Energy vulnerability in multiple occupancy housing: a problem that policy forgot

Housing in Multiple Occupancy (HMO) includes some of the UK’s worst housing stock. The tenants typically have reduced housing and social security rights, as well as reduced control and sometimes absence of basic domestic energy services. Yet, HMO is largely absent from UK policies governing energy efficiency and fuel poverty. Energy vulnerability in HMO is exacerbated by a lack of representation and recognition of HMO in official systems and statistics on the housing stock, socio-cultural preferences for traditional single-family homes, and widespread tenure prejudice based on negative imagery about HMO. An indicative typology of HMO is proposed based on five main tenancy scenarios: illegal/informal; rooms in a shared house; group of sharers; bedsits; and poorly converted flats. The purpose is to improve the recognition and inclusion of the variegated HMO sector in policy frameworks and public debates in a more nuanced fashion.


Introduction
Shared and sub-divided dwellings in the private rented sector (PRS), ranging from very informal to semi-institutional settings, are omnipresent in urban areas in the UK where the PRS is a route into affordable shelter.Statistics on Housing in Multiple Occupancy (HMO) are weak, but the UK Census implies that one to three per cent of all dwellings in England are in multiple occupancy.The figure rises to concentrations of up to 15 per cent in some areas, for example some London boroughs -a figure set to rise if current trends continue.Given its pervasiveness, contemporary energy efficiency and fuel poverty policies are strangely silent about HMO.This paper attempts to answer why the silence exists by exploring cultural perceptions of housing and tenure prejudice, beginning with the notion that "attitudes towards housing are the product of myriad social constructions" (Rowlands and Gurney, 2000: 122).Cultural messaging about housing is dominated by the idea of the English as "a nation of homeowners" (Saunders, 1990).The hegemonic media obsession and political rhetoric on homeownership and house prices mirror the middle class cultural preferences for single family owner occupancy and housing seen as a financial investment.Meanwhile, the private rented sector (PRS) is expanding owing to the decline of social housing and unaffordability of owner occupancy, all of which point to ideological support for the "transfer of UK home ownership to the investment sector" (Field, 2014: 354).Field (ibid) argues that the manner in which the PRS is expanding is creating a more socially and economically divisive housing market -arguably the growth of the HMO sector is symptomatic of such conditions.Indeed, Smith (2012: 461) posits that "the proliferation of HMO is one of the leading ways that British housing markets have been transformed over the last two decades." It can be argued that the combination of the UK housing market characterised by 'liberal market economy' (Kemp and Kofner, 2010), huge asset inequality and housing affordability crisis (Dorling, 2014), and the ongoing direction of housing and social security policy will continue to funnel more low income and disadvantaged households into the HMO sector.UK social security reforms, including those contained in the Localism Act 2011 and the Welfare Reform Act 2012, are part of this trend.The former gives local authorities the right to discharge their homeless duty via the PRS without the tenant's consent, and the latter introduced the so called 'bedroom tax' in the social rented sector which is likely to increase demand for smaller (and cheaper) privately let units.People aged under 35 now only receive housing benefit at the shared room rate, and the 'benefit cap' from April 2016 restricts the weekly support to just £296 in London (down from £350), £258 in the rest of the country, for single persons (Shelter 2015).Single person households make up almost a third of all households, and rental costs in London and in many urban centres suggest that the benefit cap is poised to increase demand for HMO, especially as the social rented sector is declining.Other factors likely to inflate HMO demand are both increasing homelessness and a rise in local authorities encouraging 'informal housing options' to reduce statutory homelessness applications (Fitzpatrick et al., 2015).In the Summer Budget 2015 (HM Treasury, 2015), the government announced that Housing Benefit entitlement of 18-21 year olds will be scrapped; leading homeless charities have warned that this will further increase homelessness among young people.The increase in net migration and asylum applications compared with recent years (ONS, 2015) is likely to result in further formal and informal HMO development (Pemberton, 2009;Perry, 2012;Robinson, 2010).Additionally, the ONS reported that "multi-family households" was the fastest growing household type in [2003][2004][2005][2006][2007][2008][2009][2010][2011][2012][2013], with an increase of 39 per cent, albeit from a low base (ONS, 2013).
In summary, the need for shared and sub-divided privately rented accommodation is likely to increase further.To remain fit for purpose, energy efficiency and fuel poverty policies should catch up with the transformations in the housing market.
Where domestic energy efficiency policies are targeted at low income households, they generally aim to empower residents to lead more healthful lives by allowing internal temperature gains in fuel poor households (Milne and Boardman, 2000).Currently these policies almost entirely overlook HMO; our intention is to identify potential reasons for why this might be.In the UK context, fuel poverty debates are broadly focused on "the poor affordability of energy for space heating (and other related domestic services) as a result of low household incomes or energy inefficient homes" (Bouzarovski and Petrova, 2015: 32).The concept of energy vulnerability is probabilistic.Whereas fuel poverty describes a state where a person is unable to obtain adequate energy services, energy vulnerability aims to draw attention to the "factors that affect the likelihood of becoming poor" (ibid: 35).In HMO, domestic energy deprivation may be a result of different factors than those identified in the traditional fuel poverty paradigm.The way in which fuel poverty is defined (Hills, 2012) is not always applicable to HMO; energy costs can be part of the rent, rental agreements can be non-existent or illegal, or energy costs can be shared between different households sharing an energy service such as heating.In HMO, reduced autonomy over energy services is commonplace due to social negotiating of energy use, and lack of access to controls, such as boilers, radiator valves or windows that can be opened to ventilate.Additionally, there are HMO dwellings where there are no basic domestic energy services available.As a result of these complex factors, HMO residents are likely to experience energy vulnerability that goes unnoticed and unrecognised by public authorities, official statistics and research efforts on fuel poverty.This paper addresses the gap in literature.
We explore inequality experienced by HMO residents via the cultural dominance of the single-family home lifestyles of the English bourgeoisie that is upheld by policy and regulatory frameworks.These cultural factors go over and above the well-recognised material deprivation and poor conditions that are endemic in the lower end of the HMO market.Our purpose is to highlight how the English policy and legislative mechanisms fail to address energy vulnerability in HMO, and what kind of socio-cultural conditions are conducive to this.To understand how this ignorance has developed in England, we deploy Bourdieusian concepts including symbolic power, middle class taste and 'tenure prejudice' (Rowlands and Gurney, 2000).Finally, we propose a tentative typology of HMO with a practical purpose of illustrating how professionals and the public could begin to have a more positive and nuanced discourse about the role and future of this highly variegated housing type.

Methods
The research was carried out in September 2013-April 2014, and its main components were a policy review, followed by 21 interviews with a range of representatives from government departments, local authorities, letting agents, professional bodies, national charities, as well as tenant and landlord representatives.Additionally, a range of stakeholders including the Chartered Institute of Environmental Health, Property Energy Professionals Association, Department for Energy and Climate Change, Department for Communities and Local Government, Building Research Establishment, Association for the Conservation of Energy and National Union of Students were consulted about the research and its findings.A full list of all consultees can be found in the technical report associated with the research (Viitanen and Weatherall, 2014).
England as a nation has an advanced system of statistical data starting with the UK Census, and specialised housing statistics are made available to researchers interested in housing conditions: the English Housing Survey, live tables, and a HMO database held by DCLG (although the latter is no longer updated).The results of consulting these statistics on the quantity, geographical spread and quality of HMO buildings are reported in full in the technical report (Viitanen and Weatherall, 2014).The Census results are briefly summarised below.

Identifying HMO
A substantial underrepresentation of HMO in English housing datasets was discovered as part of this research.The reasons are two-fold.Firstly, generic population and housing surveys tend to ignore or represent HMO inadequately in the survey design and sampling -the latter is the case for example with the English Housing Survey.Secondly, HMO households are more easily missed in population surveys such as the Census (Simpson and Middleton, 1997).This is partly explained because both landlords and tenants may benefit from informality and thus want to avoid detection by authorities (Layard, 2012).With these caveats in mind, 2011 Census data suggests that nationally between one and three per cent of English dwellings appear to be HMO, with 0.1 per cent being 'bedsits.'In terms of households, the Census question on "multi-person households" records 120,870 (0.5 per cent) full-time student households, and 666,810 (three per cent) 'other multi-person households' (unrelated adults sharing, excluding bedsits) in England.
In terms of people living in different types of HMO, the number of dwellings and households suggests that this is certainly in the millions.At a neighbourhood level, and on a street-by-street basis, concentrations of HMO can be much higher than the national and local averages imply.For illegal/informal shared housing and poorly converted self-contained flats, there are no figures available upon which to base national estimates.Our interviews and media reports suggest that some localities have high concentrations; e.g. the leader of Ealing Council has been quoted in the media to estimate that 60,000 residents in the borough could live in "beds in sheds 1 " (Express, 2015;New Statesman, 2012).
Typical HMO tenants include students (Hubbard, 2008;Rugg et al., 2002;Smith, 2008;Smith and Hubbard, 2014), young people (Ford et al., 2002;Lister, 2006;Rugg and Rhodes, 2008), vulnerable individuals including persons newly released from prison, young people leaving the care system, and people with mental health or substance misuse problems (Barratt et al., 2012 and2013;Rickley and Houghton, 2009;Smith, 2012).Since the 1980s, private sector HMO has been the preferred tenure for housing homeless people from a housing policy perspective (Stewart, 1999).Faced with high uncertainty and housing insecurity, new migrants often live in HMO (Pemberton, 2009;Perry, 2012;Robinson, 2010), and asylum seekers are placed in HMO under Home Office sub-contracts (NAO, 2014).Multi-occupancy residents in the PRS have characteristically fewer housing and social security rights, including limited or no housing benefit entitlement and limited eligibility for social housing as well as lack of choice (Ford, 2002;Kemp, 2011;Lister, 2006;Pemberton, 2009;Robinson, 2010), compared with the population living in single-family homes.Limited choice also applies to people who have steady jobs and good social status ("young professionals"), but the local housing costs are prohibitive.In the English context this is typical of the London housing market; Smith (2012: 464) estimates that almost a third of all HMO in the country are in the London boroughs.In terms of HMO geography, they are omnipresent in urban areas, but research suggests that certain places have a prevalence due to underlying social and economic trends; apart from London, university towns, coastal towns, and areas with high concentrations of migrants (Smith and Hubbard, 2014;Smith, 2012;Perry, 2012).

Social reproduction of inequality in multi-occupancy housing
Although the data is patchy, 'HMO' is a well-recognised label conferred by a number of legal and social constructs that involve the exercise of state and civic powers, such as Licensing and Planning authorities, the 'National HMO Network' -a civic campaign against HMO development, and the media debates surrounding HMO.This absence of HMO in official statistics, yet its strong discursive presence, is conducive to research that takes guidance from cultural and social symbols.
To understand cultural messaging in the housing context, we deploy a Bourdieusian framework.Previous applications of Bourdieu in UK housing research include Rowlands and Gurney (2000) on 'tenure prejudice, ' Flint and Rowlands (2003) on housing commodification and symbolic capital, and Sturzaker and Shucksmith (2011) on discursive power and symbolic violence in planning for rural housebuilding.This body of work points to the cultural and symbolic significance of housing as a commodified good in the UK, as well as the role of professional systems in representing and validating middle class tastes.The cultural stronghold of the bourgeoisie on what is preferable and normal, as well as the role of institutionalised power in social 'labelling' (Bourdieu, 1987) is foregrounded in this analysis.Symbolic power is understood as "the power to impose and to inculcate a vision of divisions, that is, the power to make visible and explicit social divisions that are implicit" (ibid: 23).According to Rowlands and Gurney (2000), tenure prejudice consists of real and fictitious images pertaining to certain housing tenures.The socialization of housing suggests that these images and attitudes towards housing are adopted through cultural messages and symbols (ibid).Negative perceptions and images of HMO are abundant in policies adopted by local authorities that aim to control and contain multifarious social, cultural, economic and environmental ills associated with HMO.Examples include anti-social behaviour, benefit dependency, litter and poor external maintenance of buildings (Layard, 2012;Smith, 2012;Rickley and Houghton, 2009), or "illegal immigration" (DCLG, 2012) and "studentification" -a term which is "used pejoratively to signify community destabilisation and decline" (Smith and Hubbard, 2014: 2).
HMO and the 'keepers of official culture' Shared (multi-occupancy) dwellings are an enduring and historically important low-rent facility in cities around the world.Studying North American cities, Groth (1999: ix) explains that official culture has taken a dislike to HMO: "For two hundred years, hotels have served a series of domestic roles in urban vernacular environments and subcultures, for at least one hundred years, the keepers of official culture have aimed at eliminating these roles."In a similar vein, researching the history of lodging houses in England, Seaton (2005: 1) writes: "Official reports and journalistic writing paint pictures of dirt, squalor, immorality, poverty and crime.Both the public and professionals perceived them as most undesirable places."Seaton's account highlights how the public perception of lodging houses were fueled by media reports as well as official enquiries into their condition, often focusing on the immorality [sic] of the inmates -pointing to tenure prejudice in the Victorian society.Yet, Seaton (2005: 97) concludes that "the biggest problem with Nottingham lodging-houses was that they were in dilapidated buildings that were unfit for human inhabitation due to defectiveness, not dirt and disease."Our research focuses on a comparable situation in England today.It is undisputed that the standard of HMO buildings is often poor, but there is reluctance on behalf of the regulators to act on the grounds of energy efficiency or fuel poverty.This could be partly explained by tenure prejudice -negative images about HMO tenants and their lifestyles or life situations.
The residential environments of HMO are "downgraded" (Smith, 2012).Available evidence, including the English Housing Survey, interviews with Environmental Health Officers (EHO)and licensing officers, points to very poor thermal efficiency and antiquated heating systems in buildings that are typically HMO.Yet, many of our interviewees did not recognise energy issues being important in HMO, in their opinion there were "bigger problems to worry about" (local government licensing manager, October 2013), such as substance misuse or other social ills -in other words, characteristics, real or imagined, associated with the occupants.Another interviewee expressed the same sentiment, with the term "neighbourhood issues" here referring to social problems, such as litter and anti-social behaviour, associated with HMO: "In terms of issues, I have to say, energy efficiency in HMOs was never a particularly high priority, there were many more pressing issues on HMOs, in terms of enforcement, repair, management, health and safety, issues in terms of neighbourhood issues emanating from HMOs" (former director of housing services, local government, February 2014).Many interviewees felt that HMO tenants themselves would not prioritise energy efficiency or affordable warmth due to their "chaotic" or "challenging" life situations.This, one interviewee suggested, was evidenced by the lack of complaints from HMO tenants to the local authority -whereas they often received complaints from 'mainstream' (single-family) PRS tenants about energy related issues, such as cold, mould or damp.Among the interviewees there was a reluctance to acknowledge that energy efficiency standards are something worth pursuing in HMO.Indeed, research with 'bedsit' residents in one English coastal town shows that for residents, personal safety is a concern, along with social stigma, lack of control and personal efficacy in this housing type (Barratt et al., 2012 and2013).Living in HMO for some represents an accumulation of multifarious disempowering circumstances, therefore discomfort due to inadequate energy services, or lack of control over them, might not be a priority.Social stigma and lack of personal efficacy referred to above may also deter HMO tenants from complaining to the local authority.However, these perceptions overlook the cumulative effects of cold homes and energy deprivation on physical and mental wellbeing, which may be confounding factors in quality of life in HMO.It is known that the HMO housing stock is extremely energy inefficient, but there is a lack of research on the impacts on the residents.However, wider research on fuel poverty has shown the detrimental impact of cold homes on quality of life, especially on adult and adolescent mental health (Liddell and Morris, 2010).More specifically, the National Union of Students (NUS) campaign on the poor energy efficiency and related problems in PRS student homes, has published research that highlights the impact of cold homes on quality of life (NUS, 2014).More generally, discussing the role of housing in wellbeing, Wade and Dixon's (2006: 203) study shows that with young people leaving the care system, "housing emerged as the life area most closely associated with mental well-being, outstripping the contribution made by involvement in education and training".There is ample evidence to suggest that housing is a key factor in the improvement or deterioration in social outcomes for young people, as well as homeless people (Warnes et al., 2013), both are groups likely to live in HMO.However, there are perceptions about HMO in the social hierarchy of housing that undermine the recognition of energy deprivation in HMO as a serious issue.
English regulatory approaches to HMO HMO refers to shared houses or flats, sub-divided dwellings including bedsits, hostels, and poorly converted flats in the PRS.The official definition of HMO varies between different policies and regulations, the main frameworks discussed here are to do with housing and planning.Rather than detailing each HMO definition, it is more insightful to look into the purpose of each piece of regulation.The main purpose of the Housing Act 2004 is to contain and manage risk associated with poor quality housing.The risk is mainly related to poor management and maintenance, and detailed guidance called "Housing Health and Safety Rating System" (HHSRS) was introduced to identify 29 different health hazards (Stewart, 2002), which include energy-related excess cold, damp and mould.However, our interviews suggest that in practice, Environmental Health Officers (EHO) are mainly focused on pragmatic collaboration with HMO landlords, enforcement being a rare, final option.Most local authority interviewees were concerned about the capacity of EHOs, compared with the (large) size and low quality of the HMO sector."When it comes to HMOs, we're just doing the paperwork now, health and safety is the only thing.We used to assess HMOs for cold and damp.They [HMO licensing enforcement] did surveys on cold and damp, and passed it back to us, and we'd try to help that person" (energy assessment officer, local authority, January 2014).Separate from general HHSRS regulations, large HMO units are subject to mandatory licensing schemes run by local authorities by virtue of the 2004 Housing Act.In contrast to the HHSRS, this part of the Act places emphasis on risk of death from fires in large HMO units.Peculiarly, the Act specifically blocks the opportunity to use the HHSRS as an HMO licensing condition, thus diverting attention away from the more fulsome "health hazards" identified in the HHSRS.Instead, fire safety and "basic standards of upkeep" have become a central focus of the mandatory licensing process as required by the 2004 Act, as also shown in the interview quote above.This regulatory approach leaves HMO tenants at a significant risk of below-standard housing with no effective recourse through the licensing route in most local authority areas.It also highlights a culture of acceptance that HMO naturally comes with low standards, and that little can be done to improve it -a form of symbolic violence.It does not have to be so, however, notable exceptions were discovered where local authorities had introduced voluntary measures to deal with poor quality HMO in their areas, including in Bath and North East Somerset, Haringey, Greenwich, Camden and Nottingham, using mechanisms such as additional licensing which local authorities have powers to adopt (see Viitanen and Weatherall, 2014 for full details).
English planning laws struggle with a reputation for red tape on the one hand, and responding to community concerns on the other.Former 'use class' restrictions on HMO were lifted and landlords can usually convert ordinary dwelling houses into HMO under permitted development rights (House of Commons Library, 2013).In response to community campaigns against HMO development, Article 4 Directions were introduced under the Localism Act 2011, which local planning authorities can deploy to remove these permitted development rights (Layard, 2012).This regulation is effective in demonstrating how established (single-family housed) residents are able to reject multi-occupancy residents on the basis of perceived lifestyle: "owner occupiers, families, individuals and couples are allowed to move into the locality, while unrelated sharers are not" (Layard, 2012: 556).This highlights the complicity of state power in catering for the tastes of the dominant social group.
Previous research highlights that migrants to the UK can be housed without appropriate tenancy agreements in constructions such as garden sheds, shipping containers, or in PRS accommodation that is sub-let to other migrants to reduce costs (Perry, 2012).The Home Office and DCLG launched a joint programme to target "rogue landlords" and "illegal immigrants" through policy guidance and a £3m fund aimed at local authorities with the most pronounced problems of informal housing known as "beds in sheds" (DCLG, 2012).Many "beds in sheds" are de facto HMO, characterised by non self-contained facilities and multiple households sharing.Related media coverage often focuses on "illegal immigration", such as Channel 4 News (22/5/2015) asking "is illegal immigration getting out of hand?", following the Prime Minister's visit to the London Borough of Ealing in the wake of the May 2015 election.Referring to the same district, the Express (3/10/2015) ran a headline "Evil, squalid slums -what immigration is doing to Britain".Robinson (2010) has shown that negative stories about immigrants and housing are frequently used for political purposes in the UK.The "beds in sheds" policy has been criticised by the Migrants' Rights Network (2013) for its thinly veiled revanchist tendencies of highly publicised "dawn raids" with little thought given to improving the often very poor quality rented accommodation of those who are in the country legally.The case of "beds in sheds" shows how the issue of migrant workers sharing squalid and overcrowded accommodation is marshalled towards the populist anti-immigration message, rather than -for example -the lack of affordable housing in certain parts of Britain that affects everyone on low and middle incomes regardless of background or migration status.

Energy efficiency and fuel poverty: legitimate loopholes
The dominant fuel poverty discourse identifies specific social groups; infants, the elderly, disabled people and those with a long-term sickness, as vulnerable to fuel poverty (Hills, 2012).Such popular imageries of the fuel poor are culturally influential.Deprivation is high also among young people and students who regularly live in HMO but are rarely recognised as a group vulnerable to fuel poverty (Bouzarovski et al., 2013).The same could be said of migrants, homeless people, and asylum seekers.HMO tenants often have reduced autonomy over basic energy services, such as heating or hot water, compared with single-family occupants.People staying in "beds in sheds" may not have access to basic domestic energy services.
The previous English policy definition of fuel poverty considered a household to be in fuel poverty where 10 per cent or more of household income goes towards energy bills; Hills (2012) suggests that the key elements instead are a combination of low incomes and requirements for high energy expenditure.Hills (2012: 14) highlights that: "90 per cent of the fuel poverty gap is accounted for by households with low incomes also living in homes that have energy ratings of E, F and G" -indeed the Hills report concludes that high energy costs in their model generally reflect building energy inefficiency (Hills, 2012: 33, emphasis added).The focus on energy (in)efficiency of the building is very useful, but it overlooks almost the entirety of multiple occupancy housing stock, as the efficiency of this housing stock is largely unknown, and the existing legislation does not require for it to be made known.
Since 2008, an EPC has had to be made available free of charge to any prospective tenant as part of the rental process and advertising in England.However, these rules do not apply to rooms let individually in shared houses (HMO type 2 -see Table 2 overleaf) or bedsits (HMO type 4).Currently, EPCs are required only where a shared HMO dwelling is let as a whole property to a group of sharers with one tenancy contract (HMO type 3), or in the case of poorly converted self-contained flats, classed as "section 257 HMO" (section 257 of the 2004 Housing Act, referring to non-compliance with building regulations -HMO type 5).The reason that the UK government cites for the exclusion of most HMO is that the European Energy Performance of Buildings Directive defines building units as "a section, floor or apartment within a building which is designed or altered to be used separately" (EP, 2010), which allows non selfcontained dwellings to be excluded.A consultation put out by a previous government (2005)(2006)(2007)(2008)(2009)(2010) proposed extending this EPC requirement to HMO (CLG, 2010a: 21) received widespread support.The subsequent government (2010)(2011)(2012)(2013)(2014)(2015) in its response to the consultation stated that EPCs for HMO unit lettings would not be taken forward as this amounted to "goldplating" the European legislation (CLG, 2010b).The omission of energy assessments in HMO have implications for understanding housing conditions and their effects on health, safety and sustainability.Keall et al (2010) outline a number of positive relationships emanating from an effective assessment mechanism, including a positive feedback loop between research, policy and frontline practice on housing conditions.Given the housing and health concerns associated with HMO, there is a case to be made for the adoption of an effective energy assessment mechanism.
The Energy Act 2011, as well as fuel poverty frameworks, relate in various ways to the existing requirements for homes to have an EPC.The PRS minimum energy efficiency standard implied in the Energy Act 2011 would not currently apply to most HMO because there is no point at which an EPC is required.This loophole effectively allows extremely inefficient HMO dwellings to be let, which would not be allowed in the rest of the PRS.Additionally, the method used to produce EPCs in HMO is a source of confusion.In our research, examples of differing advice and practices were discovered across the energy assessment and local government sectors.EHOs and some domestic energy assessors relied on RdSAP 2 , the energy assessment for domestic dwellings.However, many domestic energy assessment industry representatives argued that SBEM, 3 the non-domestic methodology, is the appropriate methodology for some types of HMO.The difficulty arises from communal spaces that are not a feature of standard single-family dwellings and not included in the RdSAP method.In the typology overleaf, an energy assessment method is proposed against each corresponding HMO type, based on the feedback from stakeholder engagement.If the EPC is provided in a format designed for a business audience (SBEM), further thought needs to be given to how it might be made more relevant for HMO tenants.

HMO typology
The HMO typology highlights how the energy practices and built environments of HMO depart from the energy industry's standard assumptions about 'energy customers', and the pathways to vulnerability recognised by mainstream fuel poverty policy.The typology aims to facilitate practical understanding of how energy efficiency policy and enforcement could be tailored for HMO.The typology was developed by a scenario method by engaging stakeholders, and it should not be taken as an exhaustive list.The typology, laid out in Table 2, brings together five primary types of HMO based on the main tenancy arrangements: 1) illegal/informal; 2) rooms in a shared house; 3) group of sharers; 4) bedsits; and 5) poorly converted flats.These types were chosen as the tenancy scenario described has a direct influence on a number of regulations, for example the Housing Act 2004, the EPC requirement, and the proposed methodology (SAP/SBEM) for producing an EPC.
The typology expands on the primary scenario 1-5 by considering the following energy characteristics described in Table 1 below.

Pathway to enforcement/ improvement
The main point(s) of contact for authorities where energy efficiency could be taken into consideration.
Illustration of numbers in each HMO type Local and/or national estimates, based on available data sources.
These characteristics allow an assessment of energy vulnerability factors, notably two types have slightly better existing rights that the rest; "group of sharers" and "poorly converted flats" currently are entitled to EPCs at the point of rental.The additional characteristics in Table 1 allow a more detailed assessment about the domestic energy arrangements -by identifying who holds the utility contract (indeed if there is one), and how energy usage is paid for by the occupants.The building typology reveals how the social arrangements relate to the built environment, especially with respect to whether the units are partially self-contained or not.The pathways to improvement identify the points of contact where energy efficiency could be taken into consideration, and the final row provides numbers of estimated HMO units either nationally or in specific localities that were discovered during the research.

Pathways to improvement
Lobbying for changes in the regulatory and policy frameworks governing energy efficiency in buildings is a key message arising from this research.The direction of UK central government policy looks at present unlikely to prioritise any domestic energy efficiency programmes after scrapping the only support mechanism for retrofit in domestic properties, the Green Deal, without plans for a replacement, as well as dropping the zero carbon homes target which was supposed to start from 2016 for new build properties, and cutting back support for solar PV feed-in-tariffs.However, a number of possibilities still exist under the current regulatory framework to improve the HMO stock in England.Additional use of HMO licensing powers offers an obvious route -this can be justified if locally there is evidence of high concentration of HMO in poor condition.For large HMO units subject to mandatory licensing (HMO type 4), a minimum EPC standard can be included as a condition of licence renewal, with landlords given two years to bring properties up to the standard.Implicit in this is that an EPC should be a licence condition.Systematic programmes of inspection, grants, support and voluntary accreditation of landlords can be used alongside licensing to promote improvement in the sector.Particularly important is that local authorities feel confident in making full use of housing health and safety powers, for example, taking robust action in requiring insulation and lower cost heating systems where an excess cold risk has been identified.The ongoing cuts to local government funding are a threat to local government capacity; however, we did encounter examples where local capacity was funded through the revenue from additional licensing -this may be appropriate in some contexts.
Many people in privately rented HMO are placed there by government agencies.
Local authorities and the central government placing homeless people, asylum seekers, and care or prison leavers in HMO should consider adopting a minimum energy performance standard as a housing assessment criterion.Better use can also be made of planning powers and public engagement in ensuring that better quality HMO is a planned part of development, rather than emerging in a stealthy and unplanned fashion to which communities and planning bodies usually react negatively.A positive planning dialogue should focus on the role of HMO in meeting housing need in the local area.This is salient in areas with a high concentration of "beds in sheds" (HMO type 1), student accommodation (HMO types 2 and 3), and bedsits (HMO type 4) which occur because of the socio-economic conditions in those localities.The HMO typology allows for a more nuanced debate about the HMO sector and the responses required locally.The typology challenges planners and communities to consider Article 4 directions more carefully, currently these are introduced as a blanket ban on all HMO.However, banning all HMO because of one type of perceived problem, e.g.student housing, constrains the ability to develop other kinds of HMO accommodation.For example, there might be a significant housing need for bedsit accommodation, or an economic case to convert e.g.disused buildings into HMO.An Article 4 Direction could also prevent "co-housing" schemes, popular in many European countries among senior citizens (Choi, 2004), and increasingly popular in England too for people of various age groups, sometimes referred to as housing cooperatives.

Conclusions
England's housing culture is, rather perversely, pre-occupied by single-family homes when the social, economic and political trends move in the opposite direction.Structural transformations in the housing market have led to a stealthy increase in privately let multi-occupancy homes.This is likely to continue under the current government's policies eroding social security and social housing, and the continued housing affordability crisis especially pronounced in London.It is imperative that energy efficiency and fuel poverty policies engage with HMO in order to avoid increasing the inequalities in the already warped UK housing system.This is unlikely to happen unless HMO tenants achieve parity with the rest of the private rented sector in terms of social status and rights.The situation as it stands is that there is less regulation when it comes to energy performance standards, and an absence of fuel poverty measures, in HMO.
Energy vulnerability is likely to be high in HMO due to circumstances that go beyond the scope of mainstream fuel poverty policies.Contributing factors include a lack of social security and housing rights among HMO tenants, and typically poor housing conditions.Reduced control and autonomy over energy services in the home, or sometimes an absence of basic domestic energy services, are also much more likely to occur in HMO compared with single-family homes.In addition to these material factors, this research provides a socio-cultural reading of how the inequalities in HMO are influenced by hegemonic cultural perceptions of housing as a single family unit, a finding which is explained through the theoretical lens of symbolic violence which propagates notions of social separation between HMO tenants and single-family housed people.Research among HMO residents shows that there is a social stigma associated with living in HMO (Barratt et al., 2013), and this stigma in housing terms amounts to tenure prejudice.We discovered that tenure prejudice, associated with assumed negative characteristics of HMO tenants, is reproduced through state mechanisms such as planning and licensing, as well as political rhetoric and mainstream media outlets.In terms of energy efficiency, the perception of social problems in HMO, alongside narrowly focused regulatory requirements on 'health and safety', were used to justify lack of priority given for energy efficiency improvements in the HMO housing stock, even when it is known to be notoriously inefficient and it is also known that many of the occupants are vulnerable on very low incomes.
The evidence from our research, and by others (e.g.Layard, 2012 about the planning system), suggests that single-family homes are the accepted and socially rewarded norm.The occupants have existing pathways to assistance to address energy deprivation, and their rights to make housing decisions based on domestic energy efficiency are recognised and supported by regulation.The obfuscation of HMO within policies and legislation, as well as official statistics and systems of evaluating the English housing stock condition, can be in part be explained by tenure prejudice.Negative perceptions consisting of images, real and imagined, associated with HMO are abundant in UK policy, legislative system and media (e.g."beds in sheds," "studentification," or "bedsits").Institutions and policy instruments are complicit in reproducing the bourgeoisie disdain for multi-occupancy homes through the control and containment of 'undesirable' HMO on the one hand, and by excluding them from progressive energy efficiency and fuel poverty policies on the other.The social stigma associated with HMO status (see Barratt et al., 2013) propagates a notion that HMO is 'naturally' associated with lower standards and expectations.The lower expectations pertain to energy efficiency, availability of fit-for-purpose energy services, and the level of comfort that can be expected.The acceptance of 'the way things are' by the socially oppressed group is a form of symbolic violence whereby the dominant group successfully imposes a vision about a 'natural' social order (Bourdieu, 1989, see also Sturzaker and Smith, 2011).The low quality of HMO buildings can be justified more readily if negative images of people who live in them are abundant.This occurs when a socially dominant group imposes a hegemonic vision about social divisions according to residential environments.In England, the hegemonic image of the home as a single family owner occupancy unit, alongside with images of students, immigrants, single males, and other marginalised social groups living in multiple occupancy, perpetuate this social division.Policymakers and legislators are complicit -multi-occupancy has been 'forgotten' when it comes to energy efficiency and fuel poverty policies.As with most inequality, a cultural and social awareness of power relations is required before action will be taken to address the injustices.The motivations and interests of the dominant groups who influence housing discourse and policy in the UK would merit further research to fully understand and address the underlying cultural processes around the social separation of HMO from the rest of the housing stock.
The research aimed to shine a light on this obfuscated tenure.An indicative typology is proposed which enables different policy actors to begin to recognise different HMO types, their relationship with existing regulation and energy practices, and pathways to improvement.The typology identifies five main HMO scenarios according to the tenancy arrangements: illegal/informal; rooms in a shared house; group of sharers; bedsits; and poorly converted flats.Additional factors identify the utility contract holder, how energy usage is paid for, the heating system and building typology.The typology also points to areas that would warrant further research, such as empirical insights into the socio-technical practices relating to meeting energy needs under each of these scenarios.
The UK government should close the existing legal loopholes, and aim to achieve a minimum EPC standard for HMO, by amendment to the Energy Act 2011.It should bring expectations in line with the rest of the privately rented stock by requiring EPCs to be issued to all HMO tenants at the point of letting.Furthermore, guidance should be developed around the preferred method in undertaking energy assessments in HMO to support a consistent industry approach, as well as how EPCs could be better utilised by a range of professionals who place vulnerable people in HMO accommodation.At the local level, a number of measures could be taken under existing legislation.These include additional licensing programmes in areas where there are concentrations of very energy inefficient HMO buildings, and introducing a minimum EPC standard, alongside a requirement for EPCs, as a condition of HMO licensing.Housing health and safety enforcement teams could take more robust action in requiring insulation and lower cost heating systems in HMO.
HMO is too often seen as a "problem" rather than a solution to housing issues in the UK.The cultural challenge to recognise and accept HMO as a necessary and even welcome part of the UK housing mix requires exposing and tackling the prejudices associated with this tenure and overcoming the inequalities experienced by HMO tenants.
Notes 1) A popular term used in media and government policy to describe accommodation in unconventional buildings.
2) Standard Assessment Procedure (SAP) is the approved national assessment method for EPCs.Because existing homes can be hard to assess (insulation is usually hidden behind walls for example) the Reduced Data SAP (RdSAP), which requires a reduced number of inputs, is approved for use in existing homes.See http://www.bre.co.uk

Table 1 :
HMO energy characteristics

Table 2
These numbers indicative only.HMO, more than any other type of home, are likely to be uncounted in official surveys; numbers are likely to be rising due to the welfare reform and affordability constraints.