In Guiste v Lambeth LBC (2019) EWCA Civ 1758, the Court of Appeal returned again to the meaning of Lord Neuberger’s eliptical phrase in Hotak v Southwark LBC that, for the purposes of the homelessness provisions in the Housing Act 1996, vulnerability meant being significantly more vulnerable than ordinarily vulnerable as a result of being made homeless.  The decision in Guiste (I’m told that it is pronounced “Geest” as opposed to “Gwist”) in some respects is one on its facts, but the Court of Appeal make a number of observations of significance in these cases and leave one point open (albeit give their penniworth on it).  As an academic interested in the field, I wonder at the amount of effort,time, and money spent in arguing the toss about vulnerability, and whether there might be better uses of that effort/time/cash, but there we go; that’s why we have our tower.

In short, the Court found that the reviewer failed to engage with the substance of an expert report provided on Mr Guiste by a distinguished consultant psychiatrist, Dr Judith Freedman, as to his mental state if he became homeless. Dr Freedman found that, if homeless, “… Mr Guiste would have increased depression and anxiety. He would be at risk for self-harm and suicide, particularly in response to his auditory hallucinations“. The review officer appeared to accept that there would be an increased risk of suicidality, but in the next sentence wrote “I do not think there is current evidence to indicate you would experience harm or deterioration as a result of homelessness“. The Court of Appeal (Henderson LJ giving the lead judgment, and with whom Rose LJ and Theis J agreed) found that the review officer had not given reasons as to why Dr Freedman’s evidence had been discounted. There was, therefore, a breach of the principles of rationality and fair decision-making (65).

One can pull out four rather key points from the judgment, which address some common issues with the Hotak approach, as expanded on in Panayiotou v Waltham Forest LBC:

(a) There has been some discussion in the cases about the identity of the ordinary person and their abilities. We know from Rother DC v Freeman-Loach (2018) EWCA Civ 368 that the reviewer does not need to define vulnerability and nor their identity, but there has been some discussion about whether the ordinary person is healthy or not. In Guiste, the Court identifies the ordinary person, with whom the applicant is to be compared, to be “in normal health” and does not have the physical or mental health or disabilities of the type that would render them vulnerable (54). It shows how far we have come in the jurisprudence on vulnerability (such that it is) that the Court of Appeal needed to re-affirm this, but there we go.

(b) The second point is about the way a reviewing officer treats medical evidence, as against the authority’s own advisors. In Guiste, Lambeth had drawn on a series of pieces of advice given to them by NowMedical’s psychiatric advisors, Drs Eskander and Wilson. In what will be familiar language to practitioners in the field, they had written, for example, that “there is no evidence of significant impairment in functioning as a result of his mental health“, “I am unable to find evidence of a severe and enduring mental disorder“, and “I acknowledge that the applicant reports occasional suicidal thinking and has a history of self harm and auditory hallucinations, although these appear related to social circumstances and substance misuse. My view would therefore be that the applicant does not have psychiatric issues of particular significance when compared to an ordinary person if homeless“. In the same paragraph as that which comments on the review officer’s reasoning, Henderson LJ said,

This evidence, from a distinguished consultant psychiatrist, and directed to the key legal point in issue, could not in my view be disregarded, and if the review officer was going to depart from it, I think it was necessary for her to provide a rational explanation of why she was doing so. … If (the reviewing officer) was intending to base her conclusion on the views of the two psychiatrists instructed by NowMedical, she needed to explain why their views should prevail over that of Dr Freedman, when they were less highly qualified that she is, and (more importantly) they had never met or interviewed Mr Guiste. Equally, I find it hard to see how [the reviewing officer] could rationally have given more weight to the report of the consultation at St George’s Hospital in September 2017 than to the more recent and much fuller report of Dr Freedman, which (unlike the earlier report) also focused on the critical question of the effect that homelessness would have on Mr Guiste’s mental health.

The Court was then reinforcing the weight to be attached to the medical evidence and adding just a little further nuance to the guidance in Shala v Birmingham CC about the way in which reviewing officers should approach the medical evidence. It is also to be noted that one of the representations made by Mr Guiste’s solicitors on review was that there should be a meeting or some ind of discussion between Dr Freedman and NowMedical (as suggested by Sedley LJ in Shala), which Lambeth did not follow up.

(c) Thirdly, the Court addressed whether vulnerability test implied some sort of “functionality” requirement. Lambeth’s counsel, Niamh O’Brien, argued that “… the relevant question is whether the particular circumstances of Mr Guiste would affect his functionality (my emphasis) so as to make a noticeable difference to his ability to deal with the consequences of being homeless” (68). This submission picked up on Lewison LJ’s approach in Panayiotou, where he had quizzically discussed Lord Neuberger’s speech in Hotak in the following terms:

One of the themes that runs through previous decisions of this court is that there must be a causal link between the particular characteristic (old age, physical disability etc) and the effect of homelessness: in other words some kind of functionality requirement. We now know that the functionality is not an ability to “fend for oneself” nor an ability “to cope with homelessness without harm”. But if it is not that, what is it? The nearest that Lord Neuberger came to providing an answer was in saying that section 189(1)(c) is concerned with: “an applicant’s vulnerability if he is not provided with accommodation”

It is worth quoting in full what Henderson LJ said (at 69-70):

I am unable to accept this submission, which would import an extra layer of complexity into a test which is already far from simple to expound. Lewison LJ’s observations on functionality were made in the context that there must be a causal link between the particular characteristic relied on under section 189(1)(c) and the effect of homelessness. They were not in my judgment intended to introduce a new and additional test, over and above the requirement for a causal link between the relevant characteristic and the effect of being made homeless. Nor is it clear to me how this supposed further requirement should be formulated, or what the minimum ingredients of such functionality would be. Ms O’Brien provided us with a list of such factors in her oral submissions, while acknowledging that the precise content of the requirement would always depend on the circumstances of the case; but she was unable to cite any authority for this approach, apart from the passage in Panayiotou which, as I have explained, goes only to the question of causation.
Furthermore, if the submission were correct, it would have some surprising consequences. Mr Westgate gave the example of a person who, by application of the Hotak comparison, is found to be likely to become seriously ill, as a direct result of being made homeless. Provided that the necessary causal link exists between the illness and the relevant protected characteristic under section 189(1)(c), it is hard to see any reason why the applicant should also have to satisfy some ill-defined test of impairment of functionality.

(d) The final point, which was left open, was as to the applicability of s. 31(2A), Senior Courts Act 1981 to statutory homelessness appeals. S. 31(2A) allows a court to refuse relief in judicial review cases “if it appears to the court to be highly likely that the outcome for the applicant would not have been substantially different if the conduct complained of had not occurred”. This is an argument one hears in the county court (quite a bit) and it would have been helpful to have an authoritative statement on its applicability. I do wonder how much it actually matters – in rationality cases, like Guiste, by their very nature, there is likely to be a “a very real chance” (73) that the authority will come to a different conclusion on review; and, in other cases, it is likely to be improbable that a court can make that judgment on the (potentially changed) facts that will be unknown to the judge but known to the reviewer after a further review. So, as much as anything else, this may be a question of principle, and the court in Guiste, expressed the view that “In provisional agreement with the submissions of Mr Westgate, I am inclined to think that there is no proper basis for extending the scope of the new test in section 31(2A), by judicial decision, to statutory housing appeals under section 204 of the 1996 Act. The question is not free from difficulty“, but there was no need to express a final opinion.