Download paper


This paper concerns the judicialisation of administrative justice and is prompted by a recent decision of the UK Supreme Court. The case arose in the context of housing, but for these purposes is relevant insofar as it raises critical constitutional questions about, among other things, the relationship between the Strasbourg Court, the domestic courts and administrative justice. A homeless refugee is offered accommodation by the local authority in the form of a flat with one round and one oblong window in the living room. She turns down the ‘final offer’ on the ground that the windows remind her of the prison in Iran where she was tortured. She claims that this particular flat will exacerbate the post-traumatic stress disorder, anxiety attacks and other conditions from which she suffers. The question goes to a statutory review by a housing officer, who rejects her argument. She has thereby rendered herself homeless and ended the local authority’s obligation to house her. What is there in these facts to require the attention not only of the housing review officer, but also of one county court judge, three Court of Appeal judges and five Supreme Court justices in addition? According to the Supreme Court in Poshteh v Royal Borough of Kensington and Chelsea, these facts raised two significant questions: First and most important, was this a situation governed by the European Convention (ECHR) and had the domestic courts been correct in their earlier rulings that it was not? Secondly, had the reviewing officer applied the right test to the situation and had her reasoning been correct? In its answer to these two questions, the Supreme Court acted to limit the inroads that the ECHR has long threatened to make into administrative law and practice, aiming to limit the judicialisation of administrative justice. This paper considers and commends the Supreme Court’s reasoning, arguing that the decision is a promising instance of a “Parliament Square Axis” for reasonable interpretation of human rights law.

Question 1: is the Convention applicable?

This case is the latest stage in what the Supreme Court politely calls a ‘continuing debate’ between the European Court of Human Rights (ECtHR) on the one hand, and the British courts and government on the other, conducted ‘against the background of the uncertain Strasbourg jurisprudence’. The dispute concerns the ambit of Article 6(1) of the ECHR, which provides that, in the determination of his ‘civil rights and obligations’, everyone is entitled to ‘a fair and public hearing’ by ‘an independent and impartial tribunal established by law’. Over the years, the ECtHR has given this provision an expansive meaning so that it has encroached on many administrative decisions in the fields of immigration, social security and urban planning that would initially not have been considered to bestow ‘civil rights and obligations’ (See Feldbrugge v The Netherlands for an example). The ECtHR’s motive is understandable: to grant to administrative decision-making a wide general immunity from human rights law would be effectively to deprive judicial protection to the beneficiaries of welfare assistance and might tempt governments to convert adjudicative into administrative decisions that would then lie outside the reach of the Convention. This is a problem that national systems of administrative law have had to face over the years and which they have largely learned to deal with (see notably, Charles Reich’s pieces here and here). Equally, the reaction at national level is understandable: adjudicative decision-making is costly and time-consuming and the judicialisation of claims to welfare services through the operation of Article 6(1) has resource implications for the state and public authorities. This is a viewpoint that the domestic courts have taken rather seriously.

This issue first arose before the House of Lords in Begum v London Borough of Tower Hamlets, where the appellant, a homeless single mother eligible for assistance and in priority need, had similarly turned down the final offer of local authority accommodation. The parties agreed that the case involved three questions, of which the first two were: (1) whether the review officer’s decision was a determination of ‘civil rights’; and (2) whether the review procedure amounted to a hearing by an ‘independent and impartial tribunal’ for the purposes of article 6(1). The Law Lords chose to focus on the second question, avoiding a direct answer to the first by making the assumption that Article 6(1) was applicable. Instead, they focused on the question whether review by a council officer amounted to ‘an independent and impartial tribunal’ and, since it clearly did not, whether the further statutory appeal to a county court gave the court sufficiently ‘full jurisdiction’ to guarantee compliance with Article 6(1). The Law Lords ruled by a majority that it did. A similar question reached the Supreme Court in R (A) v Croydon LBC, R (M) v Lambeth LBC, which involved the provision of accommodation to ‘children in need’. Here again, the Court preferred to leave the Article 6(1) issue open, deciding the case on domestic law. Nonetheless, Lady Hale and Lord Hope addressed the question at considerable length in obiter dicta with a careful exploration of the ECtHR jurisprudence. Lord Hope, identifying a somewhat unsatisfactory distinction in the jurisprudence between benefit claims that gave rise to specified rights and those involving a measure of administrative discretion, felt that it could ‘now be asserted with reasonable confidence’ that the local authority’s duty in the instant case did not give rise to a ‘civil right’ within the meaning of Article 6(1).

In the 2010 case of Ali (Tomlinson) v Birmingham City Council, the Supreme Court had to face the Article 6(1) issue of ‘civil rights’ squarely. The claim turned on a finding of fact as to whether the appellants had received a letter from the housing officer, which they denied having received, though they were disbelieved by the reviewing officer. The Supreme Court, however, chose not to dwell on this point. Instead, they chose to rule unanimously that Article 6(1) was not engaged. Lord Hope, in the leading judgment, revisited his previous distinction between benefits that are ‘defined precisely’, to which Article 6(1) is applicable, and benefits that are essentially ‘dependent upon the exercise of judgment by the relevant authority’, where no ‘civil right’ for the purpose of the ECHR is involved, concluding that the reviewing officer’s decisions did not engage Article 6(1). Lord Collins agreed that there was ‘no right to any particular accommodation’; the duty was essentially of a public nature and did not give rise to an individual economic right; nor did ‘a dispute concerning the question whether the applicant has been properly notified of the consequences of refusal of accommodation’ come within Article 6(1). Lord Kerr expressed himself as in agreement but puzzled. Noteworthy in both these decisions is the care and respect with which the Supreme Court examined the ECtHR jurisprudence.

In Poshteh, the case which prompts this paper, the matter arose once more during the legal proceedings that stretched from the County Court to the Supreme Court. In the meantime, however, Ms Ali had appealed to Strasbourg and (some five years later) the ECtHR had ruled that she did have a legally enforceable statutory right to be provided with accommodation and that a genuine and serious dispute existed over the continuing existence, if not the content, of that right. Thus Article 6(1) applied and the applicant had a right to a fair hearing before an independent and impartial tribunal. The ECtHR was, however, prepared to concede that the statutory review and appeal system provided adequate protection for her civil right. Poshteh provided the Supreme Court with the first opportunity to come to grips with this new ruling. The decision was unanimous: Article 6(1) was not engaged. Lord Carnwarth, speaking for the Court, once more trawled conscientiously through the domestic and European case law before sending a clear message to Strasbourg on the quality of its jurisprudence.

Although the ECtHR had acknowledged that weight was to be given to the interpretation of the relevant provisions by the domestic courts, it was ‘disappointing’ to find that it had failed to address in any detail either the reasoning of the Supreme Court, or indeed its concerns over ‘judicialisation’ of the welfare services and the implications for local authority resources. Instead, the ECtHR had concentrated its attention on two admittedly obiter statements, respectively by Hale LJ (as she then was) in the Court of Appeal in Adan v Newham London Borough Council, and Lord Millett in Runa Begum. However, its treatment of these two statements was open to the criticism that they were taken out of context, and without regard to their limited significance in the domestic case law. Lord Carnwarth continued:

  1. Our duty under the Human Rights Act 1998 section 2 is [to] “take account of” the decision of the court. There appears to be no relevant [ECtHR] decision on the issue, but we would normally follow a “clear and constant line” of chamber decisions (see Manchester City Council v Pinnock [2011] 2 AC 104, para 48). This might perhaps be said of some of the previous decisions referred to in the judgment, including most recently Tsfayo v United Kingdom (2006) in which the application of article 6 was conceded by the government. However, it is apparent from the Chamber’s reasoning … that it was consciously going beyond the scope of previous cases. In answer to Lord Hope’s concern that there was “no clearly defined stopping point” to the process of expansion, its answer seems to have been that none was needed. That is a possible view, but one which should not readily be adopted without full consideration of its practical implications for the working of the domestic regime.
  2. The scope and limits of the concept of a “civil right”, as applied to entitlements in the field of public welfare, raise important issues as to the interpretation of article 6, on which the views of the Chamber are unlikely to be the last word. In my view, this is a case in which, without disrespect to the Chamber, we should not regard its decision as a sufficient reason to depart from the fully considered and unanimous conclusion of the court in Ali. It is appropriate that we should await a full consideration by a Grand Chamber before considering whether (and if so how) to modify our own position.

Question 2. Standard of review

On the second question, which raises issues of domestic law that became relevant when the Supreme Court ruled that Article 6(1) was inapplicable, it is sufficient to note the deference shown by the Court to the officials tasked with difficult decision-making in the housing field. Housing officers occupy a post of considerable responsibility and have substantial experience in the housing field. There were no grounds (as counsel had argued) for ratcheting up the standard of review. It would not be proper ‘to subject the decisions of housing officials to the sort of analysis that might be applied to a contract drafted by solicitors, to an Act of Parliament, or to a court’s judgment.’ As Lord Neuberger had put it in Holmes-Moorhouse v Richmond upon Thames London Borough Council:

[A] benevolent approach should be adopted to the interpretation of review decisions. The court should not take too technical a view of the language used, or search for inconsistencies, or adopt a nit-picking approach, when confronted with an appeal against a review decision. That is not to say that the court should approve incomprehensible or misguided reasoning, but it should be realistic and practical in its approach to the interpretation of review decisions.

The legal test to be applied in considering whether it was unreasonable to refuse an offer of accommodation combined subjective and objective factors. Succinctly, it was whether a right-thinking local housing authority would conclude that it was in all the circumstances ‘reasonable that this applicant should have accepted the offer of this accommodation’. It was wrong to submit the decision letter, as counsel for the appellant had done, to an ‘over-zealous linguistic analysis’. Viewed as a whole, it read as ‘a conscientious attempt by a hard-pressed housing officer to cover every conceivable issue raised in the case… against the background of serious shortage of housing and overwhelming demand from other applicants, many no doubt equally deserving’. There is a warning here to lower courts to avoid ratcheting up reason-giving requirements to a standard of rationality that can never in practice be met. There is a warning too about departure from the ‘traditional Wednesbury tests for administrative decisions in general’. This had ‘potentially profound constitutional implications’ for which, in the instant case, no convincing reasons had been given.

Implications: European and domestic

There are two main lessons to be learned from this apparently routine administrative dispute: the first international in character, the second domestic. At the European level, Lord Carnwarth’s remarks in the unanimous Poshteh decision represent a further step in a developing new relationship between the domestic courts and the ECtHR. The first hint that the Supreme Court might be turning away from the so-called ‘mirror image’ principle enunciated in R (Ullah) v Special Adjudicator came in R v Horncastle, where Lord Neuberger said that the requirement to ‘take into account’ the Strasbourg jurisprudence would normally result in the Supreme Court applying principles clearly established by the Strasbourg Court but, he added:

there will be rare occasions where this Court has concerns as to whether a decision of the Strasbourg Court sufficiently appreciates or accommodates particular aspects of our domestic process. In such circumstances it is open to this court to decline to follow the Strasbourg decision, giving reasons for adopting this course.

This exception — which reflects the wording of s.2 of the Human Rights Act that a court determining a human rights question ‘must take into account’ ECtHR jurisprudence — was justified by Lord Neuberger on the ground that it was ‘likely to give the Strasbourg Court the opportunity to reconsider the particular aspect of the decision that is in issue, so that there takes place what may prove to be a valuable dialogue between this court and the Strasbourg Court’. In Horncastle, the Court declined to follow the ECtHR. In Manchester City Council v Pinnock where, on the other hand, the Court followed the Strasbourg jurisprudence, Lord Neuberger amplified the new approach:

This Court is not bound to follow every decision of the ECtHR. Not only would it be impractical to do so: it would sometimes be inappropriate, as it would destroy the ability of the Court to engage in the constructive dialogue with the ECtHR which is of value to the development of Convention law…. Of course, we should usually follow a clear and constant line of decisions by the ECtHR…. But we are not actually bound to do so or (in theory, at least) to follow a decision of the Grand Chamber. As Lord Mance pointed out in Doherty v Birmingham [2009] 1 AC 367, para 126, section 2 of the HRA requires our courts to “take into account” ECtHR decisions, not necessarily to follow them. Where, however, there is a clear and constant line of decisions whose effect is not inconsistent with some fundamental substantive or procedural aspect of our law, and whose reasoning does not appear to overlook or misunderstand some argument or point of principle, we consider that it would be wrong for this Court not to follow that line.

There is now a consistent line of cases that lifts the domestic courts from the position of subordination to which Ullah had reduced them and restores them to the position of authority intended by the drafters of the Human Rights Act. Taking its rightful place amongst the supreme courts of other nations, the Supreme Court has indicated pretty clearly to the ECtHR that further incursions of the Convention into the field of administrative justice require far better reasoning if they are to be accepted into domestic law. Whether this exemplifies dialogue or a dressing-down, I prefer not to say.

Secondly, at the domestic level, Poshteh raises important questions about the scale and nature of administrative justice. Extending the boundaries of Article 6(1) to enclose large areas of administrative decision-taking inevitably involves judicialisation, which in turn requires resources, something that, Lord Carnwarth noted with disapprobation, the ECtHR overlooked: it had ‘failed to address in any detail either the reasoning of the Supreme Court, or indeed its concerns over “judicialisation” of the welfare services, and the implications for local authority resources’. This is not the first time that our judges have expressed similar concerns. In Runa Begum, the House of Lords decided to focus on the question whether the composite statutory procedure of (i) internal review by a superior housing officer plus (ii) appeal on a point of law to the county court satisfied Article 6(1), avoiding the question whether Article 6(1) was applicable. The House ruled that it was. As Lord Hoffmann put it, ‘an extension of the scope of article 6 into administrative decision-making must be linked to a willingness to accept by way of compliance something less than a full review of the administrator’s decision’. Appeal to a county court was commensurate with a judicial review and both were sufficient to satisfy Article 6(1). He observed in addition that Parliament was entitled to take the view that it is not in the public interest for an excessive proportion of the funds available for a welfare scheme to be consumed in administration and legal disputes. His comments were cited with approval by Lord Hope in Ali.

Tsfayo v United Kingdom involved a claim for backdated welfare entitlements in the context of a housing claim. The appellant argued that because she had not received the relevant correspondence, her claim should not be ruled out-of-time. However, the local authority’s housing benefit review board upheld the decision to refuse the claim on the ground that there was no ‘good cause’ for the delay. The ECtHR gave two main reasons for finding a violation of Article 6. First, the decision-making process was ‘significantly different’ from that in Runa Begum, where the issues to be determined ‘required a measure of professional knowledge or experience and the exercise of discretion pursuant to wider policy aims’. Second, the housing benefit review board was not only lacking in independence from the executive, but was ‘directly connected to one of the parties to the dispute’. An adjudicative body comprised of five members of the authority responsible for paying the benefit was tainted at source; there was a ‘fundamental lack of objective impartiality’. The fact that review boards had already been replaced by a separate system of statutory tribunals may have helped the ECtHR towards this decision.

In R(A) v Croydon, there were no statutory review provisions and the remedy was by way of judicial review. Considering whether this was a sufficient safeguard, Lady Hale said:

I would be most reluctant to accept, unless driven by Strasbourg authority to do so, that article 6 requires the judicialisation of claims to welfare services of this kind… Every decision about the provision of welfare services has resource implications for the public authority providing the service. Public authorities exist to serve the public. They do so by raising and spending public money. If the officers making the decisions cannot be regarded as impartial, and the problem cannot be cured by the ordinary processes of judicial review based upon the usual criteria of legality, fairness and reasonableness or rationality, then tribunals will have to be set up to determine the merits of claims to children’s services, adult social services, education services and many more. Resources which might be spent on the services themselves will be diverted to the decision-making process. Such a conclusion would be difficult, if not impossible, to reconcile with the decision of this House in Runa Begum.

The degree of judicialisation required of an administrative decision was flexible, depending on the nature of the decision. It was this passage that Lord Carnwarth in Poshteh asserted had been taken out of context and misapplied in Ali.

As it has evolved, the case law turns on a number of unfruitful distinctions driven by the Strasbourg jurisdiction on the meaning of the term ‘civil rights’ for the purposes of Article 6; for example, between specified rights and rights dependent on the exercise of administrative discretion (Lord Hope) or decisions that involve fact-finding and those that require expertise (Tsfayo). It might be more profitable to invoke the classic distinction of English administrative law between ‘administrative’ and ‘judicial’ decision-making. Where clearly adjudicative machinery, such as a tribunal or independent review board, is set in place, there is a good case for treating the appeal as a step in a judicial process that must measure up to the requirements of Article 6(1) under the tests established by Bryan v United Kingdom. Where the process is administrative in character, as with internal review or investigation by an ombudsman, then Article 6(1) should not apply. This would have the effect of involving Parliament in the decision, for it would be for Parliament to decide whether to institute a process that was administrative or adjudicative in character. It should not, however, leave claimants devoid of procedural protections. As Lord Bingham noted in Ruma Begum:

The narrower the interpretation given to ‘civil rights’, the greater the need to insist on review by a judicial tribunal exercising full powers. Conversely, the more elastic the interpretation given to ‘civil rights’, the more flexible must be the approach to the requirement of independent and impartial review if the emasculation (by over-judicialisation) of administrative welfare schemes is to be avoided… Although I do not think that the exercise of administrative functions requires a mechanism for independent findings of fact or a full appeal, it does need to be lawful and fair (at [5]).

This is very much in line with developments in the field of administrative justice in recent years, where the emphasis has been on ‘proportionate dispute resolution’. (See here and chapter 10 in here). Core adjudicative principles that reflect the well-known Franks Committee mantra of ‘openness, fairness and impartiality’ are emerging.

In an era of dwindling legal aid, the idea that every minor case of fact-finding needs a fully judicialised body is quite simply inappropriate as is a volume of expensive and time-consuming litigation on fine points of institutional design. Moreover, it fails to take into account the polycentric nature of this type of decision-making. Housing stock and funding for housing are finite; reinstating Ms Tsfayo in the housing list would mean a changed position for others.

Thus the Supreme Court is perfectly justified in taking a firm line and, indeed, at domestic level there would seem to be much inter-institutional agreement on the issue. The appellate system of internal review plus appeal on a point of law was inserted into the Housing Acts by Parliament. The Government has intervened in all the main cases, consistently arguing that Article 6 has no application to this type of case and drawing attention to the effect on decision-making procedures of a more general extension of Article 6 into areas of government activity relating to social security and other forms of welfare. In Ali, Lord Hope stated that one reason why the House of Lords had preferred, in Runa Begum, not to decide the Article 6(1) question, was ‘the wish not to inhibit the government from developing the arguments in the Strasbourg court should it become necessary to do so’, a point noted by Lord Carnwarth in Poshteh (at [24]). As I have remarked elsewhere:

Dialogue affords the best hope of preserving the distinctive British culture of rights and reinforcing the democratic element of the parliamentary sovereignty doctrine while allowing at the same time for progress. Parliament and the courts must be prepared to engage constructively in a process of coordinate construction; equally, they must face up to the need for tough, multi-level dialogue with Strasbourg. A ‘Parliament Square axis for human rights’ is needed.

On this occasion, the ‘Parliament Square axis’ for administrative justice should be applauded.

Carol Harlow
Emeritus Professor, London School of Economics


With kind permission from Hart Publishing, we are pleased to make available Professor Harlow’s essay, which was published in N. Barber, R. Ekins, and P. Yowell (eds), Lord Sumption and the Limits of the Law, OUP, 2016. Click here to download the paper.   Hart Publishing are delighted to offer 20% discount on the book! Click here to order online and use the code CV7 at the checkout to obtain your discount.