Introduction
The changes to the homelessness legislation in Wales under Part 2 of the Housing Act (Wales) 2014 (hereafter the Act) are discussed in an earlier contribution to this themed section by Ahmed and Madoc-Jones, and examined in a growing range of articles (e.g. Connell et al., Reference Connell, Martin and St Denny2017; Mackie et al., Reference Mackie, Thomas and Bibbings2017) and reports (Ahmed et al., Reference Ahmed, Wilding, Gibbons, Jones, Rogers and Madoc-Jones2018; Madoc-Jones et al., Reference Madoc-Jones, Hughes, Dubberley, Gorden, Washington-Dyer, Wilson, Ahmed, Lockwood and Wilding2018). Building on these contributions, this article focuses on homelessness policy transfer from Wales to other contexts, exploring examples where transfer has taken place, is underway or being developed (England, Scotland and Canada respectively) (Blackman, Reference Blackman2018; Gaetz et al., Reference Gaetz, Schwan, Redman, French, Dej and Buchnea2018; Fitzpatrick et al., Reference Fitzpatrick, Pawson, Bramley, Watts, Wood, Stephens and Blenkinsopp2019) and, with clear implications for homelessness policy and service provision internationally, the conditions under which such transfers might be possible. There have been suggestions that governments worldwide could learn from the Welsh model, with Mackie (Reference Mackie2015: 41) proposing that it ‘might offer a replicable solution to the challenges of preventing homelessness across the rest of the developed world’. Given that generally there is recognition of the need for governments to adopt a rights-based approach to housing (Fitzpatrick and Watts, Reference Fitzpatrick, Watts, O’Sullivan, Busch-Geertsema, Quilgars and Pleace2010), and there has recently been considerable activity around preventing homelessness, we suggest that there is a need to further investigate the potential for policy transfer.
Policy transfer
Policy transfer can be defined as ‘the process by which actors borrow policies developed in one setting to develop programmes and policies within another’ (Dolowitz and Marsh, Reference Dolowitz and Marsh1996: 357). The literature has grown exponentially over the last three decades, and housing policy transfer has been examined in relation to areas such as choice-based lettings (Pawson and Hulse, Reference Pawson and Hulse2011); supportive housing (Parsell et al., Reference Parsell, Fitzpatrick and Busch-Geertsema2014), and homelessness prevention (Busch-Geertsema and Fitzpatrick, Reference Busch-Geertsema and Fitzpatrick2008). More generally, the policy transfer heuristic can be used for ex-post and ex-ante analysis, with the latter taking the form of prospective evaluation (Mossberger and Wolman, Reference Mossberger and Wolman2003). Accordingly, it offers a framework for investigating the extent to which governments in other national contexts can learn from housing policy in Wales.
Dolowitz and Marsh (Reference Dolowitz and Marsh1996) argue that, depending on the context, almost anything can be transferred including policy goals, content, instruments, and programs. Cut and paste transfers in which institutions and practices are completely transferred from one country to another are rare, with hybrid transfers, involving partial policy transfer, proving more common (Marsh and Sharman, Reference Marsh and Sharman2009). In relation to this, it has been suggested that policy transfer research has neglected the way that policies might mutate and be reconstituted in the process of moving (Peck, Reference Peck2011).
Significant attention has been paid to the pathologies of policy transfer (Park et al., Reference Park, Lee and Wilding2017), and so it is important to examine what might facilitate successful transfer. Here it is worth noting that policy success is a problematic concept, not least because it may be linked to policy effectiveness, efficiency and resilience, or to addressing political concerns including public satisfaction with the policy/confidence in institutions (Bovens et al., Reference Bovens, t’Hardt and Peters2001). Policy transfer requires appreciating the context of both the lender and the borrower (Park et al., Reference Park, Wilding and Chung2014). That is to say, understanding what it is that makes the policy work in its original context, and what would enable it to work in the new context, including the need for any adaptations. Through this process, transferred policies may take on different meanings as they are fitted to the borrower’s institutional context, and without working closely with stakeholders at this stage, there is a risk of distorting the original policy meaning.
While England continues to be governed by the UK Government, from 1999 onwards, devolution has enabled Welsh, Scottish and Northern Irish executives to pursue their own health, social care and housing priorities to various extents (Cairney, Reference Cairney2009). Early policy transfer examples from Wales to other parts of the UK include the abolition of prescription charges, and Children’s Commissioners; however, possibilities for policy transfer opened up after 2011 when the Welsh Government acquired primary legislative powers (Keating et al., Reference Keating, Cairney and Hepburn2012). Structural constraints can be identified that impose limits to how far devolved governments are able to realise different priorities to those of the UK Government (Connell et al., Reference Connell, Martin and St Denny2017). Still, opportunities for learning from policy differences can be associated with devolution and the potential race to the top as different administrations seek to adopt higher standards of provision compared to neighbours.
Housing contexts
Key classification systems have emerged in the welfare and housing regime literature to understand and compare welfare and housing systems. Esping-Andersen (Reference Esping-Andersen1990) famously identified three welfare regime types: Nordic social democratic regimes; continental Western Europe’s conservative-corporatist regimes; and the liberal Anglo-Saxon regimes. Subsequent efforts attempted to improve on this framework in terms of both accuracy and coverage (Abrahamson, Reference Abrahamson1999). Helpful though this classificatory system is in understanding the broader welfare system that may impact upon prospects for housing (for example, through maintaining, or not, household incomes following loss of employment), it is of limited use in understanding homelessness, because in most contexts housing need is addressed by the market (Torgersen, Reference Torgersen1987).
Housing regime theory has been subsequently developed by a number of authors (Kemeny, Reference Kemeny2002; Schwartz and Seabrooke, Reference Schwartz, Seabrooke, Schwartz and Seabrooke2009). Kemeny (Reference Kemeny2002) built on the work of Esping-Andersen to argue that there are two main housing regime types. The first is the unitary rental regime wherein public housing is offered on a universal basis in competition with the private rented sector (PRS). The lower rents produced by this regime, typically found in central Europe and Scandinavia, act as a disincentive towards homeownership. The second housing regime, typical of Anglo-Saxon systems, is the dualist rental regime. Here residual social housing may be found which does not compete with the PRS, but places emphasis on homeownership.
The relationship between welfare and housing regimes and homelessness has been explored by Stephens and Fitzpatrick (Reference Stephens and Fitzpatrick2007). Acknowledging the complexity involved, they suggest that homelessness is influenced by factors including the housing market (availability and affordability), welfare regime (levels of poverty and inequality), housing decommodification (social housing and allowances, subsidies or vouchers), and housing policy (homelessness prevention and resolution strategies). At a fundamental level, however, we suggest that there are two issues that help to shape the broader approach to homelessness policy: definition of homelessness, and right to housing. Generally, we argue, policy transfer will be more easily facilitated between similar welfare and housing regimes with broadly analogous definitions of homelessness and rights to housing.
Broad definitions of homelessness include not only rooflessness and houselessness, but also insecure and inadequate housing, while narrow definitions focus only on rough sleeping (Mackie, Reference Mackie2014). While the former view predominates in Europe, the latter is dominant in the United States so that in some contexts homelessness is recognised only in its most extreme and publicly visible form, with less provision to respond to insecure and inadequate housing.
The right to housing is particularly important when attempting to understand a government’s approach to homelessness. Outside of the UK and the weakly implemented Act Establishing the Enforceable Right to Housing (DALO) in France, there are no legal rights to housing beyond emergency accommodation for roofless people in a number of countries such as Germany (Fitzpatrick et al., Reference Fitzpatrick, Johnsen and Watts2012). This is significant as the right to housing supports a consistent response to homeless applicants, who in turn may be more likely to claim their rights.
Assessing the Act
The specific changes introduced by the Act are discussed in the introductory article to this themed section and will not be rehearsed again here. It is important to add, however, that Welsh policy was influenced by policies from other parts of the UK. Due to the high degree of interaction between the Welsh homelessness policy and practice communities, this occurred at multiple levels. For example, while the review process included an appraisal of policies from England, Finland, France, Germany, Ireland, Scotland and the US, and specific learning points relating to English prevention and the Scottish universal safety net, local authority networks also facilitated learning from various contexts (Fitzpatrick et al., Reference Fitzpatrick, Johnsen and Watts2012; Connell, Reference Connell2017).
Different ways of working make it difficult to assess the impact of the Act both pre- and post-implementation and compared with relevant legislation in other UK nations. However, although there is a higher rate of applications in Wales than throughout the UK, proportionally fewer households are owed the main duty to be accommodated (Wilson and Barton, Reference Wilson and Barton2018). Accordingly, it appears that the preventative work is making a difference.
That said, a number of serious critiques of the Act have begun to emerge, relating to the continued use of priority need, non-cooperation and local connection to exclude people from services. Concerns have been expressed about how fairly and consistently priority need is assessed, particularly in the case of vulnerable groups like prison leavers (Madoc-Jones et al., Reference Madoc-Jones, Hughes, Gorden, Dubberley, Washington-Dyer, Ahmed, Lockwood and Wilding2019). Concerns about local authority recourse to exclusion on the grounds of non-cooperation have been raised (National Assembly for Wales, 2018). Finally, use of local connection has also been found to be problematic (Ahmed et al., Reference Ahmed, Wilding, Gibbons, Jones, Rogers and Madoc-Jones2018).
The number of rough sleepers has increased following the introduction of the Act, from 240 in 2015–16 to 347 in 2018–19 (StatsWales, 2019). Although this may be connected to the exclusion of vulnerable people from services, it may also be linked to increased property sales in a buoyant housing market (and so fewer homes for rent); the introduction of Rent Smart Wales (a landlord licencing agency); and welfare reform (with difficulties following the introduction of Universal Credit and landlords evicting tenants facing payment delays).
Nevertheless, Ahmed et al. (Reference Ahmed, Wilding, Gibbons, Jones, Rogers and Madoc-Jones2018) found that the Act has influenced local authorities to offer improved information, advice and assistance, and that there is more preventative work, which is more inclusive and effective. While the Welsh legislation is not without its problems, it does go significantly further than before in the intended universality of the statutory prevention duty and widened eligibility for assistance. In this sense, it presents a learning opportunity for other governments.
Policy transfer analysis
England
Policy content has been transferred to England and into the Homelessness Reduction Act 2017. Crisis (a national homelessness charity) was instrumental in setting up an independent review of existing policy and securing political support for new legislation in England. The influence of the Act (in Wales) was instrumental and it has been suggested that the experience in Wales ‘was perfectly timed to help inform measures’ (Blackman, Reference Blackman2018). Accordingly, both Acts include a strengthened duty to provide advisory services, and duty to prevent homelessness for all eligible applicants regardless of local connection. Local authorities are required to assess all eligible applicants under both Acts, regardless of intentionality and priority need, and agree Personal Housing Plans.
Certain changes arise in England that could be said to represent a further expansion of the rights to housing and efforts to prevent homelessness. Although both Acts employ a fifty-six-day period under which applicants are threatened with homelessness, in England those with a s21 notice (i.e. those for whom the eviction process is underway) are legally eligible for services before fifty-six days. Non-cooperation must be deliberate in England, which offers further protection to applicants. The duty to refer, in place in England but not Wales, aims to encourage early intervention through the referral of service users believed to be homeless to local authority homelessness/housing options teams. On the other hand, Welsh provision for County Court challenges may help to protect legal rights in a way that is not provided for in England. Still, minimum PRS tenancies in England, at twelve months, offer more consistency for tenants, compared to six months in Wales (Ahmed et al., Reference Ahmed, Wilding, Gibbons, Jones, Rogers and Madoc-Jones2018). While this is not an exhaustive comparison of the two Acts, there has clearly been extensive policy transfer, albeit with some modification.
Such a degree of policy transfer is facilitated by the similar policy contexts, with both nations favouring prevention and the right to housing. Both nations have also attempted to introduce the changes with minimal extra spending, during large-scale public sector budget cuts. The £72.7m implementation budget in England represents a cause for concern, however, given that a number of local authorities struggled with the £11.5 million budget in Wales (Ahmed et al., Reference Ahmed, Wilding, Gibbons, Jones, Rogers and Madoc-Jones2018), where the population is less than six per cent that of England (Office for National Statistics, 2018).
Some other challenges faced by the English Act are due to the shared wider context of housing and welfare policy, and it is notable that English policymakers do not appear to have learned from the related Welsh difficulties. Although Shelter supported the aims of the Homelessness Reduction Act, the organisation has expressed concerns that without improvements to welfare policy and the availability and affordability of suitable accommodation, there could be unintended consequences, including ‘“gate-keeping” of services, unlawful decisions, increased out-of-area moves and repeat homelessness’ (Garvie, Reference Garvie2018: 17).
There also appears to have been a lack of learning from the co-productive policy processes evident in Wales. Whereas there was constant interplay between different policy actors and an opportunity for implementation training, which brought together all twenty-two local authorities with Shelter Cymru and the Welsh Government, practitioners had little input into the development of the English Act (Connell, Reference Connell2017). To some extent, this is attributable to differences in the homelessness policy and practice communities and the number of local authorities. Still, the lack of participation from English local authorities may mean that implementation is more about compliance than ownership.
Scotland
As noted above, Scottish homelessness policy influenced the Act in Wales. However, it now appears that there has been homelessness policy transfer from Wales to Scotland, albeit in a limited way, focussed on introducing a preventative duty to complement existing Scottish policy. Starting in 2003, Scotland was the first of the devolved territories to amend its homelessness legislation following devolution, with changes including abolishing priority need; the option for local authorities to discharge their duty to the PRS; an interim duty to secure accommodation while assessing homelessness; and a duty to provide advice, assistance and temporary accommodation regardless of intentionality (Wilson and Barton, Reference Wilson and Barton2018).
Despite the numerous strengths of Scotland’s homelessness policy (Watts, Reference Watts2014), prevention has not been given the requisite attention (James, Reference James2018). Indeed, a number of local authorities were identified as failing to provide appropriate advice and assistance (Scottish Housing Regulator, 2014). A commitment to a new prevention duty was introduced in Scotland in November 2018 upon the recommendation of the Homelessness and Rough Sleeping Action Group. It has been suggested that this recommendation drew on the Welsh and revised English legislation (Fitzpatrick et al., Reference Fitzpatrick, Pawson, Bramley, Watts, Wood, Stephens and Blenkinsopp2019) and so a range of prevention initiatives were introduced, not dissimilar to those found in Wales and England – for example, more flexible homelessness assessments to make it easier for people to access their right to assistance, and preventative pathways for high-risk groups.
It should be no surprise that the transfer from Wales was more selective given that significant changes to homelessness system in Scotland have already taken place. The Scottish Government has seemingly used its understanding of the functioning of the Act in Wales to adapt a key aspect of it (prevention) to operate in the Scottish context. Combined with the removal of priority need in Scotland, the introduction of a preventative duty potentially offers a more thoroughgoing approach to tackling homelessness, which, in attempting to go beyond a conditional approach, will overcome the continued use of priority need to exclude people from services, a problem which has been identified in Wales. Over time, there may also be cost savings if fewer homeless households present to local authorities in need of relief.
Canada
Outside of the UK, Canada presents as having greatest potential for policy transfer. The approach adopted in Wales has attracted interest so that The Roadmap for Youth Homelessness sets out a comprehensive systems response to youth homelessness based partly on Welsh legislation (Gaetz et al., Reference Gaetz, Schwan, Redman, French, Dej and Buchnea2018). A duty to assist young people who face homelessness is articulated including the statutory obligation for local authorities to make reasonable efforts to end a person’s homelessness or stabilise their housing (i.e. including prevention). The duty would work in broadly similar ways to Wales through taking a rights-based approach. So, the need for an enforceable right to housing for this group is acknowledged.
Significant adaptations, however, are imagined. First, duties would not be considered discharged if a young person is non-cooperative and turns down the options presented to them. This is to avoid a coercive or punitive approach. Second, a duty to refer would be included, which, as discussed above, is present in the English but not the Welsh Act. Thirdly, the appointment of a regulator/ombudsperson for homelessness services is considered necessary to ensure the duty legislation is being followed. This goes further than in Wales where, while recommended, this option was not realised (Mackie, Reference Mackie2014). The proposed Canadian duty therefore appears to be based on an understanding of experience in the Welsh context and what would enable it to be effective in Canada.
A duty to assist demonstration project is currently being designed and will be implemented in the city of Hamilton, Ontario (Gaetz et al., Reference Gaetz, Schwan, Redman, French, Dej and Buchnea2018). A key consideration is how to transform systems before changing legislation, as was the case in Wales. A further issue is how to address the needs of indigenous youth due to their over-representation among the homeless population, which will necessitate the availability of culturally appropriate, indigenous-led housing and support. This demonstration project is a way for policy transfer advocates to garner support for the proposed policy solution to Canadian homelessness issues, and it builds upon a number of Canadian government funded projects to prevent youth homelessness (Canadian Observatory on Homelessness, 2018).
The Canadian policy context differs from Wales in significant ways. Canada is a liberal welfare regime, but its welfare state evolved without the same social democratic influence as in Wales (Mishra, Reference Mishra1994). In housing, this can be seen in the high levels of homeownership (68 per cent) and low levels of social housing (6 per cent) (Housing Services Corporation, 2014). As such, crisis response has been the dominant approach to homelessness, although Housing First is beginning to make some inroads. How homelessness figures compare between Canada and Wales is not clear as data collection methods differ (Grenier et al., Reference Grenier, Barken, Sussman, Rothwell, Bourgeois-Guérin and Lavoie2016). More generally, Canada is a geographically vast decentralised federation, with a need to balance regional interests when it comes to federal spending (Suttor, Reference Suttor2011). Nevertheless, it is argued that the experience of Wales can be drawn upon to build a ‘‘Made in Canada’ Duty to Assist strategy’ (Gaetz et al., Reference Gaetz, Schwan, Redman, French, Dej and Buchnea2018: 129).
The potential for further policy transfer
There remains the potential for future policy transfer to Northern Ireland in the sense that there is a high rate of homelessness, which has been partly attributed to the lack of prevention activity (Fitzpatrick et al., Reference Fitzpatrick, Pawson, Bramley, Wilcox and Watts2016). The Northern Ireland Executive has begun to address this through a preventative model, Housing Solutions and Support, which is based on the pre-November 2018 Scottish approach to prevention (Watts and Fitzpatrick, Reference Watts and Fitzpatrick2017). Should the Northern Irish administration seek an approach more in-line with Wales and England, then the similar policy context suggests that this may be possible. Furthermore, it may be desirable due to the issues associated with light-touch prevention, Northern Ireland’s restrictive use of priority need (which does not apply to vulnerable ex-offenders or members of the armed-forces community), and the extended eligibility test for assistance which is contingent on past good behaviour as tenants (Wilson and Barton, Reference Wilson and Barton2018).
Outside of the UK and Canada, it appears that there may be limited opportunities for policy transfer unless there is (1) perceived scope to benefit from transferring the Welsh policy, and (2) strategic support in the form of advocates for policy change. This is due to factors including the general international absence of enforceable rights to housing and different definitions of homelessness internationally. More generally, there are different types of government, welfare systems, historic factors, housing systems, and different scales of homelessness.
Of course, some countries may not benefit from extensively transferring the Welsh policy. For example, although there is some variation between the Nordic social-democratic regimes, there is relatively less homelessness due in part to lower levels of poverty and inequality. With the partial exception of Sweden, the social-democratic approach to social housing represents a decommodifying influence, and there is a commitment to Housing First (Benjaminsen and Knutagård, Reference Benjaminsen and Knutagård2016). Greater proportions of homeless people in countries with lower levels of poverty and inequality have individual support needs, meaning that government takes a selective approach, focussing on marginal group access to social services and interventions (Stephens and Fitzpatrick, Reference Stephens and Fitzpatrick2007). There are also fewer potential benefits of extensive policy transfer to corporatist regimes, which have relatively lower levels of poverty and inequality, and particularly Germany with its slack housing market and effective prevention services (Fitzpatrick et al., Reference Fitzpatrick, Johnsen and Watts2012).
Definitions of homelessness mean that the preventative approach in Wales is less likely to be transferred to some contexts. In countries which use narrower homelessness definitions, such as the USA, preventative initiatives seek to address rough sleeping (Mackie, Reference Mackie2014). This represents a different conception of prevention than in Wales, one that seemingly prioritises preventing long-term homelessness, and policy transfer would not be possible without dramatic change from the borrower governments. Consequently, the Welsh policy, which is universal in prevention and focussed on the fifty-six days leading up to homelessness, would be incompatible with regimes seeking to retain narrow definitions of homelessness.
The general absence outside the UK of a right to housing is another obstacle to policy transfer, as duties to prevent homelessness and secure housing are at the core of the Act. Statutory duties aim to ensure that local authorities offer the appropriate information, advice and assistance. Without these duties, it seems unlikely that the work would continue at an appreciable level. For example, in the absence of social rights for those at risk of homelessness in the Republic of Ireland, efforts at more progressive prevention ultimately faltered (Maher and Allen, Reference Maher and Allen2014).
There is clearly a need for legal rights to housing to be adopted internationally, as campaigned for by the European Federation of National Organisations Working with the Homeless (FEANTSA), and as recognised in the Universal Declaration of Human Rights and the International Covenant on Economic, Social and Cultural Rights (Fitzpatrick and Watts, Reference Fitzpatrick, Watts, O’Sullivan, Busch-Geertsema, Quilgars and Pleace2010). As the Welsh preventative rights are universal, they may offer a more straightforward means of implementing a rights-based approach for governments internationally (Mackie, Reference Mackie2015). This could have the benefit of empowering homeless people and increasing personal responsibility in a substantive sense (Watts, Reference Watts2014).
Discussion and conclusion
This article has examined the extent to which there has been policy transfer of Part 2 of the Housing Act (Wales) 2014 to England and Scotland, and to what extent there is potential for future policy transfer to Canada and other national contexts. As such, this article represents the first review of the transferability of the Act since it came into force, and it has implications for homelessness policy and service provision across the UK as well as internationally.
The Act represents a major step change that is replicable through introducing a statutory prevention duty and widening eligibility for assistance. The Act is not a panacea, however, and we have highlighted problems relating to the available budget, priority need, non-cooperation and local connection, along with wider housing and welfare policies. It follows that there are contexts where the combination of homelessness demographics, housing market, welfare regime and homelessness policies makes extensive policy transfer less desirable, including the Nordic social-democratic regimes and Germany. More generally, though, housing is a human right, and there have long been campaigns for governments to adopt rights to housing (Fitzpatrick and Watts, Reference Fitzpatrick, Watts, O’Sullivan, Busch-Geertsema, Quilgars and Pleace2010).
In each of the cases of transfer examined, there has been some modification and adaptation. In England, adaptations include the introduction of a duty to refer, that non-cooperation must be deliberate, that homelessness can be prevented prior to fifty-six days for those with s21 notices, and minimum tenancies of twelve months when discharging duties to the PRS. However, the process of developing the English policy did not involve the same levels of co-production as in Wales, which may lead to a lack of local authority ownership and implementation issues. Other implementation difficulties will arise as have arisen in Wales related to austerity and wider housing and welfare policies. This may reflect the complexity of homelessness as a policy area, and possibly the lack of priority that it receives.
Yet, the apparent lack of learning given the relative absence of co-production in the later stages of the policy transfer process is not surprising when it is considered that the wider policies originate from Westminster, and that the Homelessness Reduction Act is expected to function as part of this institutional context. Despite the adaptations noted above, the transferred policy may be less progressive in England and there may be a risk of the original policy meaning being distorted when the policy is understood and implemented within an institutional context without a devolved policy agenda to mitigate the responsibilising of service users.
In Scotland, there has been selective transfer of the preventative duty, while retaining the broad safety net in the absence of priority need. We anticipate that this will allow the innovative preventative feature of the Act to be drawn upon, while retaining the strengths of Scottish homelessness policy. Thus, in a homeless policy context without priority need, it is possible for inclusive preventative initiatives to get much closer to their intended universal housing rights, and through this potentially offset some of the more conditional aspects of austerity and welfare reform.
It is clear that the territories of the UK are learning from each other’s homelessness policies. This is a positive development as learning between the devolved governments in the UK has not always been widespread (Keating et al., Reference Keating, Cairney and Hepburn2012), and as a form of prospective policy evaluation, policy transfer can help to improve policymaking decisions (Mossberger and Wolman, Reference Mossberger and Wolman2003). It will be interesting to observe to what extent this learning and policy transfer results in the re-convergence of homelessness policy after divergence began in Scotland from 2003 and gathered pace with the Act in Wales. It does appear at this stage that there has been some degree of re-convergence, although, as discussed above, differences in the wider institutional context may act as limiting factors.
In Canada, despite an initial focus on young people, the plans would go further than Wales through avoiding a coercive approach; a duty to refer; and appointment of a regulator/ombudsman. Although only time will tell whether these modifications will endure the political process, supporters of the policy transfer appear to understand what makes the Act work in Wales, and have a desire to understand the applicability to Canada through a demonstration project.
Narrow homelessness definitions and the general international absence of enforceable legal rights to housing beyond emergency accommodation represent stumbling blocks to further policy transfer beyond the UK, which would require a significant departure from current homelessness policy trajectories. Indeed, the historical development of welfare and housing institutions (i.e. path dependency) and associated costs may act as constraining influences (Blackwell and Kohl, Reference Blackwell and Kohl2018). Nevertheless, the Canadian case suggests that policy transfer may be possible if there is sufficient support for change.
A limitation of the present article is that the policy transfer processes are at different stages for each government, meaning that it is not possible to conduct a like-for-like comparison. Even in England where the Homelessness Reduction Act is in operation, we are only starting to see its effects in practice, and it may be some time before the effects of policy transfer to Scotland and potentially Canada can be more fully understood. Future studies should revisit the cases later in the policy process.
This article has highlighted the positive effects that could be gained from policy transfer of Part 2 of the Housing Act (Wales) 2014, through expanding the rights to housing and a more preventative approach to homelessness. There is potential for some degree of policy transfer to other national contexts that have not yet done so, although this would require appropriate adaptation and modification. Indeed, the transfer to England may yet serve as a warning of, firstly, the need to have a clear understanding of the context – not just of the lender, but also the borrower – and secondly, the ways in which homelessness policy interacts with wider housing and welfare policy.
