FYE v Middlesbrough City Council

The Upper Tribunal has a block of 34 appeals involving two local authorities. All the cases concern the provision of supported accommodation for claimants. This decision deals with the two lead cases, one from each authority. I will give a separate decision dealing with the other 32 cases.

All the cases involve the issue whether the claimant’s liability was created to take advantage of the housing benefit scheme. The evidence on this issue is common to all the cases. In every case, the local authority decided that the liability had been so created, with the effect that the claimant was not entitled to housing benefit. I am following regular usage by calling this contrivance, but it is no more than a label for the correct test, which is in the wording of the legislation. The First-tier Tribunal used a variety of words, but they were also convenient labels for the statutory language.

Some of the cases also raise the issue whether the claimants were living in exempt accommodation. In the simplest terms, this means whether they were receiving support in addition to their accommodation. In those cases, there is individual evidence relating to each of the claimants.

The support issue only arises if the claimant’s liability was not created to take advantage of the scheme. The First-tier Tribunal decided that the liability was so created. As I have found no error of law on that issue, I have decided not to deal with the support issue. First, my decision on it could not affect the outcome of these appeals. Second, it would be a disproportionate use of Upper Tribunal time to deal with all the cases in which support was in issue, given that it is immaterial. This explains why I have not dealt with appeal ground 4. This also explains why I have not dealt with some aspects of the other appeal grounds that relate to support. I deal with this in Section J below.

I have made orders protecting the identity of the claimants. They are all disabled and vulnerable to some extent. Moreover, they were not involved directly or personally in creating their tenancies; this was handled by others on their behalf.

Section 130 of the Social Security Contributions and Benefits Act 1992 creates housing benefit:

Section 137 provides for regulations:

Regulation 9 of the Housing Benefit Regulations 2006 (SI No 213) is made under the authority of section 137(2)(i):

As I have found no error of law in respect of the contrivance first issue, I do not need to set out the provisions dealing with the support issue.

The funding model used in these cases is a common one. There is a company that acquired a property and adapted it to be suitable for providing supported accommodation. The company lets the premises, in this case to a charity. The charity provides accommodation and support for housing benefit claimants. If the local authorities had not refused the claims for housing benefit, they would have paid benefit for the claimants, including if appropriate an element for their support. This would have covered the costs of the charity, including its rent to the head landlord.

There is in principle nothing objectionable about that model. I note that in April 2025, the Regulator of Social Housing published Focus report: Lease-based provision of Supported Social Housing. The Report was concerned only with issues that might undermine the viability and efficiency of the model. Otherwise, it recognised the value of the model in providing the necessary accommodation.

The ultimate issue in this case was whether the way the model was used involved a contrivance.

The tribunal gave detailed reasons for its decision in 174 paragraphs over 40 pages. No summary can do it justice. Later, I set out the tribunal’s own summary of its reasoning. For the moment, I will give sufficient background information for those reasons to be understood. Later, I refer to or quote other parts of the reasons as necessary to make sense of my reasoning.

By way of background, the owner of the properties at the relevant time was Enabling Homes Ltd. I hope I am right about that, but ultimately the name of the owner does not matter. The properties were bought by one company and immediately sold to another at a huge profit. They were then let to the charity, My Space Solutions Ltd, with no break clauses and full repairing responsibilities. Mr Paul O’Rourke was involved in the companies involved and in the charity. He was also involved in a company that provided financial support for the charity; this allowed it to continue to operate when the local authorities refused the claims for housing benefit. The charity has never engaged in what might be described as usual fund raising activities. In short, the tribunal found that Mr O’Rourke was controlling the actions of the companies and the charity, appearing in different roles at different times. He and his operations have been the subject of an article in Inside Housing and a Panorama programme called ‘The Housing Benefit Millionaire’. They have also been the subject of adverse decisions by the Regulator of Social Housing and an inquiry by the Charity Commission.

This is how the tribunal summarised its reasoning:

There is more than enough caselaw on this expression. It is sufficient to say that ‘take advantage’ is used in the sense that a writer in the romance genre might refer to the dastardly villain taking advantage of the heroine. It conveys an improper advantage being taken. The focus is on the creation of the tenant’s liability, which means the time when the tenancy was made. It is, though, permissible to use later evidence provided that it can help to show that the creation of the liability was a contrivance.

This is ground 3.

I am starting with this ground, because it is essential to understand the nature of the appeals to the First-tier Tribunal, as this affects the decision-making process undertaken by that tribunal. It also affects the approach taken by the Upper Tribunal when identifying an error of law.

The ground argues that it was procedurally unfair for Judge Barber to refuse to direct that evidence be called by the local authorities to support their decisions on contrivance. I reject this ground.

I was given a note of what the judge said during a case management hearing. This is what it says word for word:

I am sure that the judge would have expressed himself with more precision if he had put his comments into writing and honed them for dissection by counsel and scrutiny by the Upper Tribunal. This note of what he said is, though, sufficient to show both what he meant and that it does not involve an error of law. His remarks refer to two features of these appeals.

The first feature is a general one, applying to all appeals to the First-tier Tribunal in social security matters. In Subesh v Secretary of State for the Home Department [2004] EWCA Civ 56, Laws LJ described what he called the default nature of the appeal process, but he added that it was:

The nature of the appeal process in social security cases is an example of ‘a specially restricted right of appeal’. The appeal process in those cases is the closest civil example I know to the Crown Court instance described by Laws LJ. From 1948 at least, the First-tier Tribunal and its predecessors have undertaken a complete reconsideration of a claimant’s entitlement to benefit. That is now qualified by paragraph 6(9) of Schedule 7 to the Child Support, Pensions and Social Security Act 2000 (equivalent to section 12(8) of the Social Security Act 1998). The authorities of the Social Security Commissioners and now the Upper Tribunal do not require the First-tier Tribunal to have appropriate regard to the decision-maker’s knowledge and experience. This is unusual in appeals.

Taking that approach, decision-makers who act in the name of a local authority or the Secretary of State are not called to give evidence either in writing or in person. They may make submissions in support of their decisions, but they do not give evidence on their thought processes. The judge was right not to require them to give evidence.

The second feature is unique to provisions like regulation 9(1)(l). It does not depend just on matters of fact. The tribunal must first make its findings of fact on events that have taken place and states of affairs that have existed. Having done so, the tribunal has to decide whether the claimant’s liability was created to take advantage of the scheme. That requires an exercise of judgment. It might be described as an inference drawn from the actions and circumstances of the case. Or it might be described as assessing whether the facts satisfied the legal test. Either way involves an evaluation that requires the tribunal to consider all relevant facts and circumstances. Evidence has no role to play at the evaluation stage. That is part of the point the judge was making. To put it another way, evidence from the decision-maker would not be relevant.

The nature of regulation 9(1)(l) is reflected in its language. Subparagraphs (a)-(k) merely identify a fact or state of affairs that must exist. To take one simple example, subparagraph (k) applies if the claimant ‘is in residential accommodation’. In contrast, subparagraph (l) applies if the local authority ‘is satisfied that the liability was created to take advantage of the housing benefit scheme’. The reference to the authority being satisfied reflects the element of judgment involved in applying the provision.

I will give an example of the point where fact-finding moved on to evaluation in the tribunal’s decision:

The judge’s decision did not prevent evidence being given on matters of fact. All it did was to rule out evidence on how the decision-makers carried out the evaluative judgment required by regulation 9(1)(l).

Mr Manning argued that the claimants were entitled to rely, indeed should only rely, on this ground after the tribunal had made its decisions. Challenging it at the time would have caused additional cost and delay. And it would, of course, have been immaterial if the claimants had won their appeals. Mr Bhose argued that the claimants should have applied for permission to appeal at the time. I accept Mr Manning’s argument. The Upper Tribunal does not encourage appeals against case management decisions, which are usually better considered in the context of the proceedings as a whole.

There was one thing that Judge Barber was wrong about. It makes no difference, but it is important. When he said at the end of the note that it was for the judge to decide whether there was contrivance, he did not anticipate that the hearing would be before a panel consisting of a judge (himself as it happened) and a financially-qualified member. Perhaps the decision on composition had not been made at that stage. But her presence contributed to the tribunal’s understanding of the issues, the questioning of the witnesses, the assessment of the evidence, and the tribunal’s findings and conclusion. The Upper Tribunal takes that into account when assessing any criticism of the tribunal’s reasoning on financial matters.

Let me spell out what all this means for the way I have dealt with this case. Some findings of fact were not in dispute. Other findings were in dispute. The tribunal was required to give an adequate explanation of how it made those findings and it must disclose that its assessment of the evidence was rational. For those findings that benefited from the knowledge or experience of the panel members, I have made appropriate allowance in assessing whether the finding involved an error of law. I have found that the tribunal was entitled to make all the findings that it made that are relevant to contrivance.

Coming to the evaluative judgment, this required the tribunal to take account of the undisputed facts and its findings in order to decide if there had been contrivance. There is ample authority on the need to take account of the nature of the exercise involved in deciding on the combined effect of a range of factors some of which support one judgment, some support another, and the rest are equivocal. I have not needed to apply that approach in this case. Given the facts found by the tribunal, there was only one judgment that it could properly make. It made that judgment: there was a contrivance.

This ground is headlined as dealing with errors of law and begins with the burden and standard of proof. As that is a fundamental matter, it is convenient to deal with this ground next.

This is ground 2(i). It argues that the tribunal failed to direct itself correctly and to apply the burden and standard of proof. I reject this ground.

The standard of proof on the balance of probabilities only applies to matters of fact, not to evaluative judgments. See Secretary of State for the Home Department v Rehman [2003] 1 AC 153 at [56]. That said, tribunals are all familiar with the burden and standard of proof. The Court of Appeal used them as an example when it said in EJA v Secretary of State for the Home Department [2017] EWCA Civ 10 at [27] that ‘some principles are so firmly embedded in judicial thinking that they do not need to be recited.’ On appeal, a tribunal is assumed to know them unless its decision or reasons show otherwise (Piglowska v Piglowski [1999] 1 WLR 1360 at 1372).

There is also this point. The outcome of a civil case will usually be determined by the findings of fact rather than by applying the burden and standard of proof. Tribunals are under a duty to make findings of fact. Save in exceptional cases, those findings will be determinative and render any reference to the burden or standard unnecessary. See Morris v London Iron & Steel Co Ltd [1988] QB 493 at 504 and 507.

This is ground 2(ii). It argues that the tribunal was wrong to rely on the report on Exempt Accommodation from the House of Commons’ Levelling UP, Housing and Communities Committee.

I reject this argument and the various developments of it. The tribunal dealt with the report in paragraphs 89 to 93 of its written reasons. In paragraph 89, it said that the report did not reference My Space and was ‘therefore of a more general relevance’. Later, it made this more explicit:

The report does not in itself take the matter we have to decide one way or the other, but the above paragraph is relevant to our deliberations. The lease-based model can be a perfectly legitimate mechanism for providers of supported housing to increase their stock in circumstances where outright purchase might be impossible due to the capital cost of the property. That must be right. However, as the Committee alluded to, there may well be circumstances where profits associated with the lease-based model might in some way form a ‘disguised profit income stream’ for the head landlord.

So, three times in the course of five paragraphs, the tribunal said that the report did not affect its actual application of regulation 9(1)(l) to the facts and circumstances of the individual cases. It used the report merely as background briefing on issues. Mr Manning argued that ‘The general scrutiny of Parliament is in no way probative in relation to an individual case’ and accepted that ‘the parts of the report quoted do not refer to the legal tests or controls at all.’ That confirms the point I have just made.

I see no reason to doubt what the tribunal said. Mr Manning’s arguments proceed on the basis that the tribunal was treating the Report, or at least the passages it quoted, as an authoritative statement of the law. The arguments then criticise the tribunal’s reasoning and conclusions on that basis. I reject that approach and the arguments on which it is based.

As to the rest, I accept Mr Bhose’s argument that the level of the rents charged was sufficient to show that there was a profit from public funds. Whether that profit went to the charity or ultimately to the head landlord does not matter. What matters is that the claimants’ liability was created to take advantage of the scheme.

It may be that a profit will often, or even regularly, be a feature of a contrivance, but I do not see why this must be an essential feature. Provided there is a contrivance, it does not matter whether it turns out to be a profitable one or even whether it was intended to be profitable. What matters is that public funds are spent unnecessarily or inappropriately. I do not, though, have to decide that point.

This is ground 2(iii).

In paragraph 5(g) of its written reasons, the tribunal said that the charity failed to appreciate that the rent standard applied. I accept Mr Bhose’s argument that this was not put in issue before the tribunal. There was evidence that the Regulator of Social Housing was concerned about this and the claimants had not argued, or provided evidence, otherwise.

This ground is headlined as dealing with relevant and irrelevant considerations.

This is ground 1(i).

A finding on valuation is a matter of fact. These appeals can only succeed if the tribunal made a mistake of law in making its findings. A mistake of fact is not sufficient.

The tribunal described itself as expert. I regularly see tribunals using this term. At one time, an expert tribunal meant one that was entitled to make findings based solely on the expertise of the panel. The medical appeal tribunal was the classic example. It consisted of a legal chairman and two consultants. The consultants would examine the claimant and the tribunal could rely on their findings to make its own diagnosis of the claimant’s condition: see Diplock J in R v Medical Appeal Tribunal (North Midland Region) ex parte Hubble [1958] 2 QB 228 at 241. The First-tier Tribunal exercising the housing benefit jurisdiction is not expert in that sense. It must rely solely on the evidence. But it is entitled to use its experience to assess the evidence before it. In this case, the financially-qualified member will have had practical experience in dealing with valuation matters. And I know from seeing the judge’s name over the years that he is an experienced judge in social security matters, including housing benefit. In assessing whether a finding involved an error of law, the experience on which it was based is accorded appropriate respect.

The tribunal relied on its combined knowledge and experience in both cases, as it was entitled to do. At paragraph 38, it did so when accepting that the figures provided by the local authority ‘are well within in the ballpark of renting a flat’ in the relevant post code. And at paragraph 46, it did so when finding that ‘the desktop valuation broadly confirms our understanding of the … rental market as at the time of the valuation.’

This reliance is important for two reasons. First, the tribunal’s reasoning focuses on the actual valuation and tests its likely accuracy on the basis of its own experience. This reduces the significance of the factors set out in this ground. Second, this in turn is relevant in deciding whether any failure to take account of a particular consideration was material.

Having accepted evidence and explained the basis on which it did so, it was not necessary for the tribunal to go on and explain why it did not accept other evidence that did not accord with the tribunal’s knowledge and experience. Having seen the other valuations in evidence and discussed them at the hearing before me, there were other reasons why the tribunal might have rejected them, but its reasons are adequate as they stand.

This is ground 1(ii). It consists of criticisms of a number of points in the tribunal’s reasoning.

Later, I accept that tribunal’s statement in paragraphs 113-114 that one factor alone was sufficient by itself to show contrivance. None of these points relate to that part of the tribunal’s analysis. And, although it is not necessary to decide the point, I do not consider that they undermine my view that the tribunal’s evaluative judgment on contrivance was unassailable given the totality of the factors it took into account.

In paragraph 5(h) and (i), the tribunal took into account factors relating to support rather than taking advantage of the housing benefit scheme. Assuming for the sake of argument that the tribunal should not have done so, did their inclusion undermine the cogency of the other reasons?

In HK v Secretary of State for the Home Department [2006] EWCA Civ 1037 Neuberger LJ explained how to identify whether an isolated error contaminated a decision as a whole so as to make it material:

The tribunal’s reasons have to be read as a whole. I have already set out paragraph 5 of the tribunal’s written reasons. Before listing the individual reasons, the paragraph said that they collectively showed that regulation 9(1)(l) applied. But I note that the very first words of the paragraph are ‘Broadly speaking’. I also note what the tribunal said in paragraph 114, which has to be read in conjunction with paragraph 113.

Although the issue of the provision of care requires separate consideration, we also took the following into account in arriving at our general view that these liabilities were created to take advantage of the housing benefit scheme. For the reasons we set out below, care was not provided at any reasonable level in the Middlesbrough appeals and we think this was probably because importance was not placed upon the provision of care so as to channel resources to cover the costs associated with paying the head landlords and ensuring housing management standards were maintained.

Paragraph 5 refers to the reasons ‘collectively’, but paragraph 114 says one fact was ‘alone … sufficient’ and paragraph 141 says that support considerations were also taken ‘into account’. It would be easy to read these as self-contradictory or at least as inconsistent. I do not read them in that way. These paragraphs and their different wording are not unusual in tribunal’s reasons. This reflects the complexity of decision-making when multiple factors are involved. There is a range of possibilities. The following are just a sample. There may be one factor that alone is decisive. Or there may be a number of factors any one of which would be decisive. Or a number of factors may collectively lead a tribunal to its conclusion, although none on its own would be sufficient. Or there may be one main factor, with others providing additional degrees of support. Or there may be factors that merely confirm a decision made on other grounds. Paragraph 139 is an example of this in operation. This is the concluding paragraph in a section dealing with targeting support needs tenants. The paragraph begins: ‘We think this supports our conclusion on contrivance.’ I see no reason to doubt what the tribunal said. In other words, its conclusion was reached independently of this point.

The task for the Upper Tribunal is to assess the reasoning as a whole, taking account both of what the tribunal itself said as well as an objective judgment of the significance of particular parts of the reasoning. The purpose of this assessment is to decide whether some flaw in the reasoning is sufficient to make it a material error of law.

Taking the approach identified in HK, I accept the tribunal’s statement in paragraph 114 at its word. It is clear in its language and meaning. It is not inconsistent with the statement at paragraph 5 that the tribunal also relied on the collective effect of that and other reasons. The tribunal’s findings on support were not significant to its overall assessment. First, because of what the tribunal said in paragraph 114. Second, because support was not in issue in GPZ and the other Sunderland cases. The tribunal could not have relied on it in those cases. If it was an essential part of the reasoning, it would mean that the tribunal’s reasoning in FYE and the other Middlesbrough cases could not have stood on its own, without relying on a conclusion on an issue that did not arise in those cases.

For the reasons I have given, the First-tier Tribunal did not make any material error of law in either case.