A Third Return to Homelessness, the Regulator Rethinks Inspection, and Service Charges Go on Trial
Welcome to this week's Supported Housing Briefing. Below: the five developments that matter most for providers and local authorities, followed by three Deep Dives for subscribers — this week on the reversion evidence and the "real difference" test, the tightening consumer-standards regime, and service charges as the new evidential front line.
This week's Top 5
1. Almost a third of supported housing residents return to homelessness within three years
New research finds that approximately 30% of people placed in supported housing are back in homelessness within three years of placement. The finding cuts to the heart of the sector's central justification — that supported accommodation delivers stable, lasting outcomes rather than a temporary stop. It lands in three live contexts: the long-running debate over quality and accountability in the exempt-accommodation sub-sector; the evidence base MHCLG and local commissioners use to design and assess schemes under the Supported Housing (Regulatory Oversight) Act 2023; and disputes where the adequacy of "support" is contested. Whether the research disaggregates outcomes by provider type will determine how sharply it bears on the exempt-accommodation segment specifically.
2. RSH consumer judgements: East Midlands Housing Group upgraded to C1, Lancaster City Council rated C2
The Regulator of Social Housing published two consumer-standard regulatory judgements on 12 August 2026. East Midlands Housing Group — historically active in the East Midlands supported and exempt-accommodation space — was upgraded to C1, the top consumer grade, after improvements in tenant complaint handling. Lancaster City Council, a local authority landlord, received a C2, signalling the need for improvement. The judgements continue a visible post-2023 Act pattern: local authority landlords disproportionately receive C2/C3 grades while larger associations increasingly reach C1 on re-inspection. For supported housing, a strong consumer grade functions as a regulatory halo affecting how commissioners view providers; the full judgements should be read for any care, support or supervision commentary.
Source: Regulator of Social Housing
3. RSH signals its inspection regime will evolve after the four-year cycle
The Regulator of Social Housing has indicated that the proactive inspection regime introduced under the Social Housing (Regulation) Act 2023 framework will be revised once the initial four-year cycle of inspecting all registered providers is complete. The current universal sweep has already surfaced consumer-standard failures at providers with supported housing portfolios. Any redesign — particularly a move toward risk-based or thematic inspection — could change how the RSH scrutinises the care, support and supervision obligations that underpin exempt-accommodation status. Providers with complex tenure arrangements may face more targeted rather than universal scrutiny. It is an early, strategically significant signal; the sector should watch for any RSH consultation or discussion paper on the post-cycle methodology before year-end.
4. Sanctuary loses bid to strike out large parts of £650,000 service-charge claim
A major registered provider, Sanctuary, has failed in its attempt to dismiss substantial parts of a £650,000 leaseholder claim over service charges, meaning the surviving grounds proceed to be tested. Though framed as a leaseholder matter, the case sits squarely at the intersection of service-charge law and the exempt-accommodation Housing Benefit model. Tribunal scrutiny of what service charges actually represent — what may be charged, to whom, and on what evidence — increasingly informs how local authorities and the DWP approach Housing Benefit claims from supported exempt-accommodation providers, where inflated or poorly evidenced charges have been a core enforcement concern. The ruling reinforces a judicial direction of travel toward granular, evidence-led assessment of charges.
5. Home REIT completes portfolio wind-down with sale of final properties
Home REIT has sold the final properties in its portfolio, completing the wind-down of a vehicle whose holdings were heavily concentrated in supported exempt accommodation. The REIT acquired properties let to specialist supported housing providers across the UK before collapsing amid governance and valuation controversy — one of the sector's most high-profile scandals. The completion marks a corporate endpoint, but not necessarily a legal one: residual questions remain over RSH regulatory notices affecting providers that held Home REIT leases, potential Housing Benefit recovery actions by local authorities, and possible civil litigation by liquidators or shareholders. For the exempt-accommodation market, the episode continues to shape how institutional capital and lease-based models are perceived.
Deep Dives
Substantive analysis for subscribers — grounded in the primary authorities that govern supported and exempt accommodation.
Deep Dive 1 — When a third don't stay housed: outcome data and the "real difference" test
The finding that roughly one in three supported housing residents returns to homelessness within three years will be read first as a policy failure. For practitioners advising providers and local authorities, it should be read as a legal-risk signal, because outcome data of this kind bears directly on the statutory gateway that makes supported housing financially viable.
That gateway is the definition of "exempt accommodation" preserved by paragraph 4(10) of Schedule 3 to the Housing Benefit and Council Tax Benefit (Consequential Provisions) Regulations 2006. Accommodation is exempt where it is provided by a housing association, registered charity, voluntary organisation or non-metropolitan county council, and where that body — or a person acting on its behalf — provides the claimant with care, support or supervision. The consequence of falling within the definition is substantial: the accommodation escapes the Local Housing Allowance caps and the rent-officer restriction scheme, so Housing Benefit is payable on an eligible rent that includes intensive housing-management and support-related service charges. The "care, support or supervision" limb is therefore not a soft descriptor; it is the condition on which enhanced subsidy depends.
The case law has spent two decades policing that condition, and it is where the reversion data bites. In the line of Commissioners' decisions associated with Commissioner Turnbull — CH/4432/2006, CH/150/2007, CH/200/2009 and the reported R(H) 4/09 — the tribunals established that support must be more than minimal, must be genuinely provided by or on behalf of the landlord, and must be distinguishable from the general housing management any landlord performs. Signposting, routine tenancy administration and the ordinary incidents of being a landlord do not count. Bristol CC v AW crystallised the point into the "real difference" test: the support must make a real difference to the claimant's ability to occupy and sustain the accommodation. Token, nominal or box-ticking support fails.
Read against those authorities, a 30% reversion rate is double-edged. It does not by itself prove that support was minimal — homelessness has many drivers, and a resident can receive genuine, well-evidenced support and still lose their tenancy. But it invites exactly the inquiry the Turnbull decisions demand: was the support real, was it more than minimal, and did it make a real difference? Where a local authority reviewing officer, or the DWP, sees systematically poor outcomes at a provider, that data becomes circumstantial evidence in a supersession or a First-tier Tribunal appeal that the support was never substantively delivered. The evidential burden of demonstrating that support was provided, and to the requisite standard, rests with the provider; poor outcomes make that burden heavier.
Allerdale BC v JD [2019] UKUT 304 sharpens the risk further. It reinforced that decision-makers must scrutinise who actually provides the support and whether it is genuinely delivered "on behalf of" the landlord, rather than being a paper arrangement layered onto a lease-based structure. Where support is subcontracted or nominal, the "on behalf of" limb is vulnerable. Combine Allerdale's scrutiny of the provider relationship with reversion data suggesting support is not translating into stability, and a provider faces challenge on two fronts simultaneously: the reality of the support and the identity of who delivers it.
The tension practitioners must manage is that the exempt-accommodation test is input-focused — it asks whether qualifying care, support or supervision is provided — while the emerging evidential environment is increasingly outcome-sensitive. The statute and case law do not require that support succeed; they require that it be real, more than minimal, and make a real difference. But reviewing officers, tribunals and the DWP will use outcome data as a proxy for testing those inputs, and a provider without granular records of support delivered will struggle to rebut the inference the data invites.
The data also arrives into a commissioning environment that SHROA 2023 is reshaping. The Act requires local authorities to carry out supported housing needs assessments and to develop local strategies, and it enables licensing regimes conditioned on quality. Reversion evidence will feed those assessments directly: a local authority building its needs assessment now has a defensible basis for setting licensing conditions that test support outcomes, and for declining to commission — or to treat as exempt — provision that cannot demonstrate effective support. The statutory input test under Schedule 3 paragraph 4(10) does not change, but the surrounding oversight architecture increasingly uses outcome data to decide where to look. Practitioners should also keep the distinction between "support" and "care" in view: personal care may engage Care Act 2014 and CQC obligations, whereas the exempt-accommodation limb is satisfied by care, support or supervision — so a provider defending its status should be precise about which limb it relies on and evidence that limb specifically.
Three practical implications follow. First, providers should treat support-planning and support-delivery records as legal documents, not care paperwork: contemporaneous, individualised evidence that support was offered and delivered is the single best defence to a real-difference challenge, and reversion data makes that evidence more likely to be demanded. Second, local authorities commissioning or paying HB on exempt accommodation should align their assurance processes with the Turnbull and Bristol CC v AW framework, testing the substance of support rather than the existence of a support plan template. Third, providers whose outcomes sit materially below sector norms should expect their exempt status to be probed and should be able to explain poor outcomes by reference to cohort complexity rather than absent support — because the alternative explanation, once the data is public, is that the "support" funded by enhanced HB was never more than minimal.
Deep Dive 2 — The consumer-standards ratchet meets supported housing's coverage gap
The Regulator of Social Housing's 12 August judgements — East Midlands Housing Group upgraded to C1, Lancaster City Council held at C2 — look like routine grading. For supported housing they mark the tightening of a regulatory ratchet that now runs in parallel with, and sometimes ahead of, the exempt-accommodation framework, and they expose a coverage gap that practitioners need to understand.
Since the Social Housing (Regulation) Act 2023 brought the RSH's revised consumer standards into force — Safety and Quality, Transparency Influence and Accountability, Neighbourhood and Community, and Tenancy — the regulator has moved from a reactive to a proactive, inspection-led model, grading providers C1 to C4. The regulator's signal this week that the inspection regime will itself evolve once the initial four-year cycle is complete indicates that this is not a settled framework but a developing one. For supported housing providers registered with the RSH, the consumer standards apply in full, and they interact awkwardly with the specialist realities of supported and exempt accommodation: shared facilities, high resident turnover, complex needs, and support delivered by staff whose primary remit is care rather than tenancy management.
That interaction matters because the RSH consumer regime and the Supported Housing (Regulatory Oversight) Act 2023 (SHROA 2023) are two tracks aimed at overlapping problems from different directions. The RSH regime governs registered providers as landlords. SHROA 2023 empowers national supported housing standards, a licensing regime that local authorities can operate for supported exempt accommodation, and mandatory local supported housing strategies and needs assessments — reaching providers who are not registered with the RSH at all. The two regimes are complementary in ambition but structurally disjointed: a resident's protection depends on which track their provider happens to fall under.
The coverage gap is the crux. A great deal of exempt accommodation is provided by small organisations — some registered with fewer than 1,000 homes, some not registered at all — operating lease-based models. As sector commentary this week reminded smaller registered providers, those below the size threshold are not subject to the RSH's routine inspection programme, even though the consumer standards bind them. The providers most associated with poor-quality exempt accommodation are frequently the least likely to be proactively inspected by the RSH, and are instead intended to be caught by SHROA's local licensing — which depends on individual local authorities standing schemes up. Until licensing is widespread, the worst provision can sit in the space between the two regimes: too small for RSH inspection, not yet subject to local licensing.
The judgements also illustrate a pattern with direct commissioning consequences. Local authority landlords are disproportionately receiving C2 and C3 grades, while larger associations increasingly reach C1 on re-inspection — a divergence Lancaster's C2 and EMHG's C1 exemplify. For supported housing, a weak consumer grade at a council that both provides and commissions accommodation is doubly significant: it signals internal compliance strain and may push the authority toward commissioning third-party supported provision, stimulating demand for exactly the exempt-accommodation market that SHROA was passed to discipline. A C1 grade, conversely, operates as a regulatory halo, shaping how commissioners and partner authorities view a provider's wider operations.
There is a live tension between the consumer standards' tenant-voice and quality expectations and the exempt-accommodation legal framework analysed in Deep Dive 1. The consumer regime pushes providers toward demonstrable, resident-led quality; the HB framework rewards the provision of care, support or supervision as an input. A provider can theoretically satisfy the Schedule 3 paragraph 4(10) gateway — delivering more-than-minimal support that makes a real difference under Bristol CC v AW and the Turnbull line — while still failing the consumer standards on complaint handling or safety, or vice versa. Allerdale BC v JD [2019] UKUT 304 shows the HB decision-maker probing who delivers support; the RSH probes whether the landlord is meeting consumer outcomes. Providers must now satisfy both lenses at once, with different evidence bases and different enforcers.
A further complication is measurement. The consumer regime leans heavily on the Tenant Satisfaction Measures and on tenant voice, yet supported housing residents — often in transitional placements, with fluctuating capacity and short tenures — are among the hardest cohorts from whom to capture representative satisfaction data. A provider delivering genuine, more-than-minimal support to a chaotic cohort may score poorly on standardised satisfaction metrics for reasons unconnected to quality, while the reversion data in Deep Dive 1 may drag the same provider's apparent performance down further. The RSH's enforcement toolkit — regulatory judgements, improvement plans and, in serious cases, use of its enforcement powers — will be applied against those metrics, so providers need to be able to contextualise specialist-tenure data rather than be judged against general-needs baselines. This is a concrete reason to engage early with how the post-cycle inspection methodology will treat supported housing.
Practically: registered supported housing providers should map their supported schemes against each consumer standard now, rather than waiting for an inspection whose methodology is about to change, and should assume the post-cycle regime may bring thematic, specialist-tenure inspections that scrutinise care, support and supervision directly. Providers below the routine-inspection threshold should not treat that as safety — the standards still bind them, and self-referral or a complaint can trigger engagement. Local authorities should recognise that their own consumer grades and their SHROA licensing decisions are connected levers on the same market, and should sequence licensing so that the providers falling through the RSH gap are the first to be brought into local oversight. The strategic watch-point is the RSH's forthcoming articulation of its post-cycle inspection model: if it moves to risk-based or thematic inspection of specialist tenures, supported housing will move from the periphery of the consumer regime to its centre.
Deep Dive 3 — Service charges as the front line: Sanctuary and the evidential turn
Sanctuary's failure to strike out large parts of a £650,000 leaseholder service-charge claim is, on its face, a leasehold dispute far from supported housing. Its significance for the sector lies in what it confirms about the direction of judicial scrutiny of service charges — because in exempt accommodation, the service-charge component is where enhanced Housing Benefit is made or lost, and the evidential standard the tribunals are hardening is the same standard local authorities are increasingly applying to HB claims.
Begin with why service charges are central to the exempt-accommodation model. Once accommodation qualifies as exempt under paragraph 4(10) of Schedule 3 to the Housing Benefit and Council Tax Benefit (Consequential Provisions) Regulations 2006, it escapes the LHA and rent-officer restrictions, and Housing Benefit becomes payable on an eligible rent that can include substantial service charges — for intensive housing management and for services connected with the provision of the accommodation. The financial architecture of much supported housing depends on those charges being eligible and being met by HB. But eligibility is conditional: under the Housing Benefit scheme, service charges are only met where they are connected with the provision of adequate accommodation, are services the claimant is obliged to pay as a condition of occupation, and are reasonable. The reasonableness and connectedness of each charge is a decision for the local authority, appealable to the First-tier Tribunal.
That is precisely the inquiry the Sanctuary ruling reflects in the leasehold context. The refusal to strike out means the surviving grounds — which concern whether the charged sums are properly recoverable and adequately evidenced — will be tested on their merits. The First-tier Tribunal (Property Chamber) and Upper Tribunal have for some years been moving toward granular, evidence-led assessment of what a service charge actually represents, rather than accepting a global figure. A large registered provider being unable to dispose of the challenge summarily reinforces that direction of travel: the burden is on the party levying the charge to show, item by item, what the charge is for and that it is reasonable.
The read-across to exempt accommodation is direct. Where HB is claimed on service and support charges, the same evidential logic applies, and it connects back to the Turnbull line and the "real difference" test in Bristol CC v AW. A support charge is only eligible if it corresponds to support that is actually delivered and is more than minimal; a charge for "support" that cannot be evidenced as genuine, delivered support is vulnerable both as an ineligible service charge and as a failure of the underlying exempt-accommodation gateway. In other words, the reasonableness scrutiny of service charges and the substantive scrutiny of the care, support or supervision limb are two expressions of the same demand: show, with evidence, what was actually provided. CH/4432/2006 and the associated Turnbull decisions require that support be real and more than minimal; the service-charge framework requires that the sum charged for it be connected and reasonable. A provider that fails one is likely to fail the other.
Allerdale BC v JD [2019] UKUT 304 completes the picture. Its scrutiny of whether support is genuinely provided "on behalf of" the landlord maps onto service-charge disputes where support is subcontracted and the charge is passed through opaque arrangements. Where the support charge funds a third party whose delivery cannot be traced back to the landlord's responsibility, both the "on behalf of" limb and the reasonableness of the charge are exposed. The hardening evidential standard visible in Sanctuary gives local authority reviewing officers and tribunals the confidence to demand that traceability.
The tension for practitioners is between the financial model and the evidential regime. Exempt-accommodation viability often depends on service and support charges pitched well above general-needs levels, justified by the intensity of management and support. That is legitimate where the intensity is real and documented. But the direction of travel — a property tribunal unwilling to strike out a granular service-charge challenge, local authorities emboldened to interrogate exempt-accommodation charges, and a case-law framework that ties charge eligibility to demonstrable, more-than-minimal support — means that undocumented or loosely justified charges are increasingly indefensible.
The stakes are amplified by the recovery regime. Where a local authority later concludes that a service or support charge was ineligible — or that the accommodation never qualified as exempt because the support was minimal — the consequence is not merely prospective reduction but potential overpayment recovery, often over lengthy periods and, depending on the facts, recoverable from the provider as the person to whom HB was paid. An adverse service-charge finding can therefore convert a viability assumption into a retrospective liability. That is why the evidential turn Sanctuary confirms is a balance-sheet issue for providers, not merely a compliance one.
Three practical implications. First, providers should build a charge-by-charge evidence base: for every element of the service and support charge, a record of what the service is, why it is connected to the accommodation, how the cost is calculated, and — for support charges — evidence that the support was delivered to the real-difference standard. Second, lease-based and subcontracted structures should be stress-tested against Allerdale: if support delivery cannot be traced to the landlord's responsibility, both the charge and the exempt status are at risk. Third, local authorities should align their HB service-charge scrutiny with the property-tribunal standard now emerging, testing charges item by item rather than accepting global figures — and should expect that robust scrutiny, applied consistently, is now well supported by the judicial direction Sanctuary confirms. Service charges are no longer the technical afterthought of a supported housing claim; they are the front line on which its legitimacy is decided.
