In his latest blog on the European Union (Withdrawal) Bill, Paul Craig criticises the recommendation of the House of Lords Constitution Committee (“HLCC”), at paras 70 and 93, that all retained direct EU law (defined by the HLCC to encompass all the law continued under clauses 3 and 4 of the Withdrawal Bill) should be given the status of domestic primary legislation passed immediately before exit day. He suggests, instead, a hierarchy in which some law continued in force under clause 3 should be “deemed to be a statutory instrument”. This formulation is intended, it seems, to do more than its usual job (which is confined to attracting the provisions of the Statutory Instruments 1946, which are largely irrelevant for present purposes). It appears to be intended, instead, to give the law in question the status of subordinate legislation made under legislative powers delegated to the executive. But what practical effects is it designed to produce?
This blog considers the virtues of the HLCC proposal to deem all retained direct EU law to be domestic primary legislation and questions whether a formulation of the sort suggested by Paul Craig, even if amended to deal with status rather than form, could have any useful practical effect.
As a legislative drafter, I know that there is a place for “deeming” in the drafter’s toolbox, but I have always understood it to be a tool that should be used with great caution and only rarely. Deeming something to belong to a conceptual category to which it does not naturally belong is particularly risky. Legal conceptual categories are formed by combining into one category cases with the same practical legal effects. Deeming something to fall within a particular conceptual category is only a short-hand method of attracting the same legal consequences as apply to other things categorised in the same way; but it may leave open the question what other similarities may need to be assumed in particular cases. Often, it is clearer and safer not to attract certain consequences indirectly (by applying a concept all the incidents of which may not yet be settled), but, instead, to spell out directly which of the known incidents of the concept you intend to apply.
On that basis, I have more sympathy with the approach adopted in the Bill than most commentators. There is very little domestic law (apart from the difference between ss. 4 and 6 of the Human Rights Act 1998) that depends directly on the status of legislation as such, rather than on the factors that contribute to a piece of legislation being regarded as delegated legislation of a particular type. Questions about vires and amendability are dealt with below and, as I suggest there, are not adequately addressed by “deeming”.
There is nevertheless a good case for spelling out, along the lines envisaged by HLCC, the way in which the Withdrawal Bill already gives retained EU law the status of primary legislation and dates that from immediately before exit. The notion that retained direct EU law will have a separate status of its own and the supremacy provisions of clause 5 have muddied waters that need to be made clearer. If you are adopting a practical approach, in preference to a conceptual one, it is probably wiser not to speculate about the correct way to analyse those practical effects in conceptual terms.
After exit, the law continued in force by clauses 3 and 4 will have effect in UK law by virtue of those clauses, and those clauses alone. Those clauses will be contained in a piece of UK primary legislation and come into force at exit day. Legislative drafters usually oppose provisions that are unnecessary: on the grounds that “unnecessary material in statutes, as in the human body, tends to turn septic”. In this case, though, enough doubt about status exists to justify a departure from this principle. The position is comparable to that which resulted in the enactment of s. 18 of the European Union Act 2011 (which is to be repealed by the EU (Withdrawal) Bill). A provision that implemented the HLCC recommendation as a declaration of the effect of clauses 3 and 4 would be wholly consistent with that precedent, and ought to be unobjectionable. If s. 18 of the 2011 Act is capable of being said about the European Communities Act 1972 (“ECA”), it is sayable for provisions of the Bill.
This analysis then raises a question about clause 2. HLCC says that there is no question about the status of legislation carried forward under clause 2 because ”it already has the status of primary or secondary legislation in domestic law” (para 94). The analysis set out above, though, clearly raises the question whether the effect of clause 2, as it stands, would be to promote secondary legislation to which it applies to the status of primary legislation: by giving it renewed effect in primary legislation. It may be hoped that a court could relatively easily infer that that is not Parliament’s intention, but it is nevertheless worth clarifying what is intended.
For reasons given below, status is a relatively insignificant point in relation to subordinate legislation made under s.2(2) ECA, or indeed in relation to any other subordinate legislation the power to make which is either expressly or impliedly repealed by the Bill, or disappears with the UK’s withdrawal from the EU. However, it could be significant in relation to any subordinate legislation that does not otherwise need to be saved by clause 2 but is covered by its terms. In those cases, it could be argued that promotion to primary legislation by ratification under clause 2 is enough to displace the power of amendment conferred by s. 14 of the Interpretation Act 1978 (see below). This argument would provide another reason for accepting another of HLCC’s recommendations: that clause 2 should be confined to the law that needs to be preserved under the Bill and should not, as it currently does, extend unnecessarily to law that does not need to be saved (paras 21 and 22).
So, what would be the effect of deeming any form of retained direct EU law to be contained in secondary legislation? It is assumed that there are two main practical consequences to be considered (in addition to the implications for ss. 4 and 6 of the Human Rights Act 1998, which would best be dealt with expressly). The first relates to more general challenges on grounds of vires, which HLCC regards as the principal distinguishing feature of secondary legislation (para 39). The second is amendability. However, both these assumptions rely on misconceptions.
The ability to challenge domestic subordinate legislation on the grounds that it exceeds the power under which it was made is something that necessarily falls to be tested as at the time when the power was exercised. Also, it is possible only if the powers under which it is treated as made are capable of being identified, together with the Parliamentary intention with which they were conferred. That may be a far from straight-forward process in the case of something that is only deemed to be subordinate legislation, especially if it requires the power itself to have been a deemed power conferred by a provision that must itself be deemed to have been contained in domestic primary legislation. The hypotheses would need to be piled up on each other, like Pelion on Ossa, at the price of clarity and certainty.
Even if it is possible to identify a power under which the deemed subordinate instrument was or is deemed to have been made, it is clear that a provision treating something that is in the nature of secondary legislation to be continued in force – even as primary legislation – could have the effect of continuing no more than was previously the law. If it had contained a vires defect from its inception, that too would necessarily be carried forward. In its simplest form, nothing ultra vires would be carried forward because it would not exist at exit to be carried forward.
There may be – in my view, there certainly is – a strong policy case, on rule of law grounds and in the light of the inherent uncertainty in many features of EU law, for bolstering the presumption of regularity in relation to retained EU law as it applies to new cases after exit. This is particularly true of EU derived domestic legislation. Doing that would enhance the rule of law certainty and predictability that continuity is designed to secure; but for it to be effective it would require something different from deeming retained law to have the status of primary legislation.
Similarly, it is a misconception that the amendability of subordinate legislation by other subordinate legislation derives from its status as such. It does not. It depends, instead, on s. 14 of the Interpretation Act 1978 (“s. 14 IA 1978”), which reads—
“Where an Act contains power to make—
- rules, regulations or byelaws; or
- Orders in Council, orders or other subordinate legislation to be made by statutory instrument,
it implies, unless the contrary intention appears, a power exercisable in the same manner and subject to the same conditions and limitations, to revoke, amend or re-enact any instrument made under the power.”
There are two things to say about this.
The first is that the s. 14 IA 1978 power is parasitical on the power under which any subordinate legislation is made, and so is repealed if that power is itself repealed. It follows that the amendability of e.g. regulations under s. 2(2) of ECA disappears with the repeal of the primary power; and – even if it were saved – it could only be exercised subject to the “same conditions and limitations”. That would be impossible after UK withdrawal. So, s. 2(2) regulations might as well be continued under clause 2 of the Withdrawal Bill as primary legislation, as HLCC recommends for retained direct EU law. Subordinate legislation under s. 2(2) of ECA would, except so far as it is otherwise provided for, retain any vires defects it contained at making; and the repeal of s. 2 of ECA would make it unamendable except under the sunsetted powers conferred by the Bill.
The second thing to say about the s. 14 IA 1978 power is that its existence makes it unnecessary, in practice, for any power to make subordinate legislation to include power to amend subordinate legislation made under other such powers. Even Henry VIII powers do not often need to provide for the amendment of subordinate legislation as well as primary legislation, or to be construed as allowing that. It is usual to rely only on the s. 14 IA 1978 power to amend secondary legislation in response to subordinate legislation made under powers in different Acts. Exceptions to that tend to be made only where it would be desirable to use one form of instrument to make amendments across the board in instruments in different forms (e.g. so regulations can amend provisions in orders, etc., as well as in regulations) or to by-pass a more stringent form of Parliamentary control (e.g. affirmative resolution procedure) when making essentially technical consequential amendments.
For these reasons, it would also be sensible to extend HLCC’s recommendation for allowing Henry VIII powers, other than those in the Bill, to be available for amending the retained direct EU law that they are recommending should be deemed to be primary legislation (para 64): so that such powers could also be used to amend subordinate legislation made under s.2(2) of ECA.
More generally it follows from the analysis in this blog that, while there is a good case for clarifying that after exit the effect of using the Withdrawal Bill to give continuing effect to retained direct EU law is to give it the status of primary legislation, there is nothing practical that can be achieved by deeming anything to be subordinate legislation. Instead, such a provision would give rise to some potentially unanswerable questions and create considerable uncertainty.
Stephen Laws was the First Parliamentary Counsel 2006-2012 and is an Honorary Senior Research Associate, University College London, and Senior Associate Research Fellow, Institute of Advanced Legal Studies.
(Suggested citation: S. Laws, ‘Giving “Deemed” Domestic Law Status to Retained EU Law’, U.K. Const. L. Blog (28th Feb. 2018) (available at https://ukconstitutionallaw.org/))