The roles of social housing providers in creating ‘integrated’ communities

ABSTRACT Cohesion and integration agendas in Britain can be characterised by localisation of ‘race relations’ responsibilities and the importance of local institutions in shaping neighbourhoods has been acknowledged. However, little is understood about the roles of housing providers in integration initiatives. Indeed, research on housing and race has experienced a lull in the 2000s. Thus, this paper aims to examine how social housing providers negotiate their positions and are complicit in constructing a certain vision of community. It draws on interviews from the ESRC Centre on Dynamics of Ethnicity (CoDE)’s work in the ethnically diverse neighbourhoods of Cheetham Hill (Manchester), Newham (London), Butetown (Cardiff) and Pollokshields and Govanhill (Glasgow). The paper makes three arguments: first, that race and ethnicity as facets of ‘integration’ have been subsumed into broader agendas, yet remain implicit in community building; second, that housing organisation practices are often detached from local meanings of community and prioritise exclusionary activities focusing on behaviour change and, third, that the roles of housing organisations in constructing ‘integrated’ communities are highly variable and localised, influenced by the history and contemporary dynamics of place.


Introduction
Policy discourse about 'racial' and ethnic relations in neighbourhoodsthat is concerned with 'integrated', 'cohesive' or 'engaged' communitieshas at its core the notion that inter-ethnic mixing is desirable and separation, or segregation, is undesirable. This is on the basis of the presumed socially negative causes and consequences of social and residential segregation and, conversely, the assumed socially positive causes and consequences of social and residential mixing. It is within this context that social housing providers, and housing systems more generally, have been operating for more than a decade. Although a great deal of research over the past 15 years has been framed by community cohesion and integration agendas, relatively little attention has been paid to institutional roles in shaping neighbourhoods. This paper aims to address this deficiency by examining how social housing providers are complicit in constructing a certain vision of community.
It does this by drawing on selected interviews with residents and with housing and regeneration workers from the ESRC's Centre on Dynamics of Ethnicity (CoDE)'s local case study work. This was undertaken between 2013 and 2016 in the ethnically diverse neighbourhoods of Cheetham Hill (Manchester), Newham (London), Butetown (Cardiff) and Pollokshields and Govanhill (Glasgow).
After outlining the framing of our study in terms of community cohesion, integration and engagement agendas, and the changing position of housing organisations within this landscape, we draw upon interview material to make three arguments. First, in a context of resource cuts and a shifting remit of housing organisation towards generic practices of community engagement, race and ethnicity as facets of 'integration' have been subsumed into broader agendas and sometimes uncomfortably so given evident ethnic tensions and increasing diversity. Indeed, community building agendas and practices have been racialised such that issues of race are implicit if not explicit; they inflect institutional approaches even when race and ethnicity are not an identified remit. Second, housing organisation practices are often detached from local meanings of community and activities, instead focusing on behaviour change that stigmatises people and places, undermines the agency of local residents and delimits the boundaries of community inclusion. Third, these processes and the roles of housing organisations in constructing 'integrated' communities are highly variable and localised, influenced by the history and contemporary dynamics of place.

Community and integration initiatives: localising responsibilities
Over the course of the last decade and a half, questions of 'cohesion' and 'integration' have assumed primacy within UK policy. Important characteristics of these policies have been the primacy of race and ethnicity as the basis of 'communities'; the privileging of communication and contact to overcome tensions; and the importance of the 'local' with neighbourhoods and local government and service providers seen as the key actors in the production of integrated and cohesive communities (Kalra and Kapoor 2009;Ratcliffe 2012). The latter point can be seen as part of broader shifts in governance to neoliberal forms of localism (McKee 2015). Amidst concerns about increased and diversified flows of immigration, interethnic and interracial conflict, terrorism and populism, there has emerged 'a growing interest in the neighbourhood as a setting for everyday intercultural encounters and meaningful exchanges that might promote conviviality and provide building blocks for better social relations' (Phillips et al. 2014, 43). While the desirability of social mixing has permeated neighbourhood development policy since the New Labour government took power in the late 1990s, it became formalised in the 'community cohesion' agenda, which emerged in the wake of the 2001 'riots' in Burnley, Bradford and Oldham. Despite a complex array of forcesfrom institutional and systemic racism, deprivation, unemploymentbeing cited as significant (Robinson 2005;Kalra and Kapoor 2009;Thomas 2011), the official responses into the disturbances identified minority ethnic segregation as the key contributory factor (Cantle 2001;Denham, 2001).
The proposed solution to such problems -'community cohesion'emerged as the dominant policy prescription under the New Labour government, rolled out across various domains of social life including employment, education and housing (Robinson 2005). 'Community cohesion' advocated the need for greater contact cohered around 'a greater sense of citizenship, based on (a few) common principles which are shared and observed by all sections of the community' (Cantle 2001, 10). As Harrison et al. (2005, 90) has suggested, the policy is premised upon a conception of community which is laden with normative intent, implying that perceived 'micro-communities' at very local level should 'mesh into' or have counterparts at or across some higher level of community, not least because otherwise they may come into tension with each other.
While reference was made within the various policy documents to the importance of entrenched racism and material inequalities (Cantle 2001;Denham, 2001), 'community cohesion' has been criticised for placing most focus on a 'culturalist perspective', targeting principally racially and ethnically defined 'communities' and the promotion of more effective and harmonious modes of contact and communication across social and spatial divides, to the detriment of addressing more structural forces (Ratcliffe 2012; see also Kalra and Kapoor 2009;Jones 2013). 'Cohesion' was also to be enacted chiefly at the local level through the fostering of more positive forms of interaction (Amin, 2005;Robinson 2005;Ratcliffe 2012;Jones 2013), with local authorities charged with being key facilitators of cohesive communities (LGA, 2002). This represents shifts in political administration whereby 'local communities are constructed as arenas of governance' (Flint 2006, 174;see Amin, 2005).
Over the last decade, the community cohesion policy has evolved to incorporate a wider set of concerns. In 2007, the Commission for Integration and Cohesion (COIC) published the Our Shared Futures report, which sought to expand conceptions of cohesion, partly as a response to criticisms but also to emerging challenges, notably domestic instances of terrorism, the rise of new immigration in the form of refugees and EU accession migrants and fears over growing resentment amongst established white 'working-class' communities (Kalra and Kapoor 2009;Ratcliffe 2012;Jones 2013). Housing, and social housing allocation, in particular, was one arena in which perceptions of privileged treatment of ethnic minorities and migrants took hold. Alleviating this concern amongst the ethnic majority population was one reason for the introduction of residency and 'local connection' requirements for entitlement to social housing by some local authorities.
In 2012, the Coalition government published Creating the Conditions for Integration. This report dispensed with the term 'cohesion' and opted for a focus on integration, signifying a shift from the aim of communities 'getting on' to ensuring 'new residents and existing residents adapt to one another' (COIC 2007, 9;Jones 2013). The report iterated the need for integration to be not narrowly framed around questions of race and faith (despite its emphasis on Muslim communities), to remain concentrated on integration 'within and between neighbourhoods' and to be enacted locally through community engagement to address challenges which are 'too complex for laws and powers to provide the sole solution' (DCLG 2012, 6). The latest policy intervention of The Casey Review (Casey 2016) reinforces what has gone before in terms of the framing the problem as one of the segregated communities and the solution as the need for interaction and shared British values and behaviours. Indeed, the emphasis on pride in what is 'quintessentially British' (Casey 2016, 20) is given particular prominence in the most recent intervention. It also reinforces perspectives that have emerged in cohesion policy publications of the increasing threat of immigration and religious extremism. In these ways, the problem of cohesion is racialised. In terms of place, attention is focused on neighbourhoods with 'high concentration of particular ethnic or faith groups', meaning minority groups, as concerning in terms of potential 'negative impact on community cohesion' because of assumed lower levels of social interaction between groups (Casey 2016, 54). It is against this set of discourses that housing organisations are working in diverse areas, and it is to housing and integration debates that we now turn.

Housing and communities
Given the emphasis placed on residential settlement and neighbourhoods within policy discourses around cohesion and integration, housing has been identified as key to the production of cohesive and integrated places (Harrison, 2005;Robinson 2005;Flint 2006;Beider 2012;Bloch, Neal, and Solomos 2013). This has been evident in integration policy documents, including The Casey Review, which recommends improving understanding of 'how housing and regeneration policies could improve integration or reduce segregation' (Casey 2016, 17). This most recent intervention devotes less attention to housing than previous ones yet continues to pursue the theme of fairness in housing allocation that was evident in Our Shared Future: 'In the area of social housing we want to see more transparency, justification and fairness' (COIC 2007, 123). Social housing and social housing providers have been particularly targeted as, given both its public accountability and size, the sector represents 'an existing institutional framework (political, legal and organisational) for governmental intervention that does not exist in other tenures' (Flint 2006, 176). Once again, however, the ways to ensure this are to be decided at the local level informed by the particularities of specific areas, illustrating a tension of scale described by Finney, Clark, and Nazroo (2019) in the Introduction to this Special Issue. The COIC report (2007, 123) stated how, 'Housing issues are complex and vary between areas, so just like integration and cohesion, there cannot be a one size fits all approach. Local Authorities need to consider the mix and churn of their local area'. This necessitates a focus upon the ways in which housing institutions are both interpreting and shaping calls for more 'cohesive' and 'integrated' neighbourhoods at the local level, exploring how differences and similarities are manifest across locales, particularly as both local conditions, commitments and approaches to cohesion and integration are likely to vary (Flint 2006;McKee 2015).
While housing has been identified as a key tool through which cohesive and integrated communities can and should be built (Harrison et al. 2005;Flint 2006;Hills 2007;Robinson 2007), housing systems and providers have been subject to significant changes over the last decade, changes which have clearly impacted upon their ability to promote and deliver such residential neighbourhoods. What has been termed the 'housing crisis' represents a major shift not only in the balance of demand and supply of housing but the operation of housing markets and the roles of housing providers (see Lukes, de Noronha, and Finney 2019). Social housing provision has shrunk at a time when demand is high, in part due to the economic recession and rising private housing costs. The Housing Association sector has seen more large and commercially oriented housing associations emerge. This has had particularly negative impacts upon black and minority ethnic housing associations, who historically have been central to the gains made amongst black and minority ethnic groups' in relation to social housing (Harrison et al. 2005;Robinson 2007;Beider 2012). Indeed, when social housing providers (both local authorities and housing associations) have been charged with assisting in the creation of cohesive and integrated communities, they increasingly do so in a context in which their relative strength has been significantly curtailed.
The role of housing providers has also shifted with changes in legislation, including most recently, the 2016 Immigration Act, which casts housing providers as agents of state surveillance with landlords expected to ensure that provisions are not being extended to those ineligible for residence (Crawford, Leahy, and McKee 2016). The changes in resources and emphasis place social housing providers in an ambivalent and challenging position as agents of integration and cohesion.
Despite the policy entanglements of race, housing and cohesion/integration, recent years have seen surprisingly few studies of this domain reflecting a relative neglect of race in housing studies (Bloch, Neal, and Solomos 2013;Markkanen and Harrison 2013). Bloch, Neal, and Solomos (2013, 89) have noted how 'housing policy is often a marginalised part of the race and housing debate despite being very much a part of the housing opportunities and closures offered to BME communities'. Similarly, until recently, housing has been neglected in other social science disciplines. This paper places housing institutions centrally as agents of the definition and production of 'cohesive' and 'integrated' communities.

Methods
The paper draws from interview data collected as part of the first round of qualitative fieldwork led by the ESRC funded research Centre on the Dynamics of Ethnicity (CoDE). These data have been analysed with an interdisciplinary perspective as part of CoDE's interdisciplinary working project (see Finney, Clark, and Nazroo 2019). The qualitative work was carried out in four localities across the UK: Butetown (Cardiff), Cheetham Hill (Manchester), Govanhill and Pollokshields (Glasgow) and Newham (London) from 2014 to 2016. The semi-structured, qualitative interviews explored a loosely predefined set of themes related to key dimensions of social inequality including housing, education and histories of migration. The interviews also sought to establish narratives of living and working in these places. Participants included residents of the areas and workers in key third sector, public sector and community organizations based in the areas whose work addresses issues of ethnicity and ethnic inequality to some degree. Participants were recruited purposively, following an initial period of ethnographic and historical research. For the purposes of this paper, we have analysed the interviews with residents and interviews with workers involved in housing in some way. In the main, the latter includes representatives from housing organisations and local authority bodies that have some involvement in housing issues, such as local regeneration teams. In total, we draw from 60 interviews, involving 22 housing and/or local authority representatives and 38 residents.

The role of social housing providers in community integration
This section seeks to establish the aims of housing associations and the role that conceptions of 'cohesion' and 'integration' play within this. In what ways, and to what extent, do ideas of community cohesion and integration shape their work? How are the contradictions between the rhetoric and the reality of integrated/cohesive communities expressed and responded to by Housing Associations? How do local dynamics shape practices?
Across all four of the locations -Cheetham Hill (Manchester), Newham (London), Butetown (Cardiff) and Pollokshieds and Govanhill (Glasgow)it was notable that the exact terms 'community cohesion' and 'integration' were largely absent from the accounts of housing association workers. In terms of the language used, what we see instead is an increased emphasis on the need to 'engage' communities. However, as we will go on to show, many of these activities relate closely to the government's understanding of what makes an integrated society. Also of interest here is how in parallel to the 'traditional' understanding of cohesion and integration agendas, there has been an expectation that housing associations should also address wider issues including antisocial behaviour, employment initiatives and environmental and sustainability concerns (Flint 2006).
The link between the two agendas and the social housing sector's role in addressing them is indicated in the way in which, whilst interethnic tension and conflict did not always figure as a prominent concern, anti-social behaviour and community 'behaviour' more broadly interpreted did. In Cheetham Hill, for instance, 'crime and grime' (Regeneration officer, Manchester) was identified as the principle challenge facing the housing association. As the officer himself stated, 'It's the stuff they get most complaints about and it's the stuff that our officers spend an inordinate amount of time trying to resolve'. The matter of rubbish and dirt is of course a longstanding trope through which to racialise communities, inferring that uncleanliness, disease and danger are associated with (nonwhite) ethnic minority people (see, for example, Douglas 1966). These ingrained racialised discourses were reinforced in the recent Casey (2016) review as agencies were encouraged to 'teach' migrants to deal with rubbish. Similarly, aims to 'responsibilise' citizens through housing allocation have also shifted the focus of housing providers in England. The allocation of homes on the basis of 'community contribution'such as the amount of hours worked, voluntary activities, employment within the armed forceshas seen efforts of housing associations focused on enacting 'behaviour change' amongst residents. Here, housing providers are often involved in a range of initiatives aimed at helping to improve job readiness, promoting fire safety, recycling, energy efficiency and property maintenance. In Cheetham Hill, a local regeneration officer identified aging, health, environmental blight and economic development as particular strategic challenges. In fact, he argued that policy orientation had shifted in the contemporary climate of austerity and that the key strategic aim of local regeneration activity was to increase employment and reduce welfare dependency. He argued that even where racism and conflict existed, the policy emphasis was directed elsewhere: In this climate, it's about not being a burden on the system in terms of benefits and all the rest of it. It's about being socially and economically active. And therefore you're not dependent on the services, which means it costs services less. (Regeneration officer, Manchester) Similar observations were made by a housing association officer in Newham, who argued that welfare reform and austerity now dominated the local policy landscape; Welfare reform is council tax benefit changes. It is sanctions, it's conditionality, it is bedroom tax, it is the benefit cap. It is personal independence payments, it's a whole heap of things. And it's not just generally one thing that's causing people difficulty, it's a range of different things. (Housing association employee, Newham) It appeared then that the shift in emphasis as a result of wider welfare and local governance reforms somewhat marginalised an emphasis on cohesion and integration in a number of locales. Alongside this, however, reputational geographies and sensibilities of place also contributed. Similar to the situation Jones (2013) observed in Hackney, the reputation that Cheetham Hill has as a 'superdiverse' space was drawn upon to downplay the need for interventions in cohesion and integration. In contrasting the area to the problems in Oldham, an area also covered by the housing association, a local worker claimed that the 'mixed' nature of Cheetham Hill contributed to greater levels of contact and respect: 'it's mixed, it's mixed ethnicities and probably in living together and living alongside each other that they're respecting each other's cultures' (Housing association worker, Cheetham Hill). The local regeneration officer recognised these types of ideas of convivial mixing as commonplace but also recognised the existence of racism, and the danger that the emphasis on mixing and 'successful' diversity can erase experiences of racism: I think, because you tend to think Cheetham is very diverse and its people do tend to get on fairly well, I think. So we don't tend to think of Cheetham as being a hotbed of racism and stuff like that, but actually, it doesn't mean it doesn't go on. And it doesn't mean people aren't affected and that has an impact on other things. (Regeneration officer, Manchester) In Newham, also, it was other concernsnotably housing provisionthat took precedence over all other activities, although not completely divorced from them. As the worker for a housing association stated: The challenges for Newham are around, there's something like 27,000 people on Newham's housing waiting list, which is an awful lot of people on one housing waiting list. (worker, housing association, Newham) Here, ethnic and racial tensions were again not identified as significant, as the officer explained: 'we are bizarrely quite a harmonious borough considering the diversity that we have', pointing to the small amount of racially motivated crime she had noted in the borough as evidence of this. As in Cheetham Hill, many of the activities of the housing association in Newham had been focused on increasing employment and addressing poverty. The housing officer stated that their approaches were directed by 'individual need' and it was poverty rather than ethnicity that was its most significant marker. In the above accounts then, more apparently fundamental and functional concerns relating to the housing and socio-economic, and environmental conditions of the local areas under study dominated housing association agendas, but they nevertheless play a role in shaping notions of what a 'community' should be and ethnic diversity is not abstracted from that, nor is racism absent.
Different pictures were presented in both Glasgow and Cardiff, where there existed more prominent engagements with concerns over racial and ethnic conflict and disadvantage. In Pollokshields, the housing association worker interviewed was involved in a local homelessness network, where he was particularly focused on attempting to house members of the local BME population that lacked accommodation. He discussed how in the previous decade, the enrolment of outreach workers to engage with local 'Asian' and 'Arab' populations had enabled the expansion of the local BME population within social housing and as a result, 'any kind ofany suggestion of inequality of service provision was gettingwas eroded over quite a short period of time.' For the worker, the fact that '20-30 per cent' of the organisational staff was of BME background had helped to ensure that questions of ethnicity and inequality were promoted.
In Govanhill, Glasgow, concerns over 'community' conflict along racial and ethnic lines were also more prominent than in Newham or Cheetham Hill. For example, the local housing association comprised part of an 'Integration Network', which assembled a series of multiagency partnerships designed to assist in the settlement of asylum-seekers and refugees in Scotland. This was coordinated at a national level by the Scotland Refugee Council. Alongside this, housing association workers identified tensions directed towards and within local Roma communities. As an officer working for the local housing association explained, in the context of overcrowding, environmental degradation, and poverty, the Roma population had become a target of hostility; But in the last couple of years in particular the focus has very much been on the Roma. And there are issues with the Roma community, literacy issues, overcrowding, you have people living in substandard housing. Roma people can be on the street a lot more because they're in overcrowded housing. So people just see this kind of gathering on street corners and they see it as gangs and they see it as a threat, when the reality is it's just people trying to get a bit of space and to commune (Housing Association Office, Glasgow) The result was that in Govanhill, workers identified wider community tensions as a key concern for housing associations and central to their activities. When asked to describe the main challenges facing the area, the same housing officer summarised this as follows; I think community tensions, environment, overcrowding, community relations generally and cultural misunderstandings I think, definitely misunderstandings. (Housing Association Office, Glasgow) Indeed, the activities of one of the housing associations in Govanhill had been heavily oriented towards engaging black and minority ethnic residents. The organisation had outreach workers who spoke both Urdu and Punjabi who were able to offer advice to and engage with local residents, and who worked closely with the local registered tenants organisation with over 100 members. Here the aim was to address ethnic mix in terms of housing provision but also community governance. Through this, residents had established their own social enterprise, and educational projects had also been in operation as part of, 'A lot of one-door-shop type of support given to the BME community. So instead of just dealing with a housing problem, somebody will be taken from A to Z' (housing association worker, Govanhill, Glasgow).
In Butetown in Cardiff, the regeneration officer for a local housing association saw community engagement as central to his workparticularly the engagement of 'ethnically' or 'racially' defined communities. As a minority ethnic resident of the area, who had grown up through previous phases of urban regeneration there, he saw community consultation as key to the work his organisation was involved in. Indeed, he identified community tensions as a key challenge: There were so many issues within the communities because unfortunately what we have seen is as more community members come in, new community members, I felt personally myself that there was a breakdown within different communities. Even within the same communities, you had a breakdown of different communities because of different generations and different ideologies and different cultures and things like that, where the whole community used to be as one; because of theyou know, the current climate, people are always looking for people to blame. (Regeneration officer, housing association, Cardiff) It is worth noting here, however, that whilst the potential for conflict within the locality was identified this housing association did not follow a path advocated by cohesion agendas such as that detailed in the Casey Review (2016). Instead they came up with more creative forms of engagement, including free cinemas, film workshops and regular consultation events about changes to the neighbourhood that the housing association was involved in.
The presence of interethnic tensionsor at least the possibility for itwas alluded to in all of the local neighbourhoods according to the officers, although this assumed different textures in the specific localities. The greater prominence of the narrative of 'community tensions' in Butetown and Pollokshields/Govanhill could be associated with the more explicit focus on ethnicity in these locatities compared to Newham and Cheetham Hill where discourses of convivial diversity prevail.
In addition to differing imaginaries of diversity in the four localities, the attempts that housing providers might make in constructing 'integrated' communities were shaped by public and third sector funding cuts in the context of austerity. As mentioned above, questions of employability, blight and poverty often took precedence, as did 'enforcement' of housing codes. However, the types of services housing associations offered, and the ways they were delivered had also shifted. In Cheetham Hill, for instance, the local regeneration officer lamented the move to online rather than physical, face-to-face resources, which he saw as both exclusionary and less effective, driven as it was for him by cost saving. As he commented: We're having to work in a different way, which is about bringing in investment and residential growth and stuff like that. But I think we do do our very best to engage with and talk to people. It's at the heart of what we try and do. There might be limitations on that in terms of the resources that we have to do that. I mean, go back a few years, I spent a lot of my time out and about in communities, talking to people, going to meetings, doing stuff quite actively. And a lot of my time now is sitting at a desk because we don't have the officers anymore (Regeneration officer, Manchester).
The officer himself had moved to the organisation from elsewhere in Manchester as a result of restructuring due to cuts. In reflecting upon his previous position at a regeneration agency he stated how: We were in a team with aboutlots of different people, different disciplines, and we had about over a hundred people sitting in there. And within a space of six months, with all the first round of cuts in Manchester, we went down to fourteenjust a regen team sitting in there … our ability to work in the way we did, we just can't do that anymore. And if you lose three/four thousand colleagues over the space of a couple of years, well, all that expertise, that experience, that knowledge, those relationships that we had either with colleagues or other organisations or with those community groups, it's gone. And that's the environment we're having to work in, and we're still trying to work in that way as best we can, but it's very hard.
He lamented what he saw as an organisational incapacity to engage with larger scale projects, with time taken reacting to largely crime and environmental concerns, to the detriment of more constructive policies. The scarcity of resources was exacerbated by the increasing complexity of the diversity encountered by housing associations. Here both population churnthe speed of migration, as well as the 'diversity of diversities' (Vertovec, 2007) posed challenges to already strained organisations. As the local regeneration officer in Manchester explained, it's not easy because there are languages and, you know also, you know, you could speak to a bunch of people one week and then the next week a bunch of those people might have moved out and there will be new residents there. And it's a constant kind of cycle then A housing association officer in Pollokshields identified similar concerns regarding being able to offer provisions that could meet changing and complex demands: I mean the only thing that struck me is we used to be more angst-ridden about ethnicity, you know, aboutwell, making sure we had bilingual stuff and we had translators. But do you know what? Maybe its kind of you just get more cynical or you get lazy, but the problem is you know, that used to be primarily focused on communities from South East Asia. But when it starts to become more diverse, you kind of think to yourself, you know, at the end of the day I have to watchlike I pay for any translation services out of tenants' rent monies. So if I'm having to get Polish translators, and we've had quite a lot of Lithuanian and Latvian immigrants in the South Side, and they-so you need to go and get a Russian translator perhaps. It costs a fortune. (Housing association officer, Pollokshields, Glasgow) Here, in a context where public support for housing associations was declining, such services had to be paid for through rents generated. For the housing association officer, such funds should be reserved for questions of housing maintenance and there was a feeling that unless the public purse is going to pay for it, it's really difficult to justify … I kind of think in some respects it's very challenging to expect local agencies who have their income stream coming from service provision to be taking on that burden.
For him, this financial dilemma reflected also a policy dilemma around what constitutes the most appropriate model of integration in a context of increasing diversity.
Is the solution to beus to become more and more (pause) segmented that I need to have, you know, more translation and interpretive capacity or should we be driving a much harder, you know, integration model which tries to say, you know, we're only going towe need to all work to come in line with -I'm not expressing that -I'm expressing that quite crudely.
In summary, race and ethnicity clearly emerged as the key markers of 'communities' within the areas in which housing associations operated (see Harries et al. 2019). However, an explicit commitment to 'cohesion' and 'integration' per se varied, shaped by the particularities of places, their own reputational geographies, and the resources available to them. As we have seen, officers in Newham and Cheetham Hill, for instance, drew upon notions of these areas as 'superdiverse' to de-emphasise problems of racism and conflict, while this remained a clearer concern in Glasgow and Cardiff. More broadly, 'community engagement' was seen as key to building cohesive or integrated communities. However, the impacts of austerity, diminishing resources and widening policy remits meant that a focus on fostering 'cohesion' or 'integration'and any associated concentration on ethnic tensionsoften was subsumed within emphases on housing provision, employment, environmental concerns and 'behaviour change'.

Engagement that excludes
In this section, we consider how exclusions and marginalisations manifest in the way in which housing institutions navigate the 'engaged and integrated communities' funding and policy context. Particularly, we discuss the ways in which residents understood the roles of housing associations in shaping 'community'. Rather than being neutral arbiters in this process, public institutions were commonly viewed suspiciously, seen by residents as being not simply organisations to 'remedy' community conflict but as agents of exclusion themselves.
The experiences of many people working and resident in the four areas in which these interviews took place are shaped by a much longer history of housing and regeneration. Neighbourhoods in cities across the UK have periodically gone through major physical restructuring including mass housing clearances and regeneration of land for economic development. These changes have reshaped local populations, and this in turn has led to concerns that certain populations have been 'built out' by design (see Rhodes and Brown 2019). Taking account of the historical context has salience for the aims of this paper because it is directly related to the way in which current exclusions and 'community' itself are understood, often through the lens of race and ethnicity. Even when policy initiatives do not explicitly target matters relating to ethnicity, the historical narratives of localities are imbued with understandings of race, migration and race relations that can affect how broader contemporary community policies are enacted, received and experienced.
The way in which historical (including recent) housing and regeneration programmes affect the very notion of community is usefully illustrated in the following quote when one man describes earlier housing clearances as a means to break up the pre-existing community and disperse black people across Cardiff.
That trauma [of house clearances] broke up longstanding tribal family relationships in the area because, in order to knock a street of houses down, people had to be re-housed external to the community. And they were trying to get us to move to all the different suburbs, so we [black people] could disappear as flies in bowls of milk, so to speak. Similar kinds of changes have occurred in other post-industrial neighbourhoods, including in the other three neighbourhoods where we have done our research. These kinds of links between physical restructuring and the demise of community are made across all four sites. The links that residents draw reveal an embedded lack of trust in local authorities and housing providers and 'the establishment' more generally. They also convey a sense that housing institutions have little understanding or regard for what community means for residents. Here for example, another resident in Butetown illustrates this when he makes the link between earlier cycles of regeneration and the way that community is currently engendered: But you must understand I blame [the current lack of 'community spirit'] on the initial demolition, which was a great upheaval of the community. The community spirit, the community way of life was torn asunder by this act. A terrorist act, we call it municipal vandalism.
Housing institutions that want to go about trying to develop 'integrated communities' must first work with and sometimes against this context of mistrust and memories of exclusion. Although housing associations (i.e. as opposed to council housing) are wellregarded by the people we interviewed for the quality of their new housing, as one interviewee says, 'new houses don't necessarily make new communities'.
In the above discussion, we have outlined various ways that housing associations have extended their remit well beyond solely the provision of housing. We now see a whole variety of activities ongoing in neighbourhoods that are supported by housing associations either through direct funding, organising or by providing space for other groups and organisations to host activities. For example, as discussed above, in our four localities (as elsewhere), there is a significant focus on getting people back into work by providing space for job training and advice. Activities also include parent and toddler groups and group activities, typically targeted at women who are not working, such as cooking and sewing groups. Housing associations also host events that are intended to bring the wider community (not just their own tenants) together. These include 'love your neighbourhood' days in Cheetham Hill in which residents are encouraged to get together and 'celebrate' diversity, or community 'gatherings' in Govanhill. The main housing association in Butetown, perhaps rather uniquely, offers the opportunity for everyone to get together by putting on a free cinema night that does not have any specific 'integration' agenda, beyond showing mainstream films at no cost.
Each of these activities is, in their different ways, intended to engage communities and to encourage participation and opportunity. However, whilst many residents do attend activities, the aim of events is not always clear or is somewhat ill-matched to understandings of how communities work. Here, for example, a resident and member of a local community-based organisation in Newham describes the mismatch between events put on for the community to 'integrate' and the way in which he understands community, suggesting that rather superficial bringing together of neighbours does not tackle fundamental challenges of racism: Communities, individuals don't work like that. So sharing a cupcake with somebody along the street party doesn't make you friends for life, doesn't mean you still don't have racist views or you still learn to speak English any better … . This is a view sometimes shared by ground-level housing officers. A group of housing officers in Cheetham Hill, for example, describe how community engagement events are often conceived as nothing more than 'box ticking exercises' because there's little activity beyond the event itself. To an extent we have to question whether they are driven by a general anxiety around migration and diversity and political pressure to 'be seen to be doing something' about 'integration', and about resentments emerging from measures of austerity and the 'housing crisis'. Such drivers are inferred by this housing worker: We've a hub which brings the police and social workers and housing and all that together. They do some good work and they're pretty good, but … It's been more abouta typical political response about let's be seen to be doing something rather than addressing it.
From an institutional perspective, across all four sites, it is not unusual to hear people who work in housing (and indeed other public services) complain that residents are not interested in their communities because they do not turn up to events, at least not 'unless you put food on'. In these interviews, there is seldom reflection on the event itself and whether it serves a purpose of interest to local people. With little consultation around the organising of events and with little follow-up activity beyond the event itself it is perhaps not surprising that residents are suspicious about the role played by housing associations in trying to engage community. This can backfire with more serious consequences when events are organised without proper engagement from a broad section of the community. This was illustrated in 2013 when a housing association in Glasgow facilitated an event that was requested by a relatively small group of long-term and predominantly white residents. The 'gathering' itself gave voice to a very particular group with an anti-immigration agenda. This was particularly troubling in the local context in which many Roma and people from various parts of central and Eastern Europe live and experience racism. The voice of the small group was not representative of the broader section of the community as one resident who attended explains: I just sort of walked in and it was like there's two hundred people here and a hundred and ninety five of them are white. How's this a Govanhill gathering? And there was about three Asian people and then the other two was this older Romanian Roma man and the lass who'd come to translate for them, and that was it. And it was justit was horrific.
It is certainly common for public sector institutions to conceive of integration and diversity celebrations as having priority in neighbourhoods that are stigmatised and to problematise diverse neighbourhoods in ways that do not always resonate with residents. This is perhaps most clearly illustrated by the way in which many of these 'integration' activities are really about behaviour change. This has the effect of problematising local residents and seeking to solve these problems in ways that risk further stigmatisation and marginalisation. Here, for example, a housing association representative explains: We want to be seen as a community based organisation. So it's not about always people coming to us. We [are] launching [a campaign]. So that's a range of activities … and that's looking at really to promote behaviour change … Raising awareness around looking at the appearance of the neighbourhood and the environment. Recycling, energy efficiency, fire safety, being neighbourly and a range of activities but with the -not a hidden message as such, but educational as well. So trying to make it fun and interactive and promote tenant engagement but trying to look at changing behaviour.
The quote above illustrates the way in which housing associations can often conceive the people who live in the neighbourhoods in which they work. Their emphasis is on making people more neighbourly and environmentally responsible, which of course works on the assumption that people are not neighbourly, nor environmentally responsible. What is more, often the activities and events that are organised are not really open to everyone. Often they are targeted towards families, and perhaps more specifically, explicitly or not, to mothers. Furthermore, the kind of activities that are organised do not necessarily reflect local interests or priorities. Here a housing worker admits there is often a problem with a mismatch of interests: I think a lot of theand this is a very controversial thing for [me] to saybut I think a lot of our community activities are very middleclass as well. I think a lot of it's about lifestyle expressions, about growing your own food and about shopping local. A lot of is … It's a bit show offy, you know. There's a lot of people who are bit up their arse here, who kind of want to be seen to be living an equal lifestyle and having a baby in a sling and stuff like that. It's just all a bit arsey, you know (Laughs). Important to bear in mind is that all four of the localities in which we have based this research have long been stigmatised places; stigmatisation that is typically articulated through race and class. The decision-making behind prioritising lifestyle and behaviour change activities, therefore, has to be understood in this context. There is firstly an assumption that without these activities there is no 'community' and that local residents themselves do not operate as a community. The embedded lack of trust residents feel towards 'the establishment' is directed generally towards economic regeneration plans in which discussions about housing are incorporated. For example, one woman in Cheetham Hill told how she and her partner had been forced out of their first home together under a compulsory purchase order only to then struggle to find alternative affordable housing because they did not receive payment for their house for 2-3years. On finishing this story, we commented on how she must have 'witnessed quite a bit of regeneration'. In response she pulled her face awkwardly and so we asked: Would you call it regeneration or how would you describe the changes to the kind of physical environment?
Disgusted with the changes actually, there is nothing … . You haven't really got, you haven't got a shoe shop, fruit and veg place unless you go into the big supermarkets. Lots of people don't like big supermarkets for fruit and veg. I no longer shop up here.
'Regeneration' is not a concept that is positively regarded by residents in the localities, especially when it is seen as something done to a community, rather than benefiting a community. This is a topic that recurs frequently across the four sites but perhaps most poignantly in Newham and Butetown where the landscape has changed most significantly. Residents expressed the suspicion that the ways in which housing providers have arrived to fill new housing gaps and are seeking to shape communities through particular activities and emphasise addressing worklessness reflect a scheme to 'build out' less desirable sections of the population, including the racialised and the poor. This is more keenly felt in Newham where housing allocation policies include employment, community contribution and residency requirements. Here a resident and community worker in Newham explains: It's like we can't attract business if the poor are there [mimicking voice of local authority]. I don't think stereotyping -I mean, really, it is stereotyping, but I think the same rationale applies to social housing. It's less extreme, but if there is a move to bring in business, then they'll not want the poor there.
There is then a sense that many of the decisions around housing and the reshaping of communities and what it means to be an integrated community is a purposeful intervention to further fragment communities. Here a Butetown resident describes how he thinks that this will allow housing institutions to continue to make changes with increasing ease and less protest from local residents.
Why do you think [housing associations] do that then?
Because they can control … When there's no community, there is no demonstration. There is no petition. There is no voice. So they know that, but the people don't know that. They don't realise that sticking as a community, being a community, being members of a community where you live and you grow and you integrate with each other, you communicate with each other every single day, no matter what religion or race you are, this is a community and we help each other.

Discussion and conclusion
This paper has been concerned with how organisations with responsibility for social housing respond to cohesion, integration and engagement agendas, how this represents certain ideas of community and how this engenders processes of marginalisation. We have drawn upon qualitative case study data from four diverse localities: Cheetham Hill (Manchester), Newham (London), Butetown (Cardiff) and Pollokshields and Govanhill (Glasgow). We draw three conclusions from our analyses: race and ethnicity as facets of 'integration' have been subsumed into broader 'integration' agendas yet issues of race remain implicit if not explicit; housing organisation practices are often detached from local meanings of community and thus stigmatise rather than bond; the roles of housing organisations in constructing 'integrated' communities are highly variable and localised, influenced by the history and contemporary dynamics of place.
The analyses have shown that in a context of resource cuts and a shifting remit of housing organisations towards generic practices of community engagement (Robinson 2007;Beider 2012), race and ethnicity have been subsumed into broader agendas. Although housing associations retain a responsibility for creating and engaging diverse communities, the language of 'cohesion' is absent. Yet, the work being undertaken aligns with integration policy narratives, with an emphasis on behavioural change such as reducing anti-social behaviour, learning English, engaging in employment (Cantle 2001;COIC 2007;Casey 2016). However, this is not to say that race and ethnicity have become irrelevant to integration initiatives or the work of housing organisations, but that they operate in more implicit than explicit ways, through for example, what and whose behaviour is defined as needing to be changed.
This emphasis on initiatives of behaviour change in the operation of housing organisations as agents of 'integration' caused concern to residents who felt that housing organisation practices were often detached from local meanings of community. The activities focusing on behaviour change were seen to stigmatise people and places, undermine the agency of local residents and, through the prioritisation of certain events and activities, delimit the boundaries of community inclusion. Frustrations were also evident at the limited and isolated effect of 'integration' initiatives which were often seen as misaligned to longstanding understandings and practices of community; and superficial attempts to create bonds rather than meaningful actions to address fundamental problems including racism.
However, it would be creating a false dichotomy to suggest a clean opposition of the views of residents and those of housing organisation workers towards community 'integration'. Indeed, residents were welcoming of some approaches and initiatives and housing workers were critically self-reflective on their 'uncomfortable positions' (Jones 2013) and how they reconcile personal and organisations ideals with institutional possibilities constrained by political and economic context.
The situated, spatially and temporally embedded nature of housing providers' role in creating 'integrated' communities was clearly evident from this study, following Flint (2006), Jones (2013) and McKee's (2015) arguments. This reflects the localisation of race relations responsibilities (Kalra and Kapoor 2009;Beider 2012) and illustrates Thomas's (2011, 93) point that 'policy approaches to race relations and ethnic diversity have always been 'worked out' at the local level'. For example, although the possibility for conflict was an undercurrent in all four localities there was variation in the degree and manner to which responses followed cohesion and integration agendas. In Cheetham Hill and Newham, for example, we have illustrated how discourses of convivial multiculture downplayed the need for attention to be explicitly placed on ethnicity; in contrast, in Butetown, Pollockshields and Govanhill engagement of minorities in community governance was actively pursued. To understand the reasons for these locational differences in the significance of race in integration initiatives of social housing organisations we have suggested the importance of differing historically embedded imaginaries of local diversity and public and third sector funding cuts that have reduced the operational capacity of housing organisations.
We can draw from these findings broader implications for understandings of ethnicity and place. The question of how the agenda to integrate or engage communities conceptualises fairness (or equality) is raised, requiring us to consider whose vision of community the national agendas encapsulate. The policies place emphasis on behaviour and societal contribution; we hear from neighbourhoods more about desire for social support, recognition of shared history and investment in local places. Initiatives focusing on the former risk superficial and short-term impact if they are misaligned with the lived concepts and practices of local community.
In the relative recent neglect of race in housing studies, and housing in race studies (Bloch, Neal, and Solomos 2013;Markkanen and Harrison 2013) and the rise of 'integration' agendas (Cantle 2001;COIC 2007;Casey 2016) the emphasis on (individual) behaviour has been at the expense of attention to matters material. Yet, this study has shown that the material neighbourhood matters for how people live and shape community; the material circumstances of individuals matter for their well-being; the material matters for defining the role of housing organisations whose baseline priority is meeting the basic needs of the impoverished. In common with Harries, Byrne, Rhodes, and Wallace (2019) and Rhodes and Brown (2019), we conclude that it is the poor minorities, and their places, that are particularly excluded, materially and otherwise, in housing organisations' (and others') practices of integration.
The emphasis on integration as behaviour has also been at the expense of discussions of racism. There is a tension in racial and ethnic difference being acknowledged as the undertow to integration yet a reluctance to place race and ethnicity centrally within solutions, which prefer to talk about being economically active and keeping neighbourhoods tidy. Yet, in these priorities the tropes of racism are alive, as they draw on associations between blackness, (un)deservingness and dirt.
Housing providers continue to be important actors in supporting and creating communities. How they negotiate their position in enacting initiatives within dominant policy and funding narratives and responding to local populations warrants further interrogation because it has implications for who is represented and who is marginalised and stigmatised, and for how race is (or is not) present in constructions of community.