quality of the environment and the perceived nature of rural communities and the ‘neighbourliness’ of rural people. But rural housing markets are also shaped by tangible, hard factors (that may have softer underpinnings): by demographic shift, economic restructuring and consumption patterns (for example, second home buying since the 1960s). However, rural markets are not homogenous. These forces act out differently on different rural areas and markets are locally differentiated. For example, patterns of consumption will be driven by convenience: by accessibility to schools and services, to market towns, to main roads (facilitating commuting flows) and so on. Markets are physically and socially differentiated: the preferences of retired households are second home seekers will differ from those of working households, families with children or young people setting up home for the first-time. It is possible to identify drivers, but whilst these may shape markets at a macro level, at a micro level, individual choices and needs will play an important part in determining patterns of demand and property prices; Hedonic price theory tries to bring together hard and soft drivers: not only macro and micro patterns, but also hard locational and property characteristics with less tangible drivers. Hedonic analyses often deal with location, neighbourhood, structure (the property) and externality. They try to mix things that people are directly paying for (number of bedrooms for example) with the benefits they come to enjoy through property purchase, which therefore may have had some (indirect) impact on price; But arguably, the psychological drivers of rural property prices are difficult to measure. In particular, the preconceived benefits of a rural lifestyle (especially if reality does not live up to expectation) may be difficult to quantify. Home buyers may compete for rural property, driving up prices, because of a preconception that has more to do with television and media portrayals of the countryside that experience. However, the idea of the rural idyll (a potentially powerful determinant of prices) cannot be discounted out of hand. For incomers with money and means, rural areas may live up to the idea of the ’idyll’. But those with less income may find rural living more of a challenge, especially if they are forced into a competition for housing with incomers who will pay a premium for a home in the country; Assembling the data for hedonic price analysis can be challenging. Some variables may be given undue weight; or there can be accidental or co-incidental relationships which provide spurious explanations for price patterns.

2.3.2

2.3.3

2.3.4

2.3.5

2.3.6

2.3.7

8

Research into Rural Housing Affordability Final Report

2.4
2.4.1

Communities, networks and priority
The meaning of community is keenly contested. It is sometimes defined as a group of individuals who share both common values / beliefs and a common space. They think similar things and their similarity in belief (in identity and culture) is a product of living in close proximity. Using this definition, what we have is a ‘place community’, community centred on a physical location. Two interpretations have accompanied the process of community change. Change is interpreted either as a ‘loss’ or erosion of community or as a ‘transition’ from one form of (traditional) community to something different and inherently more modern. The loss of place-based communities is accompanied by the rise of network-based communities. This is the essential transition of the modern age. On the one hand, this can be seen as a tide that cannot be turned; but on the other, it can be viewed as an unacceptable change in rural areas which are built around closely-knit place-based social networks. It is an anathema to many people to project 20 years hence and see a complete breakdown of place-based community networks: villages as dormitories for electronically networked city workers, who shop and socialise online. From the rejection of this vision flows a belief that action must be taken to preserve the social characteristics of village England. This moves the debate full-circle. Strong local social networks are seen to be the realm of local households, whose ties and attitudes are inherently local. By protecting the interests of such groups, protection and enhancement of local community is the inevitable outcome (or so the argument goes). Local rights and local priority then become issues of community preservation: local people are seen as the life-blood of the community. They are thought to socialise and work locally and to have short but intense networks (and emotional ties to place) in contrast with incomers whose networks are extensive but not as anchored in the local community. Alongside arguments concerned with equity and need, this provides a justification of a kind for granting local priority. But policy-makers are then left with the challenge of setting appropriate local priorities and dealing with the thorny issue of who, exactly, is local. The setting of such priorities can have an adverse effect on the operation of local housing markets. But the case for priority must be seen in the context of delivering sustainable rural communities. Social mix and balance appears to be an important ingredient and social networks have practical benefit to society and to the national economy. These networks not only support identity, they also provide care for the young and the old. In some areas, there may be a strong case for using local housing priority policies to sustain social networks. Elsewhere, the case for preferential support may be more limited: particularly in areas where the transition to a more extensively networked community is more progressed.

2.4.2

2.4.3

2.4.4

2.4.5

2.4.6

2.4.7

2.4.8

2.5
2.5.1

Linking the three core issues
The issues summarised above have a circular relationship. Affordability ties together the key drivers of market and social change. It is an expression of the relationship between housing supply, planning, incomes and consumption habits. It is changing communities. Arguably, gentrification is a social-economic expression of the transition to extensively networked communities, but it can also intensify networks where incomers comprise retired households or down-shifters.

9

Research into Rural Housing Affordability Final Report

2.5.2

The brief analysis of market drivers here provides us with a point of departure for better understanding the impacts of consumption and supply change. Local priority and community issues are, perhaps, the area in which this study is likely to make its key recommendations. Working backwards, priority policies will not reshape housing markets, but may counter the forces of gentrification where, in particular instances, this process may be undermining the local networks that play a crucial role in sustaining communities.

2.6
2.6.1

Critical questions
The output from the literature review led to the development of a series of critical questions that formed the basis of the parish consultation and surveys. These can be summarised as: 1. Affordability and impact on communities: a) How affordability is understood b) Local debates about within and without affordability c) Affordability and mobility: to what extent can travel costs offset the gains of moving further afield; d) How lack of affordable housing may affect achievement of UK Sustainability Strategy goals; e) Gentrification / displacement and its impact on networks and social capacity The determinants of rural house prices: a) The extent to which functional characteristics can explain prices b) The extent to which preconception drives migration and prices Attitudes towards local priority: a) Local priority as a means of mustering support for development b) Priority and economic need: an essential link? c) Priority as a means, simply, of delivering social mix and therefore sustainable communities

2.

3.

10

Research into Rural Housing Affordability Final Report

3
3.1
3.1.1

Selection of districts and parishes
Background
A key part of the study is to undertake consultation and survey work within a range of rural communities. This chapter outlines the approach used to select the districts and then the villages within those districts that were to be surveyed.

3.2
3.2.1

Selection of districts
It was agreed early on that the team would select districts on the basis of a classification of rural England recently constructed by Lowe and Ward (2007). Relying on data from the 2001 Population Census (using Local Authority Districts as the spatial unit), Lowe and Ward constructed a typology comprising seven area types, each representing specific and tangible social and economic characteristics and the various ways that these currently manifest themselves. Having removed ‘urban districts’ from the Census database (using Defra’s own classification of rural areas), data for more than 100 variables were subjected to a factor analysis enabling the identification of 15 critical, uncorrelated variables, from which to develop a new typology centred on economic rurality (Figure 3.1). A map depicting the geography of rural types across England was then generated as shown in Figure 3.2. Figure 3.1: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Delimiting Rural Economies- key variables

number of residents per 100 hectares; percentage change in resident population 1991-2001; proportion of population aged 65 and over, 2001; percentage change in total employment 1991-2001; rate of economic growth 1991-2001; average total income 2000-2001; proportion of ‘knowledge workers’ to total economically active 2001; percentage change in ‘knowledge sector’ employment 1991-2001; proportion of managerial and professional workers to total economically active, 2001; number of hotel and restaurant businesses; proportion of employment in agriculture, hunting and forestry relative to total economically active; proportion of households with two or more cars 2001; net commuting 2001 (GB=100); number of national heritage sites per 1000 sq. km 2002; index of tranquillity 2001 (GB=100).

11

Research into Rural Housing Affordability Final Report

Figure 3.2:

A typology of rural England and Wales

3.2.2 3.2.3

The seven countryside types identified by the authors and their characteristics are outlined below. Dynamic Commuter Areas: Socially and economically dynamic and affluent, these areas have relatively high population densities and young to middle-age, high-income professional classes predominate. Economically, they are closely connected to adjoining urban areas with high levels of out-commuting. They can be considered as a form of wealthy outer suburb. A concentration of these areas is found in South East England. Settled Commuter Areas: Similar to the above but less economically vibrant, these areas are located on the edges of the provincial conurbations such as Birmingham, Bristol, Manchester, Newcastle and Sheffield. The fortunes of these areas are closely linked to economic trajectories of their related city regions, again due to high levels of outcommuting. Dynamic Rural Areas: These slightly lower density rural areas have fast growing economies and an increasing population. Characterised by high concentrations of professional and knowledge workers they experience lower levels of out-commuting and are less connected economically to urban conurbations than the above. The presence of major institutions such as universities or research centres tends to be a key feature of these areas which are to be found mainly south of a line between Avon and the Wash. Deep Rural Areas: These areas perhaps best reflect traditional notions of the ‘countryside’. Agriculture, particularly livestock farming and tourism form an important aspect of the local economy. Population and income levels are below rural averages. Economically and in terms of net migration, they may appear to be in steady state, but

3.2.4

3.2.5

3.2.6

12

Research into Rural Housing Affordability Final Report

recent events (such as foot and mouth disease in 2001) have revealed fragility about this situation. Commuting to urban areas is limited (due to physical remoteness or poor infrastructure) as is the ability to attract younger in-migrants and entrepreneurial activity. These areas are found, for example, in Northumberland and mid Devon. 3.2.7 Retirement Retreat Areas: These areas form popular retirement destinations and are found particularly along the coast, most notably in Southern England. Retirement-related services including leisure, social care and health are significant sources of jobs. Although income levels tend to be below average, the relatively high population density of these areas means that service provision is economically viable. With the ageing population fuelling demand, the economies of these areas are relatively vibrant. Peripheral Amenity Areas: These rural areas present some of the most significant economic and social challenges in the countryside. Located in marginal zones often in coastal locations such as the west coast of Cumbria, the economic structure is dominated by agriculture, tourism and retirement-related services. However, they are secondary tourist or retirement destinations due to their isolation, or poorer quality environment, sometimes associated with past industrial or mining activities. Income levels are well below the national average amongst both existing and incoming populations, attracted by low house prices. Generation of new economic activity tends to depend on public intervention. Transient Rural Areas: These areas are situated relatively close to urban centres and, as sources of local employment are not substantial, they experience significant levels of out-commuting. However, unlike the Dynamic and Settled Commuter Area categories, commuting does not imply high incomes. Although the majority of the population is economically active, average income levels remain very low. These areas are not attractive for entrepreneurial activity. The East Riding of Yorkshire and South Norfolk serve as examples. Like all classifications, Lowe and Ward’s typology allocates areas to a class based on best fit: the picture they generally adhere to. However, there are also differences at the very local level, between specific parishes and villages. For this reason, it was felt that for the purpose of this study a smaller number of cases could be used to represent the seven types noted in the Lowe and Ward typology, given the possibility of focusing on different villages with their own particular characteristics. Classifications capture the varied nature of communities (and markets) that characterise rural England: areas that are out of the way and dominated by those retreating to rural areas; areas under the influence of a major urban economy; areas that are economically peripheral and isolated; and areas where villages are in the sphere of market towns but where incomes are not necessarily high. Following the decision to employ the above typology as a critical frame for case study selection, the team set about collapsing the typology into four critical types: ‘Dynamic Commuter’; ‘Rural Retreat’; ‘Deep Rural’; and ‘Transient Rural’, capturing issues of commuter influence, retirement, relative isolation (from urban influence) and areas that remain economically peripheral despite proximity to apparently stronger urban economies. Four areas were selected, indicated in Table 1.1 and Figure 3.3, in order to represent this collapsed typology of rural England, drawn from a long list following discussions with the steering group.

3.2.8

3.2.9

3.2.10

3.2.11

3.2.12

13

Research into Rural Housing Affordability Final Report

Table 3.1:

Selected case study districts Regional Location West Midlands East of England South East Yorkshire and the Humber Type Deep Rural Rural Retreat Dynamic Commuter Transient Rural

1 2 3 4

District South Shropshire North Norfolk Mid Sussex Selby

Figure 3.3:

Selected case study districts

3.2.13

This selection was based on not only fit with the Lowe and Ward typology, but investigations that revealed issues of critical interest, discussed below.

14

Research into Rural Housing Affordability Final Report

South Shropshire: deep rural
Figure 3.4: South Shropshire Regional Context

3.2.14

Key characteristics of the district of interest are: Second homes and retirement pressures across the district, but concentrated in market towns such as Ludlow; Pressures concentrated in stock of smaller properties. This has two discernible effects: suppresses household formation and increases pressure on social housing; Creates pressure for more affordable housing, which has been the council’s priority in recent years (Gallent et al, 2002); Fears that housing associations are overwhelmed; would welcome more consistent way to supply local housing for local people; Retirement and commuting seen as big issues for the district, bigger than second homes; Desire to create the economic context in which communities flourish. Summary: A district of mixed rural pressures which appear to come together around Ludlow/Shrewsbury

15

Research into Rural Housing Affordability Final Report

North Norfolk: retirement retreat
Figure 3.5: North Norfolk regional context

Concentrated second home pressures in the coastal zone, especially around Cromer; Second homes result in a ‘cultural transformation’: grocery stores become upmarket delicatessens, for example; Gentrification (and displacement) is one outcome, but so too is a perceived positive benefit for the area; Affordable housing is a solution, but causes some social polarisation. General building is inappropriate, it is not possible for rural areas to ‘build themselves out’ of their problems; Social rented sector solutions are important; shared ownership is more difficult: it is suggested that this may not create opportunities that are sufficiently affordable for those in acutest need; Real resistance to trying to ‘plan out’ second homes: second homes seen to have legitimacy for a variety of reasons; The desirability of setting local priorities strongly resisted, but local people need to be ‘empowered’; Desire to avoid ‘retirement ghettoes’. Summary: key pressures focused around retirement and a desire to empower local communities to formulate their own responses.

16

Research into Rural Housing Affordability Final Report

Selby: transient rural
Figure 3.6: Selby regional context

‘Deep rural’ villages in the hinterland of a key provincial centre; Increasing influx of retirement and older households, ‘balanced’ by the outflow of the younger population and associated political concerns; Legacy of undersupply of social housing and former miner housing, relatively affordable and attractive to incomers; Resident population generally work in lower paid occupations compared to newcomers; Commuting an issue for areas with closer access to Leeds and York transport links; Second homes less of an issue in this district, which continues to haemorrhage population from some of its rural parts. Summary: an area with critical economic challenges, for which the loss of younger population members is potentially a greater issue than the arrival of newcomers.

17

Research into Rural Housing Affordability Final Report

Mid Sussex: dynamic commuter
Figure 3.7: Mid Sussex regional context

Within the influence of London commuting area as well as Brighton; Relatively low presence of pensioner households but growing retirement-related demand and associated demand for sheltered housing; Proximity of London means there are other demographic/ lifestyle factors present e.g. younger households starting families away from the capital (‘economic refugees’ from London’s housing market); Reluctance to accommodate further growth within villages due to infrastructure constraints and a fear over the further erosion of rural character; Newcomers tend to be more pro-active and well connected e.g. in resisting new development; Summary: an area facing a variety of urban influences (physical and social) and characterised by commuting pressure and an increasing degree of retirement.

18

Research into Rural Housing Affordability Final Report

3.3
3.3.1

Selection of villages/parishes
Following the selection of districts, the next stage involved the selection of specific villages/parishes1. The aim here was to identify parishes (using ward level data from the 2001 population census) that typified the broad characteristics of their district: e.g., they were characteristically ‘dynamic commuter’ or obviously ‘rural retreats’. This selection involved the examination of Census data and began with a ranking of all parishes in each district by population size and the deletion of those in urban areas and those without an active Parish Council (the decision was taken to focus on parishes where there were likely to be obvious points of contact, usually beginning with a contactable Parish Clerk). A long list of candidate parishes was drawn up for each district before being whittled down to three to four proposed consultation areas. A full description of the methodology and analysis that was undertaken and the recommendations made for the selection of consultation areas is set out in Appendix B. For South Shropshire, the aim was to identify villages (within parishes) that were deeply rural, being economically marginal (reflected in activity rates), affected by second home and retirement pressures and with high proportions of pensioner households. The parish of Bishop’s Castle and the village of Hope Bowdler (to the east of Church Stretton) were eventually selected. In North Norfolk, the clear objective was to find parishes with high proportions of retired households and correspondingly low levels of economic activity. Happisburgh, near Mundesley and the small coastal villages of Blakeney and Wiveton, just north of the market town of Holt, were selected on this basis. For Mid Sussex, the critical selection criterion was evidence of high levels of outward commuting. This was often coupled with high levels of educational attainment and a concentration of people in managerial and profession economic classes. It was also linked to strong migration inflow. Analysis of Travel to Work data provided the basis of identifying outward commuting and showed that several candidate parishes had significant numbers of residents commuting to Crawley and London. Within these same parishes, house prices far exceeded the district average. Using this combination of criteria, Slaugham Parish (comprising four small villages) and the village of Cuckfield were selected for analysis. Finally, some of the same characteristics were examined for Selby, but this time the emphasis was placed on the exchange of incoming and outgoing residents, dependence on the regional centre (Leeds in this case) and on variation in house prices, socio-economic classes, rates of economic activity and by inference, incomes. Using these criteria, the villages of Riccall and Cawood were identified and selected as this district’s consultation areas. Brief descriptors of all the selected parishes are provided in figure 3.8. Figure 3.8: Overview of selected parishes
Hope Bowdler is located in the Apedale Valley which runs eastward from Church Stretton and is within the Shropshire Hills Area of Outstanding Natural Beauty. The parish’s resident population in 2001 was 233 in 95 households. The village is entirely dependant on the services found within Church Stretton, just 1.5 miles away, with no services, apart from buses to Church

3.3.2

3.3.3

3.3.4

South Shropshire Bishop’s Castle is one of the district’s smaller market towns with a resident population of 1630 in 707 households. It is a local service hub, with two banks, a post office, three grocery stores, several cafes, six pubs and two breweries. The parish council website is run from Enterprise House, an important social and economic centre
1

The terms parish and village are often used interchangeably which in reality may not always be the case. In some instances a parish may contain a number of villages while in other instances they are in effect the same entity. The approach adopted was to use parishes due to the benefits of being able to utilise the knowledge of the local parish clerk in identifying potential consultees in an area.

19

Research into Rural Housing Affordability Final Report

for the town. An IT drop-in centre, the ‘Village Outreach Project’, several regeneration projects, a library, a suboffice of South Shropshire Housing Association, Dial-a-Ride and several private offices all operate from this location. North Norfolk Happisburgh is a larger coastal village (with 1,372 residents in 607 households) located roughly 12 miles from Cromer. It is dominated by two land-mark buildings, St Mary’s Church and the Lighthouse, both of which are open to visitors. One notable feature of the villages is its vulnerability to coastal erosion, with some houses now precariously perched on the cliff edge. The village has a single pub, a village store and post office, a fishmonger and a primary school. The nearest medical centre is in Mundesley Mid Sussex Slaugham Parish, with its 2226 residents in 951 households, comprises four villages: Handcross (the largest), Pease Pottage, Warninglid and Slaugham (the smallest). Handcross is the obvious hub of the parish, having a mix of shops, two schools, two churches, three pubs and some light industrial units. The Parish Hall at Handcross provides a social focus for the community. The other villages, including Slaugham, have some specific shops though residents tend to see Handcross as the main local service centre.

Stretton in the village itself.

Blakeney and Wiveton are separate parishes located roughly 9 miles west of Sheringham. Blakeney, the larger of the two villages (with a population of 789) was once a commercial sea port, but its harbour is now silted up. The village is popular with tourists and has two large hotels and several pubs. The historic Blakeney Guildhall also draws visitors to the village. Wiveton, on the other hand, is much smaller (with 158 residents in 74 households) and located just over a mile to the south of Blakeney. It has a single pub, the Wiveton Bell, which is popular with visitors. The village and parish of Cuckfield, with a population of 3266 in 1342 households, was the largest settlement examined in this study. It is located less a mile from the edge of Hayward’s Heath (with its 23000 residents). The village centre has a range of services including shops, restaurants and pubs. It was once the market town for the district. Today, the Queen’s Hall contains the parish council office and museum as well as a library and meeting rooms for hire. Other key services include a doctor’s surgery, a pharmacy and a post office. It has excellent transport links: a mainline rail station only 2 miles away, easy access to the M23 and Gatwick Airport is some 15 miles. Cuckfield is enveloped by Cuckfield rural parish, which was not the focus of this study. Cawood is a village of 1700 residents centred on the historic Cawood Castle and located on the Yorkshire Ouse. It was once a market town, but this function ceased in the 19th as nearby Selby (6 miles to the south) came to dominate the commercial life of the area. Today, the village has three pubs, a post office, a primary school, one shop, a hairdresser, two doctors’ surgeries, a church and a Methodist chapel. An old boy’s school at the centre of the village is used as a meeting venue and is currently rented to the parish council.

Selby Riccall in the Humberhead Levels is only 3.5 miles north of Selby and 9 miles south of York. In 2001, it had a population of 2317 in 922 households. The village has a range of services: a primary school, post office, grocery store, hairdresser, butcher, a tavern and inn, Italian and Indian restaurants, a village institute comprising two halls for hire and the ‘Regen Centre’, a conference, events and community facility. Until 2004, the Riccall Mine, part of the Selby Coalfield, was an important local employer. It is now the site of a business park.

20

Research into Rural Housing Affordability Final Report

3.4
3.4.1

The evidence base & methods
The approach set out in Table 3.2 yielded the following critical outputs, generally and linked to the specific case studies. These provide the research evidence for this report. Table 3.2: Task 1 Evidence base

2 3 4 5 3.4.2

Methodology Systematic literature review across the three issues South North Norfolk Mid Sussex Shropshire a Consultation b Consultation c Consultation reports (2) reports (2) reports (2) a Household b Household c Household survey (1) survey (1) survey (1) a Agent b Agent c Agent survey survey survey a House price b House price c House price analysis analysis analysis

Selby d d d d Consultation reports (2) Household survey (1) Agent survey House price analysis

The literature review is presented in full in Appendix A; the approach to selecting the case study areas is presented in full in Appendix B; the Household Survey findings in Appendix C; the findings of the Agent Survey in Appendix D; and the House Price Analysis (with technical description) in Appendix E. The consultation reports provide the basis of the discussions around the three critical issues presented in the next three chapters. How was this evidence base developed? The following is a summary of the methods used in this study: Parish meetings (leading to consultation reports): Two meetings were held in each of the selected parishes, with the Parish Clerk normally providing the point of initial contact. The meetings employed a ‘pinpoint facilitation’ approach with questions evolved from the initial literature review. The meetings focused on the three themes of the study, but mixed treatment of specific questions with broader discussions around general themes such ‘what it means to be local’. Meeting discussions were recorded and transcribed, but facilitation notes were also taken: recorded on large A1 sheets that provided references for the discussions. Household surveys also centred on questions evolved from the literature review and were conducted by professional survey teams in the four locations. The teams administered door-to-door interviews throughout the day and into the evening in order to gain a representative sample of household types. Estate agent survey was conducted by telephone. A total of 13 estate agents were contacted across the four locations and asked a series of questions, namely which villages tended to be the most popular or sought after and why; whether people pay a premium for these villages; and whether buyer’s could expect more for their money by purchasing a property closer to/within a town. Additional discussion was tailored specifically around the issues of critical interest raised by the rural typology of each location. House price analysis drew upon VOA (Valuation Office Agency) property sales data for the four districts. The data was used to assess the extent to which a rural location adds value to residential property in each district. This was determined through the application of the hedonic pricing technique, which uses a series of regression analyses to compare prices of properties when there are observable differences in the attributes of each property sold. The next chapters present the results of the above analysis in terms of addressing each of the three core study questions.

3.4.3

3.4.4

21

Research into Rural Housing Affordability Final Report

4
4.1
4.1.1

Housing affordability and its consequences
Introduction
This study is not concerned with quantifying levels of housing affordability in villages or across parishes. Such analyses already exist and provide the basis of local policy interventions. Rather, its purpose was to uncover the experiences that ensue from a real or perceived lack of housing that is affordable to local residents. The discussions designed to unearth these experiences and local attitudes, were rooted in the review of past studies conducted at the beginning of this project. The review highlighted a number of critical concerns: Differences in the way that affordability is understood, its geographical characteristics, issues of who benefits and issues of who is adversely affected by low levels of affordable housing: just individual residents or communities as a whole; Local differences concerning ‘within’ and ‘without’ affordability: whether measures of affordability could, or should, deal with immediate village or parish needs or whether policies should focus on wider areas. Essentially, whether there is an expectation that the housing resources of a village or parish will meet all local need, or whether access to affordable homes in nearby settlements is sufficient for dealing with local need; But related to this, local debates over mobility and employment: whether it is reasonable to expect people working in a village to move away but commute back for employment; How affordability contributes to the sustainability of rural areas and communities: whether the addition of more affordable housing contributes to (by facilitating more balanced communities) or detracts from (by consuming rural land resources) sustainable development; and linking to this; How access to housing affects social networks and social capital, whether networks have evolved, in some instances, as a result of gentrification, or whether they have been eroded and severed because of displacement attendant on inmigration. These were broadly the issues emerging from the literature review and provided a frame of reference for the parish meetings and village household surveys.

4.1.2

4.2
4.2.1

Defining affordability and understanding its outcomes
Housing affordability can be a sensitive subject within some communities. Accepting that there is an ‘affordability problem’ may be regarded as an admission that a community needs additional housing. Where such housing is not wanted, residents may prefer to talk about the different ways of ensuring that people get the housing that they need, avoiding reference to issues of ‘affordability’, which may be seen as a green light to unwanted development. We looked to disentangle some of this association in the household surveys conducted across each district. People were asked whether in their view, people on local wages can afford to buy homes in their village. The results indicated that across all four districts2 housing affordability was an issue. Just over 70% of respondents in North Norfolk and South Shropshire had the view that people on local wages could not afford local housing, whilst in Selby and Mid Sussex the figure was just over 50%, table 4.1. It is interesting to
2

4.2.2

The household survey results are presented throughout the report at a district level which is a shorthand for the two villages surveyed in each instance which in turn represent the four typologies outlined in table 3.1

22

Research into Rural Housing Affordability Final Report

note that in the former two districts a far higher proportion of people are dependent on local employment than in the later two and there are also greater second and holiday home pressures in the former. Table 4.1: Can local people working locally afford to buy homes in the village? Mid Sussex 52% 24% 24% North Norfolk 74% 15% 11% Selby 51% 35% 15% South Shropshire 71% 10% 19%

no yes do not know 4.2.3

The survey respondents were very clear in their views that there were not enough affordable homes in their village. Apart from Mid Sussex nearly two thirds of respondents reported an insufficient supply of affordable homes, as shown in table 4.2. In Mid Sussex those who felt there was sufficient affordable housing stressed the level of new housing that had recently been built and their concerns about their village growing too big. There was also a concern about attracting the “wrong type” of person to the village. Elsewhere respondents stated that either there were no affordable homes or in districts like Selby that former council homes had been brought under ‘right to buy’ legislation and had not been replaced. When respondents were asked whether housing supply by housing type was sufficient or not in the village there were some clear differences based on typologies. Table 4.2: Proportion of respondents who think the supply of housing in the village is insufficient by housing type Mid Sussex 68% 51% 44% 40% 33% North Norfolk 81% 89% 53% 70% 59% Selby 80% 81% 41% 53% 38% South Shropshire 80% 80% 43% 53% 62%

4.2.4

4.2.5

Housing type housing for young people affordable homes housing for the elderly housing for rent housing for families 4.2.6

In all cases respondents felt there was insufficient housing for young people (smaller starter homes) with the homes built locally tending to be larger. However, it was also felt that property types like flats were not appropriate to villages. In addition the employment opportunities for young people were to be found in larger urban areas. So while respondents reported that their children had had to move away from the village to find property they also recognised this move was often for employment reasons. The availability of housing for the elderly raised some interesting points. With the exception of the Selby villages, all the villages surveyed had a higher than average elderly population when compared to the district as a whole. The reason why respondents reported a lack of housing for the elderly was due to the lack of sheltered accommodation in their villages. They also raised the point that older people had left the village due to lack of local medical and care facilities. It was felt the provision of sheltered accommodation would not only enable the elderly to stay within the village but would free up housing for younger people thereby improving the age profile of their villages. For other types of housing the main issue was cost, housing was available in the village but in the views of many respondents it was not affordable due in part to pressure from second homes in some locations.

4.2.7

4.2.8

23

Research into Rural Housing Affordability Final Report

4.2.9

Interviews with estate agents in highlighted the impact of second homes in some districts. For example, in North Norfolk agents identified a unique sub-market within the district along the western coast revealing a cycle of demand for holiday homes, generated by the popularity of its character villages (stretching west of Cromer and along to Sheringham, Blakeney and Wells-next-the-Sea), which has led to considerable development and increased tourism, reinforcing demand and continual upward pressure on property prices. In the Slaugham Parish Consultation (Dynamic Commuter), affordability was defined by reference to ‘other rural areas in the country’: if housing costs in the parish compared favourably to other areas, then it could not be said that housing was unaffordable. More specifically, it was considered impossible to conclude that the parish had an ‘affordability problem’ by relating incomes to property prices. In Bishop’s Castle, however, there was a more precise interpretation of affordability in terms of local incomes. Such relationships highlight a potential for difficulty but this potential only becomes real if it can be shown that named individuals needing to live in a village are being prevented from doing so as a result of high prices. In fact ‘unaffordability’ is meaningless unless linked to need. And what was the actual need? In Slaugham, the need and desire of people not already housed in the parish to live there was questioned: such people ‘prefer more urban locations’ and so development would not meet their needs. Slaugham typifies the tendency in more ‘mobile areas’ to project the value of mobility on households that might find themselves in housing need. Moving away from the village was not seen as a threat to social networks, unless the move involved an older and less mobile person. In such instances, there is a case for housing, presumably affordable and certainly sheltered, to meet the needs of the elderly. And indeed, decanting older residents from family homes was also viewed as a means of dealing with other housing tensions (the same belief was aired in Happisburgh). Figure 4.1: Perceptions of housing affordability, Bishops Castle

4.2.10

4.2.11

Following the addition of a recent housing development in the village, it was found that up to 75% of those people in Bishop’s Castle who applied for the affordable homes in the scheme were unable to get a large enough mortgage (this was prior to the 2007 credit crunch). As a result a number of these homes went unsold and were eventually placed on the open market, while the affordable homes for rent meanwhile were well over-subscribed. This outcome was attributed to the affordability formula applied by the Housing Association which used an average income variable which did not reflect local wage levels (it was based on the district average which was considerably above that of the town). The consultees therefore felt that the price of supposedly affordable new housing in addition to open market housing remains some way out of reach of those living locally and in need of housing. 4.2.12 Similar perspectives were offered in other cases, reinforcing the view that ‘affordability’ only becomes a real issue if prices prevent people from living where they need to (otherwise, high prices have no obvious downside). In Happisburgh (Rural Retreat), property prices, combined with low turnover in social housing, had prevented young people from staying in the village and local area. But, was the real barrier homes or jobs? Young people went to Norwich, ‘where the work is’, so what is the real need for affordable housing in the village? The desire to innovate solutions (and which were being pursued by the District Council), rather than building new affordable homes (for ‘people on benefits’), was also strong in Happisburgh. Might it be possible to relax restrictions on barn conversions if they are to be used as social housing, rather than solely for rich incomers? Conversion costs are likely to be prohibitive for social providers, but such sentiments underscored the desire to limit development to that which was absolutely essential.

24

Research into Rural Housing Affordability Final Report

4.2.13

Elsewhere, a view was aired that a focus on ‘affordability’ does not necessarily result in interventions that aid local people with a legitimate need. In Hope Bowdler (Deep Rural), it was strongly felt that ‘affordable housing’ eventually ends up being unaffordable to locals. It is often sold on at market cost (a problem that government has recently recognised and sought to remedy) or is not affordable in the first place to those who really need it. In the same village, it was argued that ‘affordability’ is held up as a reason to build more homes (perhaps with the aim of addressing an imbalance in supply and demand), resulting in large developments that will detract from the environmental qualities of the area and be intrinsically problematic owing to the lack of infrastructure investment in this and similar rural areas. Similar sentiments were expressed in Riccall, although there was a feeling that inevitably further new developments would be built and a risk of the village ‘growing out of all proportion’. Still that new developments may eventually help in meeting local need meant that there was generally less outright resistance to new housing. Two issues emerge from this and from the discussion above. Firstly, ‘affordability’ is not necessarily the trigger that local communities think should drive development; rather, this should be a measured response to evidence of very local need, with solutions led by the community (a view strongly held in Slaugham and Bishop’s Castle). Secondly, where affordability concerns (presumably tracked by the principal authority) do trigger house building, there is a concern that this building is often inappropriate, goes to the” wrong people”, mainly someone who is not local and is not accompanied by the right infrastructure investments. The normative concept of affordability is understood, but it is contested as a basis for development. There was no sense that affordability (as a tension) locks lower income residents out of social networks, but there is consensus around the need to help specified local households access housing: essentially, to formulate local solutions that are not driven by top-down economic analysis.

4.2.14

4.2.15

4.2.16

4.3
4.3.1

The geographical immediacy of local need
This issue boils down to something very simple: is there belief in the need for more development within villages, or a view that local need should be dealt with in the wider area, particular in nearby market towns. Bramley (2009) talked about ‘within’ and ‘without’ affordability. He argued that many smaller rural villages are unaffordable if they are seen as discrete markets; but, that if these markets are extended to cover larger centres then levels of affordability rise. Furthermore, it makes sense to concentrate development in bigger centres where needs can be more easily and more efficiently serviced. In essence, ‘within’ affordability is frequently low in many parts of rural England, whereas ‘without’ affordability is high. If villages are seen as discrete markets, then their housing problems become intractable: if they are seen as components of a network of settlements, including larger centres, then solutions seem nearer at hand. So do villages see needs as being ‘geographically’ immediate or as something to tackle across the wider area? The consultation at Slaugham delivered an interesting story. Participants were firstly asked to rate the need for affordable housing in the borough. Contradicting some earlier views on affordability as a guiding concept, they were unanimous in the view that the supply of affordable housing in the parish was ‘inadequate’ (using the parish’s recent housing needs survey as a point of reference). But when asked whether this meant that they would accept new development, the response was to backtrack and to claim that the supply of affordable housing was adequate, perhaps ‘not in the parish’, but certainly in the ‘nearby towns that are local to the parish’.

4.3.2

4.3.3

25

Research into Rural Housing Affordability Final Report

4.3.4

Still, was the topic of discussion ‘local housing’ or ‘affordable housing’? It was accepted that there could be a case for local housing for named individuals and that there might be a need in Handcross, Slaugham, Pease Pottage or Warninglid for this. But affordable housing was seen as a more general term and affordability could be addressed across a wider geographical area. This view was echoed in Hope Bowdler and Riccall, where it was expected that future development would be directed to brownfield sites in Church Stretton and the towns of Selby/York: ‘expansion should occur in the larger centres’. Even where housing is needed for older residents, this should be provided in the market towns where the elderly are more likely to have access to the things they need. Interestingly, a countryside that is often dominated by retired people is not always seen as a suitable place in which to grow old, or rather it is viewed as sensible for the frail elderly to move up the urban hierarchy. Again, there was a sense that a mobile group of consultees had the expectation that others would share their level of mobility. Furthermore, they were clearly averse to development within their rural ‘oases’. The view at Happisburgh was rather different. The village is one of only five coastal service villages in Norfolk and there is an expectation that it will experience some additional housing growth in the years ahead. There was, in July 2009, a proposal for 15 houses in the village, of which half would be affordable. This was considered to be insufficient (in the face of market pressures, especially from second home buyers). Unlike the situation in Slaugham, there was a belief that Happisburgh should take more growth, even though it was only 12 miles from Cromer. However, it was highlighted that a lack of affordable housing was a problem for the whole of North Norfolk, not just in the smaller settlements. The village is more of a centre than Handcross in relation to its neighbouring villages (mainly due to the area being more rural) and therefore an acceptance that Happisburgh is a likely candidate for development. There is some willingness amongst the retired consultees to countenance more housing, but coupled with more efficient use of the existing stock. For example, by discouraging second home ownership (through amendments to the national tax regime), bringing empty properties back into use, allowing the re-use of commercial buildings (e.g. barns) as housing and/or extracting contributions from those converting buildings into holiday/second home use. Figure 4.2: Local need, Cawood

4.3.5

4.3.6

4.3.7

4.3.8

Cawood is situated a few miles west of Riccall in the district of Selby and which, like its neighbour, had experienced an influx of commuters over many years. In contrast to Riccall it had not seen any major additional housing development and any new development which had appeared in the village was usually too small to require the developer to include affordable housing. The consultee group which was mainly made up of an established core of active residents and those in positions of responsibility (e.g. headteacher, doctor) generally took the view that this situation meant that young people wanting to remain in the village and those now with families who had grown up in the village and with a desire to return to Cawood were unable to, with many living in nearby Selby. It was felt that the relatively small number of social and other supported housing such as agriculturally tied homes, Alms houses, those owned by charities and council housing could be directed towards this group. However, over time most of these had been sold onto the open market which only exacerbated the problem. 4.3.9 It is generally true that consultations in near-urban areas concluded that additional housing should be directed to a nearby town, whereas those held in slightly bigger centres and in deeper rural areas concluded that more land for housing should be

26

Research into Rural Housing Affordability Final Report

allocated within the village or parish. This was clearly the case in Bishop’s Castle (Deep Rural): a ‘poor town’ in which affordable homes for rent are ‘massively over-subscribed’ and ‘starter homes are not particularly affordable’. Here, the dominant view was that there has already been a considerable amount of new build, but usually of the wrong type. There is a ‘within’ affordability problem that cannot be addressed by building elsewhere or by prevailing models of affordable provision. These do not deliver homes that are affordable to people on local wages. 4.3.10 In many respects, Bishop’s Castle fitted the mould as a village willing to address its own housing difficulties through development that could be locally controlled and that would be ‘truly affordable’. Conventional policies were seen to have failed the community: there was not enough social rented provision, intermediate tenures were unaffordable and exception policies were unable to overcome the desire of landowners, who held onto land in the hope of achieving full development value. Local control was seen as the answer, channelling development through a Community Land Trust (CLT). There was even an expectation that landowners would be more willing to deal with a CLT, which could offer iron-clad guarantees that local housing would benefit local people. In Blakeney similar actions have been taken to address housing problems through the creation of a housing association supported by the Parish Council. Settlements across the coastal area were too small to support affordable housing and the district wide allocation of affordable housing in larger centres too often ignored the needs of the smaller villages particularly in recognising the needs of those who also worked locally. The problem of ‘within affordability’ was therefore felt across all settlements. Whilst some analyses, therefore, point to the inherent difficulties of dealing with discrete markets, in Bishop’s Castle and Blakeney a very local focus was seen to offer clear opportunities, by virtue of its capacity to muster support for affordable housing development. The consultations with parishes revealed the trade-off between accepting the ‘affordability problem’ and local need and the resistance to affordable housing development within residents’ immediate community. The household surveys conducted in these areas looked to shed more light on the availability of housing and who should get housing priority. The survey results also highlighted differences between districts in terms of the acceptability or otherwise of residents having to move out of the village to obtain housing. The results varied across the districts; as shown in 5.4. Table 4.3: If housing for local people is not available in the village, it is acceptable to expect them to move.. Mid Sussex 92% 86% 73% 66% North Norfolk 74% 58% 49% 30% Selby 66% 67% 61% 60% South Shropshire 80% 66% 53% 47%

4.3.11

4.3.12

4.3.13

to a village within the parish to surrounding villages to the closest town within the district 4.3.14

In Mid Sussex and Selby over 60% of respondents believed it was acceptable for local people to have to move to the closest town or elsewhere within the district to obtain housing. While in South Shropshire and Norfolk the proportion fell as low as 30%. A sizeable minority, from a third to a fifth of respondents felt it was not appropriate for people to move away from their own village, in the villages outside Mid Sussex. . If respondents do not feel it is appropriate that people should have to move away from their village where should new housing be provided? The vast majority of respondents (between 61 to 79% depending on district) felt that housing should be provided within or

4.3.15

27

Research into Rural Housing Affordability Final Report

on the outskirts of the village to meet the needs of local people. There was also little contradiction in peoples’ views. So, for example, North Norfolk residents believed it was least acceptable for people to have to move elsewhere in the district for their housing needs and the most accepting of having new housing built locally. Whilst the residents of Mid Sussex were more accepting of people having to move to neighbouring towns and the most likely to prefer new housing to meet local needs being built in the closest town. 4.3.16 Overall the surveys found a greater acceptance to new housing in their village than in the consultation events although this acceptance is related to the housing being provided for local people. Table 4.4: If additional housing is built for local people, where should it go? Mid Sussex 41% 20% 21% 18% North Norfolk 40% 39% 13% 9% Selby 23% 53% 9% 14% South Shropshire 38% 30% 18% 14%

on the outskirts of the villages in the village itself in the closest town in the surrounding villages 4.3.17

When asked to explain the reasoning behind their answers there were a number of common themes. Those who did not want more homes built in their village itself stressed the view that the village should be kept as it is, that more housing would change its character, there was no room for more housing, that greenfield land should not be used and that housing should be built in locations where there were a greater number of facilities. The contrary view was that local people should be able to obtain housing in their own village and that there was plenty of space available both brownfield and greenfield.

4.4
4.4.1

Mobility, employment and affordable housing
The literature review highlighted the gap between local wages and local affordability. The speculation in the consultations was that these additional travel costs accentuated the economic disadvantage of local workers who had already been priced out of village networks. But the evidence supporting such a view was very limited. It was perhaps only in the North Norfolk case studies that a clear concern for the ‘pricing out of village workers’ was aired. In Happisburgh this was linked to providing services particularly in meeting the needs of the retired population, whereas in Blakeney it was the implications for the wider economy, such as the loss of skills. The difficulties faced in this part of Norfolk were characterised in the following way: on the whole, there is a high proportion of retired or near-retired households, who have arrived from outside of the county to buy their homes outright; they tend to stay in the area and do not move again, resulting in low market turnover; over time, the services they need tend to cease as the young people needed to operate them is reduced. The picture in Happisburgh was not one of displaced workers commuting back to villages to run local shops, but of workers leaving the area entirely, often heading to Norwich and taking up higher quality jobs there. This pattern of change was common across most of the case study parishes: the lack of quality jobs in the villages was a recurring theme, as was the view that employment is as big a driver of out-migration amongst the young as the cost of housing and in fact bigger in many instances. It is a common thesis in studies of rural housing to link migration pressure to the relative employment opportunities in urban and rural areas and to explain movement early in the life-cycle in terms of economic push and pull factors. Rural depopulation has been explained in these terms for a hundred years, but it is only the last

4.4.2

4.4.3

28

Research into Rural Housing Affordability Final Report

40 years that depopulation has been balanced by counter-urbanisation. 4.4.4 However, the prevailing view today is that these life-cycle movements, rural to urban movement of younger households and urban to rural movements of older households, are no longer in balance: more people want to move to rural areas than want to leave, resulting in ‘aggressive displacement’. An important question is whether younger people would stay if there were jobs or whether there are jobs but the lack of local housing is the real ‘push’ factor. Two different pictures emerged and were again linked to area type. In Slaugham (Dynamic Commuter), jobs and homes were thought to be readily available in nearby centres contesting the suggestion that younger people would wish to stay in the parish: these people are rational and the rational action is to move to areas of better employment. In Riccall and Cawood (Transient Rural), however, it was suggested that lifestyle incentives, family-ties and the availability of some local employment meant that young people had a reason to stay, though were being prevented from doing so because of a lack of affordable housing for purchase and rent. Their problems had several underlying causes. Firstly, there was a local resistance to social housing, which, in the case of Riccall, was believed would bring in ‘the wrong sort of people’. Secondly, the market had delivered a glut of larger family homes (and planning had not prevented this from happening) unsuited to the needs of young singles and clearly far out of their price range. And thirdly, housing associations had failed to provide homes for young people and were perceived to show a lack of consideration to their needs in their allocation policies. These factors had conspired to produce a perverse situation in which family homes remained empty and unsold, social housing went to people from outside the village and young people were finding it increasingly difficult to stay. The view in the consultation was that some young people at least had decent jobs and wanted to stay, but on a ‘minimum wage’, there was ‘no chance’ for this to happen. But even in Riccall, where there was strong support for more housing for young people, it was more broadly accepted that rural areas cannot be expected to retain all of their population. It is in the nature of some areas to be attractive to younger people and in the nature of others to attract the retired or those who are highly mobile and have no employment need in the area in which they reside. The banner ‘local homes for local workers’ was held up as an aspiration, but it was also accepted that there is an economic reality that needs to be faced. If that reality is not faced then villages will be forced to grow in pursuit of an unachievable fit between access to homes and low-paid jobs. How much housing will be needed to achieve this fit and will the village still be sustainable in terms of its land take and its service profile? Despite an acceptance of local housing, one conclusion from the consultation in Riccall was that there are ‘plenty of derelict sites in Leeds’ as well as empty homes, all of which should be brought back into effective use. Generally, there was little evidence that local workers ‘back-commute’ from more urban locations. When they are gone, they are gone (as in the case of Happisburgh). Local wages simply do not support this additional cost and the idea of ‘back commuting’ suggests that housing is a bigger push factor than employment, which is simply not the case. Some displaced households may move in order to access a more plentiful supply of social housing (indeed, their move may well have been due to the allocation of a home elsewhere), but they also take advantage of employment opportunities. In fact, the move may well bring them closer to the job they already had.

4.4.5

4.4.6

4.4.7

4.4.8

29

Research into Rural Housing Affordability Final Report

4.5
4.5.1

Local housing and the sustainability of rural communities
Explicit reference to what might constitute a ‘sustainable (rural) community’ was made in only one consultation: in Riccall (Transient Rural), it was thought that excessive growth would cause an imbalance between population and service provision, diminishing the sustainability of the village. There was no faith that either new employment or new services would accompany that growth. A recurring view across rural England is that planning often turns communities into dormitories, without due regard to their future economic or social viability. This view was shared in nearby Cawood (Transient Rural). Tightly drawn settlement boundaries mean that the Cawood is a ‘village without growth’, being designated as such in the Local Development Framework. The lack of large housing developments in or adjoining the village has perhaps contributed to a dearth of affordable homes, but at the same time it has meant the retention of the village’s identity. New housing is viewed here as a potential threat to local character and something that would further undermine the sustainability of the community. It would do this not because of the consumption of farming land, but because new housing is rarely accompanied by an investment in services. Participants at the Cawood consultation based this belief on ‘observations’ made in the wider area. This would mean that new housing would not ‘generate’ new services and the people moving in would be unlikely to frequent existing village shops. Rather, commuting households tend to do their shopping on the way home from work, or use the internet and have groceries delivered to their doorsteps. It was felt that such shopping habits have become ‘ingrained’ and, when combined with service-free development, mean that rural communities are becoming dormitories. Such patterns of development, combined with these consumption habits, reinforce cardependency and in this respect additional housing does not provide a means of making villages more sustainable: it merely amplifies existing problems. On paper, new housing might provide local bus services with a critical mass of potential customers, but in practice the new residents moving into this housing are unlikely to use the bus: and the service will only begin to improve once people start to make use of it. The lack of belief in the propensity of housing to solve basic community challenges, centred upon services but also linking to social capacity and vibrancy, fuels a reluctance to countenance ‘substantial’ development and a preference for more affordable housing on small plots. Local housing can contribute to a community’s sustainability if it is ‘controlled and targeted’. In Cawood, the preference was for development targeted at helping those ‘starting out, youngsters etc’. Critically, it is important that affordable housing ‘will benefit the community’: the view was that the village is ‘growing old’ and that the ‘life-cycle’ of the community was broken and needed fixing. Development polices in neighbouring villages had not remedied similar problems. Nearby Thorpe Willoughby, for example, had become a dead dormitory for Leeds and York. General development did not understand or meet the needs of the communities: homes were quickly filled either by commuters or by ‘people out in the sticks with no cars, where they can’t function’. Thorpe Willoughby was a particular settlement highlighted in the interviews with estate agents in Selby, as a typically bad example of inappropriate development and growth that has detracted away from the village’s original character and has made it less popular with new buyers as a result. Agents confirmed this has been the case across a number of villages situated just outside of Selby town centre. Appropriate, sustainable development, happens when the needs of a village are fully understood: villages have a life-cycle and development must acknowledge and support this. Interestingly, it is across this life-cycle that social networks were thought to operate: of particular importance are the intergenerational ties between older residents and their

4.5.2

4.5.3

4.5.4

4.5.5

4.5.6

30

Research into Rural Housing Affordability Final Report

children. The lack of housing for younger households obviously breaks or weakens these ties and has an impact on the community as a whole. 4.5.7 The life-cycles of social networks were explored further in the household surveys. Village residents were asked to state how long they had lived in their village, the results of which presented a stark contrast between the districts. Our results suggest that the life-cycle of most districts is very similar irrespective of their rural typology, with the exception being ‘dynamic commuter’ areas. In Selby, North Norfolk and South Shropshire, the majority of residents had resided in their village for at least 20 years. In Mid Sussex however, residents tended to have only lived in their village for 1 to 5 years. This is further emphasised by looking at the combined proportion of residents that have lived in their village all their lives or for at least 20 years: on average accounting for 40% of responses in Selby, North Norfolk and South Shropshire. In Mid Sussex the opposite is true, where in fact 40% of residents have lived in their village for less than 5 years, Figure 4.3. Figure 4.3:
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Mid Sussex North Norfolk District Selby South Shropshire All my life 20+ years 10 to 20 years 6 to 10 years 1 to 5 years Less than a year

Time lived in the village

4.5.8

The surveys also looked to reveal the strength of familial ties within rural villages and networks generally across all four districts and how this could vary by life-cycle. Residents were asked whether they had any other family members (aside from those living at home) living in their immediate village. The proportion who answered yes varied from 12% in Mid Sussex to 35% in South Shropshire reflecting the different characteristics of each district. When assessing responses by life-cycle, a notable finding was that the strongest family-ties within a village were for those residents who had lived there more than 20 years. For this group a third had other family members living in the same village, a figure that falls to a sixth for those who have lived for less than five years in the village. So as mobility increases over time local family-ties are likely to reduce. The figures are even more pronounced in terms of respondents’ future expectations. When asked how long they expect to remain in the village the respondents in Mid Sussex were far more likely to anticipate a move. Almost a third of respondents expected to move within ten years compared to less than a tenth in North Norfolk, table 4.5.

4.5.9

31

Research into Rural Housing Affordability Final Report

Table 4.5:

How long do you expect to stay in this village? Mid Sussex 6% 14% 12% 68% North Norfolk 3% 5% 2% 91% Selby 2% 8% 6% 84% South Shropshire 5% 7% 5% 83%

less than a year up to 5 years 5 to 10 years plus I am not thinking of moving

4.5.10

Across all the districts housing issues (e.g. a desire to move to a bigger or smaller property or wanting to buy) were the main reason for people considering a move (making up a third of reasons). This would suggest that they did not expect their future housing needs could be met locally. The next biggest reason, in a fifth of cases, was employment followed by a desire to move closer to family members.

4.6
4.6.1

Housing access and social networks
To be sustainable, communities need to be balanced. Specifically, this means that a village needs a ‘life-cycle’: a balance between younger residents, families and older people. Across this life-cycle, family-ties are important, sometimes focused around caring for elderly relatives and providing childcare. In Cawood, ensuring ‘community balance’ was viewed as crucially important and ‘open policies’ of allocating homes were said not to work as they often fail to understand the way communities function. Even a mix of homes, leading to a mix of people, will not necessarily lead to a balanced community. Policies need to build on what is already there, recognising the imbalances that exist and seeking to address them. It was felt in Cawood that ‘local youngsters’ were drawn back to the community once they had children of their own and that, ideally, these were the sorts of household that should be assigned priority access to new housing. But the perceived practices of housing associations were claimed to impact negatively on communities. In Riccall, they were said to lack concern for creating a working community. As well as paying insufficient attention to how ‘new residents will get around’ and it was felt that they display scant concern for ‘interaction’ between residents. It was felt that by giving some thought to allocation procedures it would be possible to engineer more harmonious communities. It was clearly felt in Riccall that the needs of an ‘active community’ would be best served by assisting young local people. This would be possible if the power to grant access to affordable housing were transferred from housing associations to the local community: a view shared widely across the consultations. Indeed, in order to make a community ‘work’ it is necessary to understand its existing dynamics. Where there are imbalances, the vibrancy of a community may be undermined. However, what constitutes balance is highly contested. In Bishop’s Castle (Deep Rural), there was considered to be an imbalance caused by holiday homes that at least brought a spending stream to the village and second homes that it was felt did not. This view was shared in nearby Hope Bowdler, though no obvious remedy emerged. In Slaugham, population dynamics created what was generally thought to be a working community built on ‘enlightened self-help’ with strong support networks, perhaps not built on ‘social balance’ but on the energy of people whose interest lay in being part of village life. This was also reflected in Cuckfield, where the range of voluntary and sporting/leisure activities as well as the presence of schools had enabled strong social connections to be built up over time between established families and more recent incomers. Interestingly in Slaugham, there was a desire to assist ‘truly local people’ by allowing the Parish Council

4.6.2

4.6.3

4.6.4

32

Research into Rural Housing Affordability Final Report

to set priorities. 4.6.5 Evidence for the importance of social networks in sustaining communities emerged more strongly from the village surveys, in terms of the extent to which newcomers were considered active in village life and how this was maintained throughout the life-cycle. The surveys provided a useful comparison of this concept from the perspective of residents who have lived in the village all their lives with the perspective of recent arrivals. The first question asked whether people who had lived in the village for more than five years whether they socialised with more or fewer people than five years ago, table 4.6. Those residents who had lived in a village between 5-20 years on balance now socialised with more people than five years ago. This suggests that as people become more settled in the village their circle of acquaintances increases within a fairly stable population as one might expect. The main way people had expanded their acquaintances in the village was through their children going to the village school or to baby and toddler groups. However, amongst the longer standing residents, those who had lived in the villages more than 20 years, over a fifth reported that they socialised with fewer people and by a small proportion more reported socialising with fewer rather than more people. Older residents reported friends moving away or dying and finding few opportunities to meet newcomers, with the loss of local facilities such as pubs and lower church going among younger residents. In some villages it was clear that the school gate was the main place where people met and once children left school the opportunity to meet new people even within a small community was limited. In some villages newcomers were seen as being very different with little in common with existing (longer-term) residents (wealthier, second home owners, younger). Other long-standing residents though looked forward to new people moving in as they felt their village was too cliquey. Table 4.6: Compared to 5 years ago, do you feel you socialise with: Length of time living in village Up to 5 5 to 20 Over 20 years years years 3.4% 11.9% 21.1% 47.1% 49.4% 48.1% 40.0% 60.0% 18.9%

4.6.6

4.6.7

Less people in the village About the same More people in the village

4.6.8

Villagers were then asked if people who they knew well had left the village in the last five years. Not surprising in Mid Sussex where there was a far higher turnover of people a higher proportion answered yes than in the other districts, table 4.7. Table 4.7: Over the past 5 years, have people you know well moved out of the village? Mid Sussex Yes No 41% 59% North Norfolk 21% 79% Selby 26% 74% South Shropshire 32% 68%

4.6.9

In the cases of Mid Sussex and Selby the main reason reported why people moved was

33

Research into Rural Housing Affordability Final Report

for employment reflecting the nature of these districts as commuter locations while in North Norfolk and South Shropshire the reason was more because of housing issues such as a desire to move to a bigger property or to buy. Other reasons for moving were related to health and a desire to live closer to members of their family as residents grew older. 4.6.10 When asked what impact that people they knew moving out of the village had had, there was a mixed view. Around half said it had no impact while the other half reported a negative impact in terms of losing friends who they had less contact with or with whom they had lost contact. The latter seems to apply mostly to older people who also see a much higher proportion of their friends dying. This was a factor commented upon in the consultation where there was a concern raised about a number of older residents living in near isolation. When residents who had lived in the villages for more than five years were asked about how well they had got to know newcomers the answers were perhaps surprising, table 4.8. Table 4.8: In the past 5 years, have people moved into the village who you have come to know well? Length of time living in village Up to 5 years No Yes 69.6% 30.4% 5 to 20 years 50.4% 49.6% Over 20 years 55.5% 44.5%

4.6.11

4.6.12

Less than half of residents felt that they had got to know new residents well and where they had it was because the incomers were immediate neighbours. A lack of facilities in some villages meant that people did not get the chance to meet fellow villagers. However, there were plenty of examples of residents making friendships with newcomers and friendliness within villages was still seen as a strong positive attribute. Why do people stay in their villages? This was one of the questions in the household survey addressed to those who had been living in the village for more than five years and is later compared to what attracted people to the village in the first place. It is clear from table 4.9 that it is attributes of rurality that makes the villages attractive places to live rather than the facilities available or their accessibility. Many of the “rurality” measures were rated as very or fairly important by over 90% respondents across all the villages.

4.6.13

34

Research into Rural Housing Affordability Final Report

Table 4.9:

Proportion stating the following factors are very or fairly important in influencing their decision to stay in the village Mid Sussex 98% 97% 96% 95% 91% North Norfolk 97% 96% 95% 88% 92% Selby 95% 95% 95% 93% 97% South Shropshire 99% 94% 97% 92% 90%

Factor quality of the surrounding environment safety and security peace and tranquillity sense of neighbourliness and atmosphere in the village the personal appeal of my own house e.g. physical/ aesthetic attributes the physical attributes of the village (nice high street, buildings etc) good road links to neighbouring towns being close to family and friends proximity to nature and recreational activities facilities available in the village local bus service provision proximity to a rail station convenient commute to work quality of local schools 4.6.14

90%

79%

89%

87%

90% 82% 82% 78% 64% 56% 55% 46%

70% 74% 85% 61% 50% 13% 41% 31%

86% 63% 92% 84% 70% 19% 53% 35%

81% 80% 87% 70% 56% 35% 47% 39%

The differing typologies of the villages show through in terms of the importance of different aspects. Accessibility to transport and work were clearly more important in Mid Sussex and Selby, with their higher proportion of people commuting out of the district for work, than in North Norfolk and South Shropshire.

35

Research into Rural Housing Affordability Final Report

5
5.1
5.1.1

Influences on house prices
Introduction
House prices and the factors that influence them provided this study with a more quantitative focus. We approached the issue of influence from several angles: investigations in parish meetings (attempting to understand why migrants were drawn to particular locations), insights from the village surveys (with a focus on the motivations of residents drawn to or choosing to remain in a certain location), a focused telephone survey of estate agents (highlighting patterns of apparent purchaser preference) and a statistical analysis of the VOA residential property sales data. The initial literature review highlighted some of the obvious influences on house prices, linked to location, to property characteristics and to the value placed on living in a rural area (for example, the utility value derived from recreational advantages, the perception of peace and quiet and the belief that a greater community spirit can be found in the countryside). These issues were investigated in the meetings and surveys, with the aim being to tie generic attributes to the case studies and to particular locations. This enabled the team to establish an expectation of what might be influencing prices. The VOA data allowed the study to test whether expectation conformed to reality. But more generally and in the light of the critical questions emerging from the literature, the study was concerned with: The extent to which ‘functional characteristics’ explain prices: for example, proximity to the services that local communities need (schools, shops, health care and so on), proximity to jobs and to the transport infrastructure that makes some areas more accessible than others and the quality of housing within particular areas, whether it is the housing that buyers want and are willing to pay for; and The extent to which ‘preconceptions’ drive migration and influence prices. For example, are migrants to the areas being examined motivated by belief in a ‘rural premium’: lower levels of crime, better schools, peace and quiet, community spirit and a generally ‘green and pleasant’ land? The expectation from the outset was that it would be difficult to gauge the effect of preconception. Methods for modelling price influences draw on measurable data: preconceptions that may have proven false are clearly going to be difficult to measure, let alone factor into a model that arrives at judgements on the relative importance of different drivers. However, the strategy adopted was to analyse the quantifiable but then qualify this with observations made during village meetings and the surveys: to at least highlight the uncertain influence of taste and perception.

5.1.2

5.1.3

5.2
5.2.1

Where do people come from
Before getting into the detail of particular drivers it is worth exploring where people have come from who have moved recently into the various study villages. From the village surveys there were clear differences between districts. The dynamic commuter and to a lesser extent the rural transient districts (Mid Sussex and Selby) attracted far more urban residents, while in North Norfolk and South Shropshire, although still attracting a sizeable 40% of new residents from urban areas, were more popular with people moving from other villages. This is shown in table 5.1.

5.2.2

36

Research into Rural Housing Affordability Final Report

Table 5.1:

Where have you moved from? Mid Sussex 31% 69% North Norfolk 57% 43% Selby 52% 48% South Shropshire 60% 40%

another village or a town/city

5.2.3

When asked how far they had moved it was notable that those moving to Mid Sussex and Selby, despite being more likely to be moving from urban areas, came from relatively close by. While North Norfolk and South Shropshire attracted around a third of new residents from more than 50 miles away. This suggests that in the former cases people still want to be within a reasonable distance of other commitments such as work while the latter districts were attracting retirees or those seeking deeper rural areas. Again this matches the typologies of all four districts. Table 5.2: How far have you moved? Mid Sussex 61% 25% 14% North Norfolk 51% 15% 34% Selby 51% 30% 19% South Shropshire 33% 38% 29%

Less than 10 miles 10 to 50 miles More than 50 miles

5.3
5.3.1

Functional characteristics
The parish consultations provide a useful starting point for understanding how area characteristics and attributes might affect home-buyer preferences and residential choices, regenerating local market pressure-points or reducing the popularity of some areas. Several attributes, making areas more or less attractive, were discussed at length during the parish consultations before being picked up again in village surveys. The first was accessibility, which may affect the attractiveness of areas in different ways. In general rural residents wish to live in areas in which shops and essential services are accessible. But views as to how ‘accessibility’ should be achieved vary. In areas of greater resident mobility, where many people are commuters (e.g. in the Mid Sussex and Selby case study villages), the consensus appeared to be that proximity to ‘good main roads’ made an area attractive. In Slaugham and Cuckfield, this meant that demand for housing in the parish was consistently high, with the area attracting households working in Gatwick, nearby Haywards Heath, Crawley and London. But for these ‘dynamic commuters’, accessibility was also seen as a ‘double-edged sword’. It presented a risk, making the area more attractive to the local authority and to developers. Seeing that more people want to live in the area and that it is well connected, could induce the authority to allocate more land for housing; and this was the last thing that the participants wanted to see happen. This was reiterated by the interviews with estate agents across Mid Sussex. Agents revealed that the most sought after villages were those in close proximity to employment centres such as Haywards Heath. However, of these highly accessible villages, it was those that had maintained a status, appearance and size of a traditional village that were most desirable. Agents in Selby further emphasised the appeal to buyers of a location that had an optimal level of service and accessibility, revealing villages that straddled the hubs of either Leeds or York as highly desirable. But whilst good roads make an area and its services, accessible to highly mobile

5.3.2

5.3.3

5.3.4

37

Research into Rural Housing Affordability Final Report

residents, a lack of buses can reduce accessibility for non-car users. This was recognised in Slaugham, but highlighted as a particular challenge in South Shropshire. In Bishop’s Castle and Cawood, for example, residents aired concerns over the state of local bus services and impending cuts. This was felt to be a real challenge for the young and for elderly households. That said, it was not felt the public transport provision was a particular draw for those moving into the countryside (and, by inference, not a driver of house prices). General accessibility, on the other hand, can be attractive to people moving into the countryside from urban areas where they have been accustomed to good service access. This makes places like Hope Bowdler attractive to home purchasers, as it affords easy access to the larger Church Stretton settlement. In Riccall and Cawood, being ‘within commutable distances‘ that is ‘within an hour’ of Leeds, York and other major towns was seen as important to those relocating to the area. 5.3.5 There was, however, a belief in Riccall that it was possible to ‘survive’ without a car: it was suggested that ‘Tesco home deliveries’ benefit mothers and the elderly though the respondents all conceded that their own lifestyles were car-dependent: bus services were thought to be good (mainly because of the proximity of the A49) and there was a view that others probably used them and reaped their benefits. Very similar views were aired at Happisburgh and in Cawood. Poor public transport services cause difficulties for a section of the local community, but it is not public transport that lures people to these areas. Cars are seen to make areas accessible: commuting to work over relatively long distances is viewed as part of rural life and having good roads enhances rural liveability. So good roads stand out as a potential price driver, as does accessibility to a good range of nearby services: homebuyers want the convenience of a nearby trunk road plus the handiness of some ‘nice’ shops. But at the same time, those moving into rural areas (or existing residents for that matter), do not want to be exposed to traffic and pollution. This undermines the ‘calm of villages’, preventing ‘peaceful walks’ and generally undermining liveability. For the residents of Slaugham, the perfect location is perhaps a stone’s throw from an A road, but shielded from noise and the risk of further development. It will also have some decent shops and a pub. The surveys highlighted considerable differences between the villages in terms of the importance of transport accessibility in the choice of village of those who had recently moved in, shown in Table 5.3 which presents the proportion stating that particular factor was ‘fairly’ or ‘very’ important in their choice of village. Mid Sussex, which attracted outcommuters, the ability to commute conveniently to work scored highly especially in relation to the other districts. This was especially true in relation to rail access highlighting the importance of being able to access the central London job market. Table 5.3: Importance of transport accessibility in choice of village Mid Sussex 88% 86% 60% 49% Selby 67% 84% 20% 57% North Norfolk 53% 67% 8% 43% South Shropshire 68% 69% 34% 53%

5.3.6

5.3.7

Convenient commute to work Good road links to neighbouring towns Proximity to a railway station Local bus service provision

5.3.8

Shops and other services also stand out on their own as a potential draw for home purchasers. Current residents in the consultations consistently cited ‘range of services’ as an attractive aspect of their community, though these services did not necessarily need to be within their village. In Slaugham, it was the case that services were spread across four

38

Research into Rural Housing Affordability Final Report

villages (though concentrated in Handcross) though all were felt to be accessible by community bus if a private car were not available. Because many new residents in rural locations are highly mobile, it is difficult to tie service locations to residential preferences. This appeared to be true in all the case study parishes. Bishops Castle was said to have a critical mass of services in the immediate area. Hope Bowdler relied on nearby Church Stretton for its shops: ‘new incoming families’ would doubtless have to do a lot of ‘running around’ (by car) though this was not thought to be a disincentive for living in the village. 5.3.9 After all, wealthy incomers, with their particular habits and lifestyles, had transformed the villages. Within their lifestyles, there was no need for village services: therefore such services could hardly be held up as a draw to similar households. But that is not to say that decent services do not play a part in retaining population. In tiny hamlets, the lack of services is unlikely to be a disincentive to ex-urban households looking to live in a ‘rural backwater’. This was further highlighted in the estate agent interviews, particularly in the most rural of districts such as South Shropshire. It was suggested that certain newcomers were particularly drawn to distinctly small villages and hamlets that often necessitate a degree of isolation. Interestingly newcomers of this sort were also identified by agents in Mid Sussex. It appeared that new buyers who were less concerned with accessibility to nearby employment centres were instead motivated by remote, picturesque villages that often did not provide even the most basic local amenities. However, across all four districts, these remote villages were still required to have some form of adequate access to nearby shops and other services by car. Agents noted that as the isolation of a location increased, this would begin to outweigh the desirability of properties in a remote rural setting. It was in fact properties in these extremely remote areas, with no amenities or activity within the village or nearby, that were consistently lower in price relative to properties in other villages. However, in larger villages and small market towns, services play a key part in the social life of communities. In Riccall, for instance, there is a scale and scope of active groups and sports clubs for youngsters and older residents. This and other social facilities including the Regen Centre makes Riccall an attractive location for families. But echoing sentiments elsewhere, there is an admission that Riccall does not have everything it needs: and why should it? There is an acceptance that smaller settlements need to look to bigger ones to fill their service gaps. Villages are never ‘fully functional’ and some, like Cawood, have become less functional in recent years as a result of the arrival of more commuters with a different needs profile. To some extent, this has meant that Cawood has taken on a dormitory feel, though it retains an ‘adequate base of services and facilities’: exactly what new rural residents seek. Table 5.4 shows the importance of facilities in the choice of village for those who had recently moved in, revealed by the household surveys. Respondents in the more rural areas did not regard the availability of good facilities than in the more “commuter” areas. Table 5.4: How important is the availability of facilities in your choice of village? Respondents answering ‘fairly’ or ‘very’ important 77% 67% 60% 59%

5.3.10

5.3.11

5.3.12

Mid Sussex Selby North Norfolk South Shropshire

39

Research into Rural Housing Affordability Final Report

5.3.13

Do villages need schools to make them attractive? Are incomers attracted to villages with schools and will they pay a premium to be near them? This perception of schools affecting property prices is perhaps inspired by well-publicised experiences in urban areas and the controversies that surround school catchments and parental choice. In some rural areas, primary schools are over-subscribed: in others, they have experienced a long term trend of falling rolls (a cited problem in Cawood). The view in Slaugham and Cuckfield was that primary schools (there are two in the parish) are critically important, drawing people to the area and helping retain families. But these schools have large catchments and parents seem more than happy to drive their children to school in the morning, sometimes from quite distant villages (without schools of their own). Reinforcing this notion, the interviews with agents revealed the importance and availability of quality primary schools was an issue that was only raised in Mid Sussex. However, this issue was not entirely village specific; the lifestyles of potential buyers in the district resulted in a willingness to travel to nearby villages for reputable schools when presented with the trade-off between good schools and other attributes that make a location desirable. A key issue in Slaugham parish is the relative shortage of 4-5 bedroom homes for families. Asked whether more family housing might increase the viability of the parish’s schools, the response was that the schools have catchments and that families do not necessarily choose to live nearby. They live elsewhere for other reasons: other factors make villages attractive. But again, this seems to underscore the difficulty in identifying locational price drivers. The types of households buying rural property tend to be highly mobile and it is in the nature of rural living that new residents, in particular, do not expect everything they need to be close by. Some established, less mobile, residents might well prefer key services to be within walking distance, but are the choices made by these residents driving property prices? Figure 5.1: Schools as house price drivers, Cuckfield

5.3.14

5.3.15

5.3.16

The role of schools as a factor in shaping the community came up frequently in villages which hosted schools. This was particularly the case in Cuckfield, a large village hosting both a primary and secondary school. The schools served families living beyond the district including those living in the nearby town of Haywards Heath. However, their role as an attraction to incoming families from a large part of the south east was consistently cited during the consultation. Many incomers were seen as wealthy professionals commuting to jobs in London and content with the character and ambience of the village, which combined with good transport links made it complete as a place to settle. Importantly, the school was crucial in acting as a gateway to existing residents, many of which were previous incomers of a similar status. This encouraged involvement in activities and informal mixing within the village and often the formation of social networks amongst incomers. 5.3.17 Equally, there was much discussion in Riccall regarding the locations of secondary schools and colleges in the area. Whilst there was no suggestion that such higher-tier services should be closer to hand, it was noted that distance to such education provision might influence the residential choices of families with older children or of independent younger people. But the absence of such provision in smaller rural locations is not a factor determining patterns of property demand, as it might be in an urban area, but a potential reason for choosing an urban over a rural location. It will not have an ‘intra-rural’ impact on prices, causing one location to be favoured over another, though access to main roads may become a factor for households requiring this provision.

40

Research into Rural Housing Affordability Final Report

5.3.18

The importance of the quality of local schools in the choice of village for those who had recently moved in to a village was revealed by the household surveys and again as can be seen from table 5.5 reflects the districts’ typologies. In Mid Sussex which is attracting families from urban areas the quality of schools was a key factor in choice of village. Selby which was attracting younger professionals commuting to Leeds and York it was less of an issue and the same is true in North Norfolk where a higher proportion of older people and retirees were moving in. Table 5.5: How important is the quality of local schools in your choice of village? Respondents Answering ‘Fairly’ or ‘Very’ Important 82% 52% 67% 75%

Mid Sussex Selby North Norfolk South Shropshire

5.3.19

Environmental characteristics, together with the character of villages, certainly add to the attractiveness of a location. Particularly in the interviews with estate agents, there was continual reference made to locations that could only be described as ‘your typical traditional village’. It appeared that many agents could not attribute this status to exact characteristics but simply to being ‘quaint’ or ‘beautiful in character’. When prompted to discuss these issues further agents often referred to the stock of period properties and the natural landscape. But when consulting with residents in some areas, the ‘environment’ was not held up immediately as a factor explaining this attractiveness: rather, it was only when discussions moved to possible future development that respondents highlighted the need to preserve local character and to keep ‘villages as villages’ and avoid the ‘urbanising effect of more housing’. In Slaugham, for instance, the four villages in the parish are seen as distinct ‘oases’ (from the urban character of the wider area beyond the parish). The green space between the villages was viewed as essential, helping preserve their identity and character. On each occasion that the discussion moved towards ‘housing need’ or development, the importance of limiting development for ‘environmental reasons’ was re-emphasised. In Cuckfield this was expressed in terms of the imbalance, both physically and socially that it was feared would come from large scale development. The village had grown gradually and organically over time which had ensured that it retained a ‘central heart’ and a distinct identity from its surroundings leading to a desire ‘to keep things as they are’. It was those villages that had not grown organically that were revealed to be much less desirable to buyers in the estate agent interviews. This was a strong theme consistent across all four districts, although to a lesser extent in South Shropshire which has experienced much less development relative to the other districts. In Selby agents stated that villages which had experienced significant development 1960s and 1970s had detracted away from their original character and had placed strong downward pressure on the price of properties across these villages. Agents in Mid Sussex indicated this was also the case. An example of the highly desirable ‘twin villages’ of HurstpierPoint and Hassocks was provided, in which the villages are similar in location and accessibility to amenities and transport, however, HurstpierPoint was consistently regarded as both more popular and more expensive. Agents suggested that the higher stock of modern housing in Hassocks meant it did not

5.3.20

5.3.21

5.3.22

41

Research into Rural Housing Affordability Final Report

enjoy the same traditional village status as HurstpierPoint. 5.3.23 The issues surrounding the nature of the housing stock along the coast of North Norfolk, however, were quite different. The sought after villages stretching west of Cromer and along to Sheringham, Blakeney and Wells-Next-the-Sea appear to have welcomed a considerable amount of modern, high quality housing development. Largely to cater for second home buyers, agents suggested this housing has complemented the west coast area. However, there was some debate over the perceived impact of this development on the local landscape and character of coastal villages among long-standing residents and those that had settled more recently. Elsewhere, the ‘beauty’ of the surroundings was highlighted immediately as a factor drawing people to the area and keeping them there. In Bishop’s Castle, the ‘unspoilt and very beautiful surroundings’ had attracted a great many second home buyers to the village, with a consequent loss of family housing and smaller ‘starter homes’. It was claimed that in some streets, virtually all properties are now second homes. It was noted that there is sometimes a social downside to having a high quality environment: those from outside the area will want to buy into that environment, investing in property, firstly for second homes and then for permanent retirement. The contribution of second home owners to the local economy and to village life was questioned: their limited residence means that they ‘cannot put their heart and soul here because they have somewhere else to live’. Their ‘financial and social capital’ is tied up elsewhere. Their effect on the housing market, however, is significant, not only because prices rise (especially in village centres and for smaller ex-agricultural workers’ cottages), but because turnover in the local market grinds to a halt. These properties rarely return to that local market. In North Norfolk and South Shropshire this is an increasing concern. Figure 5.2: ‘Access to solitude’ - Blakeney

5.3.24

5.3.25

In Blakeney, incomers were identified as a mix of second home buyers as well as late career professionals and younger suburbanites. This had occurred over a number of decades from people seeking ‘access to tranquillity and solitude’ as the primary reason and aided by the relatively easy reach from major conurbations. However, in the case of these latter groups, it was observed that, in many cases, their family and friends would also look to move to the area, with the environmental factors being less important as an attraction factor. Demand for housing therefore arose from connected waves of incomers seeking to reestablish original familial and friend networks, while the environmental attributes prevailed as an ongoing driver of demand from new incomers. 5.3.26 Some areas, however, are too close to conurbations to attract second home seekers. This was clearly the case in Slaugham. Similarly, at Hope Bowdler, the fact that the village is a ‘very nice place to live’ does not mean that it attracts a great number of holiday or second home buyers: it is too close to Church Stretton; with seasonal residents often looking for locations that are a little more off the beaten track. Happisburgh perhaps fits this particular bill. The area is generally attractive to tourists and has a significant proportion of second homes in its housing stock. This is a ‘picture postcard village’, ‘surrounded by sea and green fields’. It also has a nice beach and listed buildings, which add to the character of the village. Given the profile of consultation participants, including those with local businesses, it was not surprising that there was broad acknowledgment of the positive contribution that holiday homes (over second homes) make to the local economy, caveated with a concern over empty properties (and second homes that are less regularly used) and their economic and social impacts. Similarly estate agents in North Norfolk were equally positive about the contribution of tourism and the associated development to cater for holiday homes. However, it was clear in Happisburgh that the surrounding environment and village character play key roles in influencing home-buyer

42

Research into Rural Housing Affordability Final Report

preferences. 5.3.27 The same was true in Riccall where easy access to this countryside was viewed as a significant factor attracting people to the village: country walks and cycle paths wind around the village and local allotments, adding to the ‘sense of countryside’. Even the presence of large roads, the A19 in this case, do not detract from this sense. The road runs just to the east of the village rather than through it making a key difference to the environmental quality. Indeed, the combination of good access (noted above) and retained environmental quality makes Riccall attractive to home purchasers. The household surveys highlighted that the key issue in attracting people to the villages were their rurality in terms of peace and tranquillity, quality of the surrounding environment, safety and security and proximity to nature. These were the top three/four factors in every district. Table 5.6: Importance of rurality issues in choice of village Mid Sussex Peace and tranquillity Proximity to nature and recreational activities Quality of the surrounding environment Safety and security 94% 80% 97% 98% Selby North Norfol k 93% 81% 97% 96% South Shropshir e 90% 88% 96% 94%

5.3.28

93% 90% 93% 85%

5.3.29

Finally, in Cawood (and also in Slaugham) it was noted that the positive qualities of the village are matched by the qualities of its inhabitants: it’s a ‘nice place to live in with nice people’. This sentiment was repeated across the case studies and there appeared to be general satisfaction with the ‘community spirit’ that was said to be found within these rural communities. This spirit was reflected in two main ways: the willingness for people to get involved in local projects and to have a ‘visible presence’ in the village; and the lack of ‘anti-social behaviour’ highlighted by respondents in all of the consultations. An important point is that it is difficult to link the ‘spirit’ highlighted by consultees, including the voluntary actions supporting local bus services and active church scene, with a willingness to pay a premium to buy into such villages. Interestingly interviews with estate agents in South Shropshire highlighted this issue. Agents insisted that whilst a large number of villages have lacked, or have recently lost, local amenities (typical of many ‘deeply rural’ areas); a direct substitute for buyers is often the presence of a local village hall or church. Such facilities often play a pivotal role in village life; replacing the activity that would normally be generated by local shops and services, enabling residents to come together and reinforce a sense of ‘community spirit’. It is a persistent question in the study of rural communities: are home-buyer decisions affected by the belief that a particular village’s outward characteristics suggest a ‘community’ to buy into, or do households moving from urban areas perceive that rural communities, in general, are more ‘close-knit’ and that this will increase liveability, making them better places to raise children and so on. Table 5.7 shows the results of the surveys in terms of the importance of community spirit in the choice of village of those who had recently moved in, showing respondents who answered ‘fairly’ or ‘very’ important.

5.3.30

5.3.31

43

Research into Rural Housing Affordability Final Report

Table 5.7:

Importance of community spirit in choice of village Mid Sussex 83% Selby 79% North Norfolk 76% South Shropshire 72%

Sense of neighbourliness and atmosphere in the village

5.4
5.4.1

Social characteristics and identity
Much has been written on the ‘softer’ attributes of ‘rurality’ and how these draw people to particular locations. The belief that greater ‘neighbourliness’ makes areas safer, that there will be opportunities to be part of a ‘real community’, or a view that children will benefit from village life and a cleaner environment are all perceptions of the benefits of rural living. The village consultations were unable to test such perceptions: rather, they dealt with the experiences of a small number of current residents. These consultations, however, were able to confirm a belief that the ‘spirit’ attributed to small communities is alive and well: in Slaugham, in Bishop’s Castle and in Happisburgh and Riccall, responses suggested thriving community scenes. This picture, however, was not universal. In Hope Bowdler, it was felt that ‘people do not mix easily’ and that disinterest in local matters results in a diminished community spirit. And this spirit is not always a welcoming one. It should also be noted that the parishes included in this survey were not randomly chosen but were picked on the basis of an active Parish Council that would act as gate keeper to the consultation process. One might expect, therefore, a high degree of ‘community spirit’ which might not be replicated in other locations.

5.4.2

5.5
5.5.1

Differences between what attract new residents and keeps existing residents
Previous sections have set out what makes the surveyed villages attractive to those people who recently moved into the village and those who have been living there some time. Table 5.8 shows the differences, if any between, these two groups and their views on the attractiveness of their village. A high positive number shows the attribute was regarded as more important (by the shown number of percentage points) in the choice of village to those who had recently moved in than it is to those living there, while a negative number shows the opposite. So being close to family and friends becomes far more important to people who have lived in the village for some time than those who have recently moved in, while the importance of being able to commute to work and the quality of schools was more important in attracting people than retaining them. Most telling it is the rurality issues that are equally important in attracting people to the village in the first place and keeping them there over the long term.

5.5.2

5.5.3

44

Research into Rural Housing Affordability Final Report

Table 5.8:

Difference between what attracts and retains residents (positive number greater importance in retaining rather than attracting, negative number greater importance in attracting than retaining) Mid Sussex 17 14 12 4 2 2 1 1 0 Selby 21 7 12 3 4 2 1 0 7 North Norfolk 20 13 14 3 3 3 17 3 4 South Shropshire 22 3 20 12 -2 7 11 4 6

being close to family and friends local bus service provision sense of neighbourliness and atmosphere in the village good road links to neighbouring towns proximity to nature and recreational activities peace and tranquillity facilities available in the village quality of the surrounding environment the physical attributes of the village (nice high street, buildings etc) safety and security proximity to a rail station convenient commute to work quality of local schools

-1 -4 -33 -35

0 5 -12 -35

11 -2 -14 -16

-1 1 -21 -36

5.6

Quantifying house price influences
Introduction

5.6.2

In order to assess whether much of the findings from the preceding qualitative research conformed to reality, we performed an analysis of house prices for each district using the hedonic pricing method (HPM). Essentially this stage of analysis looked to more conclusively reveal to what extent a rural location drives residential property prices and ultimately evaluate the quantifiable ‘power of rurality’. The HPM is a statistical technique used to compare prices of a product when there is observable quality differences in a product sold: in the case of this research a rural property or dwelling. A dwelling can be described generally by the characteristics of its structure, surroundings and location and a dwelling’s price can therefore be said to reflect the value of these attributes to the buyer. Essentially, hedonic pricing allows us to describe house prices as a function of a number of different attributes. Specifying a hedonic price function is possible by undertaking a series of multiple regression analyses. Regression analysis looks to identify how much of the variation in a dependent variable, which in our case is property price, can be explained by variation in a number of explanatory variables, which in our analysis is a property’s different attributes.

5.6.3

5.6.4

5.6.5

45

Research into Rural Housing Affordability Final Report

This hedonic price function in statistical terms is represented below: Y = a + b1X1 + b2X2 + b3X3…biXi 5.6.6 Where the price (Y) of a property is a function of its attributes X1 interpreted more simply as:

to Xi. This can be

House Price = constant + structural attributes + location attributes… 5.6.7 In this research we have incorporated the rural typology of a property into its hedonic price function, which allows us to isolate its impact on house prices and essentially through this impact determine the power of rurality. Whilst many hedonic pricing studies have looked to value the amenity of farm or agricultural land in rural housing markets, to our knowledge there has not been an attempt to estimate the value of a rural location in such a holistic sense. Using Defra’s classifications of rural and urban areas3 we have piloted the HPM in order to test our central hypothesis; houses in a rural location will attract a premium over houses that are located in urban areas. However, findings from the previous stages of this research also lead us to hypothesise that extremely isolated rural properties will not attract such a premium.

5.6.8

Method
5.6.9 The HPM is an approximation of the housing market observed in reality and therefore, as with any statistical model, is underpinned by a number of key assumptions. However, above all, homogeneity of the housing product is assumed. It was therefore essential when collecting price data and developing data for quality attributes that only one market was being assessed. For an individual market, it is assumed that house prices follow the same price schedule and thus variation in price can be estimated through a unique hedonic function for that market. The assumption of a single market is particularly important for this research given that our analysis is focused on four districts each representing different rural typologies. As a result, based on the findings from previous research stages, we assumed that house prices across the four districts were unlikely to move in the same way. For example, access to the rail network in Mid Sussex which is of a ‘dynamic commuter’ typology is likely to have upward pressure on the price of a house, yet this is unlikely to be the case in ‘deep rural’ South Shropshire. In our assumption of four housing markets each with its own price schedule, it was therefore necessary to estimate separate hedonic price functions for each case study district. Nevertheless, whilst we assumed that house prices in the four districts would be driven differently by a set of quality attributes, this did not compromise our central hypothesis that in any district, a rural location will attract a premium over properties that are located in urban areas. The scarcity of available property in rural areas relative to urban areas also helps to explain divergences in average house prices between urban and rural areas. While this is no doubt a prevailing factor, the analysis did not attempt to estimate the explanatory power of supply constraints, assuming that this was an external influence having equal weight across the four areas. One of our first concerns in estimating a hedonic price function is the measurement of property price, known as the dependent variable, given that a property’s value is
3

5.6.10

5.6.11

5.6.12

5.6.13

Rural Evidence Research Centre (RERC), Birkbeck College, 2004

46

Research into Rural Housing Affordability Final Report

dependent on the quantities or qualities of its characteristics. 5.6.14 In order to measure house prices across each district, sales data was sourced from the VOA for one year4. The VOA records the structural characteristics of properties sold including the number of bedrooms and bathrooms in each property and whether the property is detached, semi-detached, or terraced, for sales corresponding to postcode level (minus one digit). Collecting sales data at this spatial scale allows us to analyse house prices at a localised level whilst maintaining a large enough dataset of property transactions required to perform regression analysis. It was accepted that the size and type of property will be a large determinant of its price. In order to control for this as much as possible, house price data was obtained as an average price per room, by property type, for all sales recorded in 2007. This unit price has been used to represent the dependent variable throughout our application of the HPM for each district. It was therefore necessary to test that there was enough variation in this average price data. The HPM consists of a process of regression analysis, looking at how variation in house price is dependent on the variation in its characteristics. Therefore for the regression analysis to assess this relationship, it was necessary to ensure that there was significant variation in the sales transactions recorded (represented as unit price of an average price per room). In addition it was also necessary to ensure that the 2007 sales recorded by the VOA extended across each district and was therefore a good representation of each rural housing market. In determining this, the average price per room of a property for each district was mapped. This mapping was performed separately for each district and differentiated between property types. In Figure 5.3 to Figure 5.6 we present the mapping results of detached property sales for each district, based on our unit measure of the average price per bedroom. These figures show that typically prices are higher in rural compared to urban areas. However, there are other factors impacting on prices within the rural area. So in North Norfolk, for example, prices are higher in the north than the south of the district and highest along the coast compared into inland. While in Selby properties in villages nearer to York and Leeds achieve higher prices than those in the east and south of the district. Finally, in South Shropshire process in more remote rural areas are lower than in those villages which are closer to the larger market towns.

5.6.15

5.6.16

5.6.17

4

The entire number of sales recorded by the VOA in 2007 was taken for each district

47

Research into Rural Housing Affordability Final Report

Figure 5.3:

Mid Sussex detached property price per room (2007)

48

Research into Rural Housing Affordability Final Report

Figure 5.4:

North Norfolk detached property price per room (2007)

49

Research into Rural Housing Affordability Final Report

Figure 5.5:

Selby detached property price per room (2007)

50

Research into Rural Housing Affordability Final Report

Figure 5.6:

South Shropshire detached property price per room (2007)

51

Research into Rural Housing Affordability Final Report

5.6.18

Having defined a measure of property price, our next concern in estimating a hedonic price function was the measurement of the many attributes, known as the explanatory variables, which are used to explain the selling price of a house. There are of course a vast number of explanatory variables that could be incorporated into a hedonic price function as the value of a property to a buyer is likely to be dependent on so many attributes. However, it is possible to categorise the attributes of a dwelling as follows: Structural (number of rooms, plot size, garage, garden, age etc); Neighbourhood (e.g. crime rates, employment rate, private home ownership etc); Accessibility (e.g. distance to schools, shops, transport etc); and Surrounding environment (e.g. quality of open space, natural and built environment, noise level etc). A full list of potential explanatory variables was developed based on an a priori expectation, informed by other hedonic studies and our previous qualitative research, that each variable represents an attribute of a property that will in part determine its sale price. However, central to our application of the HPM was the need to determine the extent to which a rural location adds value to residential property. We therefore looked to identify data that could accurately capture and quantify the rural locality of a property, allowing us to isolate the impact different rural typologies have on price. The RSS (rural services data series) was identified as an ideal source for measurement of locational variables. The RSS draws on Defra’s definition of rural and urban locations, breaking them down into four settlement types; urban, rural town and fringe, rural village and rural dwelling and assigning them to either a 'sparse' or 'less sparse' regional setting reaching eight final typologies5. These settlement types are defined through bands of population density. It was felt that this definition in particular, would provide further insight into the desirability of villages that are more developed and whether an optimal size that afforded ‘traditional’ village status did in fact exist. Figure 5.7 sets out how these rural and urban typologies are defined and aggregated: Figure 5.7: Rural/Urban aggregated definitions

5.6.19

5.6.20

5.6.21

5.6.22

Source: National Statistics
5

The Rural/Urban Definition was introduced in 2004 and defines the rurality of very small census based geographies, based on population densities

52

Research into Rural Housing Affordability Final Report

5.6.23

Having collected all necessary data, each property transaction was assigned its corresponding structural, neighbourhood, accessibility and environmental attributes. Through a series of correlation and multiple regression analyses (described in detail in the technical report presented in Appendix E), an optimal hedonic price function for each district was determined. The results of these analyses are presented in the next section.

Results
5.6.24 The relationship between property price and its corresponding structural, neighbourhood, accessibility and environmental attributes was tested through correlation analysis. Correlation analysis identifies the direction and strength of co-movement between two variables (which in our case is the price of property and an individual attribute) and therefore informed us as to whether or not it was relevant to include different attributes in our hedonic function. In addition, the relationships between property attributes were tested, since some attributes are likely to be highly correlated as they share a very similar relationship with price. From this analysis, we were able to reduce the full list of property attributes that was initially developed to a set of attributes (or explanatory variables) that could be considered for inclusion in each district’s hedonic price function. This reduced set of attributes went beyond an a priori expectation that each would in part determine a property’s sale price, to being selected on the basis of a proven statistical relationship. The drivers of property price identified for each district are presented in Table 5.9 below. Table 5.9:
Mid Sussex T_DETACH YNG_FAML TWO_CAR HEALTH_SCORE EDU_SCORE

5.6.25

Results of correlation analysis: drivers of property price
North Norfolk T_DETACH YNG_FAML UNEMPL OWN_HME South Shropshire T_DETACH YNG_FAML TWO_CAR HEALTH_SCORE EDU_SCORE CRIME_SCORE Selby T_DETACH YNG_FAML UNEMPL TWO_CAR HEALTH_SCORE EDU_SCORE CRIME_SCORE A_ROADS RAILWAY TOWN DIST_POST DIST_SCHOOL DIST_BANK

RAILWAY TOWN DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK

TOWN DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK

A_ROADS RAILWAY TOWN DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK AONB RTF_S RV_S RD_S RTF_LS RV_LS RD_LS

AONB U_LS RTF_LS RV_LS RD_LS RURAL

AONB RTF_S RV_S RD_S U_LS RTF_LS RV_LS

U_LS RTF_LS RV_LS RD_LS RURAL

53

Research into Rural Housing Affordability Final Report

RD_LS SPARSE RURAL

SPARSE

5.6.26

These variables were then modelled through a series of multiple regression analyses, finally determining an optimal price function for each district housing market6. This function denotes the price schedule for each rural market, in other words, from each function we can determine the extent to which different property attributes, including rural typology, impact on price. The hedonic price function(s) for Mid Sussex was as follows: Model IV (i) YMid_Sus_Price = aconstant + bT_DETACH + bYNG_FAML

5.6.27

+ bTWO_CAR + bDIST_SCHOOL + bDIST_BANK +

bDIST_POST + bDIST_GPall + bRV_LS + bRD_LS
Model IV (ii) YMid_Sus_Price = aconstant + bT_DETACH + bYNG_FAML

+ bTWO_CAR + bDIST_SCHOOL + bDIST_BANK

+ bDIST_POST + bDIST_GPall + bRURAL
Table 5.10: Hedonic model results for Mid Sussex
Model IV(I) 95,872 23,927 -1 0 5,961 2,301 Model IV(Ii) 98,485 23,854 -1 0 3,423 2,490 4,054 -1,799

(Constant) T_DETACH YNG_FAML TWO_CAR DIST_SCHOOL DIST_BANK DIST_POST DIST_GPall RV_LS RD_LS RURAL Observations Adjusted R Square

5,929 10,368 7,847 608 0.479 608 0.482

5.6.28

The hedonic price function(s) for North Norfolk was as follows: Model IV (i)

YNorth_Nflk_Price = aconstant + bT_DETACH + bYNG_FAML + bOWN_HOME + bUNEMPL + bDIST_SCHOOL
6

With the exception of South Shropshire, two models have been determined for each district in order to test the different classifications of a rural area. That is, to determine the impact of a rural location of property price, Defra’s definition of rural was tested in its simplest form and by detailed settlement typology. For some districts not all rural typologies were able to be tested within the model. This was because either no such areas existed or there was not a significant number of a settlement type to produce variation that could be tested statistically. South Shropshire was the only district in which all property transactions fell under a ‘rural’ typology.

54

Research into Rural Housing Affordability Final Report

+ bDIST_GPall + bAONB + bRTF_S + bRV_S + bRD_S
Model IV (ii)

YNorth_Nflk_Price = aconstant + bT_DETACH + bYNG_FAML + bOWN_HOME + bUNEMPL + bDIST_SCHOOL + bDIST_GPall + bAONB + bSPARSE
Table 5.11: Hedonic model results for North Norfolk
Model IV(I) 83,888 19,800 -1 0 -1 1,252 956 12,303 5,724 9,199 6,654 Model IV(Ii) 81,741 19,806 -1 0 -1 1,429 1,086 12,310

(Constant) T_DETACH YNG_FAML OWN_HME UNEMPL DIST_SCHOOL DIST_GPall AONB RTF_S RV_S RD_S SPARSE Observations Adjusted R Square

7,151 665 0.618 665 0.616

5.6.29

The hedonic price function for South Shropshire was as follows: Model IV (i) YSouth_Shrop_Price = aconstant + bT_DETACH +

bYNG_FAML + bDIST_SCHOOL + bA_ROADS + bRV_S +

bRD_S
Table 5.12: Hedonic model results for South Shropshire
Model IV 65,110 28,915 -1 4,291 642 4,636 -3,643 323 0.668

(Constant) T_DETACH YNG_FAML DIST_SCHOOL A_ROADS RV_S RD_S Observations Adjusted R Square

5.6.30

The hedonic price function(s) for Selby was as follows: Model IV (i)

55

Research into Rural Housing Affordability Final Report

YSelby_Price = aconstant + bT_DETACH +

bYNG_FAML + bTWO_CAR + bDIST_SCHOOL + bDIST_BANK + bRAILWAY + bDIST_GPall + bU_LS + bRV_LS

Model IV (ii) YSelby_Price = aconstant + bT_DETACH + bYNG_FAML

+ bTWO_CAR + bDIST_SCHOOL + bDIST_BANK +

bRAILWAY + bDIST_GPall + bRURAL
Table 5.13: Hedonic model results for Selby
Model IV(I) 48,748 15,688 -1 0 2,119 -1,165 1,594 4,785 2,908 Model IV(Ii) 53,751 15,618 -1 0 2,610 -1,153 1,714

(Constant) T_DETACH YNG_FAML TWO_CAR DIST_SCHOOL DIST_BANK RAILWAY U_LS RV_LS RURAL Observations Adjusted R Square

-4,609 353 0.485 353 0.480

5.6.31

Generally across all four districts the type of property was found to be the most significant and largest driver of property price. As one would expect, our results show that all else being equal, detached properties will command a higher price relative to other property types (i.e. semi-detached and terraced). On average across all four districts, a detached property increased the average price per bedroom by around 28%. In addition, a property that is detached was found to have a greater impact on price in the more rural districts, namely South Shropshire followed by North Norfolk. In terms of neighbourhood attributes, our results were fairly consistent across all four housing markets and generally conform to a prior expectation that a lower socioeconomic standing in an area will reduce the relative price of property. One variable that was tested within the hedonic model was the proportion of households owning two cars or more, which had a varying impact on price across the districts. One would generally expect a higher percentage of households owning two or more cars to imply a higher socio-economic standing in that area. The results for Mid Sussex implies a negative relationship between car ownership and property price, although the relative net impact on price was found to be zero. A possible explanation is that car ownership in rural areas could be classed as a proxy for the level of isolation of a property, with an exceptionally high level of car ownership indicating that a property is particularly remote. This conforms to previous findings, in particular the interviews with estate agents, that suggested extremely isolated villages or hamlets without basic amenities are far less desirable. The impact of a property’s access to local amenities and transport links provided interesting results. Generally across all four districts, the relative price of property rises as

5.6.32

5.6.33

5.6.34

56

Research into Rural Housing Affordability Final Report

the distance to these services increases. 5.6.35 For access to the transport network, a 1km increase in the distance to the local railway station in Selby increased the average price per room by 2%, whilst in South Shropshire a 1km increase in the distance to road network increased the average price per room by 1%. This result suggests that within rural housing markets, there is a value placed on properties that are distanced from services that bring about disbenefits such as congestion, pollution and noise. For access to local amenities however, there are exceptions to this trend. For Mid Sussex, price falls as the distance to health services increases, with a 1km increase in the distance to a GP surgery reducing the average price per room by 2%. For Selby, price falls as distance to banking services increases, with a 1km increase reducing the average price per room by 2%. As both of these districts contain a greater mix of urban and rural settlements, this result could reinforce the notion that increased distance to services will have a more positive impact in rural areas/districts and a negative impact on property prices in urban areas. Whilst the direction of the relationship between access to services and price was fairly consistent across the four housing markets, the extent or size of this impact tended to range according to the district and type of service assessed. It is possible that further testing of the relationship between these variables and price in future research would indicate that there is in fact an optimal distance from services and transport that positively impacts on price. Previous stages of our research have in fact found that both residents and potential buyers favour villages with a certain level of access to local amenities that does not compromise their ‘traditional’ village status. For example, the price of a property may be driven by access to the road network. However, as proximity to a main road increases, the price of a property is likely to increase at a slower rate, as the benefits of increased accessibility are outweighed by the associated noise, pollution and so on. In reverse, as properties are located further from a main road, price is likely to fall as the disbenefit of travelling long distances to access the road network becomes relatively more important. Strong evidence of this ‘inverted U’ relationship has emerged throughout the previous consultations and surveys. Variables that relate to the quality of the environment are generally much more difficult to define and measure than other property attributes. This is an aspect of hedonic pricing that is receiving growing attention in more recent studies. Environment variables need to reflect peoples’ perceptions of the environmental amenity, such as landscaping and views, as well as the disbenefit such as noise and pollution. However, few straightforward methods exist for quantifying these variables. A further difficulty is that environmental attributes of an area are often embodied in other attributes that will have already been measured within a hedonic price function, such as noise associated with proximity to a road. When testing the impact of the quality of the surrounding environment on price, it was only possible to include this variable in the model for North Norfolk. The quality of the surrounding landscape and natural environment was tested according to whether a property was located in an Area of Outstanding Natural Beauty (AONB). A strong and significant positive relationship with price was found with properties located in an AONB commanding a considerably higher relative price, of around 17%, compared with those that were not. Within our model, the location of a property in an AONB was in fact the second largest driver of property price in North Norfolk. This result is significant in both a statistical sense and in attempting to capture some element of qualitative rurality. The previous stages of consultation confirmed that often the most desirable villages were those afforded with picturesque landscapes. It is not

5.6.36

5.6.37

5.6.38

5.6.39

5.6.40

5.6.41

57

Research into Rural Housing Affordability Final Report

surprising therefore, that quantifying house price influences found the quality of the natural environment to be a key driver of price. However, the shortage of methods available to this research for measuring environmental quality have limited our analysis at this stage, prompting recommendations for extending this understanding in future research. 5.6.42 Finally we come to analyse the extent to which a rural location adds value to residential property, hypothesising that, all else being equal, rural properties will achieve a premium over properties in urban areas. For Mid Sussex, all property transactions analysed fell into the ‘less sparse’ regional setting. For less sparse settlements, the typologies for ‘rural village’ and ‘rural dwelling’ were able to be included in the model and tested, with analysis of these results being compared to the alternate typologies of ‘urban’ and ‘rural town and fringe’. The results of the model indicated that all else being equal, properties classed as a rural dwelling command a relatively higher average price per room of around 10% and was in fact the second largest driver of price within the model. Properties located in rural villages increase the average price per room by around 6% relative to those that are not. Testing the rural classification of a property in its simplest form again indicated that a rural location will achieve a premium over urban locations in the district, increasing the average price per room by around 8%. The analysis of Selby’s housing market presented a very different result. Once again due to its typology, all property transactions analysed for the district fell into the ‘less sparse’ regional setting. In Selby it would appear that a rural location will in fact reduce the average unit price of a property by around 7%. Selby was the only district in which our analysis found that properties in an urban location will command a higher price than properties located in rural villages. For South Shropshire, all property transactions analysed for the district were classified as rural under Defra’s classifications, therefore our analysis here compared the value of the extent of ‘rurality’. Interestingly our results indicated that properties located in a village command a higher relative price whereas properties that are particularly isolated (classed as a dwelling and located in a ‘sparse’ regional setting) will have a negative impact on price. Properties located in villages increase the average price per room by around 6% and were found to be the second largest driver of price, rural dwellings however, reduced the unit price by 4%. This reinforces our hypothesis in both senses: properties in a rural location will attract a premium over those that are located in urban areas, yet extremely isolated rural properties will not. Finally in the analysis of North Norfolk’s housing market, the majority of property transactions analysed for the district fell into Defra’s ‘rural’ classification and thus did not produce enough variation to test this classification in its simplest form. However, we were able to test the breakdown of settlement types by ‘sparse’ and ‘less sparse’, finding that properties in a sparse regional setting increase the average price per room by around 10%. The results for Selby also showed that, whilst properties in a rural setting command a higher relative price than properties in urban areas, properties located in villages achieve a premium over those properties in towns or isolated hamlets. Village properties increased the average price per room by around 13%, rural dwellings increase price by 9% and properties in rural town/fringes increase price by 8%. Table 5.14 presents a summary of the extent to which a rural location adds value to residential property in each district. Full tabulated results are presented in the hedonic analysis technical report presented in the appendix.

5.6.43

5.6.44

5.6.45

5.6.46

5.6.47

5.6.48

58

Research into Rural Housing Affordability Final Report

Table 5.14:

Quantifying the power of rurality
Impact on average price per room in the district 6% 10% 8%

Rural typology tested within each district model Mid Sussex RV_LS RD_LS RURAL Selby U_LS RV_LS RURAL South Shropshire RV_S RD_S North Norfolk RTF_S RV_S RD_S SPARSE

7% 5% -7%

6% -4%

8% 13% 9% 10%

Limitations to hedonic pricing of rural house prices
5.6.49 In interpreting our results in the previous section and in drawing conclusions in the final chapter of this report, the limitations of our application of the HPM should be considered. The technical report in Appendix E sets out a detailed discussion of the weaknesses of the HPM, many of which concern data and statistical technique. However, here we address the qualitative factors that may restrict our application of the HPM specifically to rural housing markets. Non-quantifiable drivers of house prices There are a number of qualitative factors that are likely to cause variation in price within rural housing markets that simply cannot be captured within the hedonic pricing model. Though we cannot measure factors such as the feeling of community or quality of life associated with rural living as explanatory variables, or statistically prove a relationship with price, other areas of the research project have covered this in great detail. In addition to the ‘qualitative rurality’ of an area, there lacks established primary research methods or secondary sources that can measure many of the attributes that describe both the natural and built environment. This is an area of research that is receiving growing attention in more recent hedonic pricing studies. Rural housing markets are unique Throughout the other stages of the research project it has become clear that the rural housing market is relatively inactive. Often residents in rural areas have lived there for long periods of time, regardless of the different attributes of the housing stock but instead

5.6.50

5.6.51

5.6.52

59

Research into Rural Housing Affordability Final Report

due to the attraction of rural life. More specifically, the rural housing market has specific supply and demand parameters, namely limited supply and heightened demand driven by external buyers. The number of transactions, therefore, is limited. 5.6.53 This analysis is a complex exercise; whilst we can infer some form of qualitative value of rurality here, we cannot capture this within a hedonic price function. Furthermore, if such market conditions exist, this will have implications for our application of the HPM to rural housing markets in this research. This could compromise the assumption that house prices adjust instantaneously to changes in supply and demand, as house prices could in fact be driven by a shortage in the supply of rural housing stock, rather than being a reflection of people’s willingness to pay for the different attributes of a property. It should be noted that over-arching supply constraints in influencing the distinction between average rural property values and average urban property values were treated as a given and excluded from the analysis in all of the case study areas.

5.6.54

Representativeness and transferability of the analysis Finally our application of the HPM has estimated the impact of a rural location on property prices, with the aim of this research to produce estimates that essentially conclude some kind of ‘power of rurality’ that can be transferred to other rural areas across the country. This can only be achieved if our sample of properties was diverse in terms of the property types; socio-economic areas; accessibility; environmental and rural typologies covered. As previously explained in the report, the diversity of our sales data and representativeness of our sample was, to some extent, determined through mapping exercises. However, it is important to note that in reality the results of this analysis are only representative of the range of data given in the sample and that care should be taken in extrapolating well beyond that range. This is particularly relevant to our application of the HPM to rural housing markets, which are often inactive and characterised by a low turnover of the housing stock. Our sample of sales data is thus somewhat biased in the types of properties and the types of buyers it represents. Sales data is of course focused on the willingness to pay of more mobile buyers, who may have very different preferences for (and thus place very different values on) the attributes of a property compared to people that have lived in a village all their life. In addition, our analysis has been formed on house price data from one year only and therefore cannot determine whether the value of rurality is constant or changes over time.

5.6.55

5.6.56

5.6.57

60

Research into Rural Housing Affordability Final Report

6
6.1
6.1.1

Local priority
Introduction
The third focus of the study, investigated mainly through parish meetings and village surveys, was the potential policy response: the case for interventions that support ‘local’ needs, whether existing interventions have local support or how communities would like to see them adjusted to better reflect local realities. The literature review identified four areas of critical concern surrounding local needs and the case for assigning priority to local households, whether priority to new social or market housing: Persistent debates and uncertainties surrounding the definition of ‘localness’ and of what constitutes need, creating a ‘rocky foundation’ for local policies and interventions; The extent to which ‘local priority’ is a means of winning support for development which does not then subsequently tackle the needs of household who local residents would deem local; The extent to which the local need for new homes should be tied to economic necessity: whether ‘local’ housing should only be supported where there is a clear economic case, i.e. where new development would support established or new economic activity. This might be supporting additional housing in some areas and less development in others. But in contradiction to this, another important debate concerns; The extent to which local priority policies might be viewed as a means of addressing social polarisation, of creating more socially balanced communities and be seen as a stand-alone objective, separate from labour market concerns. This might involve building more retirement homes in some rural areas, for example. Taking a cue from existing debate concerning housing rights, the suggestion is that local priority is a means of ‘engineering’ more balanced communities with a view to retaining or restoring social networks and cohesion by creating opportunities for rural families to remain living in close proximity. Because this study is primarily concerned with the understanding and attitudes of local populations, our analysis here draws entirely on interview and survey material, supplemented and framed by the original review.

6.1.2

6.2
6.2.1

Localness from a local perspective
Normative definitions of what it means to be ‘local’ rely on characteristics that are easily measurable: length of residence, or whether someone was born or now works in an area for example. These are used as proxies for judging a resident’s claim on local resources. Views as to whether these are the right kinds of criteria vary between different types of rural area. In some areas, such judgements are seen to work in favour of the settled, long-standing population and so enjoy broad support. Elsewhere, especially where there are concentrations of commuting households whose direct employment connection to an area is weaker and length of residence is shorter, there is a tendency to make more qualitative judgements on an individual’s claim to be ‘local’. This difference is revealed in the contrasting consultation responses in the Deep Rural case study area (South Shropshire) and areas which had experienced greater commuter influences (Mid Sussex and Selby). Views in Bishop’s Castle might be characterised as more ‘traditional’. Here, ‘family-ties’ were seen as critically important: having grandparents who lived in the area bestowed an ancestral right on those claiming to be local. There was also a sense that such ties generated local loyalties. Whilst having a grandparent who had lived within a 6 or 7 mile radius of Bishop’s Castle might be

6.2.2

61

Research into Rural Housing Affordability Final Report

enough for someone to claim to be local, they might not wish to do so: due to local rivalry between villages. There was a strong sense from the discussion that people were obviously local or obviously newcomers. There was broad approval of the criteria employed by South Shropshire District Council: minimum length of residence, attended school in the area, employed locally, born and raised in the area, or with parents in the area that now need caring for. 6.2.3 This perspective was shared in nearby Hope Bowdler where there was consensus around the need to be born in the area or to have lived there for a substantial period to be considered local. But in both Happisburgh (Rural Retreat) and Riccall (Transient Rural), there was greater uncertainty surrounding the notion of localness, though an insistence that residence and birth set local people apart from newcomers and especially from those who are seasonal residents. Such criteria support a traditional view of localness in ‘settled areas’ where the amount of in-migration is more limited and where the newcomers who do enter the community, for example, to buy a second home, are obviously non-local. However, in areas of more dynamic population change, these clear boundaries become blurred. In Slaugham, for instance, the discussion group, comprising mainly older residents who had spent their working lives commuting to London or other centres, concluded that being local means being ‘involved in village life’. In small villages, in particular, it is inevitable that people get involved with local activities: this adds to the ‘sense of place’ and breaks down any potential division between recently arriving and more established residents. In Blakeney, while there were similar sentiments expressed about location and birth as in more transient and dynamic rural areas, it was accepted that a person could ‘become local’, even within a few years. This depends on how they interact with the community, whether they get involved and ‘willing to muck in’ and not be presumptuous. The barriers of time and distance become more irrelevant and are broken down more quickly when incomers get involved. Figure 6.1: ‘Getting involved’, Handcross

6.2.4

Handcross in the parish of Slaugham is home to many settled commuters: residents who have spent some considerable time in the four villages of the parish but who work either in nearby Gatwick or in London. Settled commuters were well-represented in the consultation events, so it is perhaps unsurprising that the preferred measure of localness in the parish is not ‘being born and bred’ in the area, but making a contribution to the community: as one resident put it ‘showing some community spirit’. There are lots of different ways that people can partake in this spirit. The parish comprises four small villages, each performing a different service function. One has the local schools, one the pubs and one a concentration of shops and a doctor’s surgery. A community bus is run by volunteers and takes older residents shopping once a week. Some of these volunteers are newer residents, but have already shown their ‘local credentials’. This, claimed the consultees, is a more meaningful measure of localness than working in the parish of having been born in the area. Few people these days can claim to be born locally and there are few truly local jobs. Getting involved, therefore, is the one way to measure localness that is useful. It could, for example, be used to set determine local priority. The people who really should be living locally are those people who help make this community. 6.2.5 Whilst there are ‘emotional ties’ associated with family history or childhood experience in villages, these are a distinct issue and do not detract from the fact that normative measures of localness are losing relevance in modern society. For example, increased mobility is making nonsense of geographical localness. Good roads mean that Brighton is only 15 minutes from Slaugham, leading one respondent to comment that his daughter, living on the coast, still considered herself local to the village. Similarly, being ‘employed

62

Research into Rural Housing Affordability Final Report

locally’ was not seen as being a measure of localness: there are few jobs in the villages though the area is ‘encircled by employment’. It is also the case there are a great many retired people in the area, does their lack of direct employment link render them less local? In both Slaugham and nearby Cuckfield at least, the community is made by ‘enlightened self-help’ which builds on the networks that make the place worth living in. It is within the community that people are accepted and the community that sees someone as local. 6.2.6 Similar sentiments were unearthed in both Riccall and Happisburgh. In the former, ‘being involved’ is a means of ‘feeling like a local’ and those people who support local services (and particularly schools) will inevitably be seen as more rooted in the village. In contrast, the commuters who use the village as a ‘base’ and do not get involved in the community can hardly consider themselves a part of the community. Though, in fairness, there are some long-standing residents who remain ‘invisible’ for much of the time. As in the majority of the consultations, the process of getting involved was seen as a means of breaking down any possible divisions between groups, reducing the importance of time and employment as measures of localness. Almost identical views were aired in Cuckfield, Slaugham, Blakeney, Riccall, Cawood and Happisburgh: contribution to village life is critical to becoming local. This is not to say that more traditional perspectives on localness are absent from areas of greater resident mobility and population change. In Slaugham, a difference was noted between the ‘old fashioned local’ who has lived in the village for ‘50 or 60 years’ and the ‘different sort of local’ who has arrived far more recently but now lays claim to localness. The view survives, in all sorts of rural community, that it is the people who were born and bred in the area who have the strongest emotional ties and the greatest claim on resources. Of course, the local perspective varies depending on ‘who’s calling it’. This was the clear view in Cawood (Transient Rural). A community’s long-standing residents may place a premium on the traditional perspective as it serves their claim to localness. Incomers, on the other hand, may feel a need to legitimise their presence. Those who sit on parish councils may favour participatory measures. Whilst newer perspectives on localness may purport to brush away a dichotomy, they remain contested and provide evidence of continued conflict in some communities.

6.2.7

6.2.8

6.3
6.3.1

Local need as a flag of convenience
Because of these differences in understanding and accepting who is ‘local’, it was considered difficult for planning authorities to use the phrase ‘local housing’ as a rallying call for more development. It was more common for authorities to be led by normative measures of affordability, to claim that additional housing would widen access and then to release that housing to people who were often considered (by consultation participants) to have a limited claim on local resources. In many instances, local authorities were thought to be guilty of misleading communities: facilitating the production of affordable homes (that did not necessarily enjoy local support), suggesting it would be for ‘local need’ and then employing allocation policies (sometimes indirectly, through a housing association) that resulted in either the ‘wrong people’ moving into the village, or failed to provide the sorts of homes for the types of people who could make a real contribution to the community. In nearly all the consultations, it was felt that parishes and local groups needed greater say in what was built locally and who was housed. This issue is returned to below. The household surveys looked to investigate this in much greater detail in an attempt to conclusively reveal exactly who village residents consider to be ‘local’. So how is a local defined? The first question asks whether a person is local based on their link with the

6.3.2

63

Research into Rural Housing Affordability Final Report

village. Someone who lives in the village is generally regarded as local, usually followed by someone who used to live in the village and still has family relatives there. After that there are some marked differences between districts. In Mid Sussex people who are active or who work in the village but who do not live there are much more likely to be regarded as local than is the case in the other districts. While second home owners are generally not regarded as local by the majority of respondents except in Mid Sussex. Table 6.1: Would you define the following as a “local person” Mid Sussex 98% 72% Selby 78% 47% North Norfolk 92% 49% South Shropshire 93% 47%

someone who lives in the village someone who participates actively in the life of the village but does not live in it someone who has left the village but still has family/relatives living in the village someone who lives the village but not on a permanent basis someone who works locally but does not necessary live in the village 6.3.3

71%

72%

57%

69%

58% 52%

23% 46%

53% 36%

32% 40%

The second point is how long do you have to live in the village before you are considered a local. Again Mid Sussex stands out as being more accepting, the majority of respondents regard someone a local who has loved a year in the village, while in Selby 48% believe you have to be in the village for more than 10 years before you can be regarded as local.

64

Research into Rural Housing Affordability Final Report

Table 6.2:

How long do you have to live in the village to be regarded as local Mid Selby North South Sussex Norfolk Shropshire Proportion of respondents saying yes

People are considered local as soon as they move into the village At least 1 year in the village At least 5 years in the village

17%

11%

28%

15%

40% 23%

19% 18%

15% 14%

22% 23%

At least 10 years in the village Over 20 years in the village Over 50 years in the village

12% 7% 0%

12% 30% 1%

14% 20% 0%

10% 18% 0%

Born in the village / lifetime Other/do not know

1% 1%

5% 4%

5% 5%

2% 12%

6.3.4

Across all the villages it is the case that the longer a person has lived in the village the longer they believe someone should have lived in the village before they are regarded as local. In terms of distance from the village people are regarded as local who live in the closest village but not in the nearest town or the wider county. Table 6.3: Are people who live in the vicinity local if… Mid Sussex Selby North Norfolk South Shropshire 89% 33% 13%

they live in the next closest village they live in the closest town they live in the county

Proportion of respondents saying yes 83% 80% 30% 6% 34% 9%

67% 27% 4%

6.4
6.4.1

Local need and economic necessity
The extent to which working locally heightens a resident’s claim on housing is an issue closely linked to debates on the nature of localness. Two clear views emerged from the consultations. Firstly and especially in areas characterised by commuting, working locally is not viewed as a measure of someone’s local connection. Secondly, there is broad consensus around the need to support those people who need to work in the local area: not because this need makes them ‘more local’, but because having people providing essential services is crucial to the wellbeing of rural communities. With one exception the need for extra housing to support key workers within the case study villages was not seen as a significant issue. In Slaugham, the consultation group was quick to welcome help for key workers where it was necessary, but questioned this

6.4.2

65

Research into Rural Housing Affordability Final Report

necessity in the same breath: if there were essential workers in need of village housing, development would be supported (in theory), but as there is not such a need, then the question was dismissed as academic. Such views might be explained in terms of the profile of this ‘dynamic commuter’ parish, but elsewhere, responses were remarkably similar. In Bishop’s Castle, the economic power of the area was seen to be in the hands of ex-commuters who now commute less and work remotely from home. There are perceived to be too few jobs of the quality that would entice young people to stay in the village. As in Slaugham, the dominant view is that building more homes for those working locally is a fair aim, but working locally is not the objective of the majority of younger residents. 6.4.3 It is in the nature of deeper rural areas that jobs are increasingly scarce and found mostly in the farming sector. This view expressed in Hope Bowdler prompted a debate surrounding rural development and enterprise: development is the product of better transport and will not necessarily mean more jobs within rural areas, but will mean better access to jobs located elsewhere. This could of course take many rural areas on a development trajectory akin to those of the ‘dynamic commuter’ areas of southern England. In both of the South Shropshire parishes, the dislike of ‘non-contributing’ second home and ‘bolt-hole’ owners was balanced by a degree of acceptance that better transport infrastructure is a life-line to many rural communities. Their economic future is dependent on easier commuting. Indeed, accessibility of the type that speeds journeys to major centres is a major selling point for some rural areas. In Riccall, although a concentration of commuters might negatively impact on community life during the working day, it has also meant renewed support for local schools. Indeed, proximity to the A49 has a huge impact on housing demand and property prices in the wider area, with a great many young families drawn to the villages in this transport corridor. These views might seem to fly in the face of analyses which emphasise the need for rural enterprise coupled with a better supply of affordable rural housing for local workers. It is of course likely that the consultation panels comprised residents who were not part of the ‘low wage economy’ that a few flagged up. But it is the case that urban centres now provide the lion’s share of jobs for rural residents. That said, it was recognised that some ‘young working people’ are not only unable to buy homes, but also have considerable difficulty in meeting local rent demands. In Happisburgh, second homes were viewed as a direct challenge to maintaining a supply of affordable homes for local workers, eating up resources that were already scarce and forcing younger people in low-paid jobs to live with parents. This returns the discussion to the two clear views set out at the beginning of this section. Commuting is viewed as a reality of rural life, with decent roads providing access to the jobs that rural populations need. It is anachronistic to think that communities can thrive on in-situ employment. Therefore commuting is not seen as a threat to communities, but as a part of modern rural life. Many of the consultees were themselves newer residents and former if not current commuters. Few people would dispute the desirability of local workers being adequately housed, but firstly this is viewed as a minor concern in most communities and secondly, it is an intractably difficult challenge: the case for residence in a particular place is not always clear and it is often the view of mobile households that local workers should (and probably prefer to) look for housing in nearby towns.

6.4.4

6.4.5

6.5
6.5.1

Assigning local priority, delivering sustainable communities
In the light of the observations made in the above sections, it is difficult to draw any link between policies that assign local priority and the desire to deliver sustainable communities. There was no strong sense that communities are currently unsustainable (though some are unbalanced: see Section 2a), though they are not without their problems: concentrations of second homes in some instances, low wages and a loss of

66

Research into Rural Housing Affordability Final Report

young people. But local priority is not generally viewed as a means of tackling these challenges or ‘restoring balance’. Other actions might have a greater remedial effect. 6.5.2 One good example of a lack of faith in local priority and a preference for other strategies is found at Slaugham. A key challenge across the parish was judged to be the shortage of larger family homes. The possibility of ‘decanting’ the elderly into retirement units was viewed as one possible response. This would lead to more efficient use of the existing housing stock without resulting in a level of development that would ‘change the dynamic of the area’. In contrast, building new homes for general need and assigning priority was viewed as a fundamentally flawed strategy. The allocation policies employed by RSLs (registered social landlord) and particularly the ‘cascade’ approach, means that ‘nonlocals’ end up being housed in villages that accept the need for more ‘local needs housing’: the lesson being not to accept it. But this view seems to be at odds with what constitutes localness. In a general sense, localness can be earned, but for housing allocation, a normative measure that the residents of the village support is called for. In the case of Slaugham and Blakeney in particular, the critical point was that ‘local priority’ is not in fact ‘local’ but rather the priority of an RSL or an LA Housing Department: there is a lack of faith in the ability of the parish to control access, to set priority and to help local people. This is another critical point: priority is a contested concept with parishes questioning the right to set and impose priority. In Slaugham, arrival at this point led to a focusing on a broader issue of local politics, centring on power and responsibility. The discussion group, including three members of the Parish Council and two village association representatives, argued that devolving power and funding to the parishes, including the power to define local priority, is a prerequisite to real progress on rural housing issues: ‘the issue surrounding ‘priority’ is ‘whose priority’, of an external body or the community: the latter is the real priority’. The discussion at Slaugham was perhaps the most enlightening on this issue, though the sentiments were not unique. At Bishop’s Castle, it was felt that the CLT needed the power to set its own priorities, which might deviate from those of the District Council.. Hope Bowdler offered two very contrasting views. The first view was that the village is a hopeless case: it is best to ‘leave it alone’ and concentrate development in Church Stretton where there is a concentration of the things that people need. Deeply rural areas are not suited to families, so why implement priority policies in areas that offer so little? The same participant did not want his children to stay in the community: ‘why should they, when there is nothing here for them’. Perhaps it is an inconvenient truth that many rural areas experience an exodus of the young and an inflow of the retired, for exactly this reason: many rural areas have an attractiveness that is life-cycle specific. It is difficult to challenge this and perhaps not always necessary to do so. The second view was that the village of Hope Bowdler is indeed ‘dying’ and in desperate need of rejuvenation. In contrast to the ‘do nothing’ perspective, the need for significant service investment was called for, specifically to make the village more attractive to young people. Clearly, it is difficult to see how such investment would be supported or justified. And even within this call to arms there is no obvious place for priority policies, to give greater access to a village where there is so little on offer and so little prospect of this changing. Of course, Hope Bowdler is a very specific case and one can envisage instances where support for certain types of household could help ensure a greater social balance, help support local services, maintain social networks and so on. The Happisburgh case certainly brings us closest to this perspective, where such policies might help offset the impacts of second home buying, though they would inevitable be coupled with additions to the housing stock that might not find favour with the retired population of the village.

6.5.3

6.5.4

6.5.5

67

Research into Rural Housing Affordability Final Report

7
7.1
7.1.1

Conclusions
The study
This study examined three inter-related questions: local affordability as it impacts on change in communities, stability and social networks; how the attributes of a rural location affect prices / affordability; the expectation of priority for housing amongst local populations, its rationale, its benefits and drawbacks and its achievability. Primarily, this study has focused on the recent experiences of eight communities (or clusters of communities) in four English districts, selected for their representativeness of current rural economic and social conditions. The districts were judged to be ‘deep rural’ (and more cut off from urban influence), performing the role of a ‘retirement retreat’ (and characterised by an influx of migrants of retirement or near-retirement age), ‘transient’ (within the sphere of nearby urban areas subject to economic restructuring) and ‘dynamic commuter’ (where it is incomers who now seem to define communities and set local political priorities). Within these districts, communities which appeared to typify the socioeconomic characteristics of their wider area were selected for detailed analysis. This analysis had three parts: first, a desk-based, contextual evaluation of the housing markets in which the communities were situated, using HPM; second, consultations with community representatives at village meetings; and third, door-to-door surveys of local households. The latter two parts explored the wider implications of affordability and access barriers for rural households and attitudes to local priority policies. The hedonic analysis was concerned with increasing current understanding of the drivers of rural house prices. All of this analysis followed on from a review of existing work in the area (see appendix A), which formed the backdrop for the local analysis. The key messages that have emerged from this study are drawn primarily from the consultations, surveys and price analysis, though we begin each of the sections below with opening insights from the review conducted of prior literature.

7.1.2

7.1.3

7.1.4

7.2
7.2.1

Affordability
The review of literature dealing with this topic highlighted a number of critical concerns, which were used to frame community consultations and surveys. Past and recent studies highlight: Differences in the way that that affordability is understood (by those who would seek to measure it and those who experience it), its geographical characteristics, issues of who benefits and issues of who is adversely affected by low levels of affordable housing: just individual residents or communities as a whole; Local differences concerning ‘within’ and ‘without’ affordability: whether measures of affordability could, or should, deal with immediate village or parish needs or whether policies should focus on wider areas. Essentially, whether there is an expectation that the housing resources of a village or parish will meet all local need, or whether increasing access to affordable homes in nearby settlements is an acceptable response to dealing with local need; But related to this, local debates over mobility and employment: whether it is reasonable to expect people working in a village to move away but commute back for work reasons;

68

Research into Rural Housing Affordability Final Report

7.2.2 7.2.3

How affordability contributes to the sustainability of rural areas and communities: whether the addition of more affordable housing contributes to (by facilitating more balanced communities) or detracts from (by consuming rural land resources) sustainable development; and linking to this; How access to housing affects social networks and social capital – whether networks have evolved, in some instances, as a result of gentrification, or whether they have been eroded and severed because of displacement attendant on inmigration; The consultations and surveys added the following insights to this debate: The general consensus from the household surveys is that local housing is not affordable to local people on local wages. However, for many rural residents, ‘housing affordability’ is too general a concept. Conceding an affordability problem presents the risk of inviting additional development. For this reason, residents often prefer to talk about the particular needs of named individuals and have a preference for highly targeted, locally administered assistance. There is a concern for the needs of ‘local people on local wages’. Further analysis identified sub-market housing pressures in many areas, of a varying degree, though there was considerable disagreement between areas as to the appropriate response to such pressures. In communities dominated by highly mobile residents, the desire of young people to stay in a community was constantly questioned. Commuters tend to project the value of mobility on others, arguing that car-based-travel is a simple reality of rural living. More generally, it was argued across the communities that talk of ‘unaffordable housing’ is meaningless unless it is linked to need. Low levels of affordability do not present a problem unless affordability is preventing someone who needs to live in a community from doing so. In these cases, there is a clear preference for ‘innovating’ housing solutions rather than simply building more homes. Attitudes to affordable housing, which were often negative, were shaped by two fears: firstly, that all forms of development present the risk of changing the fundamental character of a settlement and secondly, that a lack of community control over affordable housing and its future management, means that it is likely to be offered to occupants not considered to have a sufficient claim on local resources. Affordability should not be a ‘trigger’ for development. Rather, there needs to be compelling evidence of named local people, with a pressing need to live locally, requiring homes in the community. Such attitudes, common across the case study communities, belie a suspicion of top-down solutions and a preference for home-grown, locally controlled, intervention. This was a critical theme running especially through the village consultations. Added to this, consultation and many household respondents had little belief in the geographical immediacy of local housing need, suggesting that needs could often be met in nearby market towns where those requiring a new home would also have easier access to jobs and services. The general view was that affordability should be addressed across the wider area and should not mean additional building in villages where issues of ‘rural character’ were paramount. Yet views varied on the case for in situ development. In those communities less dominated by commuters, there was less insistence that households in ‘need’ should be obliged to move to a nearby town. Many residents felt that additional development within the village would be acceptable, but only if this development could be controlled, perhaps through the establishment of a CLT, within the village itself.

7.2.4

7.2.5

7.2.6

7.2.7

7.2.8

69

Research into Rural Housing Affordability Final Report

7.2.9

The need for local control was often justified on the grounds that intervening agencies, local authorities and housing associations, too often got in wrong, building homes of the wrong type for ‘the wrong sort of people’. These factors had conspired to produce a perverse situation in which family homes remained empty and unsold, social housing went to people from outside the villages and young people were finding it increasingly difficult to stay. There was little evidence of displaced villagers ‘back commuting’ to jobs in the villages they had left. Given the lack of quality jobs in small village locations, it was believed that moving to a nearby market town meant that younger households were closer to work. However, it needs to be conceded that this research focused on remaining residents and not on those who had potentially been forced to move away. The survey, however, reinforced the view that movers had been motivated by employment factors primarily and that housing was an ancillary consideration. To be sustainable, communities need to be balanced. Specifically, this means that a village needs a ‘life-cycle’: a balance between younger residents, families and older people. Across this life-cycle, family-ties are important, sometimes focused around caring for elderly relatives. Ensuring ‘community balance’ was viewed in some communities as crucially important and ‘open policies’ of allocating homes were said not to work as they often fail to understand the way communities function. Even a mix of homes, leading to a mix of people, will not necessarily lead to a balanced community. Policies need to build on what is already there, recognising the imbalances that exist and seeking the address them. General development did not understand or meet the needs of the communities: homes were quickly filled either by commuters or by ‘people out in the sticks with no cars, where they can’t function’. Some communities have stronger ‘familial ties’ than others. In the dynamic commuter area, only 12% of residents had other family members in the village compared with 35% in the deep rural area. These figures are consistent with their classification and are underpinned by issues of mobility and transience. However, it is interesting to note that such differences did not dilute the preference for local control and responsibility over future development, which stood out as a key finding of this study. The reasons why people move out of the villages surveyed related both to housing (desire to move to larger/smaller properties or to buy a property) and employment matters. The former would suggest that the range of properties that people want is not available at the price they can afford. However, by their very nature villages are not going to be able to provide the range of properties that urban areas can do. Whether it is a function of affordability or generally the experience of older people more generally one consequence of people moving out of villages was to some extent to lead to a greater degree of isolation for this age group.

7.2.10

7.2.11

7.2.12

7.2.13

7.2.14

7.3
7.3.1

Influence on houses prices
Key issues emerging from the literature were the extent to which price influences can be quantified and qualified: how they are affected by measurable factors and how they are also driven by the value that individuals place on the rural environment and their own access to a rural ‘lifestyle’. The survey set the scene for the local analyses, revealing that in the commuter areas, more people had moved into the villages from nearby or more distant towns and cities. In the deeper rural areas, few people had previously lived in an urban area, though many had moved in from another village in the same district.

70

Research into Rural Housing Affordability Final Report

7.3.2

The parish consultations also provided essential context for understanding price drivers. They revealed that ‘accessibility’ is often crucial for people moving into a rural area: accessibility by road and rail and accessibility to jobs and social networks. However, accessibility is a ‘double-edged sword’. It can make an area convenient (and therefore push up prices), but good infrastructure can mean that an area shows up on the radar of local planning authorities looking for prime locations for new housing development. Being too accessible poses the risk of further development, reducing exclusiveness and potentially driving down prices. It is difficult to arrive at definitive statements regarding the price impact of shops, services and local schools. It is the most mobile new residents who often have the greatest spending power and arrive in rural areas with the perception that they will need to do a lot of ‘running around’. Hence, they may not necessarily be drawn to the most convenient locations, but may rather place a premium on seclusion. Primary schools are a draw only in those areas attracting families which tend to be the commuter areas. The location of secondary school in nearby towns does not appear to have any localised affect on prices, as its catchment is likely to cover a number of villages whose property markets will be driven by other factors. The consultations highlighted the combined effect of environmental quality and village character in determining property values. This affects prices in two ways. First, it has a honey pot effect on migrants. And second, migrants and other local households will seek to preserve this environment and character, opposing additional development and so reinforcing scarcity in the housing market. Other intangibles affect prices including a perceived sense of ‘community spirit’. Survey respondents felt that neighbourliness was an important attribute of their villages, but many placed a higher premium on perceived safety and security and on the environmental qualities noted above. Perceptions of a community spirit do not appear to play a principal role in drawing people to rural areas, at least not above other ‘perceptions’. Using HPM, it was possible to quantify some of the factors that explain house price variations across the case study districts. The following attributes had particular exploratory power: Property type is of course an important determinant of prices, with larger homes commanding higher prices. Buyers place a premium on detached homes, possibly equating this with the relative privacy that they seek in rural areas; Extreme isolation appears to depress prices: they is an optimum level of exclusion, somewhere between ‘too accessible’ and ‘too isolated’, at which prices tend to be higher; This search for optimum location means that prices rise away from major roads and rail stations (for about a kilometre or so) and then begin to fall off; Because key services are sometime located in larger centres, there is often an inverse relationship with rural prices. Buyers are choosing to be further away from these services if the area is taken as a whole. Within the built up area, buyers are choosing to be nearer the services and so prices are affected in an entirely different way; The level of service provision in a village must not ‘compromise’ the character of that village. There is an optimum service level linking to higher prices, in the same way as there is an optimum distance to transport infrastructure; The HMP approach can provide only limited insights into the price effects of environmental amenity including, for example, the measurable disbenefits of noise and pollution. Gathering data on such potential drivers is difficult, though

7.3.3

7.3.4

7.3.5

7.3.6

7.3.7

7.3.8

71

Research into Rural Housing Affordability Final Report

inferences can be drawn from the apparent effects on prices of properties close to road and rail; However, in one of the case study areas (North Norfolk), location within an AONB was found to have a significant effect on prices, pushing them up on average by 17%; This may be a straightforward environmental quality effect, though it is also likely that people moving into the AONB are aware of tighter planning restriction and place a premium on moving to a location where the likelihood of future development is less, thus helping to sustain their amenity and the value of their property. Planning restriction can be seen as a plus, helping to preserve house prices; Rural location, judged in terms of ‘sparsity’ in general, appeared to have a direct effect on property prices though this effect varied between districts. In Mid Sussex, it drove up prices. But in Selby, it had the opposite effect. In South Shropshire, extreme rurality had a negative effect on prices, with higher house prices observed in slightly more accessible locations. The pattern in North Norfolk was less clear; Rurality has an uncertain effect on prices overall. Whether buyers will pay more or less for a more or less accessible location depends on their characteristics. The effects of rurality in the four case study districts were shown to be consistent with the apparent socio-economic profiles of these areas. Commuters in southern England were looking for seclusion. In Selby, less affluent buyers wanted to be closer to jobs and services. In South Shropshire and North Norfolk, a mix of buyers – and a great many retired households in the latter location – seemed to want to strike a balance between seclusion and accessibility. Areas have different demand and need profiles and these are reflected in price drivers.

7.4
7.4.1

Local people
Again, the review of literature dealing with this topic highlighted a number of critical concerns, which were used to frame community consultations and surveys. The literature revealed: Persistent debates and uncertainties surrounding the definition of ‘localness’ and of what constitutes need, creating a ‘rocky foundation’ for local policies and interventions; it also underscored; The extent to which ‘local priority’ is a means of winning support for development which does not then subsequently tackle the needs of households whom local residents would deem ‘local’; The extent to which the local need for new homes should be tied to economic necessity: whether ‘local’ housing should only be supported where there is a clear economic case, i.e. where new development would support established or new industries. This might be supporting additional housing in some areas and less development in others. But in contradiction to this, another important debate concerns; The extent to which local priority policies might be viewed a means of addressing social polarisation, of creating more socially balanced communities and be seen as a stand-alone objective, separate from labour market concerns. This might involve building more retirement homes in some rural areas, for example. Taking a cue from existing debate concerning housing rights, the suggestion is that local priority is a means of ‘engineering’ more balanced communities with a view to retaining or restoring social networks and cohesion by creating opportunities for rural families to remain living in close proximity. The research itself provided a number of important insights, building on this literature base.

7.4.2

72

Research into Rural Housing Affordability Final Report

7.4.3

Normative definitions of ‘localness’ rely on measurable characteristics: length of residence, or whether someone was born or now works in an area, for example. These are used as proxies for judging a resident’s claim on local resources. In more ‘settled’ or deeper rural areas, residents tend to support the use of such criteria to differentiate between ‘local’ and ‘newer’ residents. This was the more ‘traditional view of localness’ and tended to see family-ties as critical to having a claim on housing. But elsewhere, especially in areas where there was a concentration of commuting households whose employment or family connection was weaker, or length of residence shorter, localness was a more qualitative judgement, concerned with ‘getting involved’ and partaking in the community spirit. For many commuters or retiring households, localness is something that is actively earned and not just passively gained over time. Where there is an acceptance of the traditional, normative measures of localness it is easier for local authorities to muster support for new housing under the banner of ‘local need’. However, where this had happened, residents recorded their dissatisfaction with actual outcomes, suggesting that ‘strategic’ allocation policies had commonly resulted in the ‘wrong people’ being allocated housing. This was viewed as evidence supporting the need for local control over local housing. In commuter areas, introducing housebuilding under the guise of ‘local housing’ was seen as a non-starter. Because localness was earned though active residence, there was no such thing as a non-resident local person needing local housing. These were people who were unable to afford housing in the wider area and would be better off in one of the nearby towns. This argument was not, however, applied to people with a legitimate need to live in a specific village who, it was thought, should be provided with housing tailored to their situation. Regarding local need and economic necessity (to live in a named location) two clear views emerged: firstly and especially in areas characterised by commuting, working locally is not viewed as a measure of someone’s local connection. Rather, being able to do so was seen as a luxury, albeit a luxury that rarely delivered a decent wage. Secondly, there was broad consensus around the need to support those people who need to work in the local area: not because this need makes them ‘more local’, but because having people providing essential services is crucial to the wellbeing of rural communities. The need for extra housing to support key workers within the case study villages was rarely seen as a critical issue. The economic power of many areas was seen to be in the hands of ex-commuters who now commute less and work from home. There was generally felt to be too few jobs of the right quality to entice young people to remain in small village locations. Local housing for young people was considered a laudable aim, but many consultees questioned the strength of this ‘market’. The economic fortunes of many rural areas depend on their accessibility to commuting households. In all case study areas, better communication infrastructure was seen as a life-line to villages, though there was general concern, especially in South Shropshire, over the proliferation of second home and ‘bolt-hole’ owners whose contribution to local economies was questioned. Commuting is seen very much as a reality of rural life, with decent roads providing access to the jobs that the rural population needs. That said, it needs to be recognised that a great many consultees were themselves newer residents and former if not current communities. However, these views do express the social reality of modern rural communities.

7.4.4

7.4.5

7.4.6

7.4.7

7.4.8

7.4.9

7.4.10

7.4.11

7.4.12

73

Research into Rural Housing Affordability Final Report

7.4.13

Because of this context, many residents question the need to assign local priority, probably because many question the underlying need for additional housing. Yet, where additional housing is needed, communities seem to have little faith that new homes will go to those who are ‘truly local’. ‘Local priority’ is not in fact ‘local’ but rather the priority of an RSL or a local authority housing department: there is a lack of faith in the ability of the parish to control access, to set priority and to help local people. Priority is a contested concept with parishes questioning the right to set and impose priority. Priority is fundamentally a matter of power and responsibility. It was suggested that devolving power, responsibility and funding to the parishes, including the power to define local priority, is a prerequisite to real progress on rural housing issues: the ‘issue surrounding priority’ is ‘whose priority’: of an external body or the community? The latter is ‘really priority’. ‘Community priority’ needs to work with the economic reality and the ‘life-cycle’ of villages, accepting the ‘inconvenient truth’ that rural areas experience a natural exodus of the young and an inflow of the retired. However, it was clear in some areas that support for this life-cycle, through interventions aimed at creating new housing opportunities, might offset the impacts of second home buying, by restoring social balance and increasing the viability of services for those who depend on them.

7.4.14

7.4.15

7.4.16

74

Research into Rural Housing Affordability Final Report

Appendix
Appendix A: Literature Review
Part 1 - Local Affordability and its Impact on Communities Part 2 - Housing Markets and Market Determinants Part 3 - Communities, Networks and Local Priority

Appendix B: Shortlisting Case Study Areas Appendix C: Household Survey Appendix D: Estate Agent Survey Appendix E: Hedonic Analysis of House Price Data in each case study area

75

Research into Rural Housing Affordability Final Report

Part 1: Local Affordability and its Impact on Communities
Gentrification and displacement
Gentrification: equally applicable to rural areas
The term ‘gentrification’ is used to describe the process whereby the influx of wealthier individuals into residential area progressively replaces the working or lower-classes already living there. There are two principal explanations for the gentrification process. The first, the production or supply side theory, sees the gentrification process as the movement of financial capital back to the inner-city stimulated by the opportunity for profit. The second, the demand or consumption-side theory, regards the process as a movement of people back to the inner-city motivated primarily by residential preference. Gentrification is widely perceived as exclusively urban phenomenon, as bulk of research and literature focuses on urban settings, however rural researchers (Phillips: 1993, 1998; Darling 2005) have demonstrated the process occurs in rural areas.

Rural gentrification: supply vs. demand side theories
Within rural studies is the demand and consumption-side explanation of gentrification that has generally been adopted. The gentrification process has been depicted as the product of demand for rural residential locations by ex-urban or suburban middle class households driven primarily by lifestyle preferences and quality of life considerations – the lure of a scenic location and low density living, with a gentler pace of life and a strong sense of local ‘community’. An ideal space in which to retire or raise a family. The process is also fuelled by ‘push’ factors, the strong anti-urban motivations, such as the desire to escape crime, congestion and pollution (Mitchell, 2004). However supply side theories are equally relevant in a rural context. Wealth in ‘escalator’ areas such as the south east in particular, due to rising disposable incomes and growing housing assets, have enabled the ‘export’ of urban wealth to lower cost housing markets in ‘importing’ regions. This has manifested itself primarily in the form of increasing numbers second homes/holiday cottages in rural areas which are owned by affluent households from outside the area. Phillips (1993; 2002) and Darling (2005) identify that there are significant returns to be made on refurbishing and extending rural dwellings and the re-valorisation of defunct agricultural and other rural non-residential buildings when converted to residential accommodation. This same process was observed by Buller and Hoggart (1994) in their study of the motivations and experiences of British home buyers moving to rural France in the 1990s.

Historically protected rural settlements: more susceptible to gentrification?
Phillips (2000) identified that the development histories of particular rural settlements influenced their susceptibility to the gentrification process. Those with a history of domination by a sole estate owner were often ‘closed’ to new residential development, unless it was to meet the needs of the estate. After the modern planning system was introduced in 1947, these settlements, due to their size and ‘traditional’ character, often continued to be protected from development. It is this perpetuating protected status has resulted in them becoming progressively exclusive and predisposed to gentrification. In contrast, in open parishes, where the estate owner exerted less control, or perhaps there were numerous landowners, the villages were allowed to expand. The planning system often permits incremental in-fill development in these types of rural settlement, or, in some cases, even extensions

76

Research into Rural Housing Affordability Final Report

to them. The lack of certain protection from future development makes these less exclusive than the historically ‘closed’ settlements.

Rural gentrifiers: The defence of the rural idyll
Phillips (2002) also identifies that once ex-urban in-migrants have moved into a rural settlement, they seek to defend it against perceived threats from physical and social change. Their desire is not only to preserve the rural identity of the wider locality, but also to protect the value of their property. There is a rich literature on the NIMBY tendency to ‘raise the drawbridge’ against further development or inmigration, by taking control of local political structures and actively campaigning against further change. In their study of second homes in rural England, Gallent et al (2002) found evidence of seasonal residents in the South Cotswolds mobilising against social housing schemes, arguing that the villages they had bought into should be protected from development that was more suited to an urban setting. This is a common argument: that the countryside is an inappropriate place for all forms of development and should be maintained – frozen in aspic – for the benefit of those who make the appropriate financial investment (Woods, 2005). In areas where there has been significant migration / gentrification, an ‘environmental’ representation of the countryside frequently prevails. In areas where working communities are still strong, the tendency is to promote further development that will support the future development of social capacity. The ‘rural idyll’ is often a preconception of what the countryside should be, introduced by gentrifiers, rather than an accurate description of what it actually is.

Adverse effects of gentrification: residential displacement
One of the principal negative consequences of the process of gentrification is community change, socio-economic and demographic, caused by in-migration of professional middle class groups and the subsequent exiting of indigenous groups with fewer resources. The way in which these indigenous groups exit the community is commonly cited, in both academic (Hartman, 1979); (Atkinson, 2000) and media circles, as being as a result of being directly displaced by in-migrants. Displacement can be instigated in a number of different ways. From an economic perspective, it is due to a state of disequilibrium between property supply and demand (Sumka, 1979; Lee & Hodge, 1984). Displacement reflects the ability of professional middle class groups to acquire properties within the sought after neighbourhoods at the expense of indigenous residents. From a social perspective, it is due to the contrasting priorities, perspectives and lifestyles of the in-migrants (Atkinson, 2000); Newman & Wyly (2006). These contrive to alter the social composition of an area so that existing community networks become distended. This, combined with a (perceived) higher turnover of residents, becomes a significant push factor.

Types of residential displacement
Lyons (1996) identifies three ways in which residential displacement is deemed to occur: Where rents are increased to a level where the occupant is forced to vacate the property (often aided by financial incentives) or where individuals have insufficient financial means to enter a specific residential property market Where the existing social network is distended by the incursion of different higher income groups to the extent the area becomes alien and unrelated to their own lifestyle When the cost of living in an area is increased beyond the means of the locals, reflecting the increased disposable income and spending power of higher income groups.

77

Research into Rural Housing Affordability Final Report

Reduced/adjusted service provision which becomes more orientated towards higher income groups. Economic and environmental changes are to some extent interlinked; as the economic status of the area grows, environmental change begins to manifest. As the social class and wealth within the neighbourhood rises, the level and type of service provision, both private and public sector, adjusts to accommodate new consumption preferences (Atkinson, 2000); Newman & Wyly, 2006). The displacement of sections of rural society may act to conceal pockets of surviving deprivation (masking poverty with introduced wealth), introduce resistance to specific (if not all) types of development, and create a social structure with an entirely different set of service needs. Existing studies suggest that gentrifiers: Bring wealth to the countryside which may conceal remaining poverty and make the countryside as a whole, or at least some villages, seem capable of meeting its own needs (Gallent et al, 2008); community change can also Introduce a resistance to some forms of development – especially new housing – which may shape political reaction (at the level of parish or district councils) to new development and cause opposition to development by a majority who ignore the needs of a minority; and Create a social structure with different needs, especially where retired households, commuters or second home owners come to dominate a community: a concentration of retired people or seasonal residents will reduce demand for school places and other services; commuters may have less need for public transport or local services including shops. In some communities the balance between permanent and temporary residents has been altered and ‘changes in the number of people living permanently in the countryside or making use of its resources on a temporary basis’ (Clout, 1972: 8) may have a major impact on services.

Gentrification-induced displacement: a successive process?
Processes of gentrification can create a continued upward movement in the status of those being displaced. Lyons (1996) states that gentrified areas experience outflows by successively higher status households. Such processes have been confirmed by the research conducted by Dangschat (1991) in Hamburg where he identified continuum of gentrification waves from ‘pioneer’ to ‘ultra’ gentrifiers.

Gentrification-induced displacement – myth or reality?
On the whole, there no conclusive evidence that gentrification-induced displacement is the engine of neighbourhood change. Atkinson (2000a; 2000b) documented substantial residential displacement and community change in certain areas of London, which he attributed to the gentrification process sweeping through central areas of the city. However contradictory evidence exists as the research undertaken by Bridge (1993), also within London but focussing on other neighbourhoods, indicated that the impact on community composition and cohesion was minimal. Neither conclusion can be viewed as definitive and perhaps indicates that gentrification does not have a generic end result. It would be reasonable to assume that the extent of the impact of gentrification will depend on its totality, penetration and extensiveness, as well as the degree of community cohesion beforehand.

Displacement: a problematic research topic
There a number of problems that renders research into the process and effects of gentrificationinduced displacement problematic. The first problem lies with the conceptualisation of displacement. No unanimous definition has been agreed besides the fundamental notion of displacement as mobility that is not a voluntary outcome of household decision-making. The difficulty with this basic definition is that it fails to specify which uncontrollable circumstances constitute grounds for displacement

78

Research into Rural Housing Affordability Final Report

(Atkinson, 2000a). If strictly defined, only acts of harassment and eviction are deemed to result in forced displacement. But broader, more liberal, interpretations take account of the pricing out of residents from property markets, loss of social networks and the transformation of shops and services. The second problem with displacement research stems methodological problems. With regards to quantitative research, the suitability and availability of statistical data often presents a number of difficulties. The way in which socio-economic data and information on migration patterns is collected and collated is not always conducive to displacement research. Often concessions must be made and proxy variables utilised. From a qualitative perspective, it is difficult to identify people who have been displaces as, by definition, displaced residents have disappeared from the very places where researchers go to look for them (Newman & Wyly, 2006) and consequently the trajectory and exits routes of displaces is unknown. The third problem with displacement research is that it can only be at best a ‘snap shot’ - a static point on a continuum of change. How far a community is along this continuum can be determinative. Reflections on change may be provided by residents who, clearly, have not been displaced but are potential gentrifiers. For this reason, community reaction to displacement may appear muted. In a rural context, Bollom (1978) has observed that communities which seem unconcerned about second homes, retirement and other forms of migration are often those subject to greatest degree of change: local voices have been displaced and silenced. On the other hand, the vociferous communities are those witnessing the tide of change lapping over their neighbours but which have, as yet, been largely unaffected.

Affordability
Defining housing affordability
Affordability is a measure of whether housing can be afforded by certain groups of households. It links the cost of housing with the amount of money that households are able to spend on a home. Analysing affordability provides information about the accessibility of particular types and locations of housing to local people. Areas with poor affordability can experience unsustainable travel patterns together with high dependence on social housing, overcrowding and homelessness.

Understanding housing affordability
The Government’s central objective for housing policy is to ensure that everyone can live in a decent home, at a price they can afford and in a community where they want to live (ODPM, PPS3, 2005). Unbalanced housing markets and affordability problems are preventing this objective from being achieved in high demand areas, where the supply of new homes falls short of demand. However, balancing housing markets is not simply a case of ensuring that the total stock of housing (supply) meets the estimated the number of households (demand). Only a proportion of dwellings are actually available for sale or for rent at any given time, and it is this plus newly built dwellings coming on stream that constitutes effective supply. Similarly only a proportion of households plus newly forming households are actively seeking a home at any one point in time and it is these households constitute effective demand. In addition there may be ‘pent-up’ (also described as ‘latent’ or ‘hidden’) housing demand, from people who would like to rent or buy a home but are unable to do so. The main reason for this is usually affordability, although there can also be housing on the market that is affordable to a range of households but is not desirable or suitable. There can be a number of reasons for this, such as unfitness, poor location, or simply the wrong size e.g. one bedroom flats when demand is for larger units.

79

Research into Rural Housing Affordability Final Report

Measuring affordability
There are two main approaches to measuring affordability: a general indication of whether house prices are affordable in relation to incomes and a more specific measure of whether particular housing is affordable to certain groups. In addition, there are a couple of more detailed ways of measuring affordability such as establishing the income required to purchase a home, looking at a household’s ‘residual income’ or the income that remains once housing costs have been met and whether an individual or household has access to finance in order to purchase a house (Whitehead et al, 2008)

House prices to income ratios
This approach to measuring affordability gives a general indication of whether house prices are affordable in relation to incomes. This indicator is often termed the house price to income ratio, and the government’s preferred version is the ratio of the lower quartile house price to annual lower quartile earnings. This simple indicator can be used at national, regional, sub-regional and local spatial scales and is particularly useful in making comparisons over time or between areas.

Ability to afford home ownership/ private renting
The second approach to affordability is measuring whether housing is affordable to different household groups. This measure is used when determining the proportion of new homes that need to be affordable. Assuming that the local authority has details of the income distribution (e.g. From a household survey), then the proportion who cannot afford local market housing can be estimated.

Annual income necessary to purchase a house
This measure is a general indicator of affordability in a district or region and is useful for making comparisons. It is commonly used by Homebuy agents and housing associations in setting the affordable price and rent levels for shared ownership products and assessing eligibility. Clearly housing associations do not wish to encourage home ownership for households on the very borders of affordability, because they will be vulnerable if interest rates rise. This measure helps to avoid such situations.

Residual incomes
The residual income is the income a household has left over after they have paid housing costs. It gives what is in some ways a more accurate picture of affordability than price income ratios since it recognises that lower income households are only able to afford smaller proportions of their income on housing without facing difficulties.

Access to finance
When an individual or a household is making a decision to purchase a property one constraint may be their access to the mortgage they require. Mortgage lenders will factor in many individual characteristics when deciding whether or not to lend, but a key factor will be the size of the borrower’s deposit and the associated loan to value ratio.

What has caused an affordability problem in rural areas?
Several factors have conspired to reduce the apparent affordability of housing in some areas and to some groups in rural areas. These include:

80

Research into Rural Housing Affordability Final Report

Increased demand for rural housing
The general demand for housing is growing underpinned a changing age structure of the population, increased longevity, inward migration and smaller households. However this demand has been amplified in the most desirable residential locations and because of the value placed on rural lifestyle, rural areas generally constitute such locations. A rural settlement location is consistently found in surveys to be the preference of the majority of English people. In accessible rural areas, demand from higher income commuters has resulted in sustained rises in house prices, taking owner occupation beyond the reach of many people. In other attractive, but remoter, areas prices have risen due to retirement migration and/or demand from holiday and second homes. The consequence of this buoyant demand has been a rise in house prices well above the rate of inflation (Halfacree, 1994)

Constrained supply of rural housing
The principal focus of rural planning has been, arguably, concerned with the protection of the visual quality of the countryside rather than facilitating development to meet rural housing needs (Taylor, 2008). The economic and social consequences of restricting the supply of rural land for housing, in conjunction with the increased demand for it, has had a profound effect. The supply factors have been identified as general absence of speculative building, high construction costs and tight planning restrictions on new development (Shucksmith 1990), who suggests that these supply factors are interrelated. The difficulty in obtaining large sites, due to planning restrictions limit new building to in-fill sites within existing village boundaries, has had two effects on supply. Firstly, there is less speculative building as the sites are too small to interest the volume builders. Secondly, rural housing developments do not enjoy the economies of scale associated with larger urban developments resulting in unit costs being higher. Construction costs are often increased further because of planning condition requiring the use of traditional and expensive construction materials. These factors have led builders in many rural areas to cater for the most affluent, by building expensive luxury houses which justify the higher land costs and construction costs. Consequently, low income groups often find themselves priced out of the rural property market. One of the consequences for rural England of high demand, coupled with a lack of housing supply, has been worsening affordability. Houses are, on average, much less affordable in areas of villages, hamlets and isolated dwellings compared to small towns and larger urban areas. There is a worsening imbalance in supply and demand for rural housing, especially in the smaller rural settlements. There are also reasons why housing has become less accessible: reasons which link to falling levels of affordability but also to reductions in the availability of different forms of housing. These include the loss of public housing through the right to buy since 1980 (and the inadequacy of mechanisms designed to reduce sales in designated rural areas), the reduction in the supply of housing association homes resultant on reduced grant funding from the late 1980s, and the loss of private renting opportunities tied to rural employment. It is this reduction in tenure choice that has meant home ownership has become the principal point of housing access for many rural households. The lack of alternatives means that there is huge pressure for the market to deliver enough homes to meet the needs of a wider range of households with hugely different incomes. This lack of choice has forced many households into an increasingly competitive private market, when arguably their needs could have been more easily met had there been a sustained supply of alternatives. The rural housing debate is not only about increasing the affordability of private ownership, but also increasing accessibility to alternatives.

The spatial scale of affordability – an analytical problem
One of the barriers to analysis and understanding of rural affordability is the problem of geographical scale and at which scale the issue should be both measured and addressed. Strategic housing markets, which are defined at regional level and used by local authorities, can vary considerably in terms of geographical scale. In rural districts these sub-regions can cover large geographical areas

81

Research into Rural Housing Affordability Final Report

incorporating rural areas of relatively low and high demand. Therefore an affordability problem in one area may be offset by a low demand in the rest of the sub-region. The question of how far a household is expected to look outside of their local area is a pertinent one. Research and experience suggests that people willing to consider a range of settlements within the market area, providing the property meets their needs. On this basis it would therefore make sense to look at the dynamics of the wider housing market rather than focus on the individual settlement level. However in practice, local planning authorities are tasked with making land available either within or adjoining existing rural communities in order to meet identified local needs. This is indicative of the standpoint that affordability should be addressed at the local level. There is a need to be able to appraise affordability at a more localised level. But if we are to measure affordability at smaller scales, there is a need to relate these smaller areas to their wider market context. Bramley (2009) forwards the concepts of inward and outward affordability as a mechanism to achieve this: ‘Within area’ affordability: the percentage of households in a neighbourhood who can afford to enter the housing market in that neighbourhood and highlights discrepancies between neighbourhood incomes and house prices ‘Outward’ affordability: the percentage of households in a neighbourhood who can afford to enter the housing market in the wider HMA/LA and highlights those neighbourhoods which have the highest incidence of the social problem ‘Inward’ affordability: the percentage of households in the wider HMA/LA who could afford to enter the housing market in this particular neighbourhood and highlights those neighbourhoods that offer the greater or lesser opportunity to overcome the problem. Research by Bramley (2009) indicates that ‘outward’ affordability is greater than ‘inward’ affordability at ward level in rural areas. This means that the people living in rural areas are less likely to be poor, whilst the housing located there is less likely to be affordable. This tends to suggest that poorer people will have a tendency to drift towards more urban areas, so tending to reinforce this pattern.

The spatial scale of affordability – a sustainable development issue?
The spatial scale of affordability is also raised when considering the ‘ease of access’ to housing. The viability of affordable housing schemes on the fringes of market towns may be undermined if residents have to travel significant distances to get to work or visit friends/relatives. In this sense, the overall cost of housing, transportation and access to services may mitigate against affordability. Affordable homes built in smaller rural settlements allows households to remain close to places of work, sustain social networks ties and access local services, both formal and informal. This links the notion of affordability with environmental sustainability and argues that true housing affordability must take into account a wider range of costs than just rent or mortgage costs (Australian Conservation Foundation and the Victorian Council of Social Service, 2008)

Adverse effects of the affordability problem in rural areas – unbalanced and unsustainable communities
The terms ‘social mix’ and ‘social balance’ cover numerous overlapping characteristics of a population such as age, tenure, class, income and ethnicity. ‘Social mix’ suggests that the community in question has a blend of all of these factors. ‘Balance’, on the other hand, represents an external reference point with which comparisons can be made. A neighbourhood may be more or less ‘balanced’ when certain characteristics are compared with those of a wider district or of equivalent neighbourhoods (Cole & Goodchild, 2001)

82

Research into Rural Housing Affordability Final Report

Balance, however, has another connotation: sustainability. To date the categorisation of rural settlements, by Local Planning Authorities, into those regarded as sustainable and therefore suitable for new housing and investment, or unsustainable and unsuitable, was conducted on the basis of service availability criteria (shop, post office, regular public transport service). In reality this approach has served to further undermine the sustainability of rural settlements (Cole & Goodchild, 2001). Taylor (2008) makes this same point: he argues that the ‘restrictive nature of many planning practices’ (Taylor, 2008: 8) contributes to the pressures faced by many rural communities. Smaller villages and hamlets are caught in a ‘sustainability trap’ and only more flexible planning with a ‘real sense of vision that is based on recognising how our rural communities can be rather than writing them off as unsustainable’ (ibid, 8) will allow them to escape. Taylor’s major finding was that many rural communities are written off by the planning system as inherently unsustainable (the same point made earlier by Cole and Goodchild, and that has been made by previous generations of researchers).

Reworking sustainable communities: the integration of societal dimensions
The narrow criteria used by planners are not reflective of the step change in opinion as to what constitutes a sustainable community. The UK Sustainable Development Strategy (2005) indicates that sustainable communities are those which are well balanced in economic, environmental and social factors. Traditional economic and environmental factors are carried forward such as local employment and services, affordable housing, green space, biodiversity and reducing CO2 emissions. However, the Sustainable Development Strategy introduced social dimensions of sustainable communities referring to factors such as social capital, community cohesion and support networks. Reflecting national priority, communities are denied the chance to grow because of the desire to limit car-based travel (ibid, 45). Inflexibility fails to safeguard jobs or provide the homes that are needed by these communities (ibid, 16). For this reason, Taylor’s primary recommendation was that there should be a new accommodation between the different strands of sustainability in rural areas (essentially, that a rebalance is needed between environmental and social perspectives / rationales). The first two recommendations of the Review dealt with planning. The first called on government to provide greater coherency in planning policy, creating this balance and, by inference, placing greater weight on community need. The second suggested a new requirement that local authorities take into account ‘all three strands of sustainability [again] in a balanced way’ and achieve this through a long term vision that splices together these strands. Many of the Review’s recommendations touched on the need to reform planning practice. The central message was that planning practice (and policy) is getting it wrong, creating many of the problems that rural communities currently face. It is certainly not delivering government’s goal of building sustainable communities. Definition and components of sustainable communities: UK sustainable development strategy (2005) The UK Sustainable Development Strategy identified eight components which collectively contribute to the creation of a sustainable community Active, inclusive and safe: fair, tolerant and cohesive with a strong local culture and other shared community activities - It suggests a diverse, vibrant and creative local culture encouraging pride in the community and cohesion within it. It also suggests an active voluntary and community sector. Well run, with effective and inclusive participation, representation and leadership - Essential if a community is to respond positively to change. Effective engagement and participation by local people, groups and businesses is vital especially in the planning, design and long term stewardship of their community. Environmentally sensitive providing places for people to live that are considerate of the environment - It requires a safe and healthy local environment with well designed public and green space.

83

Research into Rural Housing Affordability Final Report

Well designed and built – featuring a quality built and natural environment - A community must be of sufficient size, scale and density and have an effective layout to support basic amenities in the neighbourhood and minimise use of resources (including land). Buildings both individually and collectively must meet different needs over time, and minimise the use of resources. A sustainable community requires a well-integrated mix of decent homes of different types and tenures to support a range of household sizes, ages and incomes. The community should have a 'sense of place'. Well connected, with good transport services and communication linking people to jobs, schools, health and other services - Good public transport and other transport infrastructure is needed both within the community and linking it to urban, rural and regional centres. Thriving with a flourishing and diverse local economy – To generate a wide range of jobs and training opportunities. Well served, with public, private, community and voluntary services that are appropriate to people’s needs and accessible to all - Good quality, local public services should be available including education and training opportunities, health care, community and leisure facilities. Fair for everyone now and in the future - All our individual and communal choices may impact adversely on others especially in terms of the overall need for sustainable development.

The impact of affordability on the components of the sustainable development strategy
The affordability of housing in rural areas and the supply of affordable housing have long been identified as essential to the vitality and sustainability of rural communities. A plethora of studies since the 1970s have drawn a link between access to housing, the retention of a stable and permanent population base, and the social and economic viability of rural communities. Noteworthy amongst these studies are the works of Cloke (1979) on key settlements, Moseley (1979; 2000) on service access and community change, Shucksmith (1981; 1990) and Capstick (1987) on local access to affordable housing, and Clark (1984) on strategies for securing this access. These various works, together with more recent studies, pinpoint the impacts of a lack of affordable housing, which can now be linked to achieving the goals of the UK Sustainable Development Strategy.

Impact on local services
The consumption behaviour of the in-migrants to rural areas has negative impact on the provision of local services. Although this can be partially attributed to the rationalisation of both public and private sector service-providers and retailing, the high mobility of in-migrants and their connection with nearby urban employment centres means they are less dependent on local services and public transport, thereby reducing the demand for these services to be delivered locally and justifying the argument for centralised services.

Impact on community cohesion
There is a trend of general ‘delocalisation’ of rural life (McCarthy, 2008). Delocalisation occurs when people obtain goods and services and conduct social and leisure activities at various extra-local places (e.g. Nearby market town or urban centre) but living in a preferred locality (i.e. Village). Inmigrants of rural villages have typically been leap-frogging the locality, travelling to acquire services: work, shopping, and leisure. This trend has caused the gradual decline of community-based activities and has undermined social cohesion.

84

Research into Rural Housing Affordability Final Report

Impact on social networks
The subsequent closure of local shops, post offices, schools and other community facilities not only causes problems for the less mobile residents, but also removes the locations for regular interaction between community members. The loss of these locations for community interaction assists the fragmentation of communities as residents interact only with their immediate neighbours and personal social networks (Liepins, 2000).

Impact on ‘fairness’
Clearly affordability impacts on low income individual/households who reside in high demand areas. When affordability creates ‘exclusionary’ displacement, whereby local people are ‘locked out’ of the local property market and are forced to look further afield to meet their housing needs, it could be argued that such a situation is unfair. There are also connotations for the future viability and vitality of rural communities if low income locals are unable to access their local housing market in perpetuity.

85

Research into Rural Housing Affordability Final Report

Part 2: Housing Markets and Market Determinants
Understanding rural housing markets
Defining the rural: official definitions
The primary definition of ‘rural’ is that provided by the Office of National Statistics (ONS). It defines settlements with over 10,000 residents as ‘urban’. Smaller settlements fall into one of the additional four categories shown in Figure 1.1. In addition to morphology, settlements are defined in context as to whether they are in ‘sparse’ or ‘less sparse’ areas. The ONS Definition cannot be applied to large geographical areas such as Local Authorities. Defra has produced a supplementary classification of Local Authority Districts and Unitary Authorities. This classification specifies six categories: • • • • • • Major Urban: districts with either 100,000 people or 50 percent of their population in urban areas with a population of more than 750,000. Large Urban: districts with either 50,000 people or 50 percent of their population in one of 17 urban areas with a population between 250,000 and 750,000. Other Urban: districts with fewer than 37,000 people or less than 26 percent of their population in rural settlements and larger market towns. Significant Rural: districts with more than 37,000 people or more than 26 percent of their population in rural settlements and larger market towns. Rural-50: districts with at least 50 percent but less than 80 percent of their population in rural settlements and larger market towns. Rural-80: districts with at least 80 percent of their population in rural settlements and larger market towns.

86

Research into Rural Housing Affordability Final Report

Figure 1.1: England’s Rural Areas

Defining the rural: concepts of rurality
Cloke et al (2006) argue that social constructions of the rural are critically important (reflecting understanding of rural spaces, attitudes towards them and driving local political priorities as well as priorities for the planning process). But these ‘social constructions’ sit alongside two other ways of

87

Research into Rural Housing Affordability Final Report

understanding, and thinking about, the countryside. In all, ‘Rurality’ is often conceptualised through one of three ‘theoretical frames’: Functional concepts of rurality The delimitation of rural areas based on land use mix (notably farming and forestry), settlement structure (small settlements of low order and an ‘extensive landscape’), a way of life characterised by cohesive identity linked to an extensive landscape; Political economy concepts Expressions of broader political-economic process creating an observable ‘rural dimension’ leading to a ‘spatial debate’ surrounding population change, employment patterns and so forth. The rural political economy is seen to be characterised by a particular political ideology (favouring private and voluntary action over public intervention), by different patterns of production, by dispersion and difficulty in providing public services, and by low consumption and economic inactivity (especially retirement); Social constructions of rurality Drawing on post-modern and post-structural thinking, and concerned with ‘sociospatial distinctiveness’, rurality may be defined by the ‘interconnections between socio-cultural constructs of rurality and nature […] and the actual lived experiences and practices of lives in these spaces’ (Cloke et al, 2006: 21). This frame sees conceptions of rural areas as being social-constructed, with many different perspectives on and meanings attached to ‘rurality’. Rural areas are understood to have functional characteristics, as having particular ways of getting things done (a ‘community spirit’ manifest in voluntary action, sometimes understood as greater freedom from public influence) and also through a personal lens. That latter may mean that many aspiring rural dwellings have a ‘chocolate box’ image of the countryside, which as MacGregor (1976: 524) points out, is ‘[…] much nearer to the jolly village green on the pantomime stage than reality’. However, the popular appeal of and belief in this image has a propensity to trigger migration, and to impact on rural property prices.

Defining the rural: notions of community
Tonnies (1957) considered that rural communities were characterised by social relationships founded on locality, community and kinship – which he termed ‘Gemeinschaft’. Conversely, urban communities were typified by an absence of these characteristics and relationships were described as transitory, superficial and impersonal – termed ‘Gessellschaft’. Tonnies’ notion of rural and urban communities being poles apart facilitated the development of the concept of the rural-urban continuum. Pahl’s (1970) study of rural communities redefined the continuum concept and introduced a greater complexity. He identified that Gemeinschaft and Gessellschaft characteristics can be found in both urban and rural spaces. Pahl used the term ‘urban villages’ to describe the districts of cities where Gemeinschaft qualities were clearly evident and ‘metropolitan villages’ for rural settlements Gessellschaft qualities were clearly evident. He reasoned that urban values had begun to permeate the countryside (and vice versa) and therefore it was unrealistic to ascribe homogenous characteristics to any particular spatial area. As such, the rural-urban continuum must be perceived as a transitional social process, rather than a spatial typological definition.

88

Research into Rural Housing Affordability Final Report

Defining the rural: social constructions
Pahl’s conclusion that community characteristics cannot be linked to spatial areas has endured and has not been significantly developed until recently by Soja (1996). Soja argues that there are two ‘spaces’ that are perceived, which, together, create a notion of community. The first, ‘firstspace’ is the perceived space derived solely from sensory perception. The second, ‘secondspace’ is the interpretation of the environment. These real and imagined spaces which when brought together form a combined ‘thirdspace’ that helps to imagine a community’ into being, based on physical surroundings, individual experiences, cultural bindings and real and imagined awareness. Soja’s theory is similar, in some regards, to those of ‘habitus’ advocated by Bourdieu (1985). Habitus is best described as patterns of thought, behaviour, and taste acquired through social positioning, personal history and upbringing. The notion of habitus can explain the different ways in which spaces and places are constructed and interpreted by different people and therefore neatly dovetails with Soja’s concept of secondspace.

Perceptions that create housing market demand in rural areas
The majority of recent urban–rural migrants have been drawn to the countryside by real and perceived ‘characteristics of the surroundings’ (reference). The ‘pleasant and green character’ of an area, especially ‘living close to nature’, ‘living outdoors’, having access to ‘a beautiful landscape’ and ‘the fresh air’ appear to be key motivating factors. This predominant appeal of the characteristics of rural areas matches the findings of Halfacree (1994) and reflects the idyllic, chocolate box image, of the English countryside derided by MacGregor in the 1970s. This ‘romanticised’ construct / image fuels the desire for a pure and simple style of living, close to green and natural amenities. In the early 19th century, a myth of the countryside built up around the Romantic Movement, with rural England seen as a refuge from the grime and noise of the Industrial Revolution. Perceptions (and understanding) of the countryside became increasingly one-dimensional: the poverty was forgotten, and so too was the reality of a working countryside with an economic life of its own. This misperception and misunderstanding has survived to the present day. The countryside has become a refuge from modernity, where one can (apparently) escape the hectic ‘urban life’ (with all its negative connotations) of big towns and cities, but continue (because of better communication links) to reap the benefits of an urban salary. People in the UK have an extremely positive image of the countryside and Halfacree (1994) point to the influence of this conceived ‘rural idyll’ in decisions to move to the countryside. Individuals may not always act on the basis of a rounded understanding of rural life (comparable crime rates, inaccessibility to jobs, declining service standards, social deprivation and so on), but sometimes on the basis of their preconception, interpretation or mental construct. This is significant for understanding the drivers of some rural housing markets. To some extent, the mass media has peddled and reinforced this image of the idyllic countryside. It is virtually impossible to recall the number of films or television serials aired since the late 1960s that have promoted rural over urban living, especially for families.

Understanding rural housing markets
In the UK there has been a growing national interest in rural housing markets. For the majority of the public, and particularly the media, it is due to the rapid rise in house prices over the last few years. For policy-makers, it has been this factor coupled with the divergent performance of the English housing market. The characteristics of regional and sub-regional markets within the country vary substantially. Localised forces often result in great variations within these sub-regional markets themselves. These variations have made it especially difficult to analyse residential mobility and housing market demand. The characteristics of rural housing markets are determined by a combination of their demographics, strength of local economy and employment market, quality of local schools and educational facilities, accessibility, quality and composition of the housing stock and residential amenity (Champion et al.

89

Research into Rural Housing Affordability Final Report

1998). Perception (above) also has a powerful role to play, with rural housing – in what is viewed as an idyllic rural setting – seen as a ‘positional good’ (Schmied, 2002). Demographic factors: The impact of the life-cycle on residential location Migration varies in a regular way with age in most developed countries at all scales. Rates peak at around the age of16-34 and then decline to a minimum in the 60s (apart from a modest rise around retirement age). These differences in migration rates by age have implications for the overall level of migration in a population as its age structure changes. Age, by itself, has no direct influence on the ability of a person to migrate and is a combination of a set of conditions which has been termed the ‘life course’. Migration is often triggered by changes to household size and circumstances (e.g. Birth/raising of children and retirement). In the child-rearing years households seek locations with secure jobs, spacious housing, good play groups, nurseries and schools and so select suburban/rural neighbourhoods where there are other families like themselves. The relationship is very different for the age group 16-29, there is a large outflow of late adolescents to areas of high population density (typically underpinned by further education studies). This age group also includes some childless households who often prefer urban areas, underpinned by sociocultural motivations. Philips (1993) demonstrates a relationship between household employment structure and relative rates of mobility and residential preference. Households with more than one breadwinner show a propensity to be less mobile than do traditional male-breadwinner households because of the need to consolidate at a fixed residential location in order to sustain two incomes or careers. Dual income household may also prefer an urban residential location due to the convenient/support services and amenities available. Economic factors: Increase in rural-based industry and increased mobility There has been a transformation in the geography of employment, which has become increasingly biased towards rural areas. Fielding (1982) linked urban-rural migration trends with economic restructuring and the changing spatial division of labour. The late 1970s and early 1980s was a period of de-concentration. The increase in the number of retirees to rural areas and the overall trend of the countryside becoming a space of consumption has lead to the significant expansion of service sector employment. Champion (1989) suggested that rural inmigration was potentially a cyclical phenomenon closely bound up with the health of the economy and that urban-rural flows would slow down in times of economic boom as urban centres became revitalised and rural areas less secure. Although there is evidence of the renewed migration, particularly amongst young people, into city centres, rural population expansion has continued during the 1990s even within the more remoter areas (Champion et al., 1998). Research suggests that long-distance migration flows are more likely to be driven by employment factors. There are clear spatial and regional patterns to job related moves with the rural areas surrounding urban conurbations being most prone to commuting flows and more remoter rural areas more characterised by in-migrants being employed locally. In many examples, families 'trade-off" housing and lifestyle advantages against employment accessibility (Halfacree, 1994). The level and type of employment available in an area determine the levels of wages that will be paid and will influence the tenure, size and location of housing. Studies suggest that as incomes rise, demand for neighbourhood quality increases together with house

90

Research into Rural Housing Affordability Final Report

size. Ultimately demand for market housing is driven by how much housing people can afford, rather than how much they might be considered to ‘need’. Macroeconomic factors such as interest rates, inflation and national economic growth affect new house building and investment on the supply side and household finances on the demand side. For example, higher interest rates increase the cost of borrowing, thereby reducing demand for housing and affecting employment levels by increasing the cost of investment. Mortgage rationing by lenders (as during the current credit crunch) affects new house building and investment as potential buyers are unable to access funds. Second homes and vacant dwellings: inverse drivers of demand Second homes and vacant dwellings can both be drivers of demand (Hickman et al., 2007). While second homes are quite a small proportion of the total stock at national level, in some areas the proportion is very high (Gallent et al, 2005). Six Wards in Cumbria record a percentage of second/holiday homes higher than 20% with the highest, Grasmere, recorded at 45.8% (State of the Countryside Report, 2005). The demand for second homes in desirable holiday areas fuels demand, pushing up house prices and creating affordability problems for local households. Although high vacancy rates are usually an indication of low demand, vacancies can be high in high demand areas as a result of absentee landlords and ‘buy to leave empty’ on the part of investors. In the case of the latter scenario this can displace demand elsewhere – the ripple effect - as any remaining potential demand will seek out the next ‘available’ location. Accessibility and school catchment areas: A locational determinant Accessibility, to employment, services, shops and schools, is an important driver of demand (Hickman et al., 2007). Proximity to railway stations adds to house prices while a lack of access to public transport reduces values. School catchment areas can also influence demand. Families are known to move home so that their children can attend popular schools. This puts additional pressure on house prices within the catchment area of the school. Again, poorly achieving schools can have the opposite effect, as people try to ensure that their children do not have to attend an unpopular school. Residential amenity: Counter-urbanisation and the penchant for rural locations For over thirty years migration research confirmed and (re-confirmed) the overall demographic growth of rural England both in terms of extent and incursion into the more remote rural areas of the country. Studies by Champion (1989) and Champion et al (1998) identify that rural Britain overall benefited from substantial in-migration, largely from urban areas during the 1970 to 1990 period. Boyle et al (1998) identify that the perceived quality of the physical and social quality of the environment are principle reasons for moving into the countryside. Buying into the 'rural idyll' has become a major element in the residential and lifestyle trajectories of the British middle classes, as a rural location is usually equated with improved quality of life and enhanced social status (Phillips, 1998): rural housing, in some instances and locations, becomes a positional good. The quality of the residential environment is also a dominant factor in migration decision-making. The most scenic and/or accessible rural locations are those which experience the greatest demand.

91

Research into Rural Housing Affordability Final Report

The urban exodus is generally continuing but is increasingly differentiated. Migration continues to be the dominant force in the changing demography of rural areas but recent analyses suggest that principal sources for that in-migration are changing; no longer the metropolitan centres and the industrial cities but increasingly the high density non-metropolitan counties (Champion et al, 1998), a different pattern from the previous decade when metropolitan counties recorded the highest rate. The greatest net losses and gains are recorded by the two ends of the settlement hierarchy, there is a clear temptation to link the two and conclude that the main element in counter-urbanisation comprises migrants moving directly from large cities into rural areas. However, rather than moving from central cities to remote rural areas in a single displacement, migration patterns associated with urban-rural shifts are often, in reality, more ‘progressive’ - the reinforcing the notion of a counter-urbanisation 'cascade' (Champion 1989). This cascade is depicted in the diagram overleaf. Although overall the population of rural Britain is expanding, largely due to inmigration down the urban hierarchy, rural out-migration remains a persistent issue in some locations. Overall rural population loss is concentrated in the northern regions, in the north west and north east. Rural districts characterised by high rates of in-migration such as the south west are also characterised by high rates of outmigration amongst the younger age groups. In many rural areas affected by the inmigration of the professional and managerial classes, a net out-migration of skilled and unskilled workers is also identified (State of the Countryside, 2007).

Figure 1.2:

The Counter-urbanisation Cascade (Champion, 2000: 14)

City Suburb Fringe Town Rural Area

Traditional assessment of housing markets: The mono-centric model Economic approaches have traditionally been used to analyse residential mobility and housing market demand. There has been a long-standing assumption that residential preference is related principally to employment and financial opportunity (e.g. Barker, 2004). This is referred to as the mono-centric model and is based on the notion that the stronger the regional economy and labour market, the greater the housing market demand within that region. However this economic approach has

92

Research into Rural Housing Affordability Final Report

limitations, as it does not have appreciation for the complex geographies of supply and demand of local markets with regional and sub-regional areas. This simplistic distinction denies the fact that low demand and affordability problems can co-exist at different spatial scales (district, sub-region and region). For this reason, talk of a ‘north-south divide’ in the national housing market – with low demand in the northern regions and an overheating market in the south (before the current economic downturn) – is often disputed. Regional markets are complex with many diverse and contrasting patterns. Critics of the mono-centric model (Janssen et al. 2001) argue that housing market analysis should take account of lifestyle and aspirational factors, which also drive demand. Changes in household structure and the different life-cycle stages are can be determinants of residential preference. Equally residential preference can be influenced by the ability of an area to provide the assets (social, cultural, environmental) deemed to be important. The latter suggests that residential mobility is more than a matter of rational choice and can be equally informed by the nature, character and outlook of individuals.

Alternative methods of housing market assessment: Hedonic price theory
Hedonic price models provide a framework for the analysis of market value of differentiated goods possessing features that, in themselves, do not have a clearly definable market price. In other words, their impact on the overall price of a good is uncertain or at least not easily quantifiable. The traditional use of hedonic estimation in housing analysis has been for the purpose of drawing ‘inferences’ about the price effects of different attributes such as air quality, airport noise, commuter access (railway, subway or highway) and neighbourhood amenities (Janssen et al. 2001) which are not directly linked to the utility of a property or affect prices in uncertain ways. Four types of price-determining attributes are frequently considered in hedonic analyses (and also in property valuation): locational, structural, and neighbourhood attributes (relatively ‘hard’ drivers) are set against externalities that can lift or lower prices. These can be described as ‘soft’ drivers. Locational attributes Accessibility is measured in terms of ease of commuting to and from amenities, and is measured by travelling time, cost of travel, convenience, and availability of different transport modes as buyers tend to trade-off housing costs against transport costs, and are willing to pay more for properties with easy accessibility to work (Janssen et al. 2001). View is sometimes considered a residential amenity usually associated with the location of a dwelling site. With regards to green amenities Garrod and Willis (1992) determined that proximity to a river, canal or woodland has a strong, positive influence on house value. The presence of a canal or river would raise the value of the average house by 4.9% while the proximity of at least 20% woodland cover would raise it by 7.1%. Location in or near a rural settlement, such as a village or country town, also has a positive effect.

Structural attributes Prices of properties are frequently related to their structural attributes. Numerous studies reveal that the number of rooms and bedrooms, the number of bathrooms and the floor area are positively related to the sale price of houses. This is because buyers are willing to pay more for more space, especially functional space. Residential properties with bigger floor areas are desired by big families and buyers who can afford a better standard of living. (Jansen et al., 2001) Neighbourhood attributes

93

Research into Rural Housing Affordability Final Report

The ‘social class’ of a neighbourhood or area has an impact on property values. The quality of public schools has also been found to have a significant impact on real house prices. School quality is more important to local residents than either crime or environmental quality, with higher test scores and better exam results having a positive impact on property prices. Proximity to council housing has been shown to lower prices. Proximity to hospitals and health centres is not desirable due to the commotion that ensues including the nuisance effect of ambulance siren and the general congestion in the vicinity of hospitals (Garrod and Willis, 1992). Proximity to other types of land use will also impact on prices. In rural areas, these can include main roads or motorways, airports (or being close to a flight path), proximity to neighbourly uses such as power plants, incinerators, kennels, catteries, piggeries, or other noise / smell generating uses. There is a documented tendency of buyers to shy away from such uses or to object to planning applications that might, for instance, cause an intensification of nearby smell, perhaps emanating from a pig farm. Externalities such as crime rates and traffic noise, Obviously, home buyers do not favour areas associated with high rates of crime or vandalism. Reaction towards noise, or quiet, is dissimilar among different groups of people. Affluent groups give noise much greater weight when considering residential locations. External benefits, including pleasant landscape, unpolluted air, serenity, quiet atmosphere, and the presence of urban forests has a significant positive effect on property values (Garrod and Willis, 1992). There is a clear potential for rural areas to have a range of (perceived) externalities that would attract potential buyers - landscape and environmental quality, peacefulness, greenery: these are all aspects of the ‘rural idyll’. Therefore, how close an area is seen to fit with this (urban middle class) representation of the countryside will impact on prices. Schmied (2002) has provided some evidence of this impact in the Cotswolds, an area that might be seen to represent archetypal English countryside. Other areas of low-land England – Cornwall, the Chiltern hills, the south downs, and the Norfolk broads – may be seen to share these characteristics, with the opportunity to buy into the rural idyll (the repository of English values, according to Margaret Thatcher in the 1980s) lifting property prices. This uplift may partly be explained in terms of strong demand pressure, but also in terms of the subsequent planning constraint associated with and amplified by the arrival of conservative and articulate middle class incomers. Upland areas, including the peaks of Derbyshire and the Lake District, offer something slightly different – majestic landscapes. This ‘majesty’ joins the list of externalities driving demand.

Shortcomings of the Hedonic Price Model
However, this approach is subject to criticisms arising from potential problems relating to fundamental model assumptions and estimation such as the selection of independent variables and market segmentation. In theory, hedonic price studies do not require the segmentation of housing markets and consequently housing markets are often pooled together. However, in practice, submarkets are likely to exist. This is because housing markets are not uniform. Hence, it is unrealistic to treat the housing market in any geographical location as a single entity (Jansen et al., 2001) making it difficult, for example, to undertake analysis on the basis of administrative boundaries merely because this is the level at which data is available. There may be factors affecting prices beyond the boundary which may lift or lower prices. Any hedonic analysis needs to be alive to this possibility.

94

Research into Rural Housing Affordability Final Report

Another issue frequently associated with the hedonic price model is the ‘misspecification’ of variables. Misspecification is the situation where an irrelevant independent variable is included (overspecification), or where a relevant independent variable (attribute of a product) is omitted (underspecification). The latter problem is perhaps more common. As hedonic price models deal with the implicit prices of quantities of attributes of a product, the problem of misspecification of variables is inevitable. Over-specification gives estimated independent variables that are both unbiased and consistent, but inefficient because of the inclusion of the irrelevant variable, whereas underspecification results in estimated coefficients that are both biased and inconsistent (Jansen et al. 2001). It is also the case that soft drivers may, by their very nature, be difficult to quantify. How does one, for example, factor ‘perception’ into a hedonic model? Most past studies of the determinants of rural house prices have merely drawn inferences from local surveys. Housing surveys, for example, may ask residents why they chose to live in a particular village or move to the area. Responses are interpreted as determinants of migration (see earlier discussion) with migration seen as a factor in pushing up property prices. But a hedonic model has to make judgements about landscape and environmental quality and infer a link between this quality and the effect it might have on the desire of people to purchase property in an area. It cannot measure the impact of preconception on relocation decisions. For example, ex-urban migrants to rural areas may expect to find a community spirit, a greener lifestyle, greater tranquillity and so on. But their may be a wide gap between expectation and reality. The reality may not explain prices to the same degree as the preconception, but the preconception can only be understood through face-to-face contact with movers, not through proxies that try to connect with lived reality. Similarly, tastes and values may drive migration and prices: how are these to be factored into a hedonic analysis. The best we can hope for is that such an analysis tells part of the stories, with any gaps filled by existing local studies or future surveys.

Dealing with shortcomings of the Hedonic Price Model
In the US, analysts have taken different approaches to overcoming the potential problems of pooling housing markets by identifying sub-market boundaries within housing areas. Post code districts and census tracts (wards) have frequently been used to identify submarkets (Goodman, 1981). Goodman and Thibodeau (1998) identified housing sub-market boundaries by developing and estimating the parameters of a hierarchical mode, based on the notion that all homes within a spatially concentrated area share amenities associated with the property’s location (i.e. the housing characteristics that determine a property’s market value are nested in a hierarchy — properties within neighbourhoods, neighbourhoods within school catchment areas, school catchment areas within boroughs/towns etc). With regards to the problems relating to the misspecification of variables, Butler (1982) states that since all estimates of hedonic price models are mis-specified to some extent, those which use a smaller number of key variables are to be preferred. Butler suggested that only those attributes that are costly to produce and yield utility be considered in the regression equation. Mok et al. (1995) concluded that biases due to missing variables are small and would not necessarily compromise results. A practical solution to the problem of missing variables, which may cause bias, is to ensure that the data set used is homogeneous. When there is homogeneity, the use of the hedonic price approach is justified.

95

Research into Rural Housing Affordability Final Report

Part 3: Communities, Networks and Local Priority
Networks and community cohesion
Defining Community
Delanty (2003) identifies four broad ways in which the term ‘community’ can be applied: The first, socio-spatial, is the territorial notion of community such as urban neighbourhood or rural village. Classic community studies focussed explicitly on the territorial notion deducing that strength of community cohesion was dependent on settlement size and geographical proximity, which influenced levels of face-to-face interaction (Bridge, 2002). Studies by Bell and Newby (1976) have rejected the territorial notion citing that geographical proximity does not necessarily create community cohesion and localness does not necessarily generate close association. Bell and Newby concluded that communities can exist without close-knit and interacting social networks, and that such social networks can transcend spatial geography. This reflects the clear distinction between Delanty’s first group and the remaining three: identity-driven, politically mobilised and technological. The socio-spatial notion of community implies a ‘static’ territorially based static community, whilst the others: identity-driven (formed along sexuality, cultural beliefs or lifestyle choices), politically mobilised (formed around the need to collectively address social injustice) and technological (formed around new technologies and the internet such as chat rooms and social network websites) are more ‘fluid’ and free from spatial fixity.

Community lost – the decline of territorial communities?
Notwithstanding evidence of the existence of non-territorial communities, successive community studies have been biased towards spatially defined communities and the factors which have led to less face-to-face social interaction within them. The ‘loss of community’ thesis is a well developed theme in community studies literature. Putnam (1995; 2000) suggests that levels of social capital are deteriorating, to the detriment of individual lives and community cohesion. Putnam shows how Americans have increasingly detached themselves from one another and how social organisations have been disintegrating. He attributes this process to a number of factors: The movement of women into the labour force. Over the last two or three decades, many women have moved out of the home and into paid employment. Putnam attributes more women in the labour market to reduced time and energy available for voluntary activities and interacting with family and friends. Therefore the emergence of two-career families might be the most important single factor in the erosion of social capital. Mobility Numerous studies have shown that residential stability affects levels of social capital. Mobility tends to disrupt opportunities to develop strong, personal ties to one another and to participate in community organisations. This theory is based on the notion that when the population of an area is constantly changing: a process which undermines cohesion and social capital, with communities eventually becoming fragmented and social disorganised. Other demographic transformations. There has been a decline in the prevalence of the ‘traditional’ family unit over the last 30 years, with the growth in number single parent families, fewer children, growth in single-person households. Evidence of the loosening of bonds within the

96

Research into Rural Housing Affordability Final Report

family, both extended and nuclear, is well documented. These changes might account for some of the progressive loss of social capital. Economic change The changes in scale that have swept over the economy, illustrated by the replacement of the corner shop by the supermarket has undermined and now Internet shopping at home, or the replacement of community-based enterprises by outposts of distant multinational firms, may perhaps have undermined the material and even physical basis for civic engagement. The technological transformation of leisure. There is reason to believe that deep-seated technological trends are radically ‘privatizing’ or ‘individualizing’ our use of leisure time and thus disrupting many opportunities for social capital formation. The most obvious and probably the most powerful instrument of this revolution is television. Electronic technology enables individual tastes to be satisfied more fully, but at the cost of the positive social externalities associated with more ‘primitive forms’ of entertainment.

Community found – the rise of networked communities?
Wellman (2001) argues that contemporary community is to be found in networks, and are no longer confined to a particular place but stretched out geographically and socially. People are tied into multiple networks essentially having a ‘personal community’ - consisting of interpersonal ties of sociality, support, information and identity. Therefore he argues that commentators who proclaim the loss of community (such as Putnam) have looked for them in the wrong places. Wellman captures this shift as a transformation from door-to-door to place-to-place to person-toperson communities. First, people walking to visit each other typify door-to-door communities that were spatially compact and densely knit; ‘little boxes’ based upon geographical propinquity (Wellman, 2002). Significant others were encountered through walking about such neighbourhoods and there was overlap of family life, work and friendships. According to Wellman, door-to-door communities expired with increased speed in transportation and especially communications which facilitated a transition to ‘place-to-place’ communities. ‘Place-to-place’ community, interactions moved inside the private home. The household is visited, telephoned or emailed (Wellman, 2002). Yet this is not seen as destroying networks and social capital, because phone calls and emails connect individuals in different geographical locations and enable interaction with those not living close by. Contrary to the argument that the Internet makes social networks disembodied and virtual, Wellman’s studies suggest that new technology supplements inperson and telephone communication rather than replacing them. Frequent contact on the Internet, in this regard, is seen to complement to frequent face-to-face contact, not a substitute. ‘Person-to-person’ community highlights how it is the person rather than place that increasingly matters. The transition from place-to-place to person-to-person communities results from innovations in communications - the technological development of portable telecommunications have enabled the rise of ‘networked individualism’. Whereas the symbolic technology of place-to-place connectivity was the fixed landline telephone, the mobile phone is the technology of person-to-person communities. The mobile phone frees people from ‘spatial fixity’ (Wellman, 2001)

Social networks and rural community change
There are the two sets of fundamental changes that have affected our villages. There are the organic changes seen in all rural areas: smaller nuclear families with fewer ties to support networks, fewer people actually living and working within any village or town, and the attrition of groups that historically

97

Research into Rural Housing Affordability Final Report

formed the ‘glue’ that kept a community together – such as women’s institutes, parish and church groups, and farming networks. More recent changes affect the dynamics of an area: the introduction of technologies that change the traditional ways of doing things, an influx of people from other countries, and the migration of younger people due to higher education and employment opportunities outside the local area (Woods, 2005). Changing views of the rural have invoked a process of repopulation of many rural areas from the city. This process of repopulation often masks the continuing out-migration of young people from rural areas. The combination of in-and out-migration leaves many remote rural areas with an ageing population, while accessible rural areas become urban dormitories where during working hours almost all inhabitants move to towns and cities nearby. This will have an impact on all aspects of life – economic activity‚ community‚ and services

Implications for rural residents current and future
Traditionally rural communities provided social support because of their 'close-knit' informal networks and family/friend networks. These informal mechanisms of support are gradually being eroded: the out-migration of younger groups has implications for vulnerable older people in rural areas as the opportunity to receive support from family members is removed (Stockdale, 2004). Many older people rely on others for assistance with day-to-day tasks bringing them vital social contact as well as support, enabling them to continue to live independently. Historically, those who have aged in rural communities will find this easier as they have social networks to support them. However rapid social change in the countryside has diffused their social networks.

Enabling and facilitating social networks: creating social capital and sustainability?
The physical and social structures of rural communities traditionally create a close-knit climate wherein community members share a strong sense of community. However, these communities, where everyone know else and neighbours are always there to lend a hand are rapidly disappearing with the loss of the elder generation (Falk and Kirkpatrick, 2000). Social contact between family, friends and neighbours, and also with community groups, is important and has been proven to protect against the effects of loneliness and isolation, especially within the older sections of the community (Wenger, 2001).

Localness and local rights
Addressing the social balance problem: local needs housing
There is a concern that local people cannot access their local property market due to demand/supply factors and this is adversely affecting the social balance, composition, and vitality of rural communities (Hoggart and Henderson, 2005). Even if a need for additional growth/housing in a rural village is identified and then delivered, experience shows that it is unlikely that all the units will go to local people if sold on the open market. The ability to pay (more) typically results in some of the units being occupied by households from outside the locality meaning the local need is not fully addressed as intended. Therefore concern for local housing needs is widespread among local authorities, particularly from a political perspective. The majority of local authorities operate some form/variant of local needs policies, yet it is not clear what is meant by ‘local need’. There is a lack consistency in the definitions of ‘local’ and ‘need’ from local authority to local authority.

What constitutes local?
The first question is who is local? It is generally accepted that in the context of local needs policies, the term ‘local’ relates to people, who have a long term or ancestral/generational link to the village or

98

Research into Rural Housing Affordability Final Report

immediate hinterlands. However, is it equitable to ring-fence the definition of local and positively discriminate in this way - are the middle class incomers not locals once they become settled in the community? The second question is what is need? It is difficult to define and measure as there is a difference between whether an individual/household needs to live in a particular locality or whether it would simply prefer to do so. This is a valid argument, however it does raise the issue of social justice – does the inability to pay mean that households should accept second, third or even fourth choice residential locations? Burnett (1998) considers the social categories of ‘local’ vs. ‘incomer’ to be fuzzy and doubts whether they are meaningful or accurate. Local can have two definitions: firstly, reference to a spatially defined locality and those residents within it and secondly, reference to a sense of things indigenous. The latter is the most problematic definition and the one that commonly underpins contemporary local needs policies. What constitutes local in this regard is not easily defined as it becomes a subjective exercise – how long must an individual reside in an area before they are construed as local, what ties to the area justify being regarded as local? On the other hand, incomers are overwhelmingly constructed as a potential threat to ‘existing’ rural cultures and communities. However, there are incomers who operate as part of a ‘local community’ and who may engage with local culture, traditions and resources either for work and leisure. Therefore the complexity of incomer–local relations and what actually constitutes the local populace, place and social structures is not straightforward.

Mechanisms to widen social mix and create sustainable communities: local needs housing
The concern for the housing needs of local rural people is reflected at both local and national level (Taylor, 2008). One of the most widely-employed tactics used to address housing affordability problems in rural areas is the use of local needs policies, which seek to positively discriminate in favour of local people. These policies entail giving priority access to local people (e.g. households with a ‘local connection’ or those working locally) to new homes, whether these are for rent, for shared ownership or for outright purchase (Gallent, forthcoming 2009). Local needs policies are an attempt by local planning authorities to resolve the conflict of objectives between restraining development in the interests of the landscape and farmland protection and encouraging housing provision for the benefit of rural communities (Shucksmith, 1990). Prevailing planning policy/legislation renders the option of increasing supply to meet both normative need and demand unviable to local authorities.

Advocating local needs policies
Therefore ‘local need’ seems to relate less to the social concept of need and more to the balancing of demand with a politically acceptable rate is housing supply in rural areas. But even if the concept of ‘local need’ is elaborated to allow a more precise and operational definition of need and then combined with some geographical boundary to arrive at a workable concept, the question that eventually arises is whether local claims to rural housing should have any special priority? Rogers (1985) forwards five reasons for support for such an approach: 1. Local residence acts as a proxy for welfare objectives in housing policy since local people in rural areas are taken to be generally poorer and more deprived. However this is an over simplification, as rural areas contain many wealthy people. Therefore a policy which uses locals as a proxy variable for the rural poor will not necessarily benefit those actually in need. Outsiders have no serious grounds for being housed in the countryside. This reason can be quickly dismissed on social justice and human rights grounds.

2.

99

Research into Rural Housing Affordability Final Report

3.

Housing for local people contributes directly to local employment generation and support and maintenance of services. This argument contains a contradiction insofar as localness can therefore be justified in the case of those who move from outside of the area to work within it. This type of claim is in direct conflict with those arguments which appeal to the idea of the locally born person beleaguered in the housing market by the influx of outsiders. Local people have the moral right to priority over outsiders. This is often popular with local councillors, since local authorities are elected to reflect the interests of existing residents / voters. It makes no sense, on equity, welfare or economic grounds, to preferentially assist local people in areas where there is no labour demand. It would make more sense to direct assistance to locations where there are jobs and other opportunities. In most instances, planning policy does the latter but there is a weight of feeling in rural areas in favour of ‘ancestral rights’ – that ‘local homes’ should go to local people and that where the open market inhibits this ‘natural justice’ the planning system should intervene to sort things out Pragmatic acknowledgement that the label of ‘local needs’ is politically very powerful in that it has genuine popular appeal and demands media attention for rural housing issues. The amalgam of rural deprivation, grass roots politics and practical compromise between development and control, represented by the label ‘local need’, provides a way of getting rural housing issues onto the political agenda and mustering support around the need for some development.

4.

5.

Three decades ago, Mark Shucksmith catalogued some of this debate, looking at the different interpretations of the term ‘local’ for policy purposes. He argued that attaching acceptable meanings to the terms ‘local’ and ‘need’ is a critical challenge for planning policy (Shucksmith, 1981: 17), adding that doubts and uncertainties surrounding definitions of local (from a specific village, from a cluster of villages, from a district, and then, of course, what does ‘being from’ entail: being born somewhere, having a family connection, working in a village or area?) and of the nature and ‘ethics’ (Rogers, 1985) of need make the idea of ‘local needs’ a ‘[…] very rocky foundation on which to build policy’ (Shucksmith, 1990: 66). This practical and moral uncertainty (should the hand of planning be making such ‘personal’ judgements?) is frequently sidestepped the use if the term ‘local need’ as a sweetener, with the label often taken to denote a ‘[…] politically acceptable rate of housing development, rather than a social concept of need’ (Shucksmith, 1990: 64)

The impact of first-wave local needs policies
The Lake District Special Planning Board (the predecessor to the current National Park Authority) sought to use Section 52 of the 1971 Town and Country Planning Act to control the occupancy of all new housing built in the Lake District, restricting ‘all new development to that which can be shown to satisfy a local need’ (LDSPB, 1977). Its aim was to combat the rise in house prices attributed to inmigration and second home buying by requiring all new housing to be built for ‘local need’. However, the policy was in fact counter-productive: the amount of speculative building declined and second home buyers continued to compete for existing property in much the same way as they had done prior to the ‘locals only’ policy. The concentration of competition in the market for existing homes combined with the reduction in house building pushed up house prices, further disadvantaging local buyers.

Impact of contemporary local needs policies
Attempts have been made recently to use the planning system in a positive way to secure the provision of affordable local needs housing, in rural areas and elsewhere. Exception sites to provide affordable housing in small rural settlements

100

Research into Rural Housing Affordability Final Report

Most local planning authorities now have policies which enable them to grant planning permission for small sites within and on the periphery of villages, which would not usually be supported under the provisions of the local planning document. These dwellings provide affordable housing to meet an identified local need and would remain affordable to the local community in perpetuity. Quotas of cross-subsidised affordable housing being required in larger developments through Section 106 planning agreements Most local planning authorities use 106 agreements to secure quotas of affordable housing in larger private developments. Subject to financial viability constraints, developers normally offset the cost of providing the affordable units via cross-subsidy. Again, a demonstrated need for affordable housing is necessary to justify the Local Planning Authorities request for the affordable units. The effectiveness of the two methods of affordable housing provision have been examined. In reality very unit are delivered via exceptions sites. This is partly due to a lack of finance and poor information on local need, but the main factors limiting the contribution of this approach are “the limited supply of land, and the potential tension” between environmental and social objectives of policy. The majority of affordable dwellings are provided via section 106 agreements as quotas of large private developments on the edges of market towns as urban extensions. Therefore section 106 agreements are effective in delivering affordable rural housing, but only in larger settlements, while the exceptions policy is invaluable in the smaller settlements but delivers relatively few houses. The overall level of affordable housing provision falls far short of estimates of need (Gallent, 2009).

Different perspectives on local needs housing
Beyond this, landowners’ motives for holding land or selling land, and thus contributing to the residential development process, may be affected by personal and non-pecuniary considerations as much as by financial or investment criteria (Goodchild & Munton, 1985). Among the factors influencing his decision to release land will be the landowner’s material interests and social relations, covering kinship, social status and class. Rural exception sites have proven problematic as landowners are often reticent about releasing land either because they hope land will one day command full development value, which is the most common, or because they have no direct control over who will benefit from their ‘good will’ (Gallent, 2009) Planning consents often imposed 'local need' conditions, so that only those with a local connection would be considered when the housing was allocated. This had mixed implications for rural RSLs. It was sometimes seen in a positive light, involving more sensitivity to local opinion and a move away from purely needs-based allocations (a 'community housing' approach which puts rural RSLs in the forefront of national trends in allocations policies). On the other hand, the imposition of occupancy restrictions creates inflexibility for RSLs in finding tenants. It may make it more difficult to attract private finance for development, as lenders' ability to recover loans through future rents or sale may be inhibited by occupancy restrictions.

Meeting the need: the case for releasing more land
The National Housing and Planning Advice Unit (NHPAU) states that the release of land for housing, and target building rates set by Regional Planning Bodies, should take account of the growth in second home demand in England (as a component of overall migration and rural population growth). This headline was read by some as an acknowledgement that second homes impact on house prices. The view of the NHPAU is that they have a potential to impact on prices and housing supply, but they should not be allowed to do so. Rather, the planning system should be releasing sufficient land for housing so as to meet the need for development implied by demography, plus this form of ‘additional requirement’. Second homes will only be a problem if enough land is not allocated for all types of housing demand. This is a hugely controversial assertion, particularly as the new homes being planned are unlikely to become second homes, but are more likely to be built in market towns and bigger centres and occupied by rural households ‘displaced’ from villages.

101

Research into Rural Housing Affordability Final Report

An estimated 15,000 homes will need to be built each year (probably in larger towns and cities) just to ‘replace’ those rural properties in villages that will, in the future, be second homes. Although second homes may not be a primary driver of the rural housing problem (although they are a component of the migration driver and as such, may have a local impact on prices – see earlier discussion), they often compound the difficulties faced by some rural households who have a need to live in a particular location but are unable to do so at a reasonable cost. As Taylor (2008) has recently argued, there is a strong case for allocating more land for housing, through the planning system, in locations where overall demand is high. And in order to address the issue of social balance and sustainability some of this housing will need to be allocated in locations – including small villages – previously redlined as inherently unsustainable locations for growth. This seems to be the nub of the rural housing debate as it currently exists. Taylor (2008) has followed in the footsteps of countless previous researchers and commentators who have called for a different approach to development planning in the countryside. But this different approach will require a vastly improved evidence base. There will need to be a better understanding of the link between housing affordability / access and the nature of sustainable rural communities, built on retained social capacity flowing through strong community networks. There will also need to be a clearer appreciation of the factors driving rural housing demand and increasing property prices: a better understanding of market process will help ensure that planning is ahead of the game, making appropriate development decisions. And finally, the debate over local priority needs to be resolved. If this priority is to be more firmly linked to the economic wellbeing of rural areas and communities, then this needs to be clearly stated in planning policy (for example), with a new emphasis on supporting the jobs-homes link necessary to nurture new economic activities in the countryside. On the other hand, it may be that planning wants to distance itself from the idea of an automatic ‘ancestral’ right to live a particular location. This may provide local politicians with an easy route to electoral support, but it may not be justifiable on wider social equity grounds. That said, such ancestral rights seem anchored in Tonnies’ (1957) understanding of rural social relationships, with tightly-knit communities wanting exclusionary protection from external influence. The fact that local rights debate is either non-existent or far more diluted in urban areas (even in Pahl’s ‘urban villages’) is perhaps evidence of sustained contrasts between rural and urban society, despite the considerable convergence between, and mixing of, the two since World War II. Sections of rural society continue to express a desire for a more restrictive planning approach that extends beyond land-use decisions into a personal arena. Should planning be prioritising personal or family interests as a means of achieving broader social goals (i.e. sustainable rural communities)? This is perhaps an interesting question for future debate.

References
Part 1: Atkinson, R. (2000) Hidden costs of gentrification: Displacement in central London, Journal of housing and the built environment, 15, pp. 307-326. Atkinson, R. (2000) Measuring gentrification and displacement in greater London, Urban Studies, 37(1), pp. 149-165. Australian Conservation Foundation and the Victorian Council of Social Service (2008) Housing Affordability: More than rent and mortgages [www.acfonline.org.au] Bridge, G. (1993) People places and networks. SAUS Publications, University of Bristol. Buller, H. and Hoggart, K. (1994) International Counterurbanisation, Ash gate: Aldershot. Butler, T. (1997) Gentrification and the Middle Classes. Ashford: Ashgate. Butler, T. and Hamnett, C. (1994) Gentrification, class and gender, Environment and Planning D, 12, pp. 477–493.

102

Research into Rural Housing Affordability Final Report

Capstick, M. (1987) Housing Dilemmas in the Lake District, Centre for North West Regional Studies: University of Lancaster. Clark, D. (1984) Rural housing and countryside planning, in Blacksell, M. and Bowler, I., (Editors) Contemporary Issues in Rural Planning, in SW Papers in Geography, pp. 93-104 Cloke, P and Little, J (1990) The rural state? The limits to planning in rural society. Clarendon: London Cloke, P. (1979) Key Settlements in Rural Areas, Methuen: London Cloke, P., Mooney, P.H. and Marsden, T. (Editors) (2006) The Handbook of Rural Studies, Sage Publications Ltd: London Clout, H. (1972) Rural Geography: An Introductory Survey, Pergammon Press: Oxford Cole, I & Goodchild, B (2001) Social Mix and the ‘Balanced Community’ in British housing policy – a tale of two epochs, GeoJournal 51 pp 351–360. Commission for Rural Communities (2007) State of the Countryside 2007 Cheltenham: CRC (Commission for Rural Communities) Darling, E (2005) The city in the country: wilderness gentrification and the rent gap, Environment and Planning A, 37, PP1015-1032 Day, G. (1998) A community of communities? Similarities and differences in Welsh rural community studies, The Economic and Social Review, 29, pp. 233-257. Day, G. et al (1989) Social change, Rural Localities and the State: the Restructuring of Rural Wales, Journal of Rural Studies, 5, pp. 227-244. England, Transactions of the Institute of British Geographers, 22, pp. 210-230. Gallent, N., Tewdwr-Jones, M. and Mace, A. (2002) Second Homes in England’s Rural Areas, Countryside Agency: Cheltenham. Harper, S. (1987) The urban-rural interface in England: a framework of analysis, Transactions of the Institute of British Geographers, 12, pp. 284-302. HM Government (2005) Securing the future delivering UK sustainable development strategy [TSO: London] Hoggart, K. (2007) The diluted working classes of rural England and Wales, Journal of Rural Studies, 23, pp. 305-317. Ilbery, B & Bowler, I (1998) From agricultural productivism to agricultural post-productivism, in B. Ilbery (ed.), The Geography of Rural Change. Harlow: Addison Wesley Longman pp. 57-84 Lewis, G & Maund, D (1976) The urbanisation of the countryside: A framework for analysis, Human Geography, 58(1), pp. 17-27 Liepins, R. (2000) Exploring rurality through community: discourses, practices and spaces shaping Australian and New Zealand rural communities, Journal of Rural Studies 16, pp.325-341. Lowe, P. et al (1993) Regulating the New Rural Space: the Uneven Development of Land, Journal of Rural Studies, 9, pp 205-222 Macgregor, B (1976) Village life: facts and Myths, in Town and Country Planning 44, 11, pp. 524-27 Mitchell, C. (2004) Making sense of counterurbanisation, Journal of Rural Studies, 20, pp. 15-34.

103

Research into Rural Housing Affordability Final Report

Moseley, M. (1979) Accessibility: the Rural Challenge, Methuen: London Moseley, M. (2000) England’s village services in the 1990s: entrepreneurialism, community involvement and the state, in Town Planning Review, 74, 1, pp. 415-433 Newby, H. (1979) Urbanisation and the rural class structure, The British Journal of Sociology, 30(4), pp. 475-499. Pahl, R (1970) Whose city? And other essays on sociology and planning. Longman: London Phillips, M. (1998) Investigations into the British rural middle classes – Part 2: Fragmentation, identity, morality and contestation, Journal of Rural Studies, 14(4), pp. 427–433. Phillips, M. (2007) Changing class complexions on and in the British Countryside, Journal of Rural Studies, 23, pp. 283-304. Schmied, D. (2002) What Price Peace and Quiet? Rural Gentrification and Affordable Housing, Bayreuther Geowissenschaftliche Arbeiten: Bayreuth. Shucksmith, M (1990) House building in Britain’s countryside. Routledge: London Shucksmith, M. (1981) No Homes for Locals? Farnborough: Gower Publishing Smith, D and Phillips, D (2001) Socio-cultural representations of greenfield Pennine rurality, Journal of rural studies, 17, pp. 457-469 Soja, E (1996) Thirdspace: Journeys to Los Angeles and Other Real and Imagined Places, Cambridge: Blackbell Spencer, D. (1997) Counterurbanisation and rural depopulation revisited: Landowners, planners and the rural development process, Journal of Rural Studies, 13(1), pp. 75-92. Whitehead, C et al. (2008) Measuring Housing Affordability: A Review of Data Sources [Cambridge Centre for Housing and Planning Research] Woods, M. (1998) Advocating rurality? The repositioning of rural local government, Journal of Rural Studies, 14, pp. 13-26 Woods, M. (2005) Rural geography: Processes, responses and experience in rural restructuring. Sage: London Part 2: Bourdieu, P (1985) The social space and the genesis of groups. Social Science Information 24 (2), pp. 195-220. Champion et al. (1998) The determinants of migration flows in England: a review of existing data and evidence [DOE: London] Cloke, P, Milbourne, P and Thomas, C. (1997) Living lives in different ways? Deprivation, marginalization and changing lifestyles in rural England, Transactions of the Institute of British Geographers, 22, pp. 210-230. Commission for Rural Communities (2007) State of the Countryside 2007 Cheltenham CRC Day, G. et al (1989) Social change, Rural Localities and the State: the Restructuring of Rural Wales, Journal of Rural Studies, 5, pp. 227-244. Fielding, A. (1982) Counterurbanisation in western Europe, Progress in Planning 17, pp. 1–52

104

Research into Rural Housing Affordability Final Report

Garrod, G. & Willis, K. (1992). Valuing the goods characteristics – an application of the hedonic price method to environmental attributes, Journal of Environmental Management, vol. 34, no. 1, pp. 59-76. Halfacree, K (1994) The importance of "the rural" in the constitution of counterurbanisation: evidence from England in the 1980s, Sociolgica Rurali. 34 pp164-89 Harper, S. (1987) The urban-rural interface in England: a framework of analysis, Transactions of the Institute of British Geographers, 12, pp. 284-302. Hickman et al. (2007) Understanding housing demand: Learning from rising markets in Yorkshire and the Humber [Joseph Rountree Foundation] Hoggart, K. (2007) The diluted working classes of rural England and Wales, Journal of Rural Studies, 23, pp. 305-317. Janssen, C. B. & Soderberg, J. Z. (2001). Robust estimation of hedonic models of price and income for investment property. Journal of Property Investment & Finance. Vol. 19, No. 4, pp. 342-360. Lewis, G & Maund, D (1976) The urbanisation of the countryside: A framework for analysis, Human Geography, 58(1), pp. 17-27 Liepins, R. (2000) Exploring rurality through community: discourses, practices and spaces shaping Australian and New Zealand rural communities, Journal of Rural Studies 16, pp.325-341. Lowe, P. et al (1993) Regulating the New Rural Space: the Uneven Development of Land, Journal of Rural Studies, 9, pp 205-222 Mitchell, C. (2004) Making sense of counterurbanisation, Journal of Rural Studies, 20, pp. 15-34. Newby, H. (1979) Urbanisation and the rural class structure, The British Journal of Sociology, 30(4), pp. 475-499. Pahl, R (1970) Whose city? And other essays on sociology and planning. Longman: London Phillips, M. (1998) Investigations into the British rural middle classes – Part 2: Fragmentation, identity, morality and contestation, Journal of Rural Studies, 14(4), pp. 427–433. Phillips, M. (2007) Changing class complexions on and in the British Countryside, Journal of Rural Studies, 23, pp. 283-304. Shucksmith, M (1990) House building in Britain’s countryside. Routledge: London Soja, E (1996) Thirdspace: Journeys to Los Angeles and Other Real and Imagined Places, Cambridge: Blackbell Spencer, D. (1997) Counterurbanisation and rural depopulation revisited: Landowners, planners and the rural development process, Journal of Rural Studies, 13(1), pp. 75-92. Tonnies, F (1957) Community and Society Harper & Row: New York Woods, M. (2005) Rural geography: Processes, responses and experience in rural restructuring. Sage: London Part 3: Albrow, M., Eade, J., Durschmidt, J. and N. Washbourne, eds. (1997) The Impact of Globalization on Sociological Concepts: Community, Culture and Milieu, in J. Eade (ed) Living the Global City: Globalization as Local Process [London: Routledge]

105

Research into Rural Housing Affordability Final Report

Bell, C & Newby H (1976) Community studies: An introduction to the sociology of local community [Allen and Unwin: London] Bridge, G (2002) The neighbourhood and social networks [CRN Paper 4: ESRC] Burnett, KA (2008) Local Heroics: Reflecting on Incomers and Local Rural Development Discourses in Scotland, Sociologia Ruralis 38 Champion, A.G. (2000), ‘Flight from the Cities?’ in Bate, R., Best, R. & Holmans, A. (Editors) On the Move: The Housing Consequences of Migration, York: York Publishing Services, pp.10-19 Delanty, G (2003) Community [Routledge: London] Falk, I & Kirkpatrick, S (2000) What is social capital: A study of interaction in a rural community, Sociologia Ruralis 40(1) pp 87-110 Fukuyama (1995) Trust: The Social Values and the Creation of Prosperity. [New York: Free Press] Gallent et al. (2008) Introduction to rural planning [Routledge, London] Hoggart, K. and Henderson, S. (2005) Excluding exceptions: housing non-affordability and the oppression of environmental sustainability? In the Journal of Rural Studies, 21, pp. 181-196 Pretty J, Ward H (2001) Social Capital and the Environment World Development 29 (2), 209-227 Putnam R. (1995) Bowling alone: America’s declining social capital. Journal of Democracy 6(1), 65-78 Putnam, R (2000) Bowling Alone: The collapse and revival of American community [Simon and Shuster: New York] Shucksmith, M (1990) House building in Britain’s countryside. Routledge: London Stockdale, A (2002) Out-migration from rural Scotland: The importance of family and social networks, Sociologia Ruralis 42(1) pp 41-64 Wellman B (2001) Physical place and cyberspace: the rise of personalised networking, International Journal of Urban and Regional Research 25 p227-252 Wellman B (2002) Little boxes, glocalisation and networked individualism, in Tanabe, M Van den Basselaar, P and Ishina, T (eds.): Digital Cities II: Computational and Sociological Approaches [Berlin: Springer] Wenger, CG (2001) Intergenerational relationships in rural areas, Age and Society 21 pp 537-545 Woods, M. (2005) Rural geography: Processes, responses and experience in rural restructuring. Sage: London

106

Research into Rural Housing Affordability Final Report

Appendix B: Shortlisting areas for consultation
Introduction
In this appendix, we set out the methodology and analysis undertaken to produce a shortlist of villages for consultation for the following rural districts previously agreed with the client: South Shropshire: Deep rural North Norfolk: Retirement retreat Mid Sussex: Dynamic commuter Selby: Transient rural For each of these four districts, a shortlist of three to four parish areas which best reflects the respective rural classification is proposed.

Methodology
A range of population, household and economic indicators were determined that would best reveal the particular characteristics of villages relevant to their rural classification. The 2001 population census was then selected as the most comprehensive source to provide data for this profiling. In addition, migration data and travel to work patterns from the census were also collected for analysis. However, whilst travel to work by mode was available at parish level, migration data and the origin and destination of resident workers were only available at ward level. Therefore a certain proportion of this analysis was observed according to a parish’s respective ward. Finally, average house prices by property type were compiled for each of the four districts based on 2007 sales data from rightmove.co.uk. Average house prices for each rural village/parish area are again based on prices at ward level given the availability of data from our source. A ‘village’ was estimated to be a settlement that has between 500 and 2,500 households and thus a population ranging between 1,000 and 5,000 residents. Our assessment of rural villages has therefore been based on parish council boundaries; this is due to the following factors: Parish Council data is the lowest geographic level available in the 2001 population census (with the exception of lower-layer Super Output Areas) allowing a full comparison between districts as well as villages/parishes themselves; 2. Parish Councils tend to have a clerk or administrator as a central point of contact for the local community, as well as active local groups and societies, many with their own websites - ideal for arranging consultation events; 3. Parishes tend to comprise settlements of population size required for our study. For each district, a list of all parish councils by their population size and the number of households was compiled. This list was then reduced down to include only those parishes which had between 1,000 and 5,000 residents and then furthermore according to whether the parish council was currently active. From this sample, three to four active parishes representing various settlement sizes within our range were then selected to be profiled. Finally the data collected for each village profile was assessed to ensure that it accurately reflected the rural classification of its respective district, producing a shortlist of three to four possible consultation areas for each of the four districts. 1.

107

Research into Rural Housing Affordability Final Report

Whilst the 2001 population census is a comprehensive data source for profiling areas, particularly at a low geographic level, this information is of course somewhat outdated. It should therefore be kept in mind that the census data and thus our final profiles, cannot fully reflect changes in economic and housing trends that districts may have experienced in recent years, such as a shift towards affordable housing or changing demography.

Proposed consultation areas
The proposed consultation areas for each of the four districts are shown in the table below: Table A 1: Shortlist of proposed consultation areas (parishes) for each district North Norfolk Happisburgh Mundesley Holt Mid Sussex Horsted Keynes Turners Hill Slaugham Cuckfield Selby Riccall Escrick Wistow

South Shropshire Bishop's Castle Craven Arms Church Stretton

Table A 2 provides an overview of the main indicators included in our compilation and analysis of villages’ socio-economic profiles for all four districts. This table is by no means exhaustive, but these specific indicators help to reveal the key rural characteristics of the villages in our shortlist, allowing us to benchmark these against the respective district and regional levels, as well as highlight the key differences between districts themselves. In the remainder of this section a brief description of the short-listed consultation areas is provided for each district in a summary table, followed by a brief summary analysis indicating which of the villages most strongly reflects the respective rural classification. This analysis builds on the key indicators presented in the previous table, as well as others which help to specifically highlight the type of rural district being discussed.

108

Research into Rural Housing Affordability Final Report

Table A 2:

Socio-economic profile for each case study district
Mid Sussex Indicator

Horsted Keynes 19% 27% 12% 66% 32% 7%

Turners Hill 20% 33% 9% 61% 37% 7%

Slaugham

Cuckfield

Mid Sussex

West Sussex

South East

England

People of Retirement Age (65+) Total pensioner households Social rented housing Economically active Economically inactive Inflow of Residents (% of total households lived elsewhere 1 year ago)

19% 22% 13% 70% 28% 9%

17% 25% 12% 68% 31% 9%

17% 24% 10% 73% 27% -

20% 29% 13% 69% 31% -

16% 24% 14% 70% 30% -

16% 15% 19% 67% 33% -

Average House Prices: Detached % difference from district average Semi-Detached % difference from district average Terraced % difference from district average Flat % difference from district average

£541,024 24% £331,514 19% £234,592 1% £206,816 12%

£462,577 6% £256,286 -8% £274,203 19% £194,255 6%

£485,158 11% £279,958 0% £233,117 1% £199,515 8%

£529,507 21% £324,055 16% £263,979 14% £176,842 -4%

£437,469

-

-

-

£278,660

-

-

-

£231,129

-

-

-

£184,072

-

-

-

109

Research into Rural Housing Affordability Final Report

South Shropshire Indicator

Bishop's Castle 22% 29% 3% 19% 63% 37% 6%

Craven Arms 21% 30% 1% 23% 65% 35% 6%

Church Stretton 36% 46% 2% 9% 58% 42% 7%

South Shropshire 22% 17% 3% 12% 65% 35% -

Shropshire

West Midlands

England

People of Retirement Age (65+) Total pensioner households Second residence/holiday accommodation Social rented housing Economically active Economically inactive Inflow of Residents (% of total households lived elsewhere 1 year ago)

18% 15% 1% 14% 69% 31% -

16% 15% 0% 21% 66% 34% -

16% 15% 1% 19% 67% 33% -

Average House Prices: Detached % difference from district average Semi-Detached % difference from district average Terraced % difference from district average Flat

£306,140 -4% £193,800 4% £176,923 -3% -

£295,048 -7% £178,814 -4% £140,395 -23% £100,650

£321,006 1% £193,355 4% £166,950 -8% £135,500

£317,264

-

-

-

£186,103

-

-

-

£182,076

-

-

-

£154,928

-

-

-

110

Research into Rural Housing Affordability Final Report

% difference from district average North Norfolk Indicator People of Retirement Age (65+) Total pensioner households Social rented housing Economically active Economically inactive Inflow of Residents (% of total households lived elsewhere 1 year ago)

-

-35%

-13%

Happisburgh 23% 32% 7% 57% 42% 9%

Mundesley 32% 42% 16% 51% 48% 9%

Holt 37% 51% 18% 46% 52% 7%

North Norfolk 25% 34% 12% 61% 39% -

Norfolk 20% 28% 15% 66% 34% -

East of England 16% 24% 14% 69% 31% -

England 16% 15% 19% 67% 33% -

Average House Prices: Detached % difference from district average Semi-Detached % difference from district average Terraced % difference from district average Flat % difference from district average

£217,006 -18% £167,821 -10% £157,832 -1% £150,500 4%

£237,239 -11% £179,806 -3% £139,485 -12% £81,338 -44%

£299,637 13% £203,311 9% £167,187 5% £230,428 59%

£266,224

-

-

-

£186,023

-

-

-

£158,665

-

-

-

£145,260

-

-

-

111

Research into Rural Housing Affordability Final Report

Selby Indicator

Riccall

Escrick

Wistow

Selby

North Yorkshire 18% 26% 12% 69% 31% -

Yorkshire & Humber

England

People of Retirement Age (65+) Total pensioner households Social rented housing Economically active Economically inactive Inflow of Residents (% of total households lived elsewhere 1 year ago) Inflow & Outflow of Residents (% of total households lived elsewhere or moved out of area 1 year ago) Average House Prices: Detached % difference from district average Semi-Detached % difference from district average Terraced % difference from district average

13% 19% 10% 72% 27% 7%

16% 27% 3% 54% 44% 7%

15% 16% 6% 75% 22% 8%

15% 22% 13% 71% 29% -

16% 24% 21% 65% 35% -

16% 15% 19% 67% 33% -

7%

7%

6%

-

-

-

-

£320,916 14% £183,685 16% £153,546 3%

£320,916 14% £183,685 16% £153,546 3%

£320,916 14% £187,990 19% £181,924 22%

£280,752

-

-

-

£158,364

-

-

-

£149,009

-

-

-

112

Research into Rural Housing Affordability Final Report

Flat % difference from district average

£148,700 19%

£148,700 19%

£132,268 6%

£125,345

-

-

-

‘-‘ Indicates that this data was missing/unavailable from the relevant data source

113

Research into Rural Housing Affordability Final Report

Table A 3:

South Shropshire Shortlist of Villages for Consultation
Population 1,630 H/Holds 704 Description One of the smaller market towns, the town has two banks, a post office and 3 grocery stores, cafes, 6 pubs, 2 breweries. The parish council’s website is run from Enterprise House, an important centre for the town, which provides a drop-in IT Centre, the Village Outreach Project and a number of local regeneration projects, library, outpost of South Shropshire Housing Association, Dial-aRide and several private offices. Most recent addition to the District's market towns, area is famous for the home of Stokesay Castle and the new visitor attraction Shropshire Hills Discovery Centre. The parish area includes the market town of Church Stretton and the villages of All Stretton, Little Stretton, Minton and Hamperley and is set within the Shropshire Hills AONB and considered the holidaying heart of the Shropshire Hills. The area has four public/village halls and a large number of charitable/ organisations. The Mayfair Community Centre is for the communities of the Strettons and the surrounding villages and countryside. Website(s) http://www.bishop scastle.co.uk/cou ncil/index.htm

Parish Bishop's Castle

Craven Arms

2,289

940

http://www.craven armstowncouncil. org.uk/

Church Stretton

4,186

1934

http://www.stretto nparish.org.uk/ http://www.mayfai rcentre.org.uk/ http://www.church stretton.co.uk/

For the assessment of South Shropshire villages, it was necessary to draw attention to indicators that best reflected the second home and retirement pressures experienced by the district. Our assessment of age structure for each village revealed that a high proportion of the resident population in Church Stretton were aged 65 and over, at 36%, whilst the number of retirees in other villages was fairly modest. This is reflected in the large number of pensioner households in Church Stretton, which account for as much as 46% of total households in the parish area. This trend is further supported by the proportion of economically inactive residents across all of the villages assessed. Much like the district itself, the proportion of people aged 16-74 both inactive and retired exceeds regional and national averages, particularly so in Church Stretton in which 42% of residents are inactive and 2/3 of this figure is attributable to the number of retirees. As deeply rural areas, all of the villages in our shortlist face second home pressures in the housing market compared with regional and national levels. The village of Bishop’s Castle has the highest proportion of dwellings registered as a second residence or holiday home, followed closely by Church

114

Research into Rural Housing Affordability Final Report

Stretton. Our analysis of commuting patterns revealed that neighbouring Telford and Wrekin and Herefordshire were the most popular workplace destinations for residents of the villages assessed. The data showed that workers from the villages of Bishop’s Castle and Church Stretton commute mostly to Telford and Wrekin, whilst residents in Craven Arms tend to travel to workplaces in Herefordshire. When assessing house price data for the parish areas, it emerged that average house prices tend to be lower than or only marginally greater than the average for South Shropshire as a whole. Moreover prices in the village of Craven Arms fall below the district average across all property types. Terraced properties and flats across all parish areas tend to be much lower than the district average, particularly in the case of Craven Arms village. Table A 4:
Parish Happisburgh

North Norfolk Shortlist of Villages for Consultation
Population 1,372 H/Holds 607 Description The area is based around two buildings; the Lighthouse and land-mark of St Mary's Church. The coastal village suffers from severe erosion and houses that used to be over 20 feet from the sea now sit at the edge of a cliff and increasingly at risk. One of the district's 3 largest villages (after the seven main towns) and is a traditional seaside resort. The area is a mixture of rich farming land and a popular seaside resort, well served by local shops in and around its small high street. The area has a medical centre, library, post office, three pubs and various B&Bs. One of 7 main market towns/settlements of the district. A small Georgian market town with a range of specialist gift shops and galleries and a Country Park. Home to Gresham’s independent school and provides a significant level of employment opportunity in the Central North Norfolk area. Holt is considered to have a complementary role to other larger service centres in the district providing retail, employment and so on. Website(s) http://www.happisb urgh.org/compone nt/option,com_front page/Itemid,1/

Mundesley

2,695

1,266

http://www.mundes ley.org/

Holt

3550

1668

http://www.holttow ncouncil.org/

For villages within North Norfolk, indicators that conveyed retirement pressures typically experienced by the region were observed. Of the three parishes, Holt has the highest proportion of people in retirement age at 37%, as well as the highest mean and median age of population at 50 and 55 respectively. Mundesley also experiences a similar trend across each of these indicators.

115

Research into Rural Housing Affordability Final Report

When assessing household characteristics, Holt has the highest proportion of pensioner households accounting for over 50% of total households, in which 29% are lone pensioners and 22% are ‘all pensioner’ households. There is a similar scenario in Mundesley, in which a fifth of households with more than one person are pensioner households. In terms of economic activity, 52% of all people aged 16-74 in Holt are economically inactive, again reflecting the extremely high proportion of retired residents in the parish area. In Mundesley this level of inactivity is also high at 48%. A high proportion of residents were also recorded as having very low levels of formal education in all three parish areas. In Happisburgh as much as 69% of the population aged 16-74 had either no qualifications or at most held the basic 1-5 GCSEs of any grade (or equivalent). In Mundesley and Holt a similar case exists, with 61% and 59% of people respectively. With respect to the housing market, average house prices at district level are higher across all property types compared with the villages of Mundesley and Happisburgh, whereas properties in Holt are more expensive than homes in North Norfolk. Table A 5: Mid Sussex Shortlist of Villages for Consultation
Population 1,507 H/Holds 594 Description Very small village relying on larger neighbouring villages and towns for most services including health, schools and so on, however, does have a local village store and post office. The parish is located in three distinct areas, the Village, Worth Abbey (run by the Benedictine Order as a monastery and public school) and Turners Hill Park, a mobile home site which provides over 200 homes. In the centre is the village green which, together with the shops and Crown Public House forms the focal point. The village has four pubs, a various sports and recreational grounds and a new community centre The Ark. The Ark provides for various sporting activities, numerous village clubs, health services, village and social events. Parish area consists of four villages, in order of size: Handcross, Pease Pottage, Warninglid and Slaugham each with its own distinct character. Handcross High Street classed as the hub of the parish area with a good mix of shops and some light industrial units, two schools, two churches and three pubs. Also Handcross Parish Hall with a social club - the largest community facility in the parish area. In contrast Slaugham village Website(s) http://www.horsted keynes.com/

Parish Horsted Keynes

Turners Hill

1,849

714

Slaugham

2,226

951

www.slaughampc. org.uk

116

Research into Rural Housing Affordability Final Report

is the smallest of the four with just one main street and has changed little over time.

Cuckfield (not inc. Cuckfield rural-two separate Parish Councils)

3,266

1,342

Area is well served by local facilities from High Street and Whitemans Green. Village centre contains mixture of uses including shops, small offices, restaurants, pubs and residential units. Many community and recreational facilities, with the Queen's Hall at the centre of village life and host to many activities (also other halls for hire, a library, museum, meeting rooms and recreation ground, allotments and playing fields at Whitemans Green). Also doctors' surgery and pharmacy, two post offices and local bus services. The small hamlet of Brook Street lies close to the northern edge of Cuckfield. Other main facilities are linked with/relied upon through Haywards Heath.

www.cuckfield.gov. uk http://www.cuckfiel d.org/

For Mid Sussex villages, particular attention was paid to the economic activity of working age residents and heir travel to work patterns in order to highlight the high levels of outward commuting experienced by much of the district. The proportion of residents of retirement age in all four parishes assessed is relatively low, measuring on average at around 19% of the population and is lowest in Cuckfield with only 17% of people aged 65 and over. In contrast the proportion of people falling within the age category of 25 and 44 is fairly high in the villages of Slaugham and Cuckfield, with 30% and 28% of total residents respectively. Furthermore the mean and median age for all four parishes tends to be around 40 years of age, with residents recorded to be younger on average in Cuckfield. Reinforcing this is the low proportion of pensioner households, particularly so in the case of Slaugham with only 21%, followed by Cuckfield with 25% of total households. The number of people (aged 16-74) employed in each of the parish areas is generally equal to or in most cases above the rate experienced by the south east as a whole. Employment is particularly high in Slaugham and Cuckfield, measuring at 70% and 68% respectively compared with the south east average of 65%. In addition, a high proportion of working age people in these two villages have completed a form of higher education, with as much as 44% of people in Cuckfield having attained qualifications at this level.

117

Research into Rural Housing Affordability Final Report

The occupational groups of people aged 16-74 across all four villages are also at a high professional or managerial level. In Cuckfield 44% of people are in such jobs, whilst in Slaugham and Horsted Keynes rates are similarly high with around 35% of people in each village. The migration patterns of residents in Mid Sussex at the time of the census were also analysed based on ward level data. Of the villages assessed, at the time of the 2001 population census Cuckfield and Slaugham villages experienced some of the highest inflows of people compared with other areas of the district, whilst Turners Hill and Horsted Keynes experienced moderate inward migration. Our analysis of commuting patterns revealed that aside from trips within Mid Sussex itself, a significant proportion of residents from all four villages travel to work in Crawley, but particularly so for residents in Turners Hill and Slaugham. The data also revealed significant outward commuter flows to workplaces in London, with the greatest proportion of trips made to the capital by residents from the village of Cuckfield, followed by Horsted Keynes. Average house prices in all four parish areas tend to far exceed the district average. This is particularly the case for detached homes, where a property of this type in Horsted Keynes or Cuckfield for example, is around £100,000 more expensive than the average in Mid Sussex. Table A 6:
Parish Riccall

Selby Shortlist of Villages for Consultation
Population 2,317 H/Holds 922 Description Well served by local services and amenities including a school, doctors' surgery, sports clubs and two village halls, with one of them being the very new/modern 'Regen Centre' which caters for business, education, sport, leisure and community uses The village has a good range of local services and amenities, which include a post office, Car Garage and village store, as well as a number of other local private businesses, as well as a social club, restaurant, B&Bs and so on, in addition to recreational facilities and social groups/clubs. There is also a recent project funded by a lottery grant to redevelop the village community building which also serves nearby village of Deighton An increasingly busy village. The main B1223 receives a significant amount of through traffic. Over the last few years Wistow has lost several services, but does have recreational/village facilities, as well as a pub, however, there are no village stores/shops Website(s) http://www.riccallpc.freeonline.co.uk/

Escrick

1,241

358

http://www.escrick.org/

Wistow

1,135

438

http://www.wistowvillage.com/

118

Research into Rural Housing Affordability Final Report

The assessment of villages in Selby was particularly focussed upon trends in migration and commuting patterns to highlight the movement and change in its resident population and links with neighbouring districts that are typical of the area’s rural classification. The proportion of residents of retirement age is consistently low across the parishes assessed, meaning that the proportion of pensioner households is small. In contrast, levels of employment are particularly high in the villages of Riccall and Wistow and exceed regional and national averages measuring at an average rate of 72%. Of those employed, ward level data revealed that aside from working in the Selby district, workers in the villages of Riccall and Escrick commute to workplaces in York, whilst residents in Wistow village tend to travel to workplaces in Leeds. Reflecting the transient nature of Selby, the 2001 population census recorded that the villages of Riccall and Escrick both fell within wards that experienced some of the highest levels of inward and outward migration of UK residents compared to other areas of the district. Wistow experienced one of the highest inflows of residents compared with other areas in Selby; however, in all cases the actual level of migration as a proportion of the total resident population was fairly modest. Unlike the other district case studies, house prices in all of the villages assessed within Selby tend to be much more expensive than the district average. Detached homes show the largest disparity, with properties in the villages of Riccall, Escrick and Wistow around £40,000 higher than the average price in Selby as a whole. A similar case exists for semi-detached properties, whilst the average price of a terraced property or flat tends to fluctuate between villages and thus with the district average.

Recommendations
Based on our profiling and analysis of population, household and economic indicators we recommend a first and second option village for each district, which we believe best reflects the relevant rural typology and are most suitable for consultation. From the short-listed villages in South Shropshire, we suggest that Church Stretton be considered the primary area for consultation. From our analysis, this village emerged most frequently in demonstrating typical ‘deep rural’ characteristics and boasts a very active community. Church Stretton also encompasses four villages, meaning there is broad scope for considering the location and participants for the consultation process. We would regard the village of Bishop’s Castle as a secondary choice for selection. This area is one of the smaller market towns within South Shropshire but displays similar characteristics to Church Stretton and has an active parish council closely tied to the local community. For North Norfolk, we recommend the village of Holt as the most suitable area for consultation, followed by the village of Mundesley. Both villages are important settlements within the district and are distinct in character; Holt as one of the key market towns in North Norfolk and Mundesley as a traditional seaside resort. Holt in particular is considered to have a key role in complementing the larger service centres in North Norfolk, mainly in terms of employment and retail and best reflects the typical age structure associated with a ‘retirement retreat’. Holt has a high proportion of pensioner households and an ageing population that is on the whole fairly inactive. The same scenario exists in Mundesley across these indicators, but to a lesser extent. In the case of Mid Sussex, we consider that the village of Cuckfield best reflects the ‘dynamic commuter’ classification owed to the area, followed by the village of Slaugham. When assessing characteristics of the working age population and economic activity, both villages represent areas with an educated and relatively ‘young’ population. The villages also experience fairly high inward migration and outward commuting of its residents to neighbouring employment centres. However, Cuckfield in particular has a much higher proportion of its working age population in higher paid

119

Research into Rural Housing Affordability Final Report

occupations and experiences higher outward commuting to London compared with Slaugham. Both villages have active parish councils closely tied with their local communities. From our shortlist of villages in Selby, we propose that Riccall is the most appropriate area for consultation, followed by the village of Wistow. From our profiling and analysis, both villages represent the transient nature typical of rural areas in Selby and both appear to have very active communities. The main factors to separate these two villages are settlement size and commuting patterns. Riccall is a much larger village and lying to the south of York, experiences a significant proportion of outward commuting to this area. Wistow is similar in age structure and like Riccall, has a high proportion of residents in employment, however, the parish area is fairly small and workers tend to commute to Leeds.

Going forward
Following our recommendations of parishes for consultation detailed above, a finalised number of parishes will need to be confirmed with the client team. Upon agreement of these parishes, the project team will look to formally contact parish clerks to discuss stakeholder and resident involvement in the study. Whilst an initial criterion for profiling these was an active parish council, it is recognised that there is still the risk that some parishes may not wish to participate or be available for the time of the study. In the event of this situation, we suggest one of the following actions to be taken to identify alternative consultation areas: Return to an alternative parish originally profiled, as set out in table 3.1 previously. (e.g. for North Norfolk, in the even that either Holt or Mundesley are deemed unsuitable, we look to contact Happisburgh parish council instead). Often within the settlement boundaries of parish councils, smaller villages exist that will be similar in their socio-economic characteristics (but were not able to be profiled given census data limitations). It is possible that parish clerks will know alternative community representatives in these villages that can be contacted to participate in the consultation. The profiling process revealed that many neighbouring parishes share very similar socio-economic characteristics. Where this is the case, a nearby parish will look to be contacted to participate in the consultation.

120

Research into Rural Housing Affordability Final Report

Appendix C: Household survey
Introduction
This appendix sets out the methodology and findings of the door-to-door household surveys that were conducted for the four district case studies agreed for the research: 1. South Shropshire: Deep rural 2. North Norfolk: Retirement retreat 3. Mid Sussex: Dynamic commuter 4. Selby: Transient rural Undertaking these surveys looked to validate the findings of previous stages of the research, particularly in terms of the views expressed in the consultation meetings held with parishes.

Village survey locations
The 8 villages that were selected for each of the four case study districts are listed below. These villages generally correspond with the locations selected for previous stages of consultation in order to align the results of the surveys with findings from the remainder of the research and help to reveal key underlying trends. 1. 2. 3. 4. Selby: Riccall, Cawood and possibly Wistow South Shropshire: Bishops Castle and Hope Bowdler/ Ticklerton/ Acton Scott/ Hatton Mid Sussex: Cuckfield and Slaugham North Norfolk: Happisburgh/ Mundesley and Wiveton/ Blakeney

Timescale and sampling approach
The surveys were conducted over a period of 2-3 days in each village. If there was no answer at the first attempt the interviewer returned later in the day, most commonly in the early evening to gain a response. If unsuccessful upon a second attempt, the survey was dropped through the household’s letterbox with a prepaid envelope. The surveys were required to be a random sample of a minimum 100 households in each parish, consisting of either a single village or a group of small villages. Whilst it was hoped that the selected villages would produce a sufficient level of responses, there were circumstances in which it was necessary to look beyond the immediate village to be certain of attaining a good sample size. A random sample of households in each street was taken, by approaching every third home. This sample was then scaled to the demographic structure of the village as given in the 2001 population census as detailed in the table below. It was important to ensure that a good response rate was achieved in each age category, thus there were again circumstances in which an additional round of ‘door knocking’ was required.

121

Research into Rural Housing Affordability Final Report

Table A 7:

2001 population census Age Profile by Village Cuckfield 3,266 1,342 7% 15% 8% 28% 25% 8% 9% 100% Hope Bowdler 515 204 5% 14% 5% 25% 33% 13% 6% 100% Bishops Castle 1,630 704 4% 13% 9% 24% 28% 11% 11% 100% Happisburgh 1,372 607 4% 11% 7% 22% 33% 14% 9% 100% Wiveton/ Blakeney 947 481 2% 9% 6% 15% 29% 18% 21% 100% Riccall 2,317 922 6% 14% 10% 32% 26% 7% 6% 100% Cawood 1,135 602 6% 13% 8% 28% 31% 9% 6% 100%

Slaugham Population Households Aged 0-4 Aged 5-15 Aged 16-24 Aged 25-44 Aged 45-64 Aged 65-74 Aged 75+ Total % 2,226 951 5% 12% 8% 30% 26% 10% 9% 100%

Survey questionnaire
The questionnaire comprised approximately 35 questions and took approximately 15 minutes to complete. Aside from the ‘about you’ section the majority of questions required a closed-ended response with the interviewee asked to select from a number of pre-prepared responses. A proportion of these questions have a follow-up open-ended question which required further explanation of the interviewee’s initial response. Given the nature of the nature of the research project, the open-ended explanations were very important to the findings of the survey, as they had the ability to uncover subtle viewpoints and insights that could not be fully captured by the closed-ended questions.

122

Research into Rural Housing Affordability Final Report

Appendix D: Estate agent survey
Introduction
This appendix sets out the findings from the consultation undertaken with estate agents in the four rural case study districts: 1. South Shropshire: Deep rural 2. North Norfolk: Retirement retreat 3. Mid Sussex: Dynamic commuter 4. Selby: Transient rural This qualitative analysis looks to inform the forthcoming HPM to be undertaken for this study. At this stage our aim is to reveal the main drivers of rural house prices and how these factors may differ between each of the case study districts. Through consultation with estate agents in the local area it is possible to determine exactly what people value when purchasing a property and as a result the premium they are willing to pay.

Interview approach
The interviews undertaken were based around the following key questions: 1. Which villages tend to be the most popular/sought after and why? 2. Do people pay a premium for these villages? 3. Can a buyer get more for their money if they look at areas closer to/in a town? For the interview process it was also necessary to tailor more specific questions according to the rural classification and nature of the district. For example, in Mid Sussex, estate agents were asked about the influence of commuting and in North Norfolk were asked about how proximity to the coast influenced buyers. It was also ensured that a range of estate agents were consulted that were representative of the different rural areas within a district. For example, in South Shropshire, agents based around each of the main market towns were interviewed and in North Norfolk, estate agents both along the coast and based inland were consulted.

Interview results
The findings from the interviews are presented separately for each district below.

North Norfolk
Conversations with agents in North Norfolk revealed that rural property prices vary significantly according to: Whether the property is in the coastal area of the district or situated inland; and When the property is in a coastal location, whether it is along the west or east coast. There was a strong consensus that villages along the coastal stretch of North Norfolk were highly sought after by potential buyers. This was true for areas up to approximately five miles inland from the coast. As a result, villages in this coastal area were found to be significantly more expensive than villages much further inland. It was also suggested that moving inland; properties become cheaper in the villages nearer to the built up town centres of Fakenham, Norwich and King’s Lynn. On average, agents suggested that properties in coastal villages were priced at least £25k higher than those inland. However, it was also noted that villages further inland but within a short drive to the coast were also

123

Research into Rural Housing Affordability Final Report

very popular and as a result more expensive. Agents treated the coastal region of North Norfolk as a separate sub-market of its own, revealing that the western coast is much more desirable than the eastern coast of the district and this is reflected in house prices between the two areas. The western coast has become extremely popular with buyers of second homes, particularly those moving in from areas outside the district, who have been attracted by the picturesque and traditional character of villages on the west coast. This stretches from villages west of Cromer and along to Sheringham, Blakeney and Wells-Next-theSea. As a result a considerable amount of new, high quality development has taken place along the west coast to cater for second home buyers and has reinforced the upward pressure on house prices in these coastal villages. One agent commented on the quality of this development, a large portion of which has been completed by Norfolk Homes Ltd; a reputable developer of high specification village properties, which has been complementary to the west coast area. Agents also noted that when looking for rural property in North Norfolk, buyers are attracted to villages that have a high concentration of character properties which tend to contribute to the general feel of a ‘traditional’ village. Locations that had a good level of community activity and village amenities were also considered particularly popular. The village of Great Massingham was revealed to be a clear example of this, continuing to maintain higher property prices than other nearby villages. The village is highly sought after given its stock of period properties, vibrant village community and picturesque rural settings. Agents commented however, on the low availability of properties for sale in this type of area; suggesting a low turnover of properties as people tend to move to the village and live there for many years, often beyond retirement age. There also appeared to be a village size threshold which had a further influence on rural property prices. Agents noted that villages with almost no amenities or activity were cheaper relative to villages that did have these facilities. Contrastingly, agents suggested villages that were very large and possibly bordering market town status, were less desirable and as a result tended to be cheaper. For example, the village of Mundesley, despite its coastal location, was frequently suggested as a less expensive alternative to other coastal villages as it was larger in size, with a large number of shops and amenities and as a result considered slightly less of a traditional, attractive village to buyers.

South Shropshire
Consultation with estate agents in South Shropshire revealed there was much less variation in rural properties prices across the district compared with the other case studies. As much of the district is deeply rural, the aesthetic quality of villages tends to be very consistent and the majority of villages are characterised by an ample supply of period properties. One agent suggested that since villages were visually very similar, variation in price tends to be on a property-by-property basis rather than between entire villages. The most significant variation between villages is the level of basic amenities available to local residents, such as a village shop, post office and pub. Many villages have lost local services in recent years and are reliant on the larger villages/market towns of Church Stretton, Bishops Castle, Craven Arms and Ludlow, for such facilities. Despite this factor, agents considered property prices across villages to be fairly similar which suggests the availability of local amenities did not hugely impact on price. This was further reinforced by one agent who highlighted the typical reliance on car use in rural villages and therefore the willingness of people to travel a fair distance to access basic amenities. In contrast agents did note that the majority of villages in the district benefit from a local village hall or church which instead acts as a focal point of the village and is central to community activity. Agents revealed this was a desirable factor for some buyers; although it was not clear whether access to a local village hall/church caused property prices in such villages to be significantly higher than those

124

Research into Rural Housing Affordability Final Report

that did not. The consultation did reveal however, that property in the district’s rural villages was more expensive than property located very close to or in the larger villages/market towns. Whilst the market towns themselves are still fairly remote, agents suggested the villages dispersed around these centres are more desirable and more expensive, once again suggesting a size threshold influencing rural property prices. A key consideration raised by agents was the low turnover of properties in many of the district’s villages and that the availability of property in these areas was much lower relative to that in the market towns.

Selby
Consultation with agents in Selby revealed a significant difference in the price of property located in the district’s villages compared with the town centre. On average agents suggested properties in rural villages were at least £15-25k more expensive than those in Selby town centre. The most desirable villages were those that were characterised by period properties, such as those built in typical Yorkshire stone and that remain ‘unspoilt’ in their appearance as a traditional village. This type of village was considered to be much more expensive than those villages that were larger in size and/or those that had undergone modern development. Smaller villages such as Wistow, which were referred to as having a traditional Yorkshire character, were considered more desirable than larger villages such as Thorpe Willoughby. Thorpe Willoughby, although still possessing a traditional village centre, experienced a great deal of development on the outskirts of the village in the 1960s and 1970s. This has been the case for a number of villages situated just outside of Selby town centre. This development was considered to detract away from the village’s original character making it less popular with potential buyers and therefore cheaper, than the smaller and more desirable traditional villages. Agents further emphasised the appeal to buyers of a small village centre offering basic amenities rather than busier and larger villages that experience a much higher level of activity. This was particularly the case for villages with closer access to commuter towns, in which buyers are attracted to a rural village lifestyle whilst still being in close proximity to a town centre. As a result, agents revealed villages become more and more expensive moving away from Selby town centre and towards the areas of Leeds and York. The issue of availability of was not raised by agents in the Selby area; there did not appear to be a shortage of rural property even in the most sought after villages. Agents highlighted that the greatest factor was variation in the price of property between villages given the differences in their perceived quality, making some areas considerably less affordable than others.

Mid Sussex
Agents in Mid Sussex indicated that the majority of the district’s rural villages are highly sought after as buyers are primarily attracted to having a typical village lifestyle, whilst still being in close proximity to a busy town centre with greater amenities. As a result, agents indicated that property in the district’s villages was more expensive than property located very near to or in the town centres, such as Haywards Heath. All of the agents interviewed suggested the most desirable villages to be Cuckfield, Lindfield, HurstpierPoint and Hassocks. Their popularity was attributed in part to the visual appeal of the areas; reflecting a traditional village setting with character properties and in part to the perceived status of these villages as ‘upmarket’ places to live. Agents considered the village of Lindfield to be most desirable and as a result the most expensive

125

Research into Rural Housing Affordability Final Report

village of the four, which has consistently maintained higher property prices throughout the market cycle. The village is in close proximity to Haywards Heath, but has maintained its status, appearance and size as a traditional village, which has made it very popular to potential buyers wanting to live in a rural location but with access to town centre amenities. Agents indicated that the villages of HurstpierPoint and Hassocks act as ‘twin villages’ in the district, similar in factors such as their location within Mid Sussex and their offer of good primary schools. However, whilst both are highly sought after, HurstpierPoint was consistently regarded as more popular and more expensive. HurstpierPoint is an older village, smaller in size and has retained a typically quaint village character, with basic amenities such as a village shop, post office and pub. Agents suggested that property in Hassocks, despite benefiting from good rail links, is slightly cheaper given that it is larger in size and is characterised by a higher stock of modern housing. As a result it was considered that Hassocks did not enjoy the same traditional village status as HurstpierPoint and Lindfield. It was noted that the more affordable rural villages were those which were much more dispersed outside of central areas of the district. Although attractive and ‘traditional’ in their appearance and quality, these very small villages were considered too remote and lacking a village centre, making them much cheaper. One agent suggested Balcombe as an example of such villages. A key issue also raised by agents was the generally low availability of properties in the most sought after villages, but particularly a shortage in the number of smaller properties in these villages.

Conclusions
Whilst there are some obvious differences between the rural housing markets of the four district case studies, there appears to be some key common factors influencing rural property prices: Aesthetics – (i) availability of character properties (ii) how picturesque a village is crucially whether the village has maintained an original/traditional unspoilt appearance or whether it has undergone modern development, as well as how complementary this development is to the rural environment. 2. Size – there appears to be an optimum size for a rural village that falls between one that is so small and remote that it is lacking any real local activity and one that is so large that the level of amenities and activity detracts away from the village’s rural setting. 3. Status – some villages appear to have maintained higher prices relative to others as they are perceived to be a more desirable location than larger village areas/towns, particularly for districts that possess a mix of rural and urban settlements, it is perceived as ‘fashionable’ to live in a village rather than the town. Another consideration to have emerged is the availability of rural property, particularly in the most sought after villages, where turnover of housing stock is low. Often in the most rural villages people have lived there the majority of their lives or those that move to desirable areas tend to reside there permanently or for many years. 1.

126

Research into Rural Housing Affordability Final Report

Appendix E: Hedonic pricing analysis
Background
Our research is centred on three critical issues: Local affordability and the part that social networks play in building and sustaining rural communities; 2. The extent to which a rural location adds value to residential property (because of perceived qualities or tangible market characteristics); and 3. How the expectation of local priority to a limited rural housing stock is justified and could, in the future, be realised. This technical appendix discusses the application of the HPM, outlining our methodology, assumptions, findings and conclusions, in order to address to what extent a rural location drives residential property prices and essentially evaluate the ‘power of rurality’ as part of the rural housing research project. Hedonic pricing analysis has been undertaken for the housing markets of the four case study districts central to our research, each of which are of a different rural typology, in line with the Lowe and Ward (2007) classification of different rural areas and economies. The case studies are as follows: South Shropshire: Deep rural North Norfolk: Retirement retreat Mid Sussex: Dynamic commuter Selby: Transient rural Our analysis at this stage is derived from statistical modelling and given that there will be certain limitations to the extent to which the power of rurality can be determined from this, our findings will be supported by the subsequent analysis of local property agent interviews. 1.

The HPM
The HPM is a statistical technique used to compare prices of a product when there is observable quality differences in a product sold: in the case of this research a rural house or dwelling. A dwelling can be described generally by the characteristics of its structure, surroundings and location and a dwelling’s price can therefore be said to reflect the value of these attributes to the buyer. Essentially, hedonic pricing allows us to describe house prices as a function of a number of different attributes. Specifying a hedonic price function is possible by undertaking a series of multiple regression analysis. Regression analysis looks to identify how much of the variation in a dependent variable, which in our research is property price, can be explained by variation in a number of explanatory variables (in our case a property’s different property attributes). This hedonic function is represented below:

Y = a + b1X1 + b2X2 + b3X3…biXi
House Price = constant + structural attributes + location attributes… In this research we have incorporated the rural typology of a dwelling into its hedonic price function, which allows us to isolate its impact on house prices and essentially through this impact determine the power of rurality. Whilst many hedonic pricing studies have looked to value the amenity of farm or agricultural land in rural housing markets, to our knowledge there has not been an attempt to estimate the value of a rural

127

Research into Rural Housing Affordability Final Report

location in such a holistic sense. Using Defra’s classifications of rural and urban areas7 we have piloted the HPM in order to test our central hypothesis; houses in a rural location will attract a premium over houses that are located in urban areas. However, findings from the previous stages of this research also lead us to hypothesise that extremely isolated rural properties will not attract such a premium. In this research, we have applied the HPM to four districts of different rural typologies in order to derive an implicit value of living in these areas. Using data on the prices of properties in each district, we have estimated separate hedonic price functions for each rural typology and in turn used these functions to calculate the change in the price of a property that results from its rural location. Our methodology discussed in this chapter is common practice of any hedonic pricing study of housing markets. This chapter therefore looks only to provide a high level overview of our approach in applying the HPM within the context of this research.

Key assumptions
The HPM is an approximation of the housing market observed in reality and therefore, as with any statistical model, is underpinned by a number of key assumptions. This section sets out the assumptions that support our application of the HPM to rural house prices. Defining appropriate housing markets The application of the HPM to the housing market rests on several key assumptions, but above all, homogeneity of the housing product is assumed. It was therefore essential when collecting price data and developing data for quality attributes that only one market was being assessed. For an individual market, it is assumed that house prices follow the same price schedule and thus variation in price can be estimated through a unique hedonic function for that market. There is considerable variation in the geographical area that previous hedonic studies have considered as one market, ranging from an entire country to analysing house price data for a single census area. The assumption of a single market is particularly important for this research given that analysis is focused on four districts each representing different rural typologies. As a result, we have assumed that house prices across the four districts are unlikely to move in the same way, for example, access to the road or rail network in Mid Sussex which is of a ‘dynamic commuter’ typology is likely to have upward pressure on the price of a house, yet this is unlikely to be the case in ‘deep rural’ South Shropshire. In our assumption of four housing markets each with its own price schedule, it was therefore necessary to estimate separate hedonic price functions for each case study district. Nevertheless, whilst we are assuming that house prices in the four districts will be driven differently by a set of quality attributes, this does not compromise our central hypothesis that in any district, a rural location will attract a premium over properties that are located in urban areas. Time and market equilibrium A second assumption of any hedonic pricing model is that the housing market is in perfect equilibrium. In order for the market to be in constant equilibrium, three key conditions must be fulfilled: Households have perfect information; Transaction costs are zero; and The hedonic price schedule adjusts instantaneously to changes in supply or demand conditions in the housing market.

7

Rural Evidence Research Centre (RERC), Birkbeck College, 2004

128

Research into Rural Housing Affordability Final Report

Given the assumption of perfect equilibrium, our analysis of house prices for all four districts has been performed on cross-sectional data, rather than data from a series of years. A risk in using time-series data that we have looked to avoid is that house prices are likely to follow different hedonic price schedules at different periods in time. This would compromise our assumptions of both a single housing market and perfect equilibrium. In using cross-sectional data, our results become much easier to interpret and we avoid the risk of analysing house price data from markets that are temporally distinct.

Defining the dependent and explanatory variables
In housing markets the characteristics or attributes of a dwelling are assumed to be capitalised into the value of a house and its land. The HPM allows us to split out the value of a dwelling to its various attributes and ‘explain’ house price variation when there are observable differences in the attributes of properties being sold. In other words, we are using the HPM to demonstrate that the price of a house is dependent on its different attributes and thus variation in such attributes will explain variation in house prices. In order to specify and estimate a hedonic price function for each district, the dependent and explanatory variables for analysis needed to be identified. In selecting variables for analysis the ability to quantify variables and the availability of data, as well as the scope of the research was considered.

The dependent variable
One of our first concerns in estimating a hedonic price function is the measurement of property price, known as the dependent variable, given that a property’s value is dependent on the quantities or qualities of its characteristics. It is common place for hedonic studies to use house price data based on actual sales as it is a true representation of a buyer’s willingness to pay for the attributes that characterise a property. In addition, house prices determined through sales data is preferable to valuations from property agents that other hedonic studies have employed. Whilst the benefit of agent valuations is that data can be measured at a very local scale, much of the valuation will be opinion based and risks reflecting the preferences of the individual agent rather than those of buyers in the open market. In identifying a suitable source for property sales data, it was necessary to consider the geographic level at which data was available. Ideally we would want to analyse data at a highly localised level, for example, on a street-by-street basis. It was equally important to ensure that, given our decision to analyse cross-sectional house price data, a sufficiently large number of property transactions was available at such a concentrated level. However, above all, it was crucial that our dataset of house prices provided a break down of the size and structure of the properties sold. As this section will go on to explain in the discussion of explanatory variables, the structural attributes of a property, particularly its size, will be a fundamental driver of house prices. For example, it is highly plausible that a five bedroom detached property will achieve a much higher selling price than a one bedroom flat. It was accepted however, that given the scope of this research it would be unlikely for us to obtain detailed sales data incorporating the many structural attributes of a property, such as the building’s age, garden and plot size etc. Similarly, as with other hedonic studies, it would not be possible to determine the internal quality of a property, which in the housing market will undoubtedly be a strong driver of price. In considering all of the above there would become unavoidable trade-offs in the completeness of the data to be used. It was concluded that in selecting a suitable source for sales data the type of property that is, whether detached, terraced etc and its size, in terms of the number of rooms, should be

129

Research into Rural Housing Affordability Final Report

accounted for. Sales data provided by the Land Registry was first considered. House price sales are provided for the different types of property; namely detached, semi-detached, terraced and flats, at postcode level for an entire year. However, the size of each property sold in terms of the number of rooms is not provided. A superior alternative for sales data was then identified through the VOA. The VOA records the structural characteristics of properties sold including the number of bedrooms and bathrooms in each property and whether the property is detached, semi-detached, or terraced, for sales corresponding to postcode level (minus 1 digit). Collecting sales data at this spatial scale would allow us to analyse house prices at a local enough level whilst maintaining a large enough dataset of property transactions. Other hedonic studies that have drawn upon VOA data have used council tax bands to represent the value of properties in place of sales data. An advantage of this approach is that council tax bands closely reflect both the structure of the property and the local neighbourhood. However, since we are looking at areas confined to district settlement boundaries, assigning properties to a limited number of council tax bands would significantly reduce the variation in our dependent variable data and consequently reduce the capabilities of regression analysis. As previously discussed it was acknowledged that much of the structural quality of a property would not be captured in our analysis, but that the size and type of property will be a large determinant of its price. In order to control for this as much as possible, house price data was obtained as an average price per room, by property type, for all sales recorded in 2007. This unit price has been used to represent the dependent variable throughout our application of the HPM for each district. It was then necessary to test that there was enough variation in this average price data. The HPM consists of a process of regression analysis, looking at how variation in house price is dependent on the variation in its characteristics. Therefore for the regression analysis to assess this relationship, it was necessary to ensure that there was significant variation in the sales transactions recorded (represented as unit price of an average price per room). Similarly it was also necessary to ensure that the 2007 sales recorded by the VOA extended across each district and was therefore a good representation of each rural housing market. These tests looked to ensure that our dataset for the dependent variable was as robust as possible. In addition, it was particularly important to run such tests since we are applying the HPM to rural house price data. Rural housing markets are typically inactive as turnover of the housing stock is low. This runs the risk of limiting both the number and location of sales recorded for a given year.

The explanatory variables
Having defined a measure of property price, our next concern in estimating a hedonic price function was the measurement of the many attributes, known as the explanatory variables, which are used to explain the selling price of a house. There are of course a vast number of explanatory variables that could be incorporated into a hedonic price function as the value of a property to a buyer is likely to be dependent on so many attributes. However, it is possible to categorise the attributes of a dwelling as follows: Structural (number of rooms, plot size, garage, garden, age etc); Neighbourhood (e.g. crime rates, employment rate, private home ownership etc); Accessibility (e.g. distance to schools, shops, transport etc); and Surrounding environment (e.g. quality of open space, natural and built environment, noise level etc). For the purposes of this research, we undertook a short desktop review of existing literature in order to 1. 2. 3. 4.

130

Research into Rural Housing Affordability Final Report

develop a set of potential explanatory variables that represented attributes from each of the categories above, whilst taking into account data that was readily available. As with our selection of house price data, the spatial scale at which data was available was very important. We looked to identify data sources for each explanatory variable that in line with our house price data was available at postcode level, enabling us to closely assign specific attributes to each property transaction. A full list of potential explanatory variables was developed based on an a priori expectation that each variable represents an attribute of a property that will in part determine its sale price. Table A 8 details each potential variable identified from the literature review, along with the relevant data source and spatial scale at which data is available. As previously discussed, it was not possible to obtain the detailed structural attributes of each property transaction. However, it has been possible to differentiate the type of property representing each sale transaction and by calculating a unit price measure for our dependent variable, we have controlled for the size of a property in terms of the number of rooms. Neighbourhood variables describe the characteristics of the local area in which the property is located. This data was largely obtained from the 2001 population census, available down to Lower Super Output Area (LSOA) which encompasses on average around 400 households. Each property transaction, using its corresponding postcode, was assigned to an LSOA allowing us to derive census data for each neighbourhood variable. This data describes the neighbourhood in which a property is located in terms of ethnic composition, age composition, employment status and so on, as set out in Table A 8. As an alternative description of the local neighbourhood, data from the Indices of Multiple Deprivation (IMD), produced by the Department for Communities and Local Government (DCLG) at LSOA level for 2007, was considered. IMD data comes in the form of both a composite measure of the economic, social and environmental deprivation of an area and broken down into domains such as education, crime and health deprivation. Both forms were considered when compiling our explanatory data. Deprivation scores at LSOA level were assigned to each property transaction, whereby a higher IMD score indicated a greater level of deprivation. Census data and IMD scores were compiled as mutually exclusive measures of the local neighbourhood. The correlation analysis described in the ‘Specification of the Hedonic Model’ section would determine which of these measures shared a greater relationship with property prices and thus the more appropriate to include in our hedonic model. Accessibility variables relate to a property’s proximity to amenities, public services and transport. An important consideration in our analysis of rural housing markets was a property’s distance from ‘localised’ amenities such as a post office or local GP, as well as those services typical of many hedonic studies such as access to the transport network and distance to nearest town or commercial centre. Data was sourced through the CRC, who hold the RSS distance-to-service data completed by Defra’s Rural Statistics Unit. The RSS is compiled using GIS mapping software, which records the shortest straight-line distances between all households in England and a set of specific services. The postcode of each property transaction was matched with those held within the RSS, determining a straight-line distance to each local service set out in Table A 8. As the RSS does not consider access to transport services, our own GIS mapping team calculated the straight-line distance between the postcodes of each property sale and the nearest railway station and A-road. Using the same method we were also interested to determine the distance between each property and the nearest town centre, which the RSS does not record. For the districts of Mid Sussex and North Norfolk, straight-line distance to the coast was also identified and measured as a potential

131

Research into Rural Housing Affordability Final Report

driver of property prices within these markets. It was recognised however, that there was likely to be considerable overlap in the data sourced from the CRC and our own GIS analysis. The RSS data focuses on the location of very small service outlets, such as a post office, bank and independent shop. Our measure of distance to a local centre can thus be seen as a composite measure of access to a number of these services For more rural districts, it is possible that CRC data would be a superior measure of accessibility variables, as local services are more randomly dispersed across villages. For the more urban districts however, villages and towns are often served by amenities contained within a key centre, causing distances to small service outlets recorded by the CRC to be very similar and thus highly correlated. This notion was tested through the correlation analysis described in the ‘Specification of the Hedonic Model’ section. Variables that relate to the quality of the environment are much more difficult to define and measure. This is an aspect of hedonic pricing that is receiving growing attention in more recent studies. Environment variables need to reflect peoples’ perceptions of the environmental amenity, such as landscaping and views, as well as disbenefits such as noise and pollution. However, few straightforward methods exist for quantifying these variables. A further difficulty is that environmental attributes of an area are often embodied in other attributes that will have already been measured within a hedonic price function, such as noise associated with proximity to a road. With the capacity of this stage of the research and availability of data in mind, proxies for measuring the quality of the local environment were considered. GIS Digital Boundary Datasets held by Natural England were used in order to determine whether properties fell within a designated environmental area, such as an AONB. These designations are able to capture the natural environmental quality in a holistic sense, but cannot account for local quality differences or the surrounding quality of the built environment. Finally, central to our application of the HPM has been the need to determine the extent to which a rural location adds value to residential property. We therefore looked to identify data that could accurately capture and quantify the rural locality of a property, allowing us to isolate the impact different rural typologies have on price. The RSS was identified as an ideal source for measurement of rural variables. The RSS draws on Defra’s definition of rural and urban locations, breaking them down into four settlement types; urban, rural town and fringe, rural village and rural dwelling and assigning them to either a 'sparse' or 'less sparse' regional setting reaching eight final typologies8.

Figure A 1 sets out how these rural and urban typologies are defined and aggregated. The postcode of each transaction was then able to be matched with those held within the RSS.

8

The Rural/Urban Definition was introduced in 2004 and defines the rurality of very small census based geographies, based on population densities.

132

Research into Rural Housing Affordability Final Report

Figure A 1:

Rural/Urban Definition

Source: National Statistics

To conclude, any hedonic pricing study must account for a vast number of important explanatory variables. Failure to do so may lead to serious bias in the estimation of the parameters for the variables that are included. In many studies, the availability of GIS software enables the calculation of large quantities of spatial data rapidly and accurately. However, in compilation of explanatory variable data for this research, the resources for such measurement are restricted. It is particularly the spatial scale at which data is available, as well as the very detailed explanatory data that we have been unable to measure, that will principally limit our application of the HPM. As with any hedonic study, there will also need to be significant caution taken when specifying a final hedonic function, as some explanatory variable data is likely to be highly correlated. In the following sections we describe how each variable was tested in order to determine whether a relationship with price did in fact exist and identify potential inter-correlations between property attributes. The results from this analysis consequently determine which explanatory variables should be incorporated into the hedonic price function for each district.

133

Research into Rural Housing Affordability Final Report

Table A 8:

Potential Explanatory Variables
Variable Description Detached House Semi-Detached House Terraced House Ethnic Composition Age Composition Age Composition Age Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Socio-economic Composition Distance to Amenities Distance to Amenities Distance to Amenities Distance to Amenities Distance to Amenities Variable Code T_DETACH T_SEMI T_TERR WHITE_BRITISH YNG_FAML OLD_FAML ELD_ALONE OWN_HME UNEMPL TWO_CAR IMD_SCORE INC_SCORE EMPL_SCORE HEALTH_SCORE EDU_SCORE BARR_SCORE CRIME_SCORE ENVIRON_SCORE BANK GP POST SCHOOL TOWN Indicator Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true % local population white British % local households young families (34 years and under with dependent children) % local households old families (35 years and over with dependent children) % local households elderly living alone % local households owning their properties % local working age population unemployed % local households owning 2+ cars Index of Multiple Deprivation score Income Deprivation score Employment Deprivation score Health Deprivation and Disability score Education, Skills and Training Deprivation score Barriers to Housing and Services score Crime Deprivation score Living Environment Deprivation score Absolute distance to bank/building society Absolute distance to GP surgery Absolute distance to post office Absolute distance to primary school Distance to town centre (1km buffers) Source VOA VOA VOA ONS ONS ONS ONS ONS ONS ONS DCLG DCLG DCLG DCLG DCLG DCLG DCLG DCLG CRC CRC CRC CRC Mapinfo Geography Postcode (- 1digit) Postcode (-1 digit) Postcode (- 1 digit) LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA Postcode (-1 digit)

Structural

Neighbourhood

Distance to Services/ Accessibility

134

Research into Rural Housing Affordability Final Report

Distance to Transport Distance to Transport Distance to Coast Area of Outstanding Natural Beauty Natural (national) Nature Reserve Local Nature Reserve National Park Sites of Specific Scientific Interest Urban Sparse Rural Town & Fringe Sparse Rural Village Sparse Rural Dwellings Sparse Urban Less Sparse Rural Town & Fringe Less Sparse Rural Village Less Sparse Rural Dwellings Less Sparse Sparse Less Sparse Rural Urban

ROAD RAIL COAST AONB NAT_NATURE LOCAL_NATU NATIONAL_P SSSI U_S RTF_S RV_S RD_S U_LS RTF_LS RV_LS RD_LS SPARSE LESS_SPARSE RURAL URBAN

Environment

Distance to A Road/Motorway Junction (1km buffers) Distance to railway station (1km buffers) Distance to coast (1km buffers) Dummy variable, 1=falls within area boundary Dummy variable, 1=falls within area boundary Dummy variable, 1=falls within area boundary Dummy variable, 1=falls within area boundary Dummy variable, 1=falls within area boundary Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true Dummy variable, 1=true

Mapinfo Mapinfo Mapinfo Mapinfo Mapinfo Mapinfo Mapinfo Mapinfo CRC CRC CRC CRC CRC CRC CRC CRC CRC CRC CRC CRC

Postcode (-1 digit) Postcode (-1 digit) Postcode (-1 digit) Postcode (-1 digit) Postcode (-1 digit) Postcode (-1 digit) Postcode (-1 digit) Postcode (-1 digit) LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA LSOA

Rural/Urban

135

Research into Rural Housing Affordability Final Report

Defining an appropriate functional form
An important task in the HPM is to establish the exact nature of the relationship between the dependent variable and the explanatory variables. The relationship between the price of a property and its different attributes may not be linear, in other words, these relationships are likely to follow different functional forms. For example, the price of a property may be driven by access to the road network. However, as proximity to a road increases, the price of a property is likely to increase at a decreasing rate, as the benefits of increased accessibility are outweighed by the associated noise, pollution and so on. Other hedonic price studies generally adopt one the following simple functional forms to describe the relationship between the price of a property and its attributes; Linear; both the dependent and explanatory variables enter the regression in their linear form Semi-log; the log of the dependent variable is regressed against linear explanatory variables Log-linear; a linear dependent variable is regressed against the log of the explanatory variables Log-log; both the dependent and explanatory variables enter the regression in their log form However, given the scope of this research, independently testing for the functional form of each variable was not within the capacity of this analysis. This does not jeopardise our application of the HPM to rural house prices. In the literature there is no guidance in economic theory that would support the best possible functional form, however, a linear relationship between house prices and its characteristics is widely assumed in other hedonic studies. A principal advantage of assuming a linear relationship between variables is that it is easy to interpret. For this reason we have assumed an entirely linear hedonic function for all four districts.

Specification of the hedonic price model: Selection of variables
Testing for a relationship between dependent and explanatory variables
Having identified a full set of explanatory data, to better recognise the direction and relationship between the dependent and explanatory variables, correlation analysis was carried out. Firstly, it was necessary to individually test the correlation between our unit house price measure, the dependent variable and each explanatory variable. The results of this correlation analysis show us the direction and strength of co-movement of the two variables, in other words, whether there is any link between house price and the individual attribute and therefore whether or not it is relevant to include this attribute in our hedonic function. In addition, this importantly informs which of those mutually exclusive variables were most suitable for analysis, for example, selecting either IMD scores or census data to represent neighbourhood attributes within our hedonic function. The results of this analysis are set out in Table A 9. In statistics, the correlation between two variables can fall between -1 and 1. The closer the correlation coefficient is to plus or minus 1, the stronger is the linear relationship between the two variables. As a rule of thumb, a correlation coefficient between 0.30 and 0.70 denotes a reasonable relationship between two variables. However, it is also possible that whilst a correlation between two variables may be weak, this relationship is still very significant. This was therefore an additional consideration in our correlation testing. Explanatory variables that revealed a significant relationship with price in our correlation analysis are highlighted in table A 9.

136

Research into Rural Housing Affordability Final Report

Table A 9: Correlation between dependent and explanatory variables
Explanatory variable (xi) Mid Sussex Structural T_DETACH T_SEMI T_TERR BRITISH ETHNIC YNG_FAML OLD_FAML ELD_ALONE OWN_HME UNEMPL TWO_CAR IMD_SCORE INC_SCORE EMPL_SCORE HEALTH_SCORE EDU_SCORE BARR_SCORE CRIME_SCORE ENVIRON_SCORE A_ROADS RAILWAY TOWN COAST DIST_POST DIST_GPall DIST_GPprinc DIST_SCHOOL DIST_BANK AONB NAT_NATURE LOCAL_NATU NATIONAL_P SSSI U_S RTF_S RV_S RD_S U_LS RTF_LS RV_LS RD_LS 0.54 -0.25 -0.30 0.01 -0.01 -0.27 -0.06 0.12 -0.04 -0.04 0.16 -0.03 -0.11 -0.12 -0.16 -0.25 0.28 -0.03 0.23 -0.07 0.26 0.32 -0.04 0.29 0.24 0.24 0.30 0.31 0.22 n/a n/a n/a n/a n/a 0.13 0.26 0.15 -0.16 -0.19 -0.07 -0.07 Correlation with unit house price (y) North South Selby Norfolk Shropshire 0.57 0.72 0.53 -0.37 -0.30 -0.24 -0.21 -0.45 -0.31 -0.13 0.05 -0.25 0.13 -0.05 0.25 -0.15 -0.26 -0.38 -0.16 0.02 0.21 0.09 -0.07 -0.24 -0.15 0.14 0.17 -0.10 -0.06 -0.32 0.07 0.24 0.35 0.06 -0.01 -0.12 -0.07 -0.04 0.19 0.10 0.13 -0.05 -0.05 0.13 -0.09 0.15 0.26 0.21 0.17 0.14 0.41 n/a n/a -0.01 -0.04 n/a 0.13 0.26 0.15 -0.16 -0.19 -0.07 -0.07 -0.01 -0.18 -0.18 -0.14 -0.23 0.26 -0.23 0.20 0.22 0.19 0.15 n/a 0.34 0.35 0.32 0.37 0.25 0.21 0.01 n/a n/a 0.00 n/a -0.21 0.16 0.01 n/a -0.01 0.09 0.03 -0.34 -0.34 -0.43 -0.42 -0.43 0.22 -0.30 0.14 0.15 0.33 0.10 n/a 0.13 0.03 0.09 0.23 0.15 n/a n/a n/a n/a n/a n/a n/a n/a n/a -0.20 -0.11 0.31 -0.08

Neighbourhood

Distance to Services/ Accessibility

Environment

Rural/Urban

137

Research into Rural Housing Affordability Final Report

SPARSE LESS_SPARSE RURAL URBAN

n/a n/a 0.32 -0.32

0.36 -0.36 0.16 -0.16

-0.06 0.06 n/a n/a

n/a n/a 0.20 -0.20

Treatment of multi-collinearity In addition, some of the explanatory variables are likely to be highly correlated because they share a very similar relationship with our dependent variable: unit house price. In statistics this is referred to as multi-collinearity. This is a common scenario in the application of the HPM to house prices, as many attributes of a property are interlinked or overlap. If a high degree of inter-correlation exists between two attributes, this can lead to overestimating their significance in the hedonic price function, or in the worst case, produce a wrongly signed relationship with property price. Regression analysis that is used to estimate the hedonic price function can find it extremely difficult to tease apart the separate influences that different, but closely related, variables have on property. Correlation analysis prior to estimating a hedonic price function therefore looks to overcome some of these risks presented with multi-collinearity. It was necessary to test the correlation between each of the explanatory variables, e.g. between x1 and x2, x1 and x3 and x2 and x3 etc, in order to indicate which of the explanatory variables listed in Table A 9 should be included in the hedonic function for each district. In some cases however, there will a level of correlation that exists between explanatory variables that is almost unavoidable, as characteristics will be ‘symptomatic’ of each other. This is particularly true for explanatory variables representing neighbourhood attributes. Furthermore, in our application of the HPM to rural house prices, it is highly probable that the rural location of a property will in some way be correlated with access to local amenities, which are both likely to have a similar relationship with price. An extremely rural dwelling for example, is likely to be very isolated from local shops and services. Similarly, the rural typology of an area will often determine the nature of the housing stock. Inner-city areas and towns for example, are likely to have a high concentration of smaller properties, whereas a rural village is likely to have a much greater supply of larger, for example, detached, properties. That is not to say that these variables should be excluded from our hedonic analysis altogether; as long as the strength of the inter-correlation is moderate such variables should be included. When we come to run multiple regression analysis for each hedonic function, these correlation results will inform us as to whether our results are plausible, or instead possibly the cause of high inter-correlation. Treatment of spatial dependence One final issue in the estimation of hedonic price functions is that of spatial dependence or correlation. Spatial dependence results from the fact that properties in close proximity to each other often share very similar neighbourhood, accessibility and environmental attributes. The traditional regression analysis that is used to apply the HPM cannot account for this correlation. Whilst in general this does not lead to biased results, it does lead to greater variance in our estimation of the hedonic price function. Specification of the hedonic price model In analysing our final correlation results, a number of ‘selection criteria’ were considered in order to specify a final hedonic price function for each district.

138

Research into Rural Housing Affordability Final Report

Firstly it was important to only include explanatory variables whereby a reasonable correlation (equal to or larger than +/-0.3) existed between the attribute and unit house price. However, in reviewing this criterion, the significance of the relationship between two variables was also considered. The results of this analysis were presented previously in Table A 9. In addition, selection was based on a consistently signed relationship exists (i.e. negative or positive) between the attribute and unit house price across all four districts, unless an a priori expectation would lead us to believe the direction of such a relationship may vary across districts. For example, we would expect the relationship between unit house price and the unemployment rate in the local neighbourhood to be consistently negative, as higher unemployment implies a higher level of economic deprivation. However, the relationship between house price and distance to transport services may be positive in the more urbanised districts and negative in those deeply rural. Finally consideration was given to the existence of any inter-correlation between the explanatory variables selected for each district’s hedonic price function. It was considered that as long as the intercorrelation between variables was no larger than each variable’s correlation with price, they could be included in our hedonic price function. The results of this stage of the correlation analysis, performed separately for each district, are presented in the tables overleaf.

139

Research into Rural Housing Affordability Final Report

MID SUSSEX results for multi-collinearity Structural

Positive T_DETACH YNG_FAML OLD_FAML UNEMPL TWO_CAR RAILWAY OLD_FAML TWO_CAR, DIST_Gpall, DIST_SCHOOL, DIST_BANK TWO_CAR, DIST_Gpall, DIST_SCHOOL, DIST_BANK, RAILWAY

Negative

Neighbourhood Composition

TOWN Distance to Services/ Accessibility COAST DIST_POST DIST_Gpall DIST_SCHOOL DIST_BANK Environment AONB U_LS Rural/Urban RTF_LS RV_LS RD_LS

DIST_POST DIST_Gpall DIST_Gpall, DIST_BANK, TOWN TWO_CAR, DIST_Gpall, DIST_BANK, RAILWAY, TOWN U_LS DIST_BANK, RAILWAY, TOWN TWO_CAR, DIST_Gpall, DIST_SCHOOL, DIST_BANK, RAILWAY U_LS

140

Research into Rural Housing Affordability Final Report

NORFOLK results for multicollinearity Structural

Positive T_DETACH YNG_FAML OLD_FAML UNEMPL TWO_CAR TOWN COAST DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK DIST_Gpall, TWO_CAR TWO_CAR OLD_FAML DIST_Gpall, TWO_CAR, DIST_BANK

Negative

Neighbourhood Composition

Distance to Services/ Accessibility

Environment

AONB U_S RTF_S RV_S RD_S U_LS RTF_LS RV_LS RD_LS SPARSE LESS_SPARSE RURAL URBAN

Rural/Urban

141

Research into Rural Housing Affordability Final Report

SOUTH SHROPSHIRE results for multicollinearity Structural

Positive T_DETACH YNG_FAML OLD_FAML UNEMPL TWO_CAR RAILWAY A_ROADS TOWN DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK OLD_FAML TWO_CAR, A_ROADS DIST_BANK TWO_CAR, RAILWAY DIST_POST DIST_POST, DIST_Gpall DIST_Gpall, DIST_SCHOOL TWO_CAR, TOWN

Negative

Neighbourhood Composition

UNEMPL

Distance to Services/ Accessibility

Environment

AONB RTF_S RV_S RD_S RTF_LS RV_LS RD_LS SPARSE LESS_SPARSE TOWN

Rural/Urban

142

Research into Rural Housing Affordability Final Report

SELBY results for multicollinearity Structural

Positive T_DETACH YNG_FAML OLD_FAML UNEMPL TWO_CAR RAILWAY A_ROADS TOWN TWO_CAR, DIST_BANK

Negative

Neighbourhood Composition

YNG_FAML OLD_FAML

OLD_FAML YNG_FAML, UNEMPL

Distance to Services/ Accessibility

DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK DIST_POST TWO_CAR, DIST_GPall YNG_FAML, UNEMPL TWO_CAR, TOWN

Environment

AONB U_LS RTF_LS RV_LS RD_LS RURAL URBAN TWO_CAR, TOWN YNG_FAML, UNEMPL

Rural/Urban

143

Research into Rural Housing Affordability Final Report

Estimating the hedonic price model
Based on the principles of the preceding analysis described in chapter 2, a reduced set of potential explanatory variables was developed for each district for consideration in developing an optimal hedonic price function. This revised set of explanatory variables for each district is presented in Table A 10. Table A 10:
Mid Sussex T_DETACH YNG_FAML TWO_CAR HEALTH_SCORE EDU_SCORE

Results of correlation analysis: drivers of property price
North Norfolk T_DETACH YNG_FAML UNEMPL OWN_HME South Shropshire T_DETACH YNG_FAML TWO_CAR HEALTH_SCORE EDU_SCORE CRIME_SCORE Selby T_DETACH YNG_FAML UNEMPL TWO_CAR HEALTH_SCORE EDU_SCORE CRIME_SCORE A_ROADS RAILWAY TOWN DIST_POST DIST_SCHOOL DIST_BANK

RAILWAY TOWN DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK

TOWN DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK

A_ROADS RAILWAY TOWN DIST_POST DIST_GPall DIST_SCHOOL DIST_BANK AONB RTF_S RV_S RD_S RTF_LS RV_LS RD_LS SPARSE

AONB U_LS RTF_LS RV_LS RD_LS RURAL

AONB RTF_S RV_S RD_S U_LS RTF_LS RV_LS RD_LS SPARSE RURAL

U_LS RTF_LS RV_LS RD_LS RURAL

A number of initial hedonic price models were investigated by performing stepped regression analysis using the variables set out above in Table A 10. An optimal hedonic price function was then specified for each district through a build up of regression analysis. Central to this analysis was testing our central hypothesis; properties in a rural location will attract a premium over those that are located in urban areas. In addition we also looked to test the hypothesis that extremely isolated rural properties will not attract such a premium. In the HPM reaching an optimal hedonic price function required us to run a series of four separate models using regression analysis for each district. In each of the models, the structural variables have been included as well as the variables

144

Research into Rural Housing Affordability Final Report

representing a property’s rural typology. As each new model was tested, additional explanatory variables concerning the neighbourhood, accessibility and surrounding environment were incorporated. The aim of this ‘stepped’ analysis was to ensure that the impact of each rural typology was relatively stable across all four models. Introducing variables into the hedonic function in a progressive manner also informed us as to which variables improved the fit of the model and which of those that did not. The following summarises the four models that were tested for each district: Model I: Structural and Rural variables; Model II: Structural, Neighbourhood and Rural variables; Model III: Structural, Neighbourhood, Accessibility and Rural variables; and Model IV: Structural, Neighbourhood, Accessibility, Environment and Rural variables. The fourth and final model tested for each district is considered as the optimal hedonic price function for each respective rural housing market, the results of which are discussed in the next section.

Hedonic Price Model Results And Interpretation
Table A 11 to Table A 14 present the results from our estimation of an optimal price function for each district’s housing market. Table A 11: Mid Sussex Results
Model IV(I) 95,872 29.4 *** 23,927 18.3 *** -1 -6.5 *** 0 -2.7 *** 5,961 4.8 *** 2,301 5.2 *** Model IV(Ii) 98,485 28.8 *** 23,854 18.2 *** -1 -6.8 *** 0 -3.3 *** 3,423 1.8 * 2,490 4.5 *** 4,054 2.5 ** -1,799 -2.1 **

(Constant)

T_DETACH (Dummy) YNG_FAML

TWO_CAR

DIST_SCHOOL

DIST_BANK

DIST_POST

DIST_GPall

RV_LS (Dummy)

5,929 2.7

145

Research into Rural Housing Affordability Final Report

RD_LS (Dummy) RURAL (Dummy)

*** 10,368 4.0 *** 7,847 4.0 *** 608 0.479 608 0.482

Observations Adjusted R Square

Table A 12:

Selby Results
Model IV(I) 48,748 10.9 *** 15,688 13.3 *** -1 -3.4 *** 0 3.3 *** 2,119 2.5 ** -1,165 -3.3 *** 1,594 4.9 *** 4,785 2.3 ** 2,908 2.0 ** Model IV(Ii) 53,751 13.1 *** 15,618 13.2 *** -1 -3.8 *** 0 3.5 *** 2,610 3.1 *** -1,153 -3.3 *** 1,714 5.3 ***

(Constant)

T_DETACH (Dummy) YNG_FAML

TWO_CAR

DIST_SCHOOL

DIST_BANK

RAILWAY

U_LS (Dummy) RV_LS (Dummy) RURAL (Dummy)

-4,609 -2.2 ** 353 0.485 353 0.480

Observations Adjusted R Square

146

Research into Rural Housing Affordability Final Report

Table A 13:

South Shropshire Results
Model IV 65,110 32.2 *** 28,915 21.6 *** -1 -4.7 *** 4,291 6.6 *** 642 3.8 *** 4,636 2.4 ** -3,643 -2.3 ** 323 0.668

(Constant)

T_DETACH (Dummy) YNG_FAML

DIST_SCHOOL

A_ROADS

RV_S (Dummy) RD_S (Dummy)

Observations Adjusted R Square

Table A 14:

North Norfolk Results
Model IV(I) 83,888 17.5 *** 19,800 23.8 *** -1 -5.1 *** 0 -5.5 *** -1 -3.3 Model IV(Ii) 81,741 17.5 *** 19,806 23.7 *** -1 -5.0 *** 0 -5.3 *** -1 -3.1

(Constant)

T_DETACH (Dummy) YNG_FAML

OWN_HME

UNEMPL

147

Research into Rural Housing Affordability Final Report

DIST_SCHOOL

DIST_GPall

AONB

RTF_S (Dummy) RV_S (Dummy) RD_S (Dummy) SPARSE (Dummy)

*** 1,252 2.4 ** 956 3.9 *** 12,303 11.5 *** 5,724 4.6 *** 9,199 6.8 *** 6,654 4.0 ***

*** 1,429 2.8 *** 1,086 4.6 *** 12,310 11.5 ***

7,151 7.7 *** 665 0.618 665 0.616

Observations Adjusted R Square

The first column gives the name of each explanatory variable. Those variables which indicate a dummy variable suggests that the value of that variable can either take the value 1 when true or 0 when it is not. For example, the dummy variable for ‘T_DETACH’ takes a value of 1 for properties which are detached and 0 for other properties (semi-detached and terraced) measured within our house price data. Those variables not presented in brackets are continuous in that they can take any value. The remaining two columns in the tables present the optimal models reached through undertaking the stepped analysis described previously. With the exception of South Shropshire, two models have been determined for each district in order to test the different classifications of a rural area. That is, to determine the impact of a rural location of property price, Defra’s definition of rural was tested in its simplest form: rural/urban and by detailed settlement typology: urban, rural town and fringe, rural village and rural dwelling each of which are assigned to either a sparse or less sparse regional setting. For some districts not all rural typologies were able to be tested within the model. This was because either no such areas existed or there was not a significant number of a settlement type to produce variation that could be tested statistically. South Shropshire was the only district in which all property transactions fell under a ‘rural’ typology. In the tables, the upper number for each variable is the coefficient describing the effect of that variable has on our dependent variable; the average unit price of a property. The lower number for each variable is known as the t-statistic. This figure denotes whether our result can be considered statistically significant, which as a rule of thumb, is when the value of ‘t’ is equal to or greater than 2.

148

Research into Rural Housing Affordability Final Report

Finally, the degree of certainty we have regarding our estimates for each variable is indicated by the number of asterisks. The asterisks refer to a certainty level, in which we can be confident that our result is statistically significant and not a matter of chance. In statistics, it is generally accepted that a 95% certainty level is evidence of a significant relationship between the explanatory and dependent variable (denoted by two asterisks). It is often accepted however, that we can be confident our result is significant at a 90% certainty level (one asterisk). When three asterisks are presented this indicates a very strong statistical result, as we can be 99% confident that our coefficient is significant in describing variation in price. In the tables, the number of observations refers to the number of property transactions that were entered into the model for each district. The last row of the table presents the overall fit of the model denoted by the adjusted R2 statistic, which describes the proportion of variation in the unit price of a property that can be explained by the model. An adjusted R2 value of around 0.7 or greater is considered a very robust model for predicting variation in the dependent variable. In hedonic studies it is generally accepted that structural variables explain a large majority of house price variation. When interpreting the results of our final hedonic model and the ‘degree of fit’, it should thus be kept in mind that a significant number of structural variables were not possible to include in our application of the HPM in this research. Noticeable across Table A 11 to Table A 14 is that the degree of fit of our hedonic model is better for those districts that are generally more rural in typology. For South Shropshire approximately 67% of the variation in the unit price of a property can be explained by the model, whilst for North Norfolk 62% of the variation in price is accounted for. For Mid Sussex and Selby, the models have a lower degree of fit with an adjusted R-squared value of 0.48 and thus are weaker representations of the true housing markets for these districts. This could be due in part to the unavoidable inter-correlation between accessibility variables in these models as properties are often served by a key commercial centre containing most local amenities and transport links. In an attempt to remove some of this inter-correlation from the models, we tested a function that substituted access to small service outlets (e.g. bank, post office, GP etc) with access to a town centre. However, doing this did not produce a statistically significant result for the ‘Town’ variable and did not improve the overall fit of the model. The weakness of these models could also be attributed to the omission of structural characteristics from much of our analysis, which will be important drivers in Mid Sussex and Selby due to the mixed nature of the housing stock. In contrast for the South Shropshire housing market, there is more consistency in the quality and type of properties supplied, allowing us to control for this when estimating the model and is reflected by the ‘goodness of fit’. Aside from this, the models estimated for each district can be considered statistically robust as all of our results have met the ‘t-statistic’ criteria and in most cases, we can be 99% confident that our estimated coefficients for each explanatory variable are significant. Generally across all four districts the type of property is the most significant and largest driver of property price. As one would expect, our results show that all else being equal, detached properties will command a higher price relative to other property types (i.e. semi-detached and terraced). This variable also accounts for on average, around half of the variation in price within the model. In terms of neighbourhood attributes, our results are fairly consistent across all four districts and generally conform to a prior expectation that a lower socio-economic standing in an area will reduce the relative price of property. In all four districts, a higher proportion of young families has a small, but very significant negative impact on price. However, the proportion of households owning two cars or more has a varying impact across the districts. One would generally expect a higher percentage of households owning two or more cars

149

Research into Rural Housing Affordability Final Report

would imply a higher socio-economic standing in that area. The results for Mid Sussex implies a negative relationship between car ownership and property price, although the relative net impact on price is zero. The analysis of Selby’s housing market however, implies a positive relationship between car ownership and property price, although the relative net impact on price is again zero. A possible explanation is that car ownership in rural areas could be classed as a proxy for the level of isolation of a property, with an exceptionally high level of car ownership indicating that a property is particularly remote. Variables measuring a property’s access to local amenities and transport links provide interesting results. Generally across all four districts, the relative price of property rises as the distance to these services increases. This result suggests that within rural housing markets, there is a value placed on properties that are distanced from services that bring about disbenefits such as congestion, pollution and noise. Notable exceptions here are the results for Mid Sussex, whereby price falls as the distance to health services (GP) increases and for Selby, where price falls as distance to banking services increases. As both of these districts contain a greater mix of urban as well as rural areas, this result could reinforce the notion that increased distance to services will have a more positive impact in rural areas/districts and a negative impact on property prices in urban areas. However, the size of the coefficients of accessibility variables are relatively lower than other explanatory variables, implying a smaller impact on property price and generally range by rural district and according to the type of service assessed. It is possible that further testing of the relationship between these variables and price, for example, in a non-linear regression, would indicate that there is in fact an optimum distance from services and transport that positively impacts on price, above which price will fall as the disbenefit of travelling long distances to local amenities becomes relatively more important. It was only possible to include the impact of the quality of the surrounding environment in the model for North Norfolk (denoted by the dummy variable for AONB). As Table A 14 shows, a strong and significant result was found. Properties located within an AONB command an average unit price that is around £12,300 higher than those that are not. This result was consistent across both models estimated for North Norfolk. Finally we come to analyse the extent to which a rural location adds value to residential property, hypothesising that all else being equal, rural properties will achieve a premium over properties in urban areas. In order to assess whether our central hypothesis holds, we present a separate discussion of the results for each district below. The results for Mid Sussex are presented in Table A 11. In the district, all property transactions analysed fell into the ‘less sparse’ regional setting. In Model IV (i), ‘rural village’ and ‘rural dwelling’ were tested, with analysis of these results being compared to the alternate typologies of ‘urban’ and ‘rural town and fringe’. The results of Model IV (i) thus indicate that all else being equal, properties classed as a rural dwelling command a relatively higher average unit price, as do properties located in rural villages. Model IV (ii) presents the results from testing the rural classification of a property in its simplest form, indicating again that a rural property will achieve a premium over urban properties. Table A 12 sets out the results for Selby, presenting a very different result. Once again due to its typology, all property transactions analysed for the district fell into the ‘less sparse’ regional setting. In Selby it would appear that a rural location will in fact reduce the average unit price of a property. The results for Model IV (i) show that properties in an urban location will command a higher price than properties located in rural villages. Similarly in Model IV (ii) our results indicate that a rural location will have a negative impact on property prices. In Table A 13 we present the results for South Shropshire. All property transactions analysed for the district were classified as rural under Defra’s classifications, therefore our analysis here compares the extent of ‘rurality’. Interestingly Model IV indicates that properties located in a rural village command a higher relative price whereas properties that are particularly isolated (classed as a rural dwelling and

150

Research into Rural Housing Affordability Final Report

located in a ‘sparse’ regional setting) will have a negative impact on price. This reinforces our hypothesis in both senses: properties in a rural location will attract a premium over those that are located in urban areas, yet extremely isolated rural properties will not. Finally the results for North Norfolk are presented in Table A 14. The majority of property transactions analysed for the district fell into Defra’s ‘rural’ classification and thus did not produce enough variation to test this typology within the model. The results of Model IV (i) show that properties located in each rural setting (rural town and fringe, rural village and rural dwelling) command a higher relative price than properties in urban areas. In particular, properties located in rural villages achieve a premium over and above other types of rural areas. Similarly in Model IV (ii), properties in a ‘sparse’ regional setting command a higher price than properties in a ‘less sparse’ setting.

Strengths and weaknesses of our application of the HPM
In interpretation of our results in the previous section and in drawing conclusions in the final chapter of this report, the key strengths and weaknesses of our application of the HPM to rural house prices should be considered. The following provides a brief summary.

Strengths
Quality of secondary data sources Our analysis of rural house price data has drawn entirely upon actual transactions from the rural housing markets, rather than using estate agent or property owner’s estimates that other hedonic studies employ. Using VOA data makes our hedonic model empirically defensible as a good approximation of the true housing markets observed for in each rural district. Similarly, in the absence of sophisticated GIS software used by other hedonic studies, we have taken secondary data for our explanatory variables from comprehensive sources such as the CRC/Defra’s Rural Services Series.

Weaknesses
Omission of detailed structural attributes and internal characteristics data The structural attributes of a property will without question, be a large, if not the largest, driver of price. Our analysis has omitted a large number of structural attributes that will in reality cause variation in property prices within a housing market. In statistical terms, this is reflected in the low ‘degree of fit’ of the hedonic model for some districts. In our analysis of rural house prices, there are particular structural attributes of a property that will influence price differently in the different districts. For example, in the more urbanised districts, a wider variation in the type and quality of the housing stock exists due to waves of modern housing development. In many areas of Selby, there exists a mixed supply of properties built in the last 30 to 40 years and period properties of a traditional Yorkshire stone build. The age of a property or possibly the type of build is therefore likely to have a strong influence on the price of properties in the district. Assumption of an entirely linear model Whilst many other hedonic studies have made the same assumption and it is accepted that linear models are simple to interpret, in reality not all explanatory variables will share a perfectly linear relationship with house prices. As we have previously highlighted in the report, often the distance to amenities and transport services will have a non-linear relationship with price. With greater scope, this research could have investigated an optional distance for each accessibility variables and tested for non-linear relationships with price. Spatial scale of explanatory variable data The HPM relies on a process of regression analysis to explain variations in house prices through variation in the different attributes associated with each property transaction. Using detailed data that

151

Research into Rural Housing Affordability Final Report

can accurately measure extremely local variation will therefore dramatically improve the fit of the model and the robustness of its results. In the most comprehensive hedonic studies, explanatory variable data analysed can be as detailed as on a property-by-property basis. However, the smallest spatial scale of data available to us for property attributes was at postcode level minus 1 digit. This ‘averaging’ of data will dilute much of the variation in attributes that is observed in the true housing market and that ultimately accounts for variation in price. This presents two risks, firstly in that the significance of an explanatory variable may be underestimated and secondly, can magnify the problems associated with the presence of multicollinearity and spatial correlation. This situation will be made worse for explanatory variable data measured at LSOA level. Non-quantifiable drivers of house prices There are a number of qualitative factors that are likely to cause variation in price within rural housing markets that simply cannot be captured within the hedonic pricing model. Though we cannot measure factors such as the feeling of community or quality of life associated with rural living as explanatory variables, or statistically prove a relationship with price, other areas of the research project have covered this in great detail. In addition to the ‘qualitative rurality’ of an area, there lacks established primary research methods or secondary sources that can measure many of the attributes that describe both the natural and built environment. This is an area of research that is receiving growing attention in more recent hedonic pricing studies. Rural housing markets are unique Throughout the other stages of the research project it has become clear that the rural housing market is relatively inactive. Often residents in rural areas have lived there for the for the most part of their lives, regardless of the different attributes of the housing stock but instead due to the attraction of rural life. More specifically, the rural housing market has specific supply and demand parameters, namely limited supply and heightened demand driven by external buyers. This analysis is a complex exercise; whilst we can infer some form of qualitative value of rurality here, we cannot capture this within a hedonic price function. Furthermore, if such market conditions exist, this will have implications for our application of the HPM to rural housing markets in this research. This could compromise the assumption that house prices adjust instantaneously to changes in supply and demand, as house prices could in fact be driven by a shortage in the supply of rural housing stock, rather than being a reflection of people’s willingness to pay for the different attributes of a property. Representativeness and transferability of the analysis Finally our application of the HPM has estimated the impact of a rural location on property prices, with the aim of this research to produce estimates that essentially conclude some kind of ‘power of rurality’ that can be transferred to other rural areas across the country. This can only be achieved if our sample of properties was diverse in terms of the property types; socio-economic areas; accessibility; environmental and rural typologies covered. As previously explained in the report, the diversity of our sales data and representativeness of our sample was to some extent, determined through mapping exercises. However, it is important to note that in reality the results of this analysis are only representative of the range of data given in the sample and that care should be taken in extrapolating well beyond that range. This is particularly relevant to our application of the HPM to rural housing markets, which are often inactive and characterised by a low turnover of the housing stock. Our sample of sales data is thus somewhat biased in the types of properties and the types of buyers it represents. Sales data is of course focused on the willingness to pay of more mobile buyers, who may have very different preferences for (and thus place very different values on) the attributes of a property compared to people that have lived in a village all their life.

152

Research into Rural Housing Affordability Final Report

In addition, our analysis has been formed on house price data from one year only and therefore cannot determine whether the value of rurality is constant or changes over time.

Recommendations for further research
In light of the limitations to our approach and analysis discussed previously, there are particular gaps in the research that would support further areas of work. In addition, the results from the village surveys and agent interviews have shed even more light on this. This section therefore proposes key elements of research that should be addressed if this hedonic analysis of rural house prices was extended in the future. The spatial scale of analysis is a factor that could certainly be addressed in future work. Our analysis has been largely based on data measured at postcode level minus 1 digit. This averaging of data reduces the degree of variation in the hedonic model, increases the risk of multicollinearity and reduces the explanatory power of our hedonic price function. Ideally further research would assess the prices and attributes of properties at least on a street-by-street basis, but ideally on a property-byproperty basis. However, studies that have captured this localised data tend to rely on property agent valuations rather than actual sales transactions, producing potentially bias and thus skewed results when analysing rural house prices across different markets/districts. It is clear from our application of the HPM that more information on the quality of a property’s structure is needed. Whilst practical methods to determine the internal quality of a property may not currently exist, other studies have successfully identified the external structural quality of a property. Particularly relevant to the analysis of rural house prices is the age of a property and/or the building’s materials such as type of stone. Accounting for these attributes within further research is likely to improve the analysis and fit of the hedonic model when applied to locations that contain a mixture of rural and urban settlements with a housing stock that is more diverse in age and quality. Interviews with estate agents highlighted this issue and insisted that in districts such as Selby, house prices were particularly driven by the type of property offered. Equally important to improving subsequent research and the fit of the model will be a greater analysis of both the natural and built environment. These attributes were almost entirely excluded from our application of the HPM due to data availability, but were continually raised as important to both residents and potential buyers in the surveys. Agents in particular emphasised the premium people are willing to pay for locations with quality rural landscapes, whilst consultation with residents revealed the importance of open space.

153

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close