The Conceptual Problems with the EU (Withdrawal) Bill

This is the first of three posts on the EU (Withdrawal) Bill: see my introduction to the series here. This is a work in progress, so any and all comments are very welcome.

There are several conceptual problems with the Bill: first, it is not clear what “retained EU law” is; second, the point at which the Bill would take effect is uncertain; and, third, the Bill has apparently been written in the absence of a coherent conception of Britain’s devolution arrangements.

In general, there are two important points to make. First, calls for conceptual clarity are not designed to satisfy lawyers’ desire for order: they are designed to make the law comprehensible and to ensure legal certainty. People must know where they stand after Brexit, preferably without having to hire a team of lawyers to draw them a map. Second, if Parliament is serious about ‘taking back control’, it ought to be clear about the concepts it employs. Otherwise, the courts will be forced to develop their own concepts in order to make sense of the post-Brexit landscape — and they might well end up drawing distinctions that Parliament did not contemplate or would have rejected outright, had they been drawn to its attention.

Retained EU Law

The purpose of the Bill is, per the Explanatory Notes, to “convert[] EU law as it stands at the moment of exit into domestic law”. This is achieved by clauses 2, 3 and 4. Clause 2 provides for delegated legislation adopted under s. 2(2) of the European Communities Act 1972 to continue to have effect (it also provides for primary legislation to continue to have effect: more on this apparent tautology below). Clause 3 incorporates directly effective EU legislation. Clause 4 gives domestic effect to EU law “rights, powers, liabilities, obligations, restrictions, remedies and procedures” that are already directly effective. This corpus of domesticated EU provisions is to be known as “retained EU law” (clause 6(7)).

Unfortunately, as Professor Paul Craig observes in his written evidence to the House of Lords Constitution Committee, “an important ambiguity…lies at the heart of the present schema, which will render the law post Brexit difficult to understand, even for the trained lawyer: the legal status of the retained law is not entirely clear”. We know that Clauses 2, 3 and 4 incorporate EU law. But we do not know how they incorporate EU law — is it primary legislation, secondary legislation or sui generis?

The difficulties here are profound. Consider clause 3 of the Bill. This incorporates directly effective EU legislation without distinction. Directly effective EU legislation can take the form of regulations analogous to domestic primary legislation, but it might equally be an implementing or delegated act resembling domestic administrative action. As Professor Craig explains:

a)      First, if the provisions are conceptualized as primary legislation then it will affect the extent to which they can be challenged. [Statutory Instruments], by way of contrast, are more susceptible to challenge on non-[Human Rights Act] grounds…

b)     Secondly, if post-Brexit there is a clash between EU direct legislation that has been retained within the UK legal order and another piece of UK law, then the resolution of the conflict will depend, inter alia, on the relative status of the two norms: the outcome could be markedly different if the retained EU regulation is conceptualized as primary legislation or a SI.

Quite what grounds of judicial review — none, some or all — will be available against “retained EU law” is anyone’s guess.

Moreover, per clause 5(2), “the principle of the supremacy of EU law continues to apply on or after exit day so far as relevant to the interpretation, disapplication or quashing of any enactment or rule of law passed or made before exit day”. Though it is inartfully drafted, the objective of this provision seems to be to imbue EU law with the supremacy principle, so as to ensure that it will prevail in any conflict with pre-Brexit domestic measures and be interpreted in harmony with other EU-law measures. But even here questions will abound: for instance, will EU law that resembles domestic administrative action prevail in a conflict with a ministerial decision taken in Britain post-Brexit, and in interpreting “retained EU law”, to what extent does the framework established in clause 6 enable UK courts to take into account future developments  in EU jurisprudence?

A further conceptual difficulty is created by the existence of powers (in clauses 7-9) to modify “retained EU law” (and, indeed, any other statute or delegated legislation). It seems that where such powers are exercised, the domesticated EU provisions (as modified) will remain “retained EU law” (clause 6(7)), to which the principle of the supremacy of EU law will continue to apply, “if the application of the principle is consistent with the intention of the modification” (clause 5(3)). Over time, this will likely create a hotchpotch of provisions, some of primary legislation, some of secondary legislation, to which different interpretive principles apply. It is difficult to improve on the metaphor offered by Professor Bell:

Rather than being transformed into living law within the various national legal systems of the UK, [retained EU law] will function as a kind of zombie EU law, which died on exit day, but continues undead to govern the courts and the devolved assemblies for years to come, well beyond the sunset period for the exercise of powers under the Bill.

The Explanatory Notes suggest (footnote 5) that the reason for the tautology in clause 2 is to make it subject to the clause 7-9 powers. This is not especially convincing, however, as pursuant to the clause 7-9 powers ministers “may make any provision that could be made by an Act of Parliament” (see also the third of Professor Elliott’s “Twenty Questions” and para. 26 of the House of Lords Constitution Committee’s Interim Report).

An alternative possibility is that primary legislation passed to give effect to obligations derived from EU law will, when modified by the clause 7-9 powers, carry the principle of the supremacy of EU law. On first view, this seems strange: EU-inspired primary legislation does not at present prevail in a conflict with other primary legislation, so why would the principle of supremacy have to be introduced? The answer is, per paragraph 97 of the Explanatory Notes, that the principle of supremacy also carries with it an interpretive obligation: “domestic law must be interpreted, as far as possible, in accordance with EU law”. However, it is not at all obvious from the wording of clauses 2 and 5 that ensuring the continuity of the interpretive obligation is the intended effect.

Exit Day

Having triggered Article 50 on March 29, 2017, Britain will leave the European Union on March 29, 2019. But for the purposes of the Bill, Brexit might take place on several different days. One of Professor Elliott’s “Twenty Questions” illustrates the point:

Can there be more than one “exit day”, given that “exit day” is to be prescribed by regulations (clause 14(2)) and given that “any power to make regulations” conferred by the Bill “may be exercised so as to … make different provision for different cases or descriptions of case, different circumstances, different purposes or different areas”? Could, for instance, “exit day” mean one thing for the purpose of clause 1 (so as to effect ECA repeal on 29 March 2019) and something else for the purpose of clause 7(7) (so as to, in effect, extend the two-year sunset clause that applies to the use of the clause 7 amendment powers)? Could differential exit dates be used to enable the power under clause 9(1) to persist beyond what is, for other purposes, taken to be exit day, thereby permitting clause 9(1) to be used to amend the Bill itself (which can be done according to clause 9(2)) long after what counts as exit day for other purposes has been and gone?

A more serious point, along the same lines, is that the clause 9 power to “make such provision as the Minister considers appropriate for the purposes of implementing the withdrawal agreement if the Minister considers that such provision should be in force on or before exit day” could in theory be used to keep large parts of EU law. For instance, it might be used to keep Britain in the European Economic Area, as Dominic Cummings, he of Vote Leave fame, suggested:

The references are to Sir Jeremy Heywood, David Davis and Philip Hammond. Of course, Brexit means Brexit, but without conceptual clarity on what “Brexit” means, it may occur on several different dates (or even not at all, if Ministers never specify an “exit day”, which would no doubt lead to interesting litigation). And the possibility of a “transition” period complicates matters still further.


The provisions of the Bill have succeeded in uniting the devolved governments of Scotland and Wales in opposition to Westminster. The crux of the matter is this: many areas of devolved power — agriculture being the most commonly cited example — are regulated not by Holyrood or Cardiff but by Brussels; when Britain leaves the European Union, competence to make laws in these areas would automatically devolve to the Scottish and Welsh governments (and the Northern Irish government, though the turmoil in Stormont means that there is not a functioning executive at present); but the British government wants to ensure that UK-wide markets function effectively post-Brexit; and, fearful that the devolved administrations might, for instance, adopt different policies on support for dairy farmers (with knock-on effects on the price of milk and cheese), insists on centralising policy-making.

Clause 11(1) nicely illustrates the current approach of the British government. It would insert a new s. 29(4A) in the Scotland Act 1998 to make clear that “an Act of the Scottish Parliament cannot modify, or confer power by subordinate legislation to modify, retained EU law”. And as we have seen, “retained EU law” could represent a very broad category indeed, continuing for possibly indefinite duration. Schedule 3, Part 1 is to similar effect. To borrow from Professor John Bell: “The structure adopted is that the supremacy of EU law as a field outside the competence of the devolved assemblies is replaced by the supremacy of Whitehall-controlled “retained EU law””.

There are two important implications: first, the Bill would effect a change in the current devolution arrangements (and hence, as the British government inevitably accepts, requires legislative consent motions); second, British ministers would be able, by modifying “retained EU law”, to expand or contract the competence of the devolved administrations, or even to make law in areas of devolved competence. The troubling picture that emerges is of a post-Brexit world in which decisions are taken at Westminster about the functioning of UK-wide markets and systems, decisions that will not necessarily heed settled devolution arrangements.

Unsurprisingly, the Scottish and Welsh governments have proposed amendments to address these radical implications:

As currently drafted, clause 11 of the Bill amends both devolution acts by inserting a new restriction on the competence of the devolved legislatures which would prevent the Scottish Parliament and the National Assembly for Wales from passing legislation which modifies retained EU law, even in areas of devolved responsibility. And provision in Part 1 of Schedule 3 to the Bill has the effect that Scottish or Welsh Ministers would have no powers to make, confirm or approve any subordinate legislation so far as it modifies retained EU law. The Welsh Government and Scottish Government consider that these provisions fundamentally cut across the principles of the devolution settlements, and they are strongly opposed to them.
The problem here, it seems to me, is not simply a technical one which can be addressed by carefully crafted amendments. It is a conceptual one: no vision of devolution — of the balance of power between Westminster and the devolved administrations — inspires the Bill. Rather, there is a glimpse of the future: Westminster and British ministers deciding, on an ad hoc basis and without having to respect existing arrangements, what is necessary to preserve those markets and systems that are deemed, at some as-yet-unknown point in the future and on the basis of criteria that remain unspecified, to require UK-wide treatment. It is one thing — maybe even a laudable thing — for the devolution of powers from Westminster to the component parts of the UK to take place in a piecemeal, pragmatic fashion, but it is quite another — and far from laudable — for previous arrangements to be undone by the stroke of a ministerial pen wielded in London.

This content has been updated on September 25, 2017 at 12:36.