UK Constitutional Law Association

affiliated to the International Association of Constitutional Law

HRA Watch: Reform, Repeal, Replace? Mark Elliott: Could the Devolved Nations Block Repeal of the Human Rights Act and the Enactment of a New Bill of Rights?

Editors’ note: The blog is inviting constitutional lawyers to comment on the UK Government’s proposal to repeal and replace the Human Rights Act. We continue with a post by Dr Mark Elliott, Reader in Public Law at the University of Cambridge. You can read the other contributions in this series here. Posts on the topic are welcome.


Mark ElliottIn my last post on the proposed repeal of the Human Rights Act 1998 and the enactment of a British Bill of Rights, I considered the extent to which the House of Lords might thwart the Government’s plans. My conclusion was that the Lords might plausibly assert itself so as to delay the legislation, traditional understandings of the Salisbury Convention notwithstanding, but that the Parliament Act 1911 clearly deprives the Lords of any absolute veto. What, however, of the devolved nations? Could they block the implementation of the UK Government’s proposals?

The Scottish Government appears to think that they can. Reiterating a position adopted by the Scottish Government in October 2014, the Scottish Social Justice Secretary, Alex Neil, told the Scottish Parliament on Tuesday that:

The Scottish government’s position is that implementation of the Conservative government’s proposals would require legislative consent and that this parliament should make clear that such consent will not be given.

In this post, I explain the legal and constitutional basis for this claim and then assess its accuracy.

The law

As in relation to the House of Lords’ role, the devolved nations’ position in this regard is determined by an amalgam of law and convention. The legal position is clear and straightforward. It is that the UK Parliament is sovereign; that devolution did nothing, as a matter of law, to detract from that sovereignty; and that the UK Parliament therefore remains legally free to make whatever laws it wishes, both for the UK as a whole and for any of its constituent parts. This is so irrespective of whether the subject-matter of the relevant UK legislation is something over which one or more devolved legislatures has authority or is something that is solely within Westminster’s legislative province. Indeed, it is of the essence of devolution — as distinct from federalism — that any authority ascribed to devolved institutions is conferred on a non-exclusive basis. Legislative authority is not therefore transferred by Westminster to the devolved legislatures, but merely shared by it with them. As a matter of law, therefore, the devolved legislatures and administrations have no capacity to preclude the repeal of the Human Rights Act or the enactment of a new Bill of Rights which (if the UK Parliament so chooses) could extend to the devolved nations as well as to England.

The Sewel Convention

However, as is often the case in relation to the British constitution, a full appreciation of how it works can be acquired only when the legal position is examined through the lens of political practice, or convention. And in the present context, a clear and weighty convention effectively neutralises the the UK Parliament’s legal capacity to legislate unilaterally on matters falling within the competence of the devolved legislatures. The convention in question is the Sewel Convention (for discussion of which see Chris McCorkindale’s recent post on the Scottish Constitutional Futures Forum Blog). The Convention is stated in the following terms in the Memorandum of Understanding that shapes the relationship between the UK Government and the devolved administrations:

The United Kingdom Parliament retains authority to legislate on any issue, whether devolved or not. It is ultimately for Parliament to decide what use to make of that power. However, the UK Government will proceed in accordance with the convention that the UK Parliament would not normally legislate with regard to devolved matters except with the agreement of the devolved legislature. The devolved administrations will be responsible for seeking such agreement as may be required for this purpose on an approach from the UK Government.

Thus the Convention requires the UK Government to obtain the consent of relevant devolved legislatures if it wishes to cause the UK Parliament to legislate with respect to devolved matters. As a convention, the Sewel Convention is not legally enforceable, but it carries very considerable political weight, and a UK Government that proceeded in breach of the Convention could fairly be said to be acting unconstitutionally. It could, of course, argue that the word “normally” in the Convention permits unilateral intervention by the UK Parliament in devolved affairs in exceptional circumstances, but it would face an uphill struggle establishing that the present circumstances justify reliance upon this get-out clause; to the contrary, the constitutional significance of what is proposed militates in favour of a consensual approach across the Union, as opposed to top-down invocation of Westminster’s “sovereign” right.

The question then becomes whether the UK Government’s human-rights proposals would involve the taking of any legislative steps that would trigger the Convention. The answer to that question turns upon whether any of those steps would involve legislating “with regard to devolved matters”. For the purpose of the Sewel Convention, it is recognised that the UK Parliament will legislate “with regard to devolved matters” if it enacts legislation that (a) a devolved legislature could have enacted or (b) affects the scope of the legal authority of a devolved legislature or a devolved administration. Whether condition (a) or (b) is satisfied in any given circumstances is a question of law, since it is the devolution legislation that determines what devolved institutions can do. Ultimately, therefore, the question whether the Convention applies turns upon an analysis of the law.

Three issues

Against this background, in seeking to work out whether the Convention applies to any aspects of the UK Government’s human-rights proposals, it is necessary to distinguish between the various legislative steps that might be involved in their implementation, and to consider whether any of them would involve legislation “with regard to devolved matters”. Three possible such legislative steps are worth addressing.

It is clear that the first step would have to be repeal of the Human Rights Act. Would UK legislation repealing that Act affect “devolved matters”? Different views have been expressed on this point; my view is that it would not. Criterion (a) is not engaged, because no devolved institution is capable of legislating with respect to the Act. The effect of section 29(2)(c) of and schedule 4 to the Scotland Act 1998 is that the whole of the Human Rights Act is outwith devolved competence, meaning that the Scottish Parliament cannot amend it or (insofar as it applies to Scotland) repeal it. (Equivalent provisions exist in the devolution legislation pertaining to Northern Ireland and Wales.)

Nor is criterion (b) engaged: in other words, repeal of the Human Rights Act would not affect the extent of the devolved institutions’ competences. This may at first sight seem odd given that those competences are limited by precisely the same provisions of the European Convention on Human Rights as are given effect by the Human Rights Act. However, the fact that those competences are so circumscribed is attributable not to the Human Rights Act but to the devolution legislation, into which the ECHR rights are hard-wired as a constraint upon devolved institutions’ authority. In other words, devolved administrations and legislatures are bound by the ECHR independently of the Human Rights Act, because the ECHR rights are effectively written into the devolved nations’ principal constitutional texts, i.e. the devolution statutes. (Admittedly, the Convention rights that bind the devolved institutions are defined parasitically upon the Human Rights Act, in that the expression “Convention rights” is said in the devolution statutes to bear the same meaning as in the Human Rights Right. However, it seems to me that the Human Rights Act could be repealed in a way that would leave unaltered the range of rights that circumscribe the devolved institutions’ competences, meaning that repeal would not trigger Sewel.)

Second, what if those statutes were themselves to be amended as part, for instance, of an attempt to reduce or excise the influence of the ECHR throughout the UK? (It would, for instance, be very odd for devolved institutions to be bound by the ECHR if, as may be in the Government’s contemplation, the UK were to cease to be a party to that treaty.) That would be a different matter entirely. Amending the devolution statutes so as to affect (whether by means of expanding or reducing) the devolved institutions’ legal powers would trigger the Sewel Convention. It follows, therefore, that if the devolved institutions were (for example) to be relieved of any obligation to abide by ECHR rights, the Sewel Convention would be triggered on ground (b), because the scope of the devolved institutions’ authority would thereby be affected.

The position, then, is that if the UK Government is set upon repealing the Human Rights Act and replacing it with a Bill of Rights that gives only limited effect to the ECHR, the Sewel Convention would (absent the devolved legislatures’ consent) require the UK Parliament to leave in place those parts of the devolution legislation that give full effect to the ECHR insofar as it circumscribes the authority of the devolved administrations and legislatures. (By “full effect”, I mean the level of effect given by the Human Rights Act, as opposed to the sort of qualified effect that appears to be envisaged by promoters of the Bill of Rights.) This point is reinforced by the fact that, as far as Northern Ireland is concerned, the Good Friday Agreement requires the ECHR to be incorporated into the law of Northern Ireland and stipulates that the Northern Ireland Assembly should lack authority to infringe the ECHR.

Third, what about the proposed British Bill of Rights itself? Would its being enacted amount to legislating “with regard to devolved matters” so as to trigger the application of the Sewel Convention? As far as condition (b) is concerned, a British Bill of Rights might affect the scope of the devolved institutions’ powers. It might, for instance, require devolved institutions to comply with peculiarly British rights (whatever they are) that are not found in the ECHR, thereby imposing limits upon the devolved institutions that go beyond the obligation to comply with the ECHR contained in the devolution statutes. We cannot however be sure about this until we know what the Bill of Rights says.

The position is clearer, however, as far as condition (a) is concerned. Although the Human Rights Act is non-devolved, meaning that no devolved legislature is permitted to amend or (insofar as it applies to the relevant devolved nation) repeal it, the same is not true of human rights. It would, for instance, be open to the Scottish Parliament to enact its own Bill of Rights. This means that the UK Parliament’s enacting a British Bill of Rights would trigger the Sewel Convention, because insofar as the British Bill of Rights would apply to Scotland, the UK Parliament would be doing something (i.e. legislating in respect of human rights in Scotland) that the Scottish Parliament is competent to do. The only way around that problem would be first to deprive the Scottish Parliament of the authority so to legislate — but that, of course, would itself affect the scope of its authority, so triggering the Convention.

Where does this leave us?

The UK Parliament is free, as a matter of law and convention, to repeal the Human Rights Act. It is also free, as a matter of law, to amend the devolution statutes (so as to vary or remove the obligation to comply with the ECHR, which it would presumably seek to do if the UK were to withdraw from the ECHR) and to enact a British Bill of Rights that applies throughout the UK. However, the Sewel Convention (absent devolved consent) precludes the amendment of the devolution legislation and the enactment of a British Bill of Rights that applies to the devolved nations in respect of devolved matters. The Sewel Convention (absent consent) also effectively precludes UK withdrawal from the ECHR because such withdrawal would, on any rational analysis, necessitate amendment of the devolution statutes so as to remove the obligation upon devolved institutions to comply with the ECHR.

It follows that if the devolved institutions were to refuse to give consent, the UK Government would be limited in what it could do. One option would be to repeal the Human Rights Act as it applies in relation to English matters (and in relation to non-devolved matters elsewhere) whilst leaving the Act in place as far as it applies to devolved matters within the devolved nations. It could go further by repealing the Human Rights Act in its entirety and enacting a British Bill of Rights, but one that would only apply in relation to English matters (and to non-devolved matters elsewhere). It would then be for the devolved legislatures, to the extent that their competence permitted, to enact local bills of rights pertaining to matters falling within devolved competence. The result, far from delivering on the rhetoric of a “British” Bill of Rights, would be a patchwork of human-rights standards drawing upon the ECHR (which would continue to be hard-wired into the devolution statutes), devolved bills of rights and a British Bill of Rights enacted by the UK Government but largely applicable only to England.

It is worth closing by re-emphasising that the Sewel Convention is just that: a convention, not a law. (It has been proposed by the Smith Commission that the Convention should be recognised in statute, but, as I have argued elsewhere, it is not clear that this would fundamentally alter the status of the Convention — although it would give it added political gravitas.) The UK Government could therefore, without acting unlawfully, ignore the Sewel Convention. But however lawful that would be, it would be unconstitutional. That might sound hollow: if a court is not going to prevent the Government from proceeding, why should it be concerned about the niceties of the Sewel Convention? The answer is that conventions, to the extent that they have binding force, acquire it through the political realm. And it is hard to think of a convention that is more laden with political significance — and invested with political clout — than the Sewel Convention. At a time when no exaggeration is involved in saying that the Union hangs by a thread, it would be a foolish Prime Minister who cast aside a convention that institutionalises respect for devolved autonomy in order to implement the proposed human-rights changes. All of which suggests that, for all that the Human Rights Act is not legally entrenched, it may be far more deeply politically entrenched within our multi-layered constitutional order than the UK Government has so far appreciated.

Mark Elliott is Reader in Public Law at the University of Cambridge.

(Suggested citation: M. Elliott, ‘HRA Watch: Reform, Repeal, Replace? Could the Devolved Nations Block Repeal of the Human Rights Act and the Enactment of a New Bill of Rights?’ U.K. Const. L. Blog (16th May 2015) (available at

This post first appeared on Mark Elliott’s blog, Public Law for Everyone, and is reposted here with thanks.

5 comments on “HRA Watch: Reform, Repeal, Replace? Mark Elliott: Could the Devolved Nations Block Repeal of the Human Rights Act and the Enactment of a New Bill of Rights?

  1. truthaholics
    May 16, 2015

    Reblogged this on | truthaholics.

  2. christine Bell
    May 18, 2015

    Interesting blog and the general point is well made, that it quite depends on how the Human Rights Act is repealed and what replaces it, exactly how devolution is affected and how clear the need for a LCM is.

    However, in the spirit of debate to add a few points:

    I would take issue with the analysis on the first mode of changing (as human rights are currently partially devolved in Scotland and so any change to the Westminster regime in my view would affect the exercise of devolved powers). To give one example, the Scottish Act establishing the Scottish Human Rights Commission to ‘promote’ human rights (a devolved power), is linked to the Human Rights Act and would therefore need amended were it to be repealed – indicating how difficult it is not to construe any substantive change to the Human Rights Act s impacting on devolved powers in any but the most legalistic of senses (if even that).

    Second, even accepting the analysis, the Northern Ireland situation is significantly different from the Scottish one. The Human Rights Act itself is mentioned in the Northern Ireland Act (unlike in the Scotland Act), meaning that there would at least be immediate consequential amendments to that Act required upon repeal, which again it is difficult to analyse as not implicating devolution powers. It is also worth noting that the UK government position on a Bill of Rights in Northern Ireland has consistently been that enacting such a Bill of Rights at Westminster to supplement the Human Rights Act would require a legislative consent motion (LCM) because it would amend the Human Rights Act and therefore change the constraints on the Northern Ireland Assembly. For that reason, successive governments have sought the consent of the parties before implementing (meaning that we do not have the Bill of Rights promised in the agreement). Although it can be argued that adding to the HRA is diifferent from subtracting from it, the UK government seems somewhat pre-committed to the notion that changes to the Human Rights Act implicate LCMs.

    (Third: It is also worth noting that the commitments of the Good Friday Agreement – and the commitment to incorporate the ECHR and provide judicial remedies – at a time when the Human Rights Act was the vehcile offered – was not only crucial to agreement to the Belfast Agreement but was subject to an all-island referendum in Ireland, and written into a British-Irish treaty, raising wider public international law issues with repeal, which also implicate and constrain what is put in place. So legally technically it is simply not true to say that the UK Parliament is completely free to unilaterally change the devolution statutes – it may be (although I would dispute it), as a matter of UK doctrine of Parliamentary sovereignty, but as regards Northern Ireland it is not as a matter of public international law).

    Finally I would add: we often teach that Conventions are a source of constitutional law, but are enforced primarily politically rather than legally. International lawyers often view treaty commitments to be similarly so enforced in practice if not in law.

    So while legalistic arguments as to when LCMs and international treaties are breached can perhaps be coherently made in this case, there are other parties to the interpretation than the UK government. The political consequences of overly legalistic arguments therefore need to be born in mind as a constitutional legal matter. In both the case of Scottish and Northern Irish devolution (for quite different reasons), a lot turns on a UK legalistic argument which does not convince devolved partners as in either the letter or the spirit of how the devolved and international commitments made by the UK government have been understood to operate.

    Apologies for what has turned into a blog-like comment!

    Christine Bell

  3. Pingback: HRA Watch: Reform, Repeal, Replace? Colin Harvey: Rights in a Fractured Union | UK Constitutional Law Association

  4. Pingback: Cameron’s Human Rights Headache? | Parliaments and Legislatures

  5. Pingback: HRA Watch: Reform, Repeal, Replace? Iain Jamieson: The repeal of the Human Rights Act and the Sewel Convention in Scotland | UK Constitutional Law Association

Leave a Reply to christine Bell Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Ordinary membership

UKCLA yearly membership (ordinary)


Student membership

UKCLA yearly membership (student)


Associate membership

UKCLA yearly membership (associate)


%d bloggers like this: