Earlier today, @suesspiciousmin tweeted that a new case might have rebooted the stream of section 20 compensation claims:
I couldn’t wait for his commentary, I had to search the case out for myself, and it is this case here: London Borough of Hackney v Williams & Anor  EWCA Civ 26 (26 January 2017).
Not least, section 20 is an issue I’ve written about on a number of previous occasions, over a number of years, including:
- ‘Can I? Should I? Must I? …take this baby into care’ (May 2008)
- ‘Hobson’s Choice – Voluntary or Compulsory Removal of a Child?’ (March 2012)
- ‘What’s going on with Section 20?’ (December 2015)
A review of my previous posts will reveal that I have held to a very difficult line. On the one hand, I have defended the flexible use of section 20, against those who would argue that it is unjust to use a so-called “voluntary” provision when in truth you have powers of compulsion available, because it denies parents the opportunity to test whether the threshold is crossed. I have defended it on the grounds that it is a useful tool in the social worker’s toolkit, expanding the scope for non-compulsory intervention, since even when the threshold might be crossed, parents might prefer to work co-operatively with the local authority. On the other hand, I have not defended misuse, and have been clear in particular that it is quite right that section 20 should not be “compulsion in disguise”, and should be underpinned by informed consent.
So I sought out today’s case, with some trepidation, to see whether I would have to review and revise the line that I have taken. And my conclusion is: no, I do not need to review and revise that line. But unfortunately, that is in part because this is not really a very good decision of the Court of Appeal. Disappointingly, it is weak and loosely reasoned on the very points where it needs to be a model of clarity.
I am going to make my points by comparing and contrasting the issues that arise out of just two of the cases considered by the Court of Appeal today. Both are decisions of Munby, one when he sat as a High Court Judge, and the latter sitting in the Court of Appeal and as President of the Family Division. For reasons that will become apparent, I will consider the latter first.
In this case, the President took the opportunity to review a long line of section 20 cases. Along the way, he approved the reasoning of a number of decisions of lower courts. He also significantly extended the requirements for valid consent to section 20 accommodation, making clear that:
i) Wherever possible the agreement of a parent to the accommodation of their child under section 20 should be properly recorded in writing and evidenced by the parent’s signature.
ii) The written document should be clear and precise as to its terms, drafted in simple and straight-forward language that the particular parent can readily understand.
iii) The written document should spell out, following the language of section 20(8), that the parent can “remove the child” from the local authority accommodation “at any time”.
iv) The written document should not seek to impose any fetters on the exercise of the parent’s right under section 20(8).
v) Where the parent is not fluent in English, the written document should be translated into the parent’s own language and the parent should sign the foreign language text, adding, in the parent’s language, words to the effect that ‘I have read this document and I agree to its terms.’
These requirements are problematic. Such requirements for a valid section 20 agreement go way beyond the bare requirements of the statute that there should be no objection. Accompanied as it is by threats of damages, it smacks of judicial activism, looking more like rewriting the statute than interpreting it.
Some of us have coined the term ‘presedent’ to highlight this phenomenon:
‘Presedent’ is a hybrid of President and precedent, and is intended to reflect the tendency of the President of the Family Division to say things in the context of delivering judgments, in the expectation that they will be treated as judicially binding precedent. For the more legally pedantic among us, something is not binding precedent just because a judge has said it: it needs to form part of the ratio of the case, which means effectively that it is only that part of the judge’s reasoning that was necessary to decide the issue in the case that is binding. But this is a subtle point. For most practitioners of other professions – and it should be recalled that these remarks were directed to another profession, namely the social work profession – the key idea about caselaw and precedent is to ensure that similar cases are decided in similar ways, and they do not want to be troubled by trying to distinguish ratio decidendi from obiter dicta.
Since the President makes a habit of this, many of us myself included would have really welcomed some sort of clear judgment that explains when the President’s words matter and when they don’t and can safely be ignored. And to be fair, the Court of Appeal had a go. It repeatedly referred to the principles established in the line of caselaw on section 20 as good practice guidelines, thereby distinguishing it from law. It observed that the comments on section 20 in this case were unnecessary to this judgment. And it pointed out that the focus of today’s case was a claim for compensation, underpinned by a claim of breach of statutory duty, and that a breach of statutory duty has to relate to the wording of the statute – section 20 – rather than a later line of caselaw. So far, so good.
And yet… And yet… I’m afraid that what it says about good practice guidelines just doesn’t stack up. Let’s turn to the other case, the one that started it all off, the Nottingham Baby case:
By way of reminder, in this case, a baby was removed at birth, from a hospital, without a care order. In its defence, Nottingham highlighted that this course of action had been in the birth plan drawn up in advance, which the mother knew of, and to which she had not objected. Munby was scathing about this defence:
But the local authority here seemed to be going far beyond this. It seemed to be conflating absence of objection with actual consent – a doctrine which at least in this context is, in my judgment, entirely contrary to principle and which, moreover, contains within it the potential for the most pernicious consequences, not least because there are probably many mothers who believe, quite erroneously, that a local authority has power, without any court order, to do what the local authority did in this case.
To equate helpless acquiescence with consent when a parent is confronted in circumstances such as this with the misuse (or perhaps on another occasion the misrepresentation) of non-existent authority by an agent of the State is, in my judgment, both unprincipled and, indeed, fraught with potential danger…
Submission in the face of asserted State authority is not the same as consent. In this context, as in that, nothing short of consent will suffice.
I want to make two points quite forcefully about Munby’s reasoning in this case.
Firstly, for my part, I am in no doubt that this was the ratio of the case (the bit that even the legally pedantic approach to precedent would accept is legally binding). It is a specific rebuttal of the heart of Nottingham’s defence, that the mother had not objected. Even the Court of Appeal today seems to accept that it was the ratio of the case (“Munby J’s decision on this point may have been determinative and have led to his overall conclusion”, paragraph 69). If Munby was wrong, the way to change the precedent was to appeal it. It was not appealed, and it has instead been approved on countless occasions subsequently, including by the Court of Appeal itself.
Secondly, for my part I am equally in no doubt that it was correctly decided. The Nottingham case was not a compensation case, it was a judicial review. Judicial reviews are challenges by the citizen to the state, that the state has acted unlawfully due to some form of recognised conspicuous procedural unfairness. These forms of conspicuous procedural unfairness have been established in the case law of the Administrative Court. When Munby held for the reasons he did that consent was required and not merely the absence of objection, this was not on the basis of pure statutory construction, but on the basis of well-established administrative law principles of fairness in the relationship between the citizen and the state. When the Court of Appeal today says as it does in paragraph 77 that a breach of these principles of fairness “does not found an action for judicial review”, I think it is simply wrong: an action for judicial review is exactly what it does found.
So, in highlighting that the detailed procedural requirements established in N (Children : Adoption: Jurisdiction)  EWCA Civ 1112 (02 November 2015) are merely good practice guidelines, the Court of Appeal may have done us a favour. But in going further, and suggesting that the principles of procedural fairness that underpinned the earlier decision in G, R (on the application of) v Nottingham City Council  EWHC 400 (Admin) (05 March 2008) might equally be merely good practice guidelines, the Court of Appeal has done us no favours at all.
I cannot think that this case will be the last word on the matter.