Speech: The tortuous path to the UK's exit from the EU


This speech was given by Paul Hardy, DLA Piper's Brexit Director, to the Legal Research Foundation in Auckland on 10 July 2017

The political consequences of the referendum

On 23 June 2016, the UK voted by a 52:48 majority to leave the European Union. A day later David Cameron resigned as Prime Minister. The country was plunged into a state of political uncertainty, which, in reality, has yet to pass.

Theresa May was selected by the Conservative Party as their new leader, after her rivals fell, one by one, along the way. It was at times comical. One contender had to fall on her sword after praying in aid her motherhood as a competitive edge over Theresa May. The prospect of a Boris premiership was brought up short by an Et tu Brute? moment of the highest Shakespearean drama. Michael Gove, a friend from Oxford dining club days, fellow arch-Brexiteer and chief cheerleader in the Boris camp, at the last minute stood against him. Boris withdrew. No one was quite sure why. Michael Gove's ploy was thoroughly deprecated by the rank and file of the Conservative Party, and he was quickly eliminated in the first round of voting.

This left the way clear for Theresa May, best known for being a safe, and unusually long-serving, pair of hands at the Home Office. Oh, and incidentally she voted against Brexit.

Theresa May cleared out the Cameron Cabinet, and set about the task of constructing Government policy on Brexit—a task made less easy by the fact that George Osborne had banned the Civil Service from planning for such an eventuality. A Brexit team in the Cabinet Office soon became a new Department of State, the Department for Exiting the European Union, led by David Davis. A new Department for Trade was also established. This was the first indication that the Government was intending not to remain within the EU's Customs Union, within which trade policy is vested in the EU itself, but to become an independent trading nation.

By late Autumn, the Prime Minister was able to say that the UK would trigger Article 50 in early 2017. The Government's policy on Brexit would not become clearer, however, until then.

Let us now turn to the rules for leaving the EU under both EU and UK law.

Withdrawing from the EU under EU law

Article 50 of the Treaty on European Union

Article 50 of the Treaty on European Union (the TEU), incorporated into the TEU in 2009 by the Lisbon Treaty, sets out the rules on withdrawing from the EU. Whilst the predominant legal opinion before its incorporation was that an EU Member State could secede from the EU, Article 50 makes this plain as a matter of EU law, rather than international law. It was a significant thing for the EU to include such a provision in its Treaties, given its founding principle of closer integration.

Paragraph 1 of Article 50 states that: Any Member State may decide to withdraw from the Union in accordance with its own constitutional requirements. This, then, sets out an EU Member State's unilateral right to secede from the EU under EU law.

Paragraph 2 of Article 50 states that: A Member State which decides to withdraw shall notify the European Council of its intention. In the light of the guidelines provided by the European Council, the Union shall negotiate and conclude an agreement with that State, setting out the arrangements for its withdrawal, taking account of the framework for its future relationship with the Union. According to paragraph 2, the process of withdrawal is triggered when the withdrawing State notifies the European Council, consisting of EU heads of State, of its intention to withdraw. There is no time limit for the notification – the UK delayed nine months. Once notified, the European Council adopts guidelines which set the ambit within which the negotiations are to be conducted by the EU's Commission, with oversight by the Council (of Ministers). The objective of the negotiations is an agreement with the withdrawing State which sets out the mechanics of the withdrawal, but which also takes into account "the framework" of a future relationship.

Paragraph 3 states that: The Treaties shall cease to apply to the State in question from the date of entry into force of the withdrawal agreement or, failing that, two years after the notification referred to in paragraph 2, unless the European Council, in agreement with the Member State concerned, unanimously decides to extend this period. The words "the Treaties shall cease to apply to the State in question" are stark. EU law is based on a "hierarchy of norms", at the top of which sit the EU Treaties. All EU legislation, the competence of all of its institutions and agencies, the legal force of all of its acts, derive from the Treaties. Once they cease to apply to an EU Member State, it follows that every other aspect of EU competence and law, including the enforcement powers of the Commission and the jurisdiction of the Court of Justice of the EU, come to an end.

In terms of timing, paragraph 3 makes clear that the Treaties can cease to apply on three alternative dates:

  • Two years after the date of notification; unless
  • The withdrawal agreement comes into force beforehand; or unless
  • The two-year period is extended by unanimous agreement of the EU Member States, excluding the withdrawing State.

The two-year deadline was, in fact, inserted into Article 50 to show that the right to withdraw was unilateral: the withdrawing Member State did not, ultimately, need the consent of the other EU Member States. The drafters of Article 50 always foresaw, however, the two-year period being extended, it being an unrealistically short time given the scope and complexity of EU integration. But for political reasons on both sides of the Channel the two-year deadline has become fixed: in the UK because of a wish to see the referendum result implemented quickly; and on the EU side because there are elections to the European Parliament in May 2019, and the appointment of a new Commission in 2020.

As the enormity of the task of withdrawal becomes clearer, attention may yet focus on the possibility of extending this deadline; or, at the least, having transitional measures in place for the period between Brexit and the coming into force of a new UK-EU trade agreement.

Much attention, or more accurately hope, has focussed on whether the process of withdrawing from the EU can be reversed. It is perhaps an obvious question to ask, but Article 50 is silent on it. Views differ on whether that silence can be interpreted as implied consent or prohibition. In my own view, I think it can be interpreted as applied consent. The fact that there is no explicit prohibition is significant. By contrast, paragraph 4 of Article 50 specifies how a State re-joins after it has withdrawn (it has to go through the application procedure of any candidate country). It is relevant too that the Court of Justice of the EU takes a purposive approach to interpreting the Treaties. The overarching Treaty objective of "ever closer union" would weigh heavily in favour of a State being allowed to remain if it wished, rather than being forced to proceed on a course to withdrawal, only then to have to re-apply as a candidate country (See the views of Sir David Edward QC and Professor Derrick Wyatt QC, paragraph 10 of the Lords Report: The Process of Withdrawing from the EU).

If (and it's a very big if) the UK held a further referendum as a result of which it decided to stay in the EU, it follows that I do not think a legal objection would stop the UK from remaining in the EU. The EU's decision would primarily be a political one.

Note well, however, that the UK’s exit from the EU is determined by EU law – Article 50 – not by UK law, nor by UK politics. There could be a series of elections in the UK before March 2019, but none will effect a reversal of Brexit unless a UK Government considers it has a popular mandate to go to the EU and say it’s all been a big mistake. I wouldn’t bet on that happening.

Withdrawing from the EU under UK law

I turn now to consider how the UK's membership of the EU was given effect in national law in 1972, and the effect this has had on the Government's right to notify the EU of its intention to withdraw under paragraph 1 of Article 50.

The European Communities Act 1972

The UK is a "dualist" State. In dualist states, as many of you will know, a treaty ratified by the Government does not alter the laws of the State unless and until it is incorporated into national law by legislation. In the absence of incorporating legislation, national courts have no power to enforce the rights and obligations in the treaty in question.

In 1972, the UK Parliament gave effect to the UK's rights and duties under the former Community and now EU Treaties in national law through the European Communities Act 1972 (the ECA).

The ECA is a very clever piece of legislation: it was able to give effect to EU legal doctrines such as direct effect and supremacy without offending against the constitutional requirements of a dualist State, as well as to future-proof itself through ambulatory means to take account of developments in EU law.

It is worth dwelling on the key provisions of the ECA, because of their relevance to the Miller litigation.

By virtue of section 2(1) of the ECA, provisions of EU law that are directly applicable or have direct effect, such as all EU Regulations or certain Articles of the EU Treaties, are automatically (without further enactment) incorporated and binding in national law without the need for a further Act of Parliament. Section 2(1) applies to EU law now and as it develops in the future (from time to time) either by Treaty revision (created by) or interpretation by the Court of Justice of the EU (arising under).

Section 2(2), by contrast, applies to measures of EU law that are neither directly applicable nor have direct effect. This provision makes it possible to give effect in national law to such measures by secondary, or delegated, legislation, such as statutory instruments. Importantly, such secondary legislation can amend an Act of Parliament (section 2(4)) since the delegated legislative power includes the power to make such provision as might be made by Act of Parliament.

Section 2(4) and 3(1) give effect to the doctrine of the supremacy of EU law over national law. Where EU law is in doubt, it requires UK courts to refer the question to the Court of Justice of the EU. As a consequence, all primary legislation enacted by Parliament after the entry into force of the ECA has to be construed by the courts subject to the requirements of EU law. This obliges the courts to dis-apply legislation which is inconsistent with EU law, which is what happened in the celebrated (Factortame case Factortame (No 1) [1990] 2 AC 85; Factortame (No 2) [1991] 1 AC 603): Part II of the Merchant Shipping Act 1988 was held by the House of Lords to be inconsistent with EU law and therefore dis-applied. The same principle was followed by the House of Lords in dis-applying discriminatory provisions in the Employment Protection (Consolidated) Act 1978 (R v Secretary of State for Transport, ex p. Equal Opportunities Commission [1995] 1 AC 1).

The power given to national courts under section 2(4) is remarkable. By dis-applying provisions of primary legislation, the court refused on these two occasions to give effect to an Act—the will—of Parliament. It is also unique: it is only by virtue of the ECA that the courts have this power. Under the Human Rights Act, for example, the courts have the power to make a declaration of incompatibility, on which Parliament decides whether to act, but not to dis-apply the offending statutory provision. More recently, in a case called Benkharbouche (Benkharbouche (Respondent) v Secretary of State for Foreign and Commonwealth Affairs (Appellant) [2015] EWCA Civ 33), the Court of Appeal dis-applied the State Immunity Act 1978 in respect of claims by two employees of the Libyan embassy in London which fell within the scope of EU law. It also made a declaration of incompatibility under the Human Rights Act, but it was the disapplication of the State Immunity Act that was by far the more powerful remedy. The case was appealed to the Supreme Court, which heard it in June. Judgment is awaited.

So, in sum, the ECA provides individuals not only with a wide array of rights under EU law, but also the power to enforce them. The scheme of the ECA was a significant factor in deciding the outcome of the Miller litigation, which I now consider.

The Miller litigation

The Royal Prerogative

The power of the UK Government to ratify, or withdraw from, a treaty is one which traditionally falls within the Royal Prerogative. Royal Prerogative powers are, in crude outline, a remnant of the absolute powers of the monarch that were not vested in Parliament after the Bill of Rights and the Acts of Settlement and Union, and have not been vested in Parliament ever since. They are exercised by the Government mainly in the field of foreign relations, without consultation with, or consent from, Parliament. Because they are conventional, rather than statutory, powers.

The UK Parliament enjoys legislative sovereignty, or supremacy, meaning that it can make and unmake any law it chooses. This, as you will know, is cardinal principle of the UK's constitution. As a consequence, it has long been established by case law (For example, Attorney-General v De Keyser's Royal Hotel Ltd [1920] AC 508; Burmah Oil Co (Burma Trading) Ltd v Lord Advocate [1965] AC 75, 101; and R v Secretary of State for the Home Department, Ex p Fire Brigades Union [1995] 2 AC 513, 552) that the exercise of the Royal Prerogative cannot override or alter legislation made by Parliament. To do so would be to interfere with Parliament's legislative sovereignty.

Background to the appeal to the Supreme Court

Gina Miller, a fund manager in London, together with others, questioned whether the UK Government's right to notify the EU of its intention to withdraw from the EU could legitimately be made under Royal Prerogative powers. Her contention was that the ECA had conferred rights on individuals that would be removed by service of the Article 50 notice, in breach of the principle of Parliamentary sovereignty.

She wisely instructed David Pannick QC, a leading public lawyer, Fellow of All Souls College, Oxford, and member of the House of Lords, to act on her behalf.

The remedy she sought was a declaration from the Divisional Court that the Article 50 notice could only be served with prior authorisation of an Act of Parliament.

The Divisional Court agreed, making the declaration. It concluded that the Government could not act under the Royal Prerogative where it would result in a change in domestic law. In this case, the rights under the ECA, and other EU implementing legislation, would be changed. It was not therefore open to the Government to withdraw from the EU Treaties, and therefore to serve the Article 50 notice, without statutory authorisation.

The Government appealed the case to the Supreme Court, which gave judgment in January this year R (on the application of Miller and another) (Respondents) v Secretary of State for Exiting the European Union (Appellant) [2017] UKSC 5 (Miller).

The Supreme Court judgment

It is perhaps important to bear in mind that, for reasons of public perception, the Government did not argue in these appeals that the notification process under Article 50 was reversible. It feared being accused in the Conservative press of not wishing to implement the referendum result. It also feared giving the court an opportunity to refer the case to the Court of Justice of the EU for a binding interpretation of Article 50—perhaps the ultimate irony for a country seeking to end the jurisdiction over it of the Court of Justice. The Government not only accepted that serving notice to withdraw from the EU was not reversible, but also that it would inevitably affect the rights granted to individuals through the UK's membership of the EU, given domestic effect by the ECA.

These, in my view, were significant concessions. If the Government had argued that the notification process was reversible, it would have been harder to establish a legal nexus between the exercise of Royal Prerogative powers and the loss of individual rights. It would have reinforced that accessing to, and withdrawing from, an international treaty, including an EU treaty, sounds in international law, rather than national law.

In a majority judgment of 8-3, the Supreme Court held that an Act of Parliament was required to authorise ministers to give notice of the decision of the UK to withdraw from the EU. It gave the following reasons, the first two of which departed from the ratio in the Divisional Court judgment:

  • Section 2 of the ECA "authorised a dynamic process" by which EU law becomes a source of UK law and takes precedence over all domestic sources of UK law, including statutes. So long as the ECA remained in force its effect was to constitute EU law as "an independent and overriding source of domestic law". It operated as a partial transfer of law-making powers, an assignment of legislative competences, by Parliament to EU institutions, unless and until Parliament decided otherwise. The constitutional effect of the ECA was "unprecedented": it provided for "a new constitutional process for making law in the United Kingdom".
  • Withdrawal from the EU made a "fundamental legal change" to the UK’s constitutional arrangements, by cutting off the source of EU law. Such a fundamental legal change would be the inevitable effect of the Article 50 notice being served. The UK constitution required such fundamental legal changes to be effected by Parliamentary legislation.
  • The fact that withdrawal from the EU would remove existing rights under the ECA also rendered it impermissible for the Government to withdraw from the EU Treaties under the Royal Prerogative. This was an "additional, albeit related ground" for the requirement of an Act of Parliament.

It is worth recording the dissenting judgment of Lord Reed.

  • Lord Reed, with whom Lord Carnwath and Lord Hughes agreed, took the opposite view. He considered that "EU law was not itself an independent source of domestic law, but depended for its effect in domestic law on the 1972 Act". Furthermore, the effect which Parliament had given to EU law under the ECA—in other words the rights and obligations which flowed from the ECA—was conditional on the application of the EU treaties to the UK and therefore on the UK’s membership of the EU. It was clear that the ECA did not impose any requirement on the UK to remain a member of the EU. It did not therefore affect the Government's exercise of Royal Prerogative powers in respect of UK membership.

Article 50 and devolution

The Supreme Court was also asked to consider whether serving notice to withdraw from the EU under Article 50 would require the prior consent, through legislation, of the Devolved Legislatures in Scotland, Wales and Northern Ireland.

It decided that, whilst the devolution Acts were passed by Parliament on the assumption that the UK would be a member of the EU, they did not require the UK to remain a member. Relations with the EU and other foreign affairs matters were reserved to UK Government and Parliament, not to the devolved institutions. That said, withdrawal from the EU would alter the competence of the devolved institutions, and remove their responsibilities to comply with EU law.

The Advocate General for Scotland and the Counsel General for Wales argued, additionally, that the Sewel Convention prevented the Government from triggering Article 50 without the consent of the Scottish Parliament.

The Sewel Convention is a convention of the devolution settlement by which the UK Parliament will not normally legislate with regard to devolved matters without the consent of the Devolved Parliaments. It was incorporated into both the 2016 Scotland Act and the 2017 Wales Act.

The Supreme Court unanimously found that the Sewel Convention was a political instrument, rather than a legal one. As such, the courts were "neither the parents nor the guardians of political conventions; they were merely observers." They could recognise the operation of a political convention in the context of deciding a legal questions "but they could not give legal rulings on its operation or scope, because those matters were determined within the political world."

The fact that the convention had been put on a statutory footing did not change this analysis. The UK Parliament "was not seeking to convert the Sewel Convention into a rule which can be interpreted, let alone enforced, by the courts; rather, it was recognising the convention for what it was, namely a political convention, and was effectively declaring that it is a permanent feature of the relevant devolution settlement."


The conclusion of the majority of the Supreme Court that EU law was an independent source of UK law was pivotal to its reasoning, and unexpected. It seems to me that it risks being inconsistent with the constitutional requirements of dualist States, namely that it is national legislation which gives effect to Treaty obligations under national law, rather than the other way around. Consider the way in which the judges' views are expressed in the following statements:

"In one sense, of course, it can be said that the 1972 Act is the source of EU law, in that, without that Act, EU law would have no domestic status. But in a more fundamental sense and, we consider, a more realistic sense, where EU law applies in the United Kingdom, it is the EU institutions which are the relevant source of that law."

Professor Mark Elliot, a professor of public law at the University of Cambridge, makes the point well in his article on Miller on his blog site Public Law for Everyone:

"It is hard to see in what sense the EU’s legislative and constitutional apparatus can be an ‘independent source’ of UK law if the source of EU law’s validity in the UK is itself UK law (in the form of the ECA). The EU regime may be independently capable of generating law — and so be a source of law — viewed from an EU perspective, but it does not follow that the same is true from a UK constitutional perspective" (Analysis / The Supreme Court’s Judgment in Miller)."

I rather agree with this view, and take some consolation in the fact that it is the view expressed in the minority judgment.

Precedent effect?

In terms of the precedent of the Miller for the Royal Prerogative powers of the UK Government to withdraw from a treaty, I suggest the main conclusion to draw is that the Supreme Court's judgment reaffirms, albeit on an incidental basis, the decision of the Divisional Court that Royal Prerogative power cannot be exercised if it affects, amends or overrides legislation, because of the sovereignty of Parliament.

The Supreme Court's ratio in deciding that EU law amounts to an independent source of UK law is sui generis, specific to the nature of membership of the EU and the mechanisms of the ECA, and so of narrower relevance as a precedent for the curtailment of Royal Prerogative powers to make and unmake treaties.

I would add as a final remark that court's unanimous view on the justiciability of a political convention is a helpful clarification, and of wider relevance, particularly where a political convention has been incorporated in statute.

Implementing Miller

The European Union (Notification of Withdrawal) Act 2017 came into force on 17 March 2017. The Prime Minister sent the UK’s notification of withdrawal on 29 March 2017.

Implementing Brexit

Let me turn now, very briefly, to the Brexit legislation announced by the Prime Minister in the Queen's Speech on 21 June.

The Great Repeal Bill

The Great Repeal Bill, now simply called the Repeal Bill, will be laid in Parliament this week.

i) Three objectives

It has three objectives:

  • Repeal: The ECA which gives domestic effect to directly applicable EU law and provides a legal basis for the domestication of other EU law, will be repealed.
  • Convert: Directly applicable EU law will be converted into domestic law, while EU law already domesticated under powers granted by the ECA will be preserved, the ECA’s repeal notwithstanding. This will give rise to a body of ‘EU-derived law’ that will, from Brexit Day, exist in the UK solely as UK law.
  • Correct: The Repeal Bill will equip UK Ministers and devolved Ministers with delegated powers to ‘correct’ EU-derived domestic law and other relevant domestic law in order to ensure that EU-derived domestic law is workable in the UK post-Brexit.

ii) Replacing the ECA as the basis for the application of EU law in the UK

The Repeal Bill will be a procedural statute: it will provide the mechanism by which EU law is converted, but not the details on how each law will be converted. It will convert EU law by:

  • Providing a new legal base for directly applicable EU legislation (EU Treaties, Regulations and Decisions) to apply in the UK, replacing section 2(1) ECA
  • roviding a new legal base for indirectly applicable EU legislation (EU Directives), the majority of which are based on the parent power in section 2(2) ECA
  • Containing delegated powers which will allow the Government to amend primary legislation and secondary legislation by means of secondary legislation

iii) Case law of the Court of Justice of the EU

The case law of the Court of Justice of the EU, up to the date of Brexit, will be binding on UK courts post-Brexit in the interpretation of what the Government calls "EU-derived law". It will have the same authority as a judgment of the Supreme Court, which can only be departed from in exceptional circumstances:

"This approach maximises legal certainty at the point of departure, but the intention is not to fossilise the past decisions of the CJEU forever. As such, we propose that the Bill will provide that historic CJEU case law be given the same binding, or precedent, status in our courts as decisions of our own Supreme Court." (Government White Paper, Legislating for the United Kingdom’s Withdrawal from the European Union, para 2.16)

iv) Supremacy of EU law

Where a conflict arises between EU-derived law and new primary legislation passed by Parliament after the UK’s exit from the EU, then newer legislation will take precedence over the EU-derived law. In this way, the Repeal Bill will end the general supremacy of EU law. But if, after Brexit, a conflict arises between two pre-Brexit laws, one of which is an EU-derived law and the other not, then the EU-derived law will continue to take precedence over the other pre-exit law.

v) Scope of delegated powers

Delegated powers can be constitutionally and legally controversial. Constitutionally because they can allow Government to change primary legislation by means of secondary legislation (Henry VIII powers), thereby circumventing the full-scale Parliamentary scrutiny which accompanies the passage of primary legislation through Parliament. Legally because, unless narrowly defined, they give Government wide powers to change policy with little scope for effective judicial review. The Lords Constitution Committee put it this way:

"The challenge facing Parliament—and on which we focus in this report—is how to grant the Government relatively wide delegated powers for the purpose of converting EU law into UK law, while ensuring that they cannot also be used simply to implement new policies desired by the Government in areas which were formerly within EU competence.

"We consider that Parliament should address this challenge in two distinct ways. First, by limiting the scope of the delegated powers granted under the Great Repeal Bill, and second, by putting in place processes to ensure that Parliament has on-going control over the exercise of those powers." (Summary, The Great Repeal Bill and Delegated Powers)

The Government has said in response that it "will ensure that the power will not be available where the Government wishes to make a policy change which is not designed to deal with deficiencies in preserved EU-derived law arising out of our exit from the EU." (Government White Paper, Legislating for the United Kingdom’s Withdrawal from the European Union, para 3.17) The Government has said it does not want to change the way in which delegated legislation is scrutinised by Parliament; but Parliament may have other ideas. Come what may, you can expect the powers of delegation in the Repeal Bill to be minutely scrutinised, particularly in the Lords, with a prospect of Government defeats if the view is reached that they are illegal or unconstitutional.

In addition, delegated legislation, unlike primary legislation, can be struck down by the courts. It is possible that delegated legislation that is perceived to make adverse amendments to the regulation off a commercial sector could be legally challenged in judicial review proceedings. The Government will have to set the scope of any delegated legislation with this risk in mind.

Seven other Brexit Bills

The Queens Speech promised not just the Repeal Bill, but Bills on trade, customs, immigration, agriculture, fisheries, data protection, and international sanctions. The package of Brexit legislation shows that the UK is on course to leave both the Customs Union and the Single Market. Why else am Immigration, Customs and Trade Bill? A further indication of this is the appointment of Crawford Falconer as the Chief Trade Negotiation Adviser and Second Permanent Secretary at the Department for International Trade. It is good to know that our new trade policy is in the most experienced of hands.

It is an ambitious legislative package, and there is a real of serial amendments and defeat on the hundreds of votes that will be required to get it through both Houses. There is, in other words, no certainty that the Government will succeed.

New Devolved competences

Many of the EU policies be returned to the UK fall within the competencies of the Devolved Nations, particularly agriculture, fisheries and environmental protection. It will be interesting to see how the devolution of these policies is put into effect in Brexit legislation. There will be tension between the need to respect the Devolved Settlements right to formulate policy in their own fashion, and a Whitehall desire to have a UK single market in, for example, agriculture. This will also have a bearing on what terms the UK can offer preferential access to its markets in trade deals.

Those keen to see what type of UK might emerge from the tortuous path to its exit from the EU should keep a close eye on the passage of all these Bills. It promises to be a roller-coaster ride, and one which has to be completed in eighteen months.

EU negotiations

At the same time, the Government is negotiating a deal on its withdrawal from the EU. There will be little time to put in place replacement relationships for the hundreds of areas in which the UK currently cooperate with the EU before March 2019. Transitional measures covering the period from Brexit to the entry into force of a new UK-EU trade deal will, in my view, be critical.

Parliament will get a "take-it-or-leave-it" vote on the final deal. Beware, however: this is of political significance but of no legal consequence. The two-year cut-off in Article 50 applies whether Parliament approves the deal or not. Come 29 March 2015, the EU Treaties "cease to have effect" in the UK.

Thank you very much indeed for your time.