Shindler & Anor v Chancellor of the Duchy of Lancaster & Anor

In these proceedings, the claimants seek to challenge section 2 of the EU Referendum Act 2105 (“the 2015 Act”) which, by virtue of adopting certain franchise rules for UK Parliamentary elections, excludes from the franchise for the EU Referendum UK citizens who have moved abroad and were last registered to vote in the UK more than 15 years ago (“the 15 year rule”). They claim that their exclusion from the franchise constitutes an unjustified restriction of their EU law rights of free movement.

The Divisional Court (Lloyd-Jones LJ and Blake J) held that (i) section 2 of the 2015 Act falls within the scope of EU law, so that the rights of free movement conferred by EU law are, in principle, engaged; but (ii) section 2 is not a restriction on their rights of free movement; however, (iii) if section 2 is such a restriction, it is objectively justified as a proportionate means of achieving a legitimate objective, namely of testing the strength of a British citizen’s links with the UK over a significant period of time. The court, therefore, dismissed the claims. Finally, (iv) it rejected the defendants’ submission that the claimants have delayed in bringing their claim so as to disentitle them to a remedy on that account.

In this appeal, Mr O’Neill QC submits that the Divisional Court erred in reaching the conclusions in (ii) and (iii) above. In addition, he raises the new point that we should grant a declaration that the 2015 Act is unconstitutional in so far as it conflicts with the fundamental common law constitutional right of British citizens to vote. For the defendants, Mr Coppel QC submits that the Divisional Court reached the right conclusions on (ii) and (iii), contests the new common law claim but says that the Divisional Court erred in its conclusion in relation to (i) and (iv).

The claimants are both British nationals. They are described more fully at paras 3 and 4 of the Divisional Court’s judgment. Mr Shindler was born in London in 1921. On his retirement, in exercise of his right of free movement, he moved to Italy where he has resided ever since. His name last appeared in the UK register of electors in 1982. Ms MacLennan was born in Inverness in 1961. In 1987, in exercise of her right of free movement, she moved to Belgium where she has resided ever since. Her name last appeared in the UK register of electors in 1987.

The relevant legislation is set out in full by the Divisional Court at paras 10 to 14 of its judgment. It is sufficient to refer to section 2 of the 2015 Act which, so far as material provides:

“(1)

Those entitled to vote in the referendum are—

(a)

the persons who, on the date of the referendum, would be entitled to vote as electors at a parliamentary election in any constituency,

(b)

the persons who, on that date, are disqualified by reason of being peers from voting as electors at parliamentary elections but—

(i)

would be entitled to vote as electors at a local government election in any electoral area in Great Britain,

(ii)

would be entitled to vote as electors at a local election in any district electoral area in Northern Ireland, or

(iii)

would be entitled to vote as electors at a European Parliamentary election in any electoral region by virtue of section 3 of the Representation of the People Act 1985 (peers resident outside the United Kingdom), and

(c)

the persons who, on the date of the referendum—

(i)

would be entitled to vote in Gibraltar as electors at a European Parliamentary election in the combined electoral region in which Gibraltar is comprised, and

(ii)

fall within subsection (2).…

(2)

A person falls within this subsection if the person is either—

(a)

a Commonwealth citizen, or

(b)

a citizen of the Republic of Ireland.”

In support of their case that section 2 does not fall within the scope of EU law, the defendants rely on Article 50 of the Treaty on European Union (“TEU”) which lays down the procedure whereby a Member State may withdraw from the EU. So far as material, it provides:

“1.

Any Member State may decide to withdraw from the Union in accordance with its own constitutional requirements.

2.

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. That agreement shall be negotiated in accordance with Article 218(3) of the Treaty on the Functioning of the European Union. It shall be concluded on behalf of the Union by the Council, acting by a qualified majority, after obtaining the consent of the European Parliament.

3.

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 meaning of the phrase “in accordance with its own constitutional requirements” is not subject to elucidation in the travaux préparatoires to the Treaty of Lisbon which introduced Article 50 TEU. Article 50 has not been considered previously by the domestic courts or by the CJEU. It has, however, been analysed by the German Constitutional Court in Re Ratification of the Treaty of Lisbon [2010] 3 CMLR 13. The principal issue in that case was whether the Treaty of Lisbon represented an unacceptable infringement of Member State sovereignty. In arguing that it did not, the German Government submitted that the “free right of withdrawal” conferred by Article 50 TEU confirmed the “continued existence of state sovereignty” and that Member States “would remain the ‘masters of the Treaties’ and would not have granted the European Union Kompetenz-Kompetenz”: see para 126(2). The court accepted this submission. It stated:

“305.

The Treaty system covered by the Act approving the Treaty of Lisbon clearly shows the existing principle of association…in the system of the responsible transfer of sovereign powers and thus satisfies constitutional requirements. The Treaty makes explicit for the first time the right of each Member State to withdraw from the European Union (Art.50 Lisbon TEU). The right to withdraw underlines the Member State’s sovereignty…

306.

Any Member State may withdraw from the European Union even against the will of the other Member States…There is no obligation for the decision to withdraw to be implemented by a withdrawal agreement between the European Union and the Member State concerned. In the case of an agreement failing to be concluded, the withdrawal takes effect two years after the notification of the decision to withdraw (Article 50.3 Lisbon TEU). The right to withdraw can be exercised without further obligations because the Member State that wishes to withdraw does not need to state reasons for its decision. Article 50.1 Lisbon TEU merely sets out that the withdrawal of the Member State must take place “in accordance with its own constitutional requirements”. Whether these requirements have been complied with in the individual case can, however, only be verified by the Member State itself, not by the European Union or the other Member States”.

It is clear from this analysis that the German Court did not accept that the domestic constitutional requirements applicable to a decision to withdraw were themselves subject to validation under EU law and could be overturned on grounds of incompatibility with the EU Treaties.

The Divisional Court said at para 24:

“The United Kingdom undoubtedly has a sovereign right to determine for itself whether it wishes to remain a party to the EU treaties and to determine the constitutional procedures which shall be followed in determining this question. These are, both in EU law and in domestic law, pre-eminently matters within the competence of the United Kingdom. A natural reading of Article 50(1) TEU confirms this.”

But it continued:

“However, it does not follow that the manner in which such a competence of a Member State is exercised is incapable of engaging EU law. On the contrary, Preston, Rottmann and Tas-Hagen among other authorities demonstrate that a Member State when acting within a field of national sovereign competence must nevertheless have regard to the impact of the manner of exercise of that competence on fundamental rights in EU law. In this way, EU law may be engaged in principle. This is the case even where the most fundamental issues of national competence are concerned, such as the grant or withdrawal of nationality or determining the franchise for a national Parliamentary election. Contrary to the submission on behalf of the defendants, we do not consider that Article 50(1) goes further and confers on a Member State a total exemption from EU law in this regard. The words of Article 50(1) in their natural meaning do not support such a result. If such a striking departure from established principles of EU law were intended very clear words would be required and they are not present here. Moreover, we have not been referred to any travaux préparatoires or other admissible materials which support such a reading.”

Mr Coppel submits that the Divisional Court rightly recognised that the UK has a sovereign right to determine whether it wishes to withdraw from EU treaties and that this is pre-eminently a matter falling within the competence of the UK. However, inconsistently with this recognition and wrongly, the court then concluded that the manner in which the UK had set its constitutional procedures engaged EU law in principle and so could be subject to challenge if the decision to withdraw interfered with fundamental EU law rights. In summary, Mr Coppel submits that the UK cannot sensibly be said to have a sovereign right to decide to withdraw from EU treaties, the exercise of which is constrained by the rules of the very treaties from which it wishes to withdraw.

Mr O’Neill seeks to support the reasoning of the Divisional Court on the ground that the defendants’ argument is the same as that which was rejected by the Divisional Court and the Court of Appeal in Preston. The argument should, therefore, be rejected for the simple reason that this court is bound by the Court of Appeal decision in R (Preston) v Wandsworth London Borough Council [2013] QB 687. The short answer to this is that the issue in Preston was whether the 15 year rule as applied to Parliamentary general elections was outside the scope of EU law, in circumstances where there was no applicable provision of EU law making clear that Member States could adopt their own constitutional rules. The most that could be said for the defendant was that “the franchise exists solely by virtue of the domestic legislation of Member States; it is not the express creation of any of theArticles in the TFEU”: per Mummery LJ at para 70. In the context of legislation introduced by a current Member State, it is not at all surprising that Mummery LJ said at para 72: “even in areas falling within their national sovereignty, the exercise of competence by a member state is subject to the applicable provisions of EU law”.

In my view, the Divisional Court was wrong to reject the defendants’ submission that Preston is to be distinguished because in that case there was no specific provision of EU law making clear that Member States could withdraw from the Union in accordance with their own constitutional rules. The 2015 Act contains part of the constitutional requirements of the UK as to how it may decide to withdraw from the EU. These include the holding of a referendum the franchise for which adopts the franchise for UK Parliamentary elections, including the 15 year rule.

Unlike the CJEU cases relied upon by the Divisional Court, the present case is not concerned with the effective operation of EU law, where it is understandable that the CJEU would wish to prevent interference by the laws of a Member State with the exercise of fundamental EU rights. Rather, it is concerned with the sovereign decision of whether or not a Member State should be bound by EU law at all. By Article 50(1) TEU, EU law has expressly provided an area where Member States may adopt their own requirements. It would be contrary to that provision if articles of another EU Treaty relating to citizenship and free movement were to intervene so as to determine the constitutional requirements to be adopted by a Member State which is deciding whether to leave the EU.

The Divisional Court held that its conclusion was consistent with an established line of authority of the CJEU. One example it gave was Case C-135/08 Rottmann v Freistaat Bayern (2010) QB 761. This case concerned the withdrawal of German nationality with retrospective effect from Mr Rottmann who had previously possessed Austrian nationality, on the ground that he had obtained German nationality by deception. The CJEU held that, while it was for each Member State to lay down the conditions for the acquisition and loss of its nationality, it must nevertheless have due regard to Community law. The situation in which his loss of German nationality could cause him to lose his status as a citizen of the EU fell, by reason of its nature and its consequences, within the ambit of EU law. It was for EU law to determine the conditions in which an EU citizen may, because he loses his nationality, lose his status as an EU citizen and thereby be deprived of the right attaching to that status: see paras 39 to 47. Further examples of this approach are to be found in the other cases cited by the Divisional Court at para 22.

But a decision by a Member State to withdraw from the EU is an exercise of national sovereignty of a special kind for which the TEU has made the express provision that this may be done in accordance with a Member State’s own constitutional requirements. That is hardly surprising. It would have been surprising if the Member States had agreed that a Member who wishes to withdraw from the EU altogether could only do so if the decision to withdraw did not infringe one or more fundamental EU rules. An obvious reason why a Member State might wish to withdraw is that it found such rules unacceptable and was no longer willing to be bound by them. The right of free movement is a plain example of such a rule and one which has particular resonance in the context of the proposed UK referendum. It is one thing for Member States to agree that, while they are members of the EU, they will not infringe EU law and to that extent will accept what might be described as a loss of sovereignty. It is quite a different matter for them to agree that they may only decide to withdraw from the EU if they can do so without infringing EU law. If this had been the intention of the Member States, this would surely have been expressly agreed. But they have not done so. On the contrary, they agreed Article 50(1) TEU whose plain and natural meaning does not have this effect.

Mr O’Neill submits that all that article 50(1) does is to provide that there are no uniform procedures under EU law determining how a Member State may reach a decision to withdraw from the EU and that this is a matter for its own constitution. He also submits that the reference to the Members States’ constitutional requirements is a commonplace phrase used throughout the TEU and the TFEU, usually in provisions requiring Member States to approve possible extension of powers for the EU institutions. But neither of these submissions addresses Mr Coppel’s arguments.

He further says that the holding of the referendum is not itself “a decision to decide to withdraw from the Union”. It may be a preliminary to such a decision, but it is not in any event a necessary preliminary as a matter of constitutional law, since Parliament does not need the mandate of a specific referendum to give it the power to pass legislation mandating the withdrawal of the UK from the EU.

I accept that Parliament is sovereign and that it does not need the mandate of a referendum to give it the power to withdraw from the EU. But by passing the 2015 Act, Parliament has decided that it will not withdraw from the EU unless a withdrawal is supported by referendum. In theory, Parliament could decide to withdraw without waiting for the result of the referendum despite the passing of the 2015 Act. But this is no more than a theoretical possibility. The reality is that it has decided that it will withdraw only if that course is sanctioned by the referendum that it has set in train. In other words, the referendum (if it supports a withdrawal) is an integral part of the process of deciding to withdraw from the EU. In short, by passing the 2015 Act, Parliament decided that one of the constitutional requirements that had to be satisfied as a condition of a withdrawal from the EU was a referendum.

For these reasons, I would hold that the 2015 Act does not fall within the scope of EU law and that the claim fails at the first hurdle. But in case I am wrong about this, I shall go on to consider whether the Divisional Court was right to conclude that section 2 of the 2015 Act is a restriction on the EU right of free movement.

The claimants’ case is that the 15 year rule is a restriction on their rights of free movement in that it places them at a disadvantage or penalises them because they have exercised their right to move and reside in another Member State. The disadvantage or penalty is quite simply that they do not have the right to vote in the referendum on an issue which affects them in ways which they consider to be very important. They say that it is not necessary for them to show that the disadvantage might deter them (or anyone else in their position) from exercising their free movement rights in the future. It is sufficient that the legislation penalises on-going free movement: if they wished to vote, they would have to cease to reside in the countries in which they have been living for many years and move to the UK. The Divisional Court held that, in order to succeed, the claimants also have to show that the 15 year rule was liable to dissuade them from continuing to exercise their right of free movement. The court said that (i) it was bound by the decision of this court in Preston to hold that this was the correct approach; and (ii) an analysis of the jurisprudence of the CJEU shows that this is correct.

The first question, therefore, is whether the decision of this court in Preston is determinative of the issue in this court. In that case, the court decided that the 15 year rule, when applied to UK Parliamentary general elections, did not create a restriction on the right of free movement. In the Divisional Court Elias LJ said at para 38 that the claimant had to show that the obstacle created could “fairly be said to deter persons from exercising their rights.” He added, relying on (Case C-190/98) Graf v Filzmoser Maschinenbau GmbH [2000] ECRI-493 that the restriction must not affect the right in a way which was “too indirect or uncertain”. He considered that in Preston the restriction was too indirect and uncertain and had not been established on the evidence.

In the Court of Appeal, Mummery LJ at para 76 endorsed what Elias LJ said. At para 77, Mummery LJ said:

“As Elias LJ pointed out, ante, p 700, para 39, Mr Subiotto accepted before the Divisional Court that “it was unrealistic to suggest that the possibility of being denied the right to vote 15 years down the line would in practice deter anyone from leaving the UK to live in another member state”. Nor, as Elias LJ observed, would the rule discourage someone who has been resident overseas for almost 15 years from staying abroad in another member state: “it is inherently unlikely that the loss of the right to vote would be sufficient to cause them to up sticks and return to the UK.”

He continued:

“79.

I agree with Mr Subiotto that what is effectively a suspension of the right to vote of those British citizens who voluntarily choose to reside in another Member State for more than 15 years can be characterised as a “disadvantage”. It does not follow, however, that every disadvantage of non-residence in the UK is a restriction on or deterrent to free movement. Further, as disenfranchisement is only triggered after the passing of 15 years' residence overseas, a long term view has to be taken when considering whether the prospect of ceasing to be eligible as an overseas voter after the end of 15 years of non-residence in the UK could deter free movement.

80.

That question obviously does not have to be answered in terms of statistical evidence or specific evidence of actual cases of deterrence. In practice the claimant's assertion about the potential effect of the 15 year rule on free movement is very difficult to demonstrate by any means, because it does not square with ordinary human experience. In the course of crowded human lives over a period of 15 years inevitable uncertainties, unknowns and contingencies make it is impossible to arrive at a reliable or credible conclusion that the rule could deter free movement. No legal test, whether formulated in terms of “probability”, or “likelihood”, or “capability”, or “liability”, or “real possibility”, addresses the basic difficulty that what is asserted in the claimant's case is too speculative, remote and indefinite to establish a case. Every British citizen knows that, over a period of 15 years, he or she will be potentially affected by so many unforeseeable circumstances, combinations of circumstances and changes in circumstances that it is simply not possible for a court or anyone else to conclude that the 15 year rule could deter British citizens from going to reside and work in other Member States of the EU, or from doing so for as long as they like.

81.

Disenfranchisement by reason of 15 years non-residence in the UK is, in my view, both qualitatively and quantitatively different from those more direct, certain and immediate obstacles and barriers to basic day-to-day living that are set up by social benefits rules requiring the claimant to be present in the UK at the date of claiming the benefit and/or resident in the UK for a relatively short period before the date of claiming the benefit. ”

Mr O’Neill submits that Preston is distinguishable from the present case because it was only concerned with a rule excluding British citizens from the vote for UK Parliamentary general elections after 15 years of non-residence in the country. That is true. But it is clear that the court decided that the relevant test was whether the restriction was liable to deter an individual from exercising his right of free movement. That was part of its reasoning and is binding on this court. Mr O’Neill seeks to distinguish Preston on the facts and submits that Mummery LJ’s observations do not apply in the different context of an EU referendum. In the present case, the Divisional Court said at para 32 that, in the context of the EU referendum, the claims by Mr Shindler and Ms MacLennan are weaker than was that of Mr Preston. Mr O’Neill submits that this assessment is wrong. But his submission does not meet the point that the test enunciated and applied in Preston is part of the ratio of the court’s decision. Whether it was applied correctly on the facts of that case is not material to the question of what the test was.

But even if the Preston test is not binding on us, I am satisfied that it is correct. Mr O’Neill submits that it is inconsistent with the jurisprudence of the ECJ and the CJEU which has been clarified in the recent Case C-359/13 Martens v Minister van Onderwijs [2015] 3 CMLR 3. In a nutshell, he says that Martens shows that the requirement to justify legislation which disadvantages those who exercise the right of free movement is not dependant on whether such a disadvantage has deterred or is liable to deter the exercise of the right of free movement. All that is required is that the measure disadvantages or penalises an individual consequent upon his exercise of the right of free movement.

The facts in Martens were described by the Divisional Court at para 35 as follows:

“M, a Netherlands national, moved with her parents in 1993 to Belgium where her father was employed and where the family remained. A Netherlands law enacted in 2000 made eligibility for funding for higher educational studies pursued outside the European Netherlands dependent on the student being resident in the Netherlands for at least 3 out of the 6 years preceding the commencement of the course of study. In 2006 M enrolled on a degree course at the University of the Netherlands Antilles in Curacao for which she obtained a grant from the Netherlands authorities. When the authorities later discovered that she had not fulfilled the residence requirement they revoked the grant. The referring court asked the CJEU whether national legislation such as that in issue was precluded by EU law.”

The CJEU (Third Chamber) held that, although the organisation of national educational systems fell within the competence of Member States, they had to exercise that competence in accordance with EU law and in particular the Treaty provisions on freedom of movement. Although EU law does not impose any obligation on Member States to provide a system of funding for higher education, where it does so it must ensure that the rules do not create an unjustified restriction of the right of free movement. It continued:

“25 In that regard, it is apparent from settled case-law that national legislation which places certain nationals at a disadvantage simply because they have exercised their freedom to move and to reside in another Member State constitutes a restriction on the freedoms conferred by Article 21(1) TFEU on every citizen of the Union (judgments in Morgan and Bucher, EU: C: 2007:626, paragraph 25, and Prinz and Seeberger, EU: C: 2013:524, paragraph 27).”

26 Indeed, the opportunities offered by the Treaty in relation to freedom of movement for citizens of the Union cannot be fully effective if a national of a Member State could be dissuaded from using them by obstacles resulting from his stay in another Member State because of legislation of his State of origin penalising the mere fact that he has used those opportunities (see, to that effect, judgments in Morgan and Bucher, EU:C:2007:626, paragraph 26, and Prinz and Seeberger, EU:C:2013:524, paragraph 28)”.

The court considered that M had continued to exercise her right of free movement throughout the time she lived in Belgium. It said:

“31.

By making the continued grant of funding for studies abroad subject to the three-out-of-six-years rule, the legislation at issue in the main proceedings is liable to penalise an applicant merely because he has exercised his right to freedom of movement and residence in another Member State, given the effect that exercising that freedom is likely to have on the possibility of receiving funding for higher education (see, to that effect, judgments in D’Hoop, EU:C:2002:432, paragraph 30; Prinz and Seeberger, EU:C:2013:524, paragraph 32; and Thiele Meneses, EU:C:2013:683, paragraph 28).”

At para 38 of its judgment in the present case, the Divisional Court said that the CJEU was not laying down a test comprising alternative limbs either of which could independently lead to the conclusion that a measure constitutes a restriction on the right of free movement. It said that the test enunciated and applied was the same as had been applied in a long line of cases and showed that the potential deterrent effect of a measure was an integral element of the test to be applied. I agree.

As Mr Coppel demonstrated, the reasoning adopted by the court in Martens is the standard reasoning that had been previously adopted by the court on many occasions when analysing whether national legislation interferes with the right of free movement. It involves considering (i) whether the measure under consideration places an individual at a disadvantage or penalises him simply because he has exercised his right of free movement and (ii) whether the disadvantage or penalty is liable to deter the individual from exercising the rights of free movement. In Martens, the court expressed (i) at para 25 of its judgment. At para 26, it stated that legislation penalising an individual for exercising his right of free movement would deprive the right of full effectiveness if he could be dissuaded from exercising the right. The court brought both strands of the analysis together at para 31.

Mr O’Neill submits that the approach of the CJEU in Martens marks a shift from the court’s previous approach. In my view, the language of paras 25 and 26 of the judgment is not materially different from the language used in earlier authorities. In order to demonstrate this, it is sufficient to refer to Morgan at paras 25 and 26 and Prinz at paras 27 and 28. It is true that at para 31 of Martens, the court decided that the legislation was liable to “penalise” (rather than “deter” or “dissuade”) an applicant merely because he had exercised his right of freedom of movement. At first sight, this might seem to indicate that the question of whether an individual was liable to be dissuaded from exercising his EU right was immaterial and that, as Mr O’Neill submits, the only relevant consideration was whether the measure penalised the individual. I do not accept this. First, it is important to note that at para 31, the court said that the measure was “liable to penalise” (emphasis added), not that it “penalised” the individual. If Mr O’Neill is right, the only question is whether a measure does in fact penalise an individual. Secondly, it is striking that one of the authorities mentioned at para 31 is Prinz where the court said at para 32:

“[The national measure] is likely to dissuade nationals, such as the applicants in the main proceedings, from exercising their right to freedom of movement and residence in another Member State….”

There is no mention here of penalisation alone being sufficient. Further examples of decisions in which the liability of a measure to deter or dissuade the exercise of the right of free movement were given by the Divisional Court at para 38(1) of its judgment. Similar reasoning is employed in the case of other fundamental freedoms too: for example, C-211/08 Commission v Spain [2010] ECR I-5267 at para 72 and C-602/10 SC Volksbank Romania ECLI:EU:C:2012:443 at para 81 (free movement of services); C-291/09 Francesco Guarneri v Vandewvelde Eddy [2011] ECR I-2685 at para 17(free movement of goods); C-282/04 Commission v The Netherlands [2006] ECR I-9141 at para 29 (free movement of capital); and C-311/08 SGI [2010] ECR I-487 at para 50 (freedom of establishment).

All of these authorities are inconsistent with Mr O’Neill’s submission that it is sufficient merely to point to something which can be characterised as a penalty upon the exercise of the right of free movement. As Mr Coppel points out, if the only question was whether a measure disadvantaged or penalised an individual for having exercised his right to move to another Member State, it would not be necessary to consider whether the penalty or disadvantage was liable to dissuade the individual from exercising his right of free movement. And yet, the Grand Chamber in Morgan at para 30 decided that the impugned measure in that case was a restriction of the right of free movement not merely because it placed the applicants at a disadvantage, but because that disadvantage was liable to discourage citizens of the EU from leaving one Member State to pursue studies in another Member State.

There will be some cases where it is considered to be obvious and axiomatic that a penalising measure will be liable to deter or dissuade an individual from exercising his right of free movement and where the court may not consider it necessary to spell out that the measure will have that effect. Measures which have the effect of causing claimants to lose benefits if they have exercised their right of free movement are paradigm examples of measures which are classified as obstacles to free movement. That is because they are considered to be self-evidently “capable of hindering or rendering less attractive the exercise by Community nationals of the fundamental freedoms guaranteed by the Treaty”: see Case C-212/06 Government of the French Community v Flemish Government [2008] ECR I-1683 at para 45.

Mr O’Neill relies on C-406/04 De Cuyper [2006] ECR I-6947 in support of his submission that all that he has to show is that section 2 of the 2015 Act disadvantages individuals like the claimants for having exercised their right of free movement. This was a social security benefits case. It is perhaps, therefore, not surprising that at para 39 of its judgment, the court simply said:

“It is established that national legislation such as that in this case which places at a disadvantage certain of its nationals simply because they have exercised their freedom to move and reside in another Member State is a restriction on the freedoms…”

I do not consider that De Cuyper shows that it is a principle of EU law that there is no need to show that an impugned national measure is liable to deter or dissuade an individual from exercising his right of free movement. In the light of the weight of the EU jurisprudence and the wider point made at paras 41 and 42 below, there is no basis for holding that such a principle exists.

I should add that Mr O’Neill submits that the rules at issue in Martens could not have deterred free movement because they were enacted after the claimant had moved from the Netherlands to Belgium. Advocate General Sharpston made it clear that she considered that a measure requiring uninterrupted residence as a condition for an education or training grant was “likely to dissuade nationals from exercising their right to freedom of movement”: see AG 101 and 113.

More fundamentally, however, as the Divisional Court pointed out at para 37, the principle applied by the CJEU does not require the court to focus on the position of the individual concerned. As Advocate General Sharpston explained in the Flemish Government case at para 65 of her Opinion:

“I do not think that the Court should try to evaluate the precise extent to which such a measure affects the individual worker’s decision. Otherwise, the fact that some workers may not be daunted by a particular measure could always be used as a reason for holding that that measure’s effect on access to the labour market was potentially too uncertain and indirect. Moreover, it is difficult to see how the Court would go about conducting such an evaluation. It seems to me that, for a measure to constitute an obstacle, it is sufficient that it should be reasonably likely to have that effect on migrant workers.”

To summarise, I would reject Mr O’Neill’s submission that it is not necessary to decide whether a national measure is liable to deter or dissuade an individual from exercising his right of free movement for two principal reasons. First, there is a strong line of authority which unequivocally supports the proposition that a national measure does not interfere with the right of free movement unless (i) it penalises an individual for having exercising that right and (ii) it is liable to deter or dissuade him from exercising it or continuing to exercise it in the future. If para 31 of Martens did represent a fundamental shift of approach by the court, it is extraordinary that not only did the court not indicate that it was departing from its previous approach, but in support of what it said in para 31, it referred to earlier authority in which the previous approach was applied.

Secondly, as Mr Coppel puts it, it would make no more sense in Martens than in any other case for the CJEU to apply free movement rules to national legislation which had no dissuasive or deterrent effect on free movement. As the Divisional Court said at para 38(3);

“More fundamentally, it is the potential for deterrence which constitutes the mischief and provides the justification for the rule. It is what is capable of giving rise to a restriction. It is the fact that a national measure is capable of hindering or rendering less attractive the exercise by EU nationals of their fundamental rights of free movement and residence which renders it incompatible with the fundamental rights of free movement guaranteed by the TFEU. ”

This statement is amply supported by, for example, the judgment of the Grand Chamber in the Flemish Government case (para 45). Put shortly, if a national measure has no potential effect on the exercise by EU nationals of their EU rights, it is of no relevance in EU law.

The CJEU has been careful to say that not every disadvantage of non-residence in a Member State is a restriction on the right of free movement. The court has held that a measure whose effect is “too uncertain or indirect” or “too uncertain or insignificant” to have the requisite deterrent or dissuasive impact on the right does not interfere with it. As we have seen (para 22 above), the “uncertain or indirect” test was applied in Graf and adopted by Elias LJ in the Divisional Court in Preston at para 38 of his judgment. In Morgan at para 32, the court rejected on the facts the submission of the German Government that the restrictive effects of the national measure were “too uncertain or too insignificant” to constitute a restriction on the freedom to move and reside within the territory of another Member State. Another example is Case C-275/12 Elrick v Bezirksregierung Koln [2014] 1 CMLR 16where the court said at para 29 that the national measure was likely to dissuade citizens such as the applicant from exercising their right to free movement and that the restrictive effects of the measure “cannot be regarded as too uncertain or too insignificant to constitute a restriction”. See also the Flemish Government case at para 51.

As we have seen, in Preston, both Elias LJ in the Divisional Court and Mummery LJ in this court held that the 15 year rule was too indirect and uncertain a restriction to be liable to deter British citizens from going to reside and work in other Member States, or from doing so for as long as they liked. It is true that this conclusion was reached in the context of a UK Parliamentary general election and the issues raised by an EU referendum are different from those raised by a general election. In the present case, the Divisional Court said at para 32 that, if the inability to vote in Parliamentary elections occurring at relatively regular intervals not exceeding 5 years could not reasonably be considered liable to dissuade or deter British citizens from exercising their rights of free movement, it is difficult to see how the inability to vote in a once occurring referendum on membership of the EU could be considered to be capable of having such an effect either. I do not consider that this qualitative assessment of the restrictive effects on the right of free movement of individuals who have been residing outside the UK for more than 15 years can reasonably be said to be wrong. There is no evidence as to the likely effect of section 2 of the 2015 Act on such persons. The court must therefore make a judgment. The difficulties of making this judgment were well explained by Mummery LJ at para 80 in Preston (see para 24 above). The Divisional Court in the present case said:

“We share the view of the Court of Appeal that the 15 year rule in its application to the Parliamentary franchise is not liable to restrict free movement within the European Union. In our view it is totally unrealistic to suggest that this rule could have the effect of deterring or discouraging anyone considering whether to settle or remain in another Member State. In this regard we note that according to the Parliamentary briefing paper, referred to at paragraph 8 above, in the period 1987 to 2014 the highest number of overseas voters registered in any one year has been 34,454 and the current number is 26,918. By contrast, the number of British citizens resident in other Member States of the European Union alone is estimated at between 1 and 2 million. The 15 year rule in its application to the franchise for the EU referendum is, in our view, an a fortiori case. We are unable to accept that the prospect of disenfranchisement in a one-off referendum is a factor which could influence a decision whether to settle or remain in another Member State. We conclude, therefore, that it is not a measure which requires to be objectively justified under EU law.”

I am not persuaded that the overall assessment by the Divisional Court is wrong.  Indeed, Mr O’Neill accepted that it was fanciful to believe that anyone would be dissuaded from leaving the UK because they might lose the right to vote if they remain resident in the EU for more than 15 years.  But he contended that it was not unrealistic to believe that British nationals who had been living outside the UK for more than 15 years might feel compelled to return in order to establish their right to vote. He submitted that for this group of people the referendum was more important than a general election, given that the outcome of the referendum could affect their right to remain citizens of the EU and, as a consequence, their way of life.  There is undoubtedly force in that submission and I would accept that the Divisional Court may have been wrong to say that given the conclusion in Preston, this was an a fortiori case. But the question is still whether the fact that they are disenfranchised will be reasonably likely to persuade British nationals who fall into that category to return to the UK, to adopt a formulation derived from the observations of Advocate General Sharpston in the Flemish Government case: see para.39 above. Viewing the matter in that way, I believe that the Divisional Court was entitled to conclude that it is unrealistic to suggest that they would. I do not disagree with Mr O’Neill when he says that the right to vote is an important constitutional right. It does not follow, however, that those denied that right after 15 years residence abroad will pick up sticks and return.  The appellants have not identified anyone who has actually taken that step and, whilst that is not conclusive, it does support the Divisional Court’s assessment that there is unlikely to be any material interference with free movement.  Putting it another way, the effect on the class as a whole is too uncertain, indirect and/or insignificant to amount to a restriction on their rights of free movement.

For all these reasons, I would hold that the Divisional Court correctly concluded that the 15 year rule does not interfere with the right of free movement. In view of this conclusion and the need to deliver this judgment as a matter of urgency, I do not propose to consider the claimants’ challenge to the conclusion of the Divisional Court that, if the rule does interfere with their EU rights, it is objectively justified because (i) it had a legitimate aim, namely a relevant connection with the UK as a qualification for the EU Referendum franchise; and (ii) it constituted a rational, consistent and proportionate means of achieving that aim.

Mr O’Neill submits that the claimants’ right to vote as British citizens is “a thing of the highest importance, and so great a privilege a thing of the highest importance, and so great a privilege that it is a great injury to deprive the plaintiff of it”: Ashby v White (1703) 1 Smith’s LC (13^th ed, 1929) 253, per Holt CJ at 273. Because the right to vote is a “constitutional right” recognised as such at common law, the “principle of legality” applies in construing any statutory provision which seeks to abrogate that right: see Watkins v Secretary of State for the Home Department [2006] UKHL 17, [2006] AC 396 per Lord Rodger at para 61.

Mr O’Neill submits that the court has a discretionary power at common law to declare legislation unconstitutional where it conflicts with fundamental constitutional rights such as the right to vote, and that such a common law declaration is a possible and appropriate remedy in the circumstances of this case.

The short answer to this new point is that there is no such common law right as that for which Mr O’Neill contends which could take precedence over an Act of Parliament. That was the answer given by the Supreme Court in Moohan v Lord Advocate [2014] UKSC 67, [2015] AC 901, where a similar argument was raised on behalf of prisoners who had been excluded from the franchise by section 3 of the Representation of the People Act 1983 Act and was roundly rejected by Lord Hodge at paras 32 to 37. Lord Hodge did not exclude the possibility that:

“in the very unlikely event that a parliamentary majority abusively sought to entrench its power by a curtailment of the franchise or similar device, the common law, informed by the principles of democracy and the rule of law and international norms, would be able to declare such legislation unlawful.”

But that possibility has no application in the present case. There is nothing abusive about section 2 of the 2015 Act. I regard this argument as hopeless.

For all these reasons, I would dismiss these appeals. Section 2 does not fall within the scope of EU law. But if it does, it does not restrict the rights of free movement of the claimants or other British citizens who are in the same situation as they are.

I agree with the judgment of the Master of the Rolls on the two issues he has considered. I wish to make some observations about an argument which was, it seems to me, logically implicit in the submissions of the respondents on the Article 50 point, albeit that it was not directly advanced. The issue is whether, as a matter of domestic law, EU law could take precedence over UK law in relation to the rules adopted by the UK to determine whether or not to leave the EU. The question was raised by the court for the first time in the course of the hearing, and neither counsel had an opportunity to deal with it fully. I discuss the matter, and express some provisional conclusions on it, because I think it is of some constitutional importance and because it may need to be considered if this case were to go further.

Article 50 (1) of the TEU states:

“Any Member State may decide to withdraw from the Union in accordance with its own constitutional requirements.”

The respondents submit that this provision should be read so as to allow the UK to exercise complete sovereignty over the decision whether to withdraw from the European Union. They say that, properly construed, Article 50 is no more than a recognition of the fact that the question of withdrawal from the EU Treaties is pre-eminently a matter falling within the sovereign competence of the state itself. Having accepted that the UK had the sovereign right to determine its own arrangements for leaving the EU, the Divisional Court could not at the same time properly conclude that the rules for establishing the franchise laid down in section 2 of the EU Referendum Act were subject to the fundamental principles of EU law.

The starting point for this submission is Article 50 itself. It is, therefore, implicit in the argument that EU law is in principle binding on the UK even in relation to the rules which the UK chooses to adopt for the purpose of deciding whether or not it should leave the EU. The UK has sole jurisdiction over that question but only because EU law, in the shape article 50(1), has chosen to relinquish its own jurisdiction and transfer it back exclusively to the UK.

I have serious doubts whether EU law does take precedence in the context of the question of withdrawing from the EU. EU law is given primacy over UK law only by virtue of section 2(1) of the European Communities Act 1972; it is because Parliament has said that it should, not because EU law has said that it should.

I accept that in general the effect of section 2(1) is that EU law must take precedence whilst the 1972 Act remains in force. The courts have set their face against the notion that later legislation can impliedly repeal the 1972 Act. That is implicit in the Factortame decision (R v Secretary of State for Transport ex parte Factortame Ltd [1990] 2 AC 85). Moreover, this will be so even in relation to rules defining the franchise for the general election. In Preston both the Divisional Court and the Court of Appeal held that the 15 year rule, as applied to the franchise for general elections, was in principle subject to compliance with EU law, including the right to freedom of movement. The rule could be set aside in so far as it was inconsistent with the principles of EU law. The reason why the applicants failed in that case was because the rule did not in fact interfere with any right to freedom of movement. But in respectful disagreement with the Divisional Court, in my judgment it does not follow that the same principle should apply in the context of possible withdrawal from the EU.

Mr O’Neill accepts that EU law takes precedence over UK law only because of section 2(1) the European Communities Act. He concedes that if Parliament were expressly to dis-apply section 2(1) in relation to the EU Referendum Act, EU law would have no traction. But he submits that an express statutory provision would be necessary; nothing less will do. He agrees with the Divisional Court that Preston shows that the courts must give primacy to EU law even when applying the EU Referendum Act, at least for so long as section 2(1) remains on the statute book or unless it is expressly dis-applied. There is no implied repeal.

I doubt whether, purely as a matter of domestic law, Parliament would have intended section 2(1) to apply so as to give primacy to EU law where the very question in issue is whether the UK should remain bound by EU law. The effect of section 2(1) is to bind the UK to the rules of the club whilst it remains a member; but I do not think it can have been intended to bind the UK to those rules when the very question is whether it should be bound by those rules. Parliament agreed to join the EU by exercising sovereign powers untrammelled by EU law and I think it would expect to be able to leave the EU in the exercise of the same untrammelled sovereign power, whether the later legislation expressly dis-applies section 2(1) or not. It is not, in my view, a question of implied repeal but rather a question of the scope of section 2(1). Parliament would not have intended that the UK should give precedence to EU law when the very question to be decided is whether the UK should continue to give precedence to EU law.

This seems to have been the approach of the German Constitutional Court in Re Ratification of the Treaty of Lisbon [2010] 3 CMLR 13, discussed by the Master of the Rolls at para.14 and relied upon by the respondents. The German government in that case successfully submitted that article 50 was merely confirming the continuing existence of state sovereignty; it was not establishing it. As the German government put it in its submissions, a state “would remain the ‘masters of the Treaties’ and would not have granted the European Union Kompetenz-Kompetenz” over the question of withdrawal from the Treaties. It seems to me that the logic of this approach is that section 2(1) should not be read as extending to the very question of withdrawal itself; Parliament would not have granted that competence to the EU. Mr Coppel for the respondents, in agreement with Mr O’Neill, suggested that this would only be the case if section 2(1) was expressly dis-applied. However, that starts from the premise, which in my view is inconsistent with the reasoning in the German constitutional case on which the respondents rely, that the UK did indeed give competence to the EU on this issue, at least until that competence is expressly withdrawn. On the respondents’ case EU law would have been binding, absent the express dis-application of section 2(1), even had article 50 said that a state could only leave the EU if there was a two thirds majority in favour following a referendum. I doubt whether Parliament intended section 2(1) to have that effect.

In my judgment, the construction of Article 50 adopted by the Master of the Rolls, with which I entirely agree, simply recognises the political reality that EU law can have no part to play in the decision whether a state chooses to remain in the EU. As a consequence of this construction, nothing in fact turns on the question whether it is EU or UK law which takes precedence. Even if it is EU law as a result of section 2(1), the EU has chosen to exercise its power so as to refer sovereignty back exclusively to the UK. But the issue would become important if the Divisional Court’s construction of article 50 were correct.

I too would dismiss the appeal.

I agree with both judgments.