A Return to Purer Waters? The Supreme Court Re-Casts the Pereira Test by Jed Meers (University of York)

In his oral arguments in the recent Supreme Court decision in Hotak, Jan Luba QC argued that ‘the safest course when dealing with a statute is to drink from its pure waters.’[1] He was talking about s.189(1)(c) Housing Act 1996, which details the ‘priority need’ hurdle homeless applicants must navigate to be owed a housing duty by their Local Authority.

Another post on this blog has outlined s,189(1)(c) and the problems with its interpretation through the so-called Pereira Test. The approach taken was to assess the applicant’s vulnerability against that of an ordinary street homeless person, which in turn led many homeless applicants who plainly appeared to be in need of assistance – such as Mr Ajilore – to be denied a housing duty under s.193 Housing Act 1996. This position had come under sustained criticism for moving far beyond the ‘pure waters’ of the statute, and creating discretionary space for the infiltration of considerations absent from the original intentions of the legislation.

However, in a landmark decision, the Supreme Court has reconsidered the meaning of ‘vulnerable’ under s.189(1)(c), ruling that its interpretation should be revised.

The facts of the three claimants (outlined in detail in paras.19-33) each turn on a particular element of the Pereira Test. In his lead judgment, Lord Neuberger outlined the three key issues addressed in the case: (i) whether the assessment of vulnerability requires an exercise in comparability, and if so, to whom; (ii) whether support from others can be taken into account when assessing vulnerability; and (iii) what role, if any, does the Public Sector Equality Duty under the s.149 Equality Act 2010 play. I will deal with each of these.

Firstly, the judgment is clear that ‘vulnerability’ necessitates some exercise of comparability, as an adjective which carries ‘a necessary implication of relativity’ (para.51), akin to the assessment of being ‘tall’ (para.92). Clearly, comparison demands a comparator. Here, the Court dispensed with the ordinary street homeless person, which was considered to be against the original intentions of the legislation and lead to ‘arbitrary and unpredictable outcomes’ (para.56). In other words, in the use of a street homeless person as a comparator, the test was assessing vulnerability using an already incredibly vulnerable benchmark. Instead, the Court assessed that an ‘ordinary person who is in need of accommodation’ (para.59) was the appropriate comparator.

The question in interpreting s.189(1)(c) therefore, is – simpliciter- whether the applicant would be, when homeless, significantly more at risk of harm than an ordinary person in need of accommodation (para.52). Perhaps with some of the more ornate constructions of comparators by Local Authorities in mind, the Court was clear that this ‘ordinary person’ is: ‘robust and healthy’ (para.71), made without reference to the resources available to the Local Authority (para.39), and cannot be formed from the ‘very dangerous’ use of statistics (para.43).

Secondly, the majority of the Court considered that taking into account support from third parties was a relevant factor which flows from the ‘contextual and practical’ nature of s.189(1)(c) (para.62), albeit with the requirement that such support would – if it were to mitigate a finding of priority need status – have to be provided on a ‘consistent and predictable basis’ (para.65).

The final key issue is the role played by the Public Sector Equality Duty (PSED). The judgment highlighted it as ‘complementary’ duty, requiring a sharp focus on: (i) whether the applicant has a protected characteristic under Chapter 1 of the 2010 Act, (ii) the extent of this, (iii) the likely effect of the protected characteristic in the broader context of the applicant’s position if they were to be made homeless, and (iv) whether this results in them being ‘vulnerable’ (para.78).

The ruling of the Court is a long-awaited decision which is likely to have far-reaching consequences. There are two points worth noting here.

Firstly, the revised judicial interpretation coupled with the complementary PSED, appears to enhance obligations towards those with disabilities. Indeed, it may well be the case that the findings of fact inherent in the process to claim DLA (or PIP) would prima facie render the applicant vulnerable. Even for lower rate components, the factual tests inherent – such as being unable to walk outdoors ‘without guidance or supervision from another person most of the time’ (s.73(1)(d) SSCBA 1992) – would appear to indicate that such an applicant, when homeless, would be significantly more at risk of harm than a ‘robust and healthy’ (para.71) ordinary person in need of accommodation (para.52).

Secondly, there is clear anxiety in the lead judgment to ‘emphasise the primacy of the statutory words’ (para.59), and consequently avoid the construction of already highly vulnerable ‘street homeless’ comparators. However, this inevitably leads to the question of what characteristics – if any – such a comparator possesses, and how ‘significantly more at risk of harm’ should be interpreted with reference to them (para.51).

Though this judgment is a positive step forward from the old downward drag of Pereira, the spaces it leaves in its re-casting will result in significant further litigation. In emphasising the original meaning of the statute, the Supreme Court has attempted to return to ‘purer waters,’ but quite how the revised interpretation of s.189(1)(c) will work in practice within Local Authorities already straining under huge financial pressures, remains to be seen.

[1] Incidentally, for those looking to fill a quiet evening, it is now possible to watch all UKSC hearings on their ‘on demand’ service. The footage of Jan Luba QC’s remarks can be see here https://www.supremecourt.uk/watch/uksc-2013-0234/151214-am.html at 58:04 – 58:17.


By Jed Meers, University of York


Housing Law and the Welfare State – the role of ‘law’ in governing state-citizen relations

The Downward Drag of Pereira: Ongoing disputes with Vulnerability By Jed Meers, University of York

In 2012, Nicola Sturgeon heralded the scrapping of ‘priority need’ assessments for homeless households in Scotland.[1] She described the change – which effectively results in anybody unintentionally homeless being owed the full housing duty – as ‘the most progressive homelessness legislation in Europe.’ In the aftermath of the Scottish Referendum, a great deal has been written about policy divergences north and south of the border. Sadly, for those of us in England, recent cases before the High Court and Court of Appeal bring into sharp relief how brutal our treatment of the homeless can be in comparison.

In England and Wales, one of the hurdles homeless applicants still need to navigate to be owed the full housing duty is the ‘priority need’ requirement. They can do so by demonstrating they fall into one of the categories outlined in s.189 Housing Act 1996 – a standard poor law inspired barometer of pregnant women, those with children, or people in old age – with a separate catch-all category of being ‘vulnerable’ for some ‘other special reason.’ There has been a gradual decline in the numbers accepted under this category; a trend demonstrated in figure one.[2]



It is this construction of ‘vulnerable’ which can generate a downward drag on those accepted as being in priority need. The term is not defined anywhere in the legislation itself, which has, unsurprisingly, led to a smorgasbord of legal appeals grappling with the concept and attempts by the Courts to define and refine the term on multiple occasions. The current position – stemming from Bowers [3] and given added gloss by Pereira [4] – is whether the applicant ‘when homeless, [would be] less able to fend for himself than an ordinary homeless person so that injury or detriment to him will result when a less vulnerable man would be able to cope without harmful effects.’[5]

There are two key issues to note in its interpretation. Firstly, following Johnson,[6] the test concerns an ‘ordinary homeless person’ who is already homeless – not somebody who is non-homeless who then becomes homeless. Secondly, with particular significance to recent appeals, the ‘ordinary homeless person’ is a ‘necessarily imprecise’[7] hypothetical construction which is largely left to the imagination of the local authority. It is assumed that ‘a busy local housing authority will have a vast experience of the range of homeless persons,’[8] and as such, they are best placed to create the appropriate comparator.

The problems this test can cause have been ably demonstrated in some recent case law. The judgment in Hotak[9] serves as an example of the Catch-22 situation which can present itself when Local Authorities stretch the Pereira test outwards. Here, the appellant had learning difficulties, a history of self-harm, depression and post-traumatic stress disorder. He relied on his brother for daily support, including ensuring his personal hygiene, changing clothes, cooking meals and handling his finances.[10] The Local Authority decided that he was not more vulnerable than the ‘ordinary homeless person,’ as he had the support of his brother – something homeless persons rarely enjoy.

Their decision was upheld by the Court of Appeal, who saw support from others as being part of the ‘intensely fact sensitive and practical’[11] process of assessing vulnerability. The judgment was applied more recently in Kanu,[12] where the applicant suffered from Hepatitis B and haemorrhoids, and family support resulted in them falling foul of the Pereira test. However, could the facts of Mr Hotak’s case not be taken both ways? If he did not receive support, could it be argued that he is living independently and therefore is no less vulnerable than an ‘ordinary homeless person’?

This downward drag effect is particularly clear when Local Authorities draw on questionable evidence to support their construction of this ‘ordinary homeless person’ comparator. A good example was provided earlier this month by the Court of Appeal in Ajilore,[13] where the applicant suffered from hepatitis B, had a history of class-A drug use, and was suicidal. The Local Authority officer found that he was not in priority need, in part due to his assertion that ‘it was not unusual for street homeless people to have thoughts of self-harm and suicide,’[14] an assertion supported by a ‘report published in the British Medical Journal in 2005 [which] confirmed that homeless people do have higher self-harm incidents than the ordinary population.’ In short, a lot of homeless persons want to kill themselves, so that does not render an applicant any different from an ‘ordinary homeless person.’

However, the Court found that this ‘report’ was not a published piece of research at all. In fact, it was a letter sent to the Journal by a General Practitioner, which detailed the result of a survey of 116 of their patients. In this small sample, 7.2% had an episode of self-harm or suicide attempts. This did not prevent the Court upholding his decision, deciding instead that ‘the statistics were used by the reviewing officer in order to back up the conclusions that he had drawn from his own knowledge of an ordinary street homeless’ person.[15] Effectively, the statistics were for the purposes of illustration, instead of providing evidence per se.

The same approach has been used recently to justify denying a housing duty to those who have used drugs. In Johnson v Solihull,[16] the applicant was a recovering heroin addict who had been in-and-out of prison since the age of 16. The Local Authority did not consider him ‘vulnerable’ in line with s.189, citing in their decision letter a study by Homeless Link finding that 92% of homelessness services work with people who use drugs. The logic goes that a lot of homeless people are recovering drug users, or are still using, so that does not render an applicant any different from an ‘ordinary homeless person.’

The picture as it stands appears to be at odds with the purpose behind both the Pereira Test and s.189 Housing Act 1996. The principle is to ensure that those who are more vulnerable receive assistance, but the formation of a comparator which is already very vulnerable – and supported by what can often be shaky evidence – pushes the test downwards and denies the housing duty in questionable cases. Hotak is on appeal and will be heard at the Supreme Court on 10th December, which will be the first time the highest court has considered this issue. Hopefully they will take the opportunity to re-assess the test as it currently stands, and halt its downward drag.

Jed Meers, University of York  (@jed_meers)



[1] See Homelessness (Abolition of Priority Need Test) (Scotland) Order 2012

[2] Data taken from the Department of Communities and Local Government

[3] R v Waveney DC Ex p. Bowers [1983] Q.B. 238

[4] R. v Camden LBC Ex p. Pereira (1999) 31 H.L.R. 317

[5] Pereira [329] per LJ Hobhouse

[6] Johnson v Sollihul [2013] EWCA Civ 752

[7] Ibid [18] per LJ Arden

[8] Tetteh v Kingston upon Thames Royal LBC [2004] EWCA Civ 1775 [21] per LJ Gage

[9] Hotak v Southwark LBC [2013] EWCA Civ 515

[10] Hotak [4] per LJ Pitchford

[11] Ibid [39] per LJ Pitchford

[12] Kanu v Southwark LBC [2014] EWCA Civ 1085

[13] Ajilore v Hackney LBC [2014] EWCA Civ 1273

[14] Ajilore [22] per LJ Gloster

[15] Ajilore [39] per LJ Gloster

[16] Johnson v Solihull [2013]



Using the Social Security System to Deliver Housing Policy By Louise Cheung, University of Southampton

The interaction between housing policy and social security law has been my primary research interest for the last few years. My PhD research has looked at several overarching themes to unpick this topic, including; government measures to prevent repossession, the ideology of homeownership, and the state versus individual responsibility for housing costs. The political basis behind the ideology of homeownership which takes into account financial investment considerations in conjunction with the x -factor benefits of home is of particular interest, and it has been my aim to marry this together with the state’s approach towards social security law and policy. The idea of home in law has been an ever increasing popular research area in property law. To gain an understanding of the nebulous idea of home; I would refer to the excellent account of the conceptualisation of home in law by Fox O’Mahony as primary reference material.

In the UK, it is fair to say that we are a nation of homeowners and this is a product of a string of governmental policies throughout the 20th and 21st century. I have found particularly interesting the contrast from the UK and continental European countries where renting is the main tenure, and the reasons why. For an insightful perspective on this, I would refer to a superb article by Matthew Phillips about why most Germans rent. In particular, the development of social housing and its relative decline since the 1980s is a pet interest of mine when contemplating the notion of tenure. There are many fascinating accounts of the history of social housing, including, Cole and Furbey’s ‘The Eclipse of Council Housing’.

The key means-tested social security welfare benefits that I discuss in my research are Housing Benefit for renters, and Support for Mortgage Interest for mortgaged homeowners. The increasingly residualised nature of the government safety net housing costs is a key aspect of my thesis’ discussion of these welfare benefits for housing costs. With the upcoming Universal Credit reforms, these benefits will be shelved, but the theory driving the use of tax payers’ money for the use of these welfare benefits for housing in both tenures remains relevant. The eligibility criteria, the method of delivery and the use of bright line rules all give details about the policy direction which drives the regulation governing these social security benefits. Statistics about the number of welfare benefit recipients, analysis of the mention of housing in key political manifestos throughout the 20th century, and budget records are all primary material which are analysed to give this research further depth to the exploration of material, to come to functional and evidenced conclusions.

A great difficulty in my research has been pinpointing the exact nature of housing policy, and the key objectives which run through it. Through researching the history of housing policy, it seems to me that there are two common threads which guide housing policy in amongst some incoherent policy; (1) provide support for rented accommodation for low-income households (mostly as a safety net) (2) to promote homeownership to enhance individual responsibility for housing costs. It is these two guiding objectives which tie together the various discussions relevant to my research, alongside the key theoretical underpinning of my thesis, which is measuring the relative shifting of state towards individual responsibility for housing costs, be that in the rented or owner-occupier sector.

It seems to me that property law lends itself to discussion of policy. Property law is about governing relationships between land/structures and people, and policy drives relationships that people have with the land/structures, and the impact people have on land/structures. If you would like to read more about contemporary legal research in property, please visit our blog, where discussion pieces about property law’s relationship with sustainability, policy and social justice can be found. The blog was created through the gathering of early career academics at the postgraduate stream of the most recent Modern Studies in Property Law Conference in Liverpool. The property collective editors welcome blog posts from those interested in property research and its connection to sustainability, policy and social justice.


Louise Cheung (@MsLouiseCheung)

PhD Candidate, University of Southampton, Law School, Faculty of Business and Law

Property Graduate Trainee, Hertfordshire County Council