Scottish Government

CHAPTER TEN - CONCLUSIONS

10.1 In the first part of this final chapter we review the research findings set out above in terms of the aims and objectives originally set out in the Scottish Executive's project brief.

Mapping activities being undertaken to prevent homelessness

10.2 A wide range of activities is being undertaken by Scottish LAs in an effort to prevent homelessness. In terms of staff intensiveness, this ranges from general housing advice at one end of the spectrum, to intensive family support schemes, at the other.

10.3 Most authorities report operating some kind of scheme to help people at risk of homelessness to access private tenancies. However, whilst some of these rent deposit guarantee ( RDG) schemes have been expanded from those established under the earlier Rough Sleepers Initiative, the statistical evidence available suggests that most continue to operate on only a relatively small scale. Authorities report highly contrasting experiences in terms of the supply of suitable accommodation accessible in their area. Particularly in rural and/or more expensive semi-urban areas councils report finding it difficult to attract sufficient numbers of landlords to meet demand. At the same time, however, there is no obvious correlation between the extent to which private tenancies are used to accommodate homeless applicants and the overall amount of private rented housing in each authority. The research did not identify any councils emulating the practice of some English LAs in offering explicit bounty payments to landlords willing to accept RDG referrals. Nor were any of the case study LAs routinely attempting to negotiate with landlords for improved tenancy terms (e.g. 12 months rather than six months minimum duration).

10.4 Another contrast with England in this respect is that most RDG schemes focus mainly on helping non-priority single homeless applicants rather than families liable to be classed as priority homeless cases. However, whilst few if any Scottish authorities see RDG placements as appropriate to meet the needs of families liable to be classed as priority homeless some RDG schemes in fact accommodate significant numbers of family households to whom the council owes no permanent rehousing duty (e.g. due to their having been found intentionally homeless or having refused a final ('discharge of duty') offer of council housing).

10.5 Outreach support to promote tenancy sustainment is provided at some scale by most LAs and this absorbs very significant staffing resources usually underpinned by Supporting People funding. Such services are particularly difficult to evaluate in terms of the actual contribution to homelessness prevention - i.e. what proportion of the households assisted would have lost their tenancies in the absence of such help. Available evidence suggests that this proportion might be relatively small. For example, in 2002/03 (before the Council's tenancy support services described in Chapter 8 were fully operational), North Ayrshire recorded only 23 homelessness applications (from a total of 1,582) where the reason for homelessness was loss of a local authority tenancy ( HL1 data). The numbers of tenants subject to tenancy sustainment support in 2005/06, however, totalled an estimated 166 (see para 8.23). In Edinburgh, four 2002/03 homelessness applications involving tenants having been evicted from LA housing were recorded ( HL1 data). Direct monitoring of LA housing management activity suggests the number of ASB evictions actually carried through by Edinburgh Council in 2002/03 was two. (The same source indicates that across the whole of Scotland there were only 62 ASB evictions by LAs in 2002/03) ⁴⁹. Yet in the first 10 months of 2005/06, 27 families were accepted onto an Edinburgh scheme providing special support (at least partly) to prevent such evictions (see para 8.23).

10.6 These figures could be interpreted as suggesting that tenancy sustainment services partly justified in terms of 'homelessness prevention' are, in practice, extended across much larger groups within the council tenant population than the relatively small numbers who might in fact otherwise become homeless through failing to sustain their tenancy. This may be unavoidable because of the difficulty in pinpointing precisely those who would be critically at risk without special support. In any case, whether or not it is usually decisive in preventing homelessness, tenancy support assistance has the capacity to significantly improve many service users' lives and may also benefit the communities where they live (e.g. in helping to ensure that vulnerable people are referred to specialist services to address problems such as drug addiction or mental ill-health).

10.7 Most authorities operate one or more forms of intervention specifically designed to prevent youth homelessness. Around half of all authorities have one or more facilities providing supported transitional housing ( STH) for young vulnerable adults. Again, such schemes - usually operated by voluntary agencies under contract - are highly staff intensive and usually financed mainly through Supporting People funds. Whilst projects of this kind accommodate young people from a range of backgrounds, a significant proportion may involve care leavers and others with substantial behavioural problems and support needs. The express purpose of STH schemes is to better equip young people destined for council tenancies with the skills and habits which will maximise their chances of sustaining such tenancies. In practice, it appears that most projects experience a substantial 'fallout rate' involving young people being evicted from or otherwise leaving such accommodation ahead of being offered a council tenancy. However, a proportion of these are young people returning to the family home and for a proportion of these, at least (e.g. some of those having received family mediation), this could possibly be counted as a success.

10.8 It would appear that the Scottish picture contrasts from that south of the border in that STH projects are a more prominent aspect of homelessness prevention activity in Scotland. Conversely, family mediation activities appear generally fairly small-scale in Scotland by comparison with England. A related finding is the very small proportion of Scottish authorities with a standard policy of making home visits in cases of young people facing possible homelessness due to 'family/friend exclusions' (see paras 4.26-4.31; 6.8-6.19).

10.9 Again in contrast to England, only very recently have authorities looked at setting up 'sanctuary' schemes to prevent homelessness involving women needing to leave tenancies due to the threat of violence from outside the home.

10.10 Geographically, it appears that - as might well be expected - certain homelessness prevention measures (e.g. transitional supported accommodation projects) are more liable to be operated in larger and more urban authorities than in smaller and more rural areas. As a rule it is undoubtedly the case that the larger city authorities have developed a wider range of services than other councils. Other than this, however, it is difficult to draw many general conclusions about the spatial distribution of prevention projects or measures on the basis of this research.

Monitoring and evaluating homelessness prevention and measuring success

Monitoring effectiveness

10.11 Effectiveness is closely related to appropriateness. In this connection it is useful to note that only two of the 26 service user interviewees felt that the Council's action to address their housing problems had done nothing to resolve these (see Table 9.3).

10.12 All local authorities were, to some degree, evaluating individual homelessness prevention projects. However, most LAs saw the measurement and monitoring of prevention activities as a difficult issue, and one which has yet to be addressed by official guidance. One important issue was the extent to which the HL1 system should be seen as having potential in this area. A problem here is what some LAs perceive to be a lack of clarity in official guidance on the scope of the system - see discussion in Chapter 3 (para 3.8).

10.13 In general, authorities' monitoring of homelessness prevention activities was weak - a conclusion also reached in most Communities Scotland LA homelessness inspection reports. In part, this stemmed from the commonly experienced situation that such activities (e.g. provision of housing advice, mediation-style discussions with young people and their parents) were largely integral to the activities of homelessness caseworkers and were not logged specifically as 'prevention cases'. Similar problems were evident in relation to homelessness prevention activities on the part of generic housing advice staff - where there were statistical records of casework activity and outcomes these did not distinguish between cases specifically involving homelessness prevention and those involving other service users.

10.14 Records kept by RDG schemes tended to be more useful to the extent that these at least indicated the overall scale of caseloads (e.g. in terms of the numbers of inward referrals and the numbers of tenancies established with RDG assistance). For such schemes it seemed self-evident to practitioners that 'success' equated simply to the creation of a tenancy accessed with such assistance. Monitoring the sustainment of such tenancies was less commonly a standard practice.

10.15 The sustainment issue was even more important in relation to family mediation or similar interventions. Some LAs could enumerate the number of 'mediation referrals' and the proportion of these where the young person concerned had returned to (or remained in) the family home. However, longer term outcomes in such cases were rarely, if ever, tracked. Another important issue here was the principal criterion of success. Some LAs accorded overwhelming importance to the question of whether or not the household remained in or returned to the family home. Others focused more on whether or not the intervention was considered to have achieved family reconciliation (irrespective of whether this brought about a return to the family home).

10.16 A finding from the homelessness prevention service user interviews is also very relevant to the question of 'sustainability of solutions'. At the time of their interview, only six of the 26 respondents felt 'currently at risk of homelessness'. This is despite the fact that 20 were living in inherently 'insecure' private tenancies or had returned to the parental home following earlier ejection (see para 9.65).

10.17 Schemes with the express aim of maximising tenancy sustainment are particularly problematic in defining and measuring 'success'. Some councils refer to trend-over-time figures in the incidence of repeat homelessness here. But this is an imperfect measure of effective tenancy sustainment intervention because the definition of 'repeat homelessness' as used in the HL1 system is not limited to households assessed as priority homeless and rehoused in social housing ⁵⁰. The concept of measuring the proportion of those rehoused during a given time period and retaining their tenancies for a given amount of time (e.g. six or 12 months) was found somewhat alien to most of the case study LAs and most research team requests for such statistics met with failure (see recommendation R5 later in this chapter).

Monitoring cost-effectiveness

10.18 LAs generally recognised the complexity of monitoring the cost effectiveness of homelessness prevention and were concerned that this should not be glossed over:

"there is a real concern around the notion that you can simply do a simple analysis of cost effectiveness. It is a highly complex issue and it is essential that this message gets back to the Scottish Executive" ( LA staff member interviewed in national telephone survey)

10.19 Some LAs were able to provide data on the costs of running specific homelessness prevention services. Again, however, this was seen as impractical in relation to such activities where these were undertaken by homelessness caseworkers as an integral element of homelessness assessment. And the paucity of data on the homelessness prevention caseload and/or the proportion of service user outcomes which could be classed as 'success' meant that it was often impossible to determine prevention service unit costs, even where total cost data were available.

10.20 In these circumstances we were unable to identify any LAs holding sufficiently detailed (and comparable) data on service costs and benefits to underpin purely 'objective' judgements about the 'most cost-effective' way to make use of limited resources in preventing homelessness. In any case, the funds used to underpin homelessness prevention services are not in all instances sourced from budgets which would allow such flexibility. Under current regulations, for example, funds accessed under the Supporting People regime (e.g. for tenancy sustainment services) could be switched to paying for the expansion of RDG activities (e.g. making 'bounty payments' to landlords in return for their agreeing to accept referrals on preferential terms).

Barriers to homelessness prevention

10.21 The most frequently cited barrier to homelessness prevention as voiced by local authorities in the national survey and the case study work was the shortage of affordable housing. Most LAs saw this primarily in terms of the reducing supply of social housing, but some councils drew attention to their difficulties in identifying an adequate supply of private tenancies (see above). Beyond these, the key barriers would seem to be:

10.22 Funding: Most local authorities see homelessness prevention activities as calling for additional expenditure in staff-intensive services and, whilst there is some appreciation that stemming homelessness could be justified by 'downstream' savings in public spending, such savings are not seen as accruing directly to LAs themselves. Hence, it is seen that limited Scottish Executive funding for prevention services constitutes a significant constraint on the scale of LA activities here.

10.23 Annuality: Authorities emphasized the problems arising from the annuality of funding streams, making it difficult to forward plan with certainty. Although the homelessness strategy funding has been greatly welcomed, it was emphasised that it has remained static over the last 3-4 years. As such, this affords little scope to introduce or expand new services.

10.24 Housing Benefit restrictions: The Single Room Rent ( SRR) restrictions are widely seen as compromising the scope for local authorities to prevent homelessness involving young people aged 18-25. Whilst it is recognised that bedsit rents may be, in some cases, affordable within the SRR context, the supply of such dwellings is very small in most areas.

10.25 Use of the private sector: Some LAs believe that the homelessness legislation (and the way the current legislation is presented in the Homelessness Code of Guidance) discourages the use of private tenancies as a means of rehousing priority homeless households.

10.26 Culture change: The need to overcome ingrained attitudes among homelessness caseworkers and other LA staff which date from the era when homelessness was a strictly responsive service rather than one which emphasizes a strategic, pro-active approach.

What prevention activities work best for which groups?

10.27 As noted above, there is an absence of reliable statistical evidence to form a basis on which to judge relative effectiveness of different prevention activities. As discussed above, this is particularly true of the schemes focusing on helping social sector tenants retain existing tenancies (e.g. tenancy sustainment support, young persons transitional supported housing schemes).

10.28 It also needs to be recognised that there can be very different ways of implementing what appear, on the surface, to be standard concepts. Take family mediation, for example. LAs experimenting with 'family mediation' report widely differing views as to the effectiveness of their schemes. Some LAs have experienced 'disappointing' levels of engagement with the service. Much will, however, depend on factors such as:

• The extent to which mediation is treated as the standard approach for young people at risk of homelessness due to family/friend exclusions
• The point at which mediation referrals are made (prior to, or following on from the completion of a homelessness assessment)
• The speed with which mediation referrals can be made and meetings held
• The extent to which mediators will regard the refusal of one or other party to attend a meeting face to face with the other party and the mediator as an effective veto on the process.

10.29 The best we can really do in addressing the question at the start of this section is to 'map' groups 'at particular risk of homelessness' against the mechanisms most commonly adopted (at least in some cases successfully) to address the needs of the particular group concerned. This is shown in Table 10.1.

Table 10.1 - Services/mechanisms by service user group

Monitoring recommendations

R1. The Executive and Communities Scotland need to provide clear and consistent guidance as to the proper scope of the HL1 system with respect to people approaching local authorities for assistance with housing but where local authorities judge there is no reason to believe that they are homeless. This is necessary to redress the situation where some local authorities are recording all approaches as homelessness applications through the HL1 system, while others are recording only cases where they have reason to believe the household is homeless or likely to become so. A particular consideration should be whether the HL1 system should be used to record all approaches regardless of circumstances.

R2. The Scottish Executive needs to encourage local authorities to be more rigorous and consistent about distinguishing between potential homelessness and actual homelessness. This applies, in particular, to the significant number of authorities which currently record very few applicants as such.

R3. There should be clearer advice regarding how to record last actions on the HL1 where help in accessing alternative accommodation has resulted in a settled tenancy.

R4. Local authorities should be encouraged to use the revised HL1 to more fully record any prevention activity they may undertake, such as the type of advice and assistance provided. However, given the restricted scope for the HL1 to be used to record such activity in detail, homelessness and housing advice services should be encouraged to develop monitoring systems to record prevention activities, including information about 'presenting households' circumstances in relation to their risk of homelessness and the housing outcome of any advice and assistance provided.

R5. For the purposes of homelessness strategy Local Outcome Agreements ( LOAs), local authorities should be required to monitor the numbers of households referred for help to specific homelessness prevention services and the proportion of such cases resulting in 'success'.

R6. It would be useful to consider a national indicator for the social rented sector in relation to the proportion of tenancies created in a given year and terminated within 12 months of the tenancy date.

R7. Local authorities should oblige private landlords accepting rent deposit scheme referrals to notify the council (a) when such a tenancy is under threat, and (b) when such a tenancy is terminated (whether or not this comes about through active repossession on the landlord's part). Through these means, local authorities should assess the numbers of tenancies created through rent deposit scheme referrals which are sustained for more than six months.

R8. Local authorities should be encouraged to adopt North Ayrshire's 'matrix model' for the systematic measurement of support needs and progress over time in addressing such needs.

The future

10.30 Despite the scepticism across some authorities with respect to the value of homelessness prevention, the research found that LAs were, at the very least, speaking the language of prevention. Many are investing time and resources in consolidating and extending prevention services. Some authorities are focusing on major changes, such as restructuring posts, whilst others are developing new projects. However, much of the planned activity focuses on improving joint working arrangements and raising awareness about homelessness and the role of prevention.

10.31 In requiring local authorities to produce homelessness strategies, the Executive's main objective was to foster a more prevention-centred LA approach. It appears that this has met with considerable success. Across the country it is clear that many prevention initiatives have been developed since 2003. Most authorities can point to various recently-established schemes or practices demonstrating a commitment to 'early intervention' which is at the heart of the prevention ethos.

10.32 Most LAs also report having arrangements in place for 'early warning' where a council tenant is at risk of eviction. It is, however, less clear at what stage such notifications occur and what actions they trigger. And, as yet, only a handful of authorities are routinely screening all housing register applicants to identify those possibly at risk of homelessness.

10.33 Increasingly, authorities are seeing the necessity of making more effective use of private tenancies as a means of averting homelessness. Nevertheless, there appears to be considerable scope for further work in this area - e.g. to develop registers of reputable landlords potentially willing to accommodate tenants in receipt of Housing Benefit and committed to providing LAs with early warning of instances where established tenancies are at risk.

10.34 In most authorities, the adoption of a more 'prevention-centred' approach has had significant implications for the nature of homelessness work. Some prevention activities (e.g. family mediation or administration of rent deposit schemes) may be delivered by voluntary agencies working under contract. In many cases, however, preventative activities have been slotted in alongside the routine homelessness caseworker duties of investigating and assessing applications. This shift has managerial implications in terms of recruitment and training of homelessness staff. By comparison with their counterparts of the 1990s, such staff are now more in need of skills such as negotiation, counselling and multi-agency working.

10.35 In addition to addressing the issue of home visits, key areas where official guidance is called for include the way that 'housing options' interviews are slotted into the process of handling homelessness presentations and the monitoring issues which arise here. More broadly, LAs are eager to receive Executive advice on the monitoring of prevention activities and their effectiveness.