Thursday 29 January 2015

Who should get the vote in a Brexit referendum?


Steve Peers

“Nasty Yes won the referendum yesterday”

That was my initial response to the Brexit referendum of 1975, written precociously in my childhood diary. I rather suspect that a hundred or so miles away, in a rather posher diary, and in an infinitely posher school, David Cameron was writing exactly the same words.

Of course, we had different motivations for this youthful flirtation with the Dark Side. In my case, I was simply copying my parents’ left-wing opposition to the EU, which remained unabated even though my dad had been one of the very first British citizens to (briefly) exercise the free movement of workers to another Member State. (I remained in the UK with my mother, who was a third-country national; it’s lucky we didn’t end up as an EU law exam question).

On the other hand, David Cameron probably didn’t care much about the referendum at all, but felt he needed to write those (hypothetical) words in his diary so that right-wing Eurosceptic bullies would finally leave him alone. They didn’t.

Forty years later, it looks like we may have another Brexit referendum in the near future, depending (as things stand) upon the very uncertain result (and aftermath) of the general election due in May. I have already blogged here about the reasons why advocates of the EU should support such a referendum. My topic today is who should get the vote in a Brexit referendum; I suspect many on the pro-EU side won’t like my argument on this issue much either. But like my case for a referendum, I believe that the case I make here is a principled one – and ultimately, the pro-EU case can only legitimately be made upon principled foundations.

Although it’s not yet certain that a Brexit referendum is imminent, I am prompted to write now on the issue of the franchise due to comments on the weekend by Nigel Farage, the leader of the UK Independence Party, who said that only British citizens should get the vote in the referendum. It wasn’t clear whether he thought British expatriates in the EU (or elsewhere) should also get it.

The starting point should be the franchise rules that already exist, although they could be changed for a referendum (as they were for the Scottish independence referendum last year). Among the existing rules, it makes sense to focus on those for UK-wide elections, rather than those for local government or the devolved bodies in Scotland, Wales and Northern Ireland, since a Brexit vote would be held nationwide.

There are two nationwide templates to choose from. In general elections, all British, Irish and Commonwealth citizens can vote, as can any British expatriates if they have been away from the country (whether in the EU or elsewhere) for less than 15 years. In European Parliament elections, citizens of all European Union countries can vote.  (For more details, see here).

At first sight, it might seem attractive to argue (as many on the pro-EU side do) that all EU citizens in the UK should get the vote. The departure of the UK from the EU would certainly affect them fundamentally. Even though UKIP’s current official position is that all those legally present should get to stay, there’s no guarantee that this protection would exist in practice after Brexit. And giving all EU citizens the vote is not unprecedented for nationwide elections, as they have the vote in European Parliament elections. (Indeed, they have the vote in local elections too).

However, this argument should be rejected, for two reasons. First of all, European Parliament elections are different in principle from other nationwide votes. They determine who will be the UK’s Members of the European Parliament, but that is a multinational body with a role in EU-wide decision-making. Of course EU laws have an impact on the UK, but the European Parliament is not the place where decisions distinct to the UK as a separate state get made. In contrast, such decisions get made via means of direct democracy, in nationwide referendums, and more frequently via means of indirect democracy, via means of our vote for the national parliament. So it makes more sense for all votes on the future of the UK as a separate state to be subject to the same franchise rules. In fact, this is the practice: the Westminster voting rules (leaving aside members of the House of Lords) were applied to the 2011 referendum on a change to the electoral system, and in the recent private members’ bills (supported by the Conservative party) providing for a 2017 Brexit referendum.

The second argument is one of legitimacy. If the pro-EU side narrowly wins a Brexit referendum in which all EU citizens are allowed to vote, Eurosceptics will endlessly claim that the election was ‘stolen’ from them. I can already anticipate the reaction to this point: Eurosceptics will demand another referendum anyway. The historical parallels are legion: Quebec separatists demanded another referendum after they lost the first one in 1980; Scottish nationalists are already agitating for a second independence referendum; and the Eurosceptics of the 1970s took over the Labour party shortly afterward.

But the point is not to try to stop hardcore Eurosceptics arguing for another Brexit referendum. They are bound to do that. The point is to stop them winning the argument for another referendum in public opinion more broadly. The pro-EU side should aim to win that argument fairly, by ensuring that the upcoming Brexit referendum (if there is one) is, as far as possible, beyond reproach.  (Again, of course hardcore Eurosceptics are bound to reproach it if they lose; the battleground is mainstream public opinion). The result of a Brexit referendum is always likely to be seen as a little bit dubious in mainstream public opinion if it depends on the votes of people who don’t usually have the vote in general elections, given that EU citizens can only vote in EP elections in the first place because of EU law. A good historical parallel would be the Canadian election of 1917, which was won by a pro-conscription party in the midst of the First World War by disenfranchising conscientious objectors and enfranchising women, but only if they were related to servicemen. For good reason, this victory was regarded as illegitimate by those opposing conscription.

Furthermore it’s not impossible to convince the broader public that a fresh referendum is unnecessary, where there’s a good case that the earlier one was clearly legitimate. After all, while Quebec separatists did get to hold another referendum in 1995, they have not yet secured another one despite their very narrow loss on that occasion. The Eurosceptic takeover of the Labour party proved shortlived, since it contributed to splitting the party and its biggest electoral defeat since the 1930s.

Should the franchise be changed in some other way for a Brexit referendum? No, for the same reason: any special rule would look like an attempt to fix the result, possibly in a Eurosceptic direction. So the franchise should not be narrowed to British citizens only: Irish and Commonwealth citizens have the vote in general elections, so should have the vote for the Brexit referendum too. (Remember that Cyprus and Malta are in both the EU and the Commonwealth). So should British citizens living abroad, whether they are living in the EU or not. It can’t be assumed that they are all pro-EU voters, as my dad’s example shows. Some of them may even have left the UK because of a belief that there were too many immigrants there – displaying an obvious lack of appreciation of irony.

What if the general election mandate is changed anyway? There may indeed be a case to be made that only British citizens should vote in general elections. However, for good reasons, there is a tradition that changes to the franchise need broad support across the political spectrum, including the main opposition party. And it would clearly be obnoxious to shove through a change to the franchise in the last few weeks before a general election. If there is such broad support for changing the franchise after the next general election, the first time to apply the new rules should be the subsequent general election, not any Brexit referendum that might take place.

My argument above will disappoint those who believe strongly that EU citizens in the UK should have the vote on a matter that affects them so significantly. But the case against letting them vote in a Brexit referendum is also tactical: the complaints against it are more likely to lead to a further referendum in the near future, or even to increase the votes for the anti-EU side. Yet even if the net result of letting all EU citizens vote would be to increase the pro-EU vote, it would be wrong to give them the vote in principle. And if EU citizens are particularly keen to vote, they could always consider obtaining UK citizenship.

Postscript (10 May 2015):

The above analysis obviously becomes more relevant after the results of the UK general election, given that a referendum on EU membership now seems certain to happen in the near future. That doesn't give me any reason to change my view, but it's a good occasion to make two further points.

First of all, I would reiterate the case against changing the voting franchise at all for the EU referendum, no matter what the changes concern. So the UKIP argument that 'only UK citizens should have the vote' should also be rejected, because it is equally wrong in principle. If Commonwealth citizens and Irish citizens can normally vote on the key issues facing this country in general elections, why can they not vote on the question of EU membership? Excluding them would appear to be an attempt to bias the vote toward the anti-EU side. Similarly, while I am in favour of extending the vote to 16 and 17-year-olds, the EU referendum would be the wrong time to change the rules. Again, it would appear to be an attempt to fix the results, particularly in light of the bizarre UKIP argument that young minds have been polluted by EU colouring books featuring evil Euro-cows.

Secondly, I think there's a better way to put my argument of principle against allowing all EU citizens to vote in the EU referendum. The whole point of the referendum is to decide on the polity (ie the political identity) of the UK in the first place. Should we remain part of the EU or not? Allowing all EU citizens in the UK to participate in that decision would effectively mean that they are judges in their own case.

I can understand the concerns of EU citizens in the UK that they ought to have a vote in an issue which will affect them so directly.  One solution is to obtain UK citizenship. For those unwilling or unable to do that, they have every right to participate in the broader public debate to convince voters that EU membership makes an important contribution to the UK's economic prosperity, environmental protection. social justice, animal welfare and cultural identity - and in particular that migrants from the EU make a huge contribution to Britain's economic and social success.  

Second postscript (25 May 2015)

Press reports indicate that the UK referendum on EU membership will indeed essentially follow the general election franchise, and not extend the vote to citizens of other EU countries (besides Ireland, Cyprus and Malta) living in the UK. Some have questioned the legality of this, but there seems little ground to challenge it. The EU Treaties require Member States to give all EU citizens the vote in local and European Parliament elections, but there is no reference to other votes. Since the Treaties include the power (never used) to widen EU citizens' rights beyond those listed, it must therefore be presumed that the Treaty drafters did not intend to require Member States to extend voting rights in referenda (or in any other elections) in the absence of a decision to lengthen the list of rights attached to EU citizenship.  As for the ECHR, the First Protocol to the Convention requires States to hold 'free elections', but on the face of it this does not apply to referenda, and the case law does not suggest that voting must be extended to non-nationals in any event.     

Barnard & Peers: chapter 2

Photo credit:
Meme: Steve Peers
Jumper design: Nigel Farage :)

Tuesday 27 January 2015

The Ups and Downs of Dual Citizenship – the CJEU on Dual Turkish/EU citizens and social rights

Professors: Elspeth Guild, Queen Mary University of London; Kees Groenendijk, Radboud University; and Steve Peers, University of Essex

In recent years, an increasing number of Turkish citizens residing in the EU have become dual citizens of Turkey and an EU Member State. Like other dual citizens of a Member State and a non-Member State, they can invoke EU free movement law to move and reside in another Member State (see the CJEU’s Micheletti judgment). But as a general rule, EU free movement law will not confer rights in their home Member State. So in that case, can they still rely upon their Turkish citizenship to claim rights under the EU/Turkey association agreement?

Back in 2012, in its judgment in Kahveci and Inan, the CJEU determined that dual Turkish/Dutch nationals were entitled to rely on their Turkish citizenship to enjoy the benefits of the EU/Turkey association agreement as regards family reunification and thus escape the stricter national legislation for own nationals’ family members. So until the recent decision earlier this month in Demirci, it looked as if Turkish citizens with a second passport from a Member State were more or less free to choose which one to rely on in order to benefit from EU law. But the Demirci judgment changes the rules slightly and provides a more elaborated legal reasoning to when dual nationals of Turkey and a Member State may use one or other of their citizenship.

The finding is rather ominous so we will start some background on the facts of the case and conclude with a rather optimistic analysis where we conclude that dual nationality is still a strong source of rights in EU law and that this judgment is perhaps the exception.

The Facts:

Mr Demirci and his fellow applicants are all former Turkish workers who had worked in the Netherlands for many years fulfilling the conditions of the secondary legislation of the EU/Turkey Association Agreement, Decision 1/80 of the EU/Turkey Association Council, which regulates aspects of the immigration status of Turkish workers and their family members. They had all naturalised as Dutch nationals but kept their Turkish citizenship. They all became disabled and incapacitated for work and thus permanently left the labour force. They received a Dutch social benefit designed to provide income for the incapacitated. But this benefit is rather low so they applied for a top up benefit to bring their income closer to that of the minimum wage. They were all awarded the top up benefit.

Then a series of things happened. The men retired with their families to Turkey. The Dutch Government began to change the rules on the top up benefit to exclude anyone not resident in the Netherlands (or the EU). Mr Demirci and his colleagues first had their top up benefit reduced then it was cut off altogether on the basis that they no longer lived in the Netherlands (or EU). They appealed, relying on their Turkish nationality, on the basis that this treatment was contrary to the EU/Turkey Association Agreement measure on social security – Association Council Decision 3/80. The argument went that the Dutch authorities may be able to cut off the top up benefit to their own nationals living outside the EU as this is a matter wholly internal to one Member State. But they cannot cut off the top up benefit to Turkish nationals who have fulfilled the conditions as workers in a Member State under Decision 1/80 as this is a matter of EU law (which does not permit such an act – Article 6(1) of Decision 3/80 protects Turkish workers who retire to Turkey as regards receipt of social benefits). Several years ago, the CJEU ruled in Akdas that such rules infringed Decision 3/80 as regards Turkish nationals who returned to Turkey. But could a dual citizen of Turkey and a Member State rely on that judgment, as a Turkish citizen – or would he or she be prevented from doing so, as a national of that Member State?  

The reasoning

The CJEU bluntly tells Mr Demirci and his colleagues that they cannot rely on Decision 3/80 to object to the residence requirement imposed by the Dutch authorities (para 52). This is because, according to the Court, the objectives of the Decision and the EU/Turkey regime is to ensure the progressive integration of Turkish workers into the territory of the host Member State. The social security provisions consolidate that objective.

The Court provides two main reasons for this position. First, because Mr Demirci and his colleagues had acquired Dutch nationality they are in a particular situation as regards the Agreement. Citizenship is ‘the most accomplished level of integration’ of a person into the host state (para 54). This new citizenship means that the former Turkish worker can now enter and reside freely in the Netherlands or indeed any other EU Member State where he or she might wish to go. Conversely, in Kahveci & Inan the Court did not accept the argument of the Dutch government that naturalisation is the pinnacle of integration. In that case AG Sharpston argued that naturalisation may be an indication that an immigrant is on his path to integration, but that is not the same as saying that he has become completely integrated. .

But as Turkish nationals, Mr Demirci and his friends could only live in Turkey or their host Member State (the Netherlands) and so they have no free movement rights. Further as such, they only benefit from certain rights in the host Member State. So, says the Court, for the purposes of paying them a benefit, it is reasonable for the national authorities to make this subject to the same rules as apply to all other Dutch nationals (para 57).   

Secondly, dual Turkish/Dutch nationals would be placed in a better position than other EU citizens if they were allowed to have the top up benefit even though they did not fulfil the residential requirement (para 58). The right to export to Turkey social benefits in Decision 3/80, according to the Court, is a sort of compensation for the fact that Turkish nationals will no longer be able to return to and live in the host Member State. As the CJEU held in Bozkurt, a Turkish national ceases to be a protected person under Decision 1/80 if he or she becomes totally and permanently incapacitated for work (para 64). So there is a justification for applying different rules to those who are exclusively Turkish nationals as they have a much less secure residence status in the host Member State and no free movement rights in EU law. They therefore need the extra protection of the export right. For dual Turkish/Dutch nationals, they can always move back to the EU and fulfil the residential requirements for the top up benefit (even if they would rather not) (para 65).

 Distinguishing Kahveci & Inan

As mentioned above, this judgment takes a different approach from the CJEU’s own jurisprudence in Kahveci & Inan where it held that dual Turkish/Dutch nationals were allowed to rely on their Turkish nationality for the purposes of the EU/Turkey family reunification rules, in order to benefit from the expulsion of Turkish workers’ family members, which are more favourable than the rules applying to the expulsion of the family members of Dutch citizens in the Netherlands. However, the CJEU is anxious to protect its ruling in Kahveci and goes to some lengths to explain why the finding in Kahveci is consistent with that which they were giving in Demirci (para 66). The argument goes like this. Family reunification enhances integration, for Turkish workers who are already legally integrated into the host Member State. Article 7 of Decision 1/80 deepens the last integration of a Turkish worker by granting to that worker’s family members, after three years residence, access to the labour force (para 67). So, acquisition of national citizenship could not be used as a reason to deprive the worker of the benefit of family reunification in Decision 1/80 (para 68). By contrast with the facts in Demirci, the family’s integration would be hindered if it was denied on the basis of dual citizenship. Further in Kahveci the person was seeking to benefit family members who are also Turkish nationals (para 70). Presumably this reasoning means that dual Turkish/EU citizens can also invoke the ‘standstill’ clause in the EU/Turkey association agreement, as interpreted last year in the CJEU’s Dogan judgment (discussed here), to avoid stricter rules for family reunion that apply to a Member State’s own nationals.

But in the case of Mr Demirci he can always go back to the Netherlands (or the EU) to satisfy the residential requirement to get the top up (para 69). Further all he and his colleagues wanted was a top up benefit for themselves (para 71). Finally, the CJEU considered that if Mr Demirci could rely on Decision 1/80 to have the top up benefit while not fulfilling the residential requirement, this would put them in a better situation than that of other citizens of the Union (and thus contravene Article 59 of the Additional Protocol to the EU/Turkey association agreement, which rules out Turkish citizens being better off than EU citizens).

The Court’s approach seems to be that there is something inherently wrong about letting Mr Demirci and his colleagues have their cake and eat it too. The arguments may not be the most compelling in the world but they show a clear judicial line. The CJEU will favour Turkish nationals living in the EU even if they have taken a second citizenship so long as this improves their long lasting integration. But they cannot rely on their Turkish citizenship after naturalization when what they seek is a financial benefit which is subject to a residential requirement for EU citizens (and which they do not fulfil because they have left the EU).

The Court appears to implicitly return somewhat to its reasoning in the Mesbah judgement of 1999, where it held that the Moroccan mother of a Belgian-Moroccan worker who was living with her son in Belgium could not rely on the clause prohibiting discrimination on grounds on nationality in the EEC-Morocco Association Agreement to claim a disablement allowance that under Belgian law was only granted to Belgian nationals. The Court in Demirci, however, does explicitly point to the difference with Kahveci & Inan: “[in] the present case, by contrast, the respondents in the main proceedings are relying on the provisions of Decision 1/80 on their own behalf and in their own interest” (para 70). The Court leaves the door open for the Turkish spouse of a Turkish/Dutch (ex-) worker to rely on Decision 3/80, because the spouse, not having Dutch nationality, would be unable to return to the Netherlands.

The most important thing to remember about the Demirci decision is that it does not undermine the Court’s judgment in favour of dual rights for dual citizens in Kahveci & Inan. Yes, Turkish nationals can rely on the EU/Turkey association agreement family reunion rules even after they have naturalized in their host Member State, provided they are allowed to have dual nationality (it isn’t yet clear if they could invoke the EU’s own family reunion Directive). But they cannot rely on their non-EU citizenship after they leave the EU to get around a national residential requirement for the export of a social benefit if such a requirement applies to nationals of the host Member State.

Barnard & Peers: chapter 13, chapter 26
Photo credit:

Sunday 25 January 2015

Catharsis or catastrophe: what next for Greece and the Eurozone?


Steve Peers

Yesterday the anti-austerity party Syriza won a large victory in Greek elections and seems certain to become the government, probably in coalition with a smaller party. What is the likely impact upon the EU’s economic and monetary union?

The starting point is Syriza’s election platform. As discussed in more detail in this Open Europe blog post, that party’s aim is not to leave the EU or even the single currency, but rather to renegotiate Greece’s debts and the related austerity obligations. In particular, it wants part of Greece’s debt to be forgiven, and the terms of the remaining debt to be renegotiated, along with an abolition of the austerity demands made upon Greece as condition of previous bail-outs.

But whatever the political and economic arguments for this programme, it potentially faces some legal hurdles. There are limits on forgiving debt or ending austerity, as set out in the EU Treaties and the case law of the CJEU, which I discussed in a previous blog post (which this post updates).

In particular, according to Article 136(3) TFEU, any financial assistance to a eurozone Member State must be subject to ‘strict conditionality’. This is consistent with the CJEU ruling in Pringle, which stated that the ‘no-bailout’ rule in the EU Treaties (Article 125 TFEU) allowed Member States to offer each other financial assistance on the condition that it took the form of loans, rather than a direct assumption of Greek government debt by other Member States. Moreover, the CJEU pointed out, the ESM Treaty (the treaty between eurozone Member States which governs bail-outs) required that in the event of non-payment, the loans would remain payable, and had to be charged an appropriate level of interest.

So it’s not possible for Member States to drop all conditionality as regards loans to Greece, to forgive debt as such or to loan money interest-free. But it is open in principle to reduce the stringency of the conditions somewhat, to reduce the interest rates payable and to lengthen the repayment period – although there is always the risk that some litigant will try to convince a national court or the CJEU that this is going too far. Moreover, the rules in the EU Treaties only bind EU institutions and Member States, not private parties, third States or international organisations (although it might be argued that Member States are constrained as members of the IMF not to violate the no-bailout rule indirectly). So any renegotiation or default as regards such creditors is not subject to EU law rules in principle, although of course other legal rules might be applicable.  And as the Open Europe analysis points out, the bulk of the debt is owed to the Eurozone.

The case law does not rule out a short period of non-repayment of principal or interest, as long as the loans remain payable and subject to interest. Nor does it specifically address the possible conversion of loans into bonds, as some in Syriza have suggested.

Overall, it’s hard to see how the relatively modest renegotiation which EU law would permit would do enough to reduce Greece’s mountain of debt significantly, or to satisfy the voters which supported a Syriza-led government.

The renegotiation of loans might not be the only possibility to help out Greece. For example, arguably the Treaties do not rule out a form of (supplementary) unemployment insurance system as between Eurozone Member States, or support for another Member State’s social spending, as long as it would not take the form of paying off another State’s debts as such.

There is the ultimate possibility of leaving the euro, either at the behest of Greece itself or the other eurozone Member States. As I pointed out in the previous post, it isn’t legal to leave the Eurozone (or to force a Member State out) without that State leaving the EU. On that point, while it’s open to any Member State to leave the EU, it’s not legal to force a Member State out. At the end of the day, though, the European Central Bank holds the trump cards, since it could force a Member State to leave monetary union in practice by stopping the supply of money to that State. The independence of the ECB prevents politicians from ordering the bank to take such a radical step, but it might act on its own initiative.

Quite apart from its very dubious legality and severe economic effects, such a move would be a huge political mistake. The result might not be an increase in support for those moderate parties that reluctantly supported austerity, but rather for the far-right neo-Nazi Golden Dawn party, which came third in the elections.

The better course for the EU is to take this opportunity to re-engage with the millions of EU citizens who are affected or angered by austerity, and re-orient the EU towards ending that austerity, instead of generating more of it. Although this is more easily said than done, it should never be forgotten that the initial rationale for the EU was not austerity, but economic growth which raised living standards for the population as a whole. So in voting for a party which promised the latter, Greeks have reaffirmed, not rejected, the Union’s traditional raison d’etre, reminding it that the Union cannot maintain its social or political legitimacy if it becomes no more than a mechanism for enforcing austerity.

Barnard & Peers: chapter 19

Friday 23 January 2015

More research is needed: the EU’s attempt to attract more non-EU researchers and students


Steve Peers

The economies of the EU’s Member States, like other advanced developing countries, increasingly need highly-skilled workforces. But since the EU’s current laws have only had a modest impact on attracting more students and researchers to the EU, the Commission proposed an overhaul of the legislation concerned in 2013. The European Parliament (EP) and the Council have each now agreed  their position on the proposed new law (see here and here), and will soon start negotiations on a final text. This is a good time to examine what the new law should include, if the EU genuinely wants to increase the numbers of researchers and students coming from third States.


The existing legislation on this issue is the researchers’ Directive, adopted in 2005, and the students’ Directive, adopted in 2004. The latter Directive also applies to the admission of school pupils on exchange programmes, unpaid trainees and volunteers, although Member States have an option to apply it to the latter three groups of migrants. The Commission issued reports (see here and here) on the two Directives in 2011, concluding that they had not done much to accomplish their objective (there were only 7000 researchers admitted in 2010) and probably needed to be reformed – hence the 2013 proposal. The CJEU has ruled twice on the interpretation of the students’ Directive, first of all ruling in Sommer that Member States could not apply a labour-market preference test for students, and secondly ruling in the Ben Alaya case (discussed here) that Member States must admit students who comply with the rules on admission in the students’ Directive. The same logically applies to the current researchers’ Directive. The UK and Denmark opted out of both Directives, while Ireland opted in to the researchers' Directive.

The proposed new law

The Commission proposal would merge the students’ and researchers’ Directives, making major changes to them both. In order to attract more researchers and students to EU territory, there would be stronger rules on their equal treatment and their movement (‘mobility’) between Member States for the purpose of their studies and research; those rules would also be extended to paid trainees. They would be able to stay after their research or study for a period of 12 months to look for work or self-employment, although after 3 months Member States could check on the genuineness of this search, and after 6 months they could ask the migrant to prove that they have real prospects. Students would be able to work for 20 hours a week (the current rules provide for 10 hours), and the option to ban students from working during their first year of studies would be dropped.

Also, the proposal would replace the weak rules on family reunion in the current researchers’ Directive with a fully-fledged right to family reunion, which would moreover waive some of the restrictions in the EU’s Directive on family reunion for third-country citizens. In particular, researchers could obtain family reunion without a minimum waiting period, and would not have to show that they have a reasonable prospect of permanent residence. Integration requirements for family members could only be applied after entry, and there would be a shorter deadline to process applications. Family members would have a longer period of authorised stay, and no wait before they could access the labour market.

The 2013 proposal would also introduce a 60-day deadline to decide on applications for admission, shortened to 30 days for those benefiting from EU mobility programmes. (The current laws have no deadlines to decide on applications at all).  

Finally, the proposal would require Member States to apply the rules relating to school pupils, unpaid trainees and volunteers, and would furthermore add rules on two new groups of migrants: au pairs and paid trainees. It would also limit Member States’ current power to apply more favourable rules for students and researchers, confining that power to only a few provisions relating to the rights of migrants, while fully harmonising the rules on admission.

The European Parliament and Council positions

On some key points, the EP and the Council are going in quite different directions. The EP wants to oblige Member States to consider in-country applications for researchers, while the Council (like the Commission) wants to retain the existing rules, which give Member States an option to allow this. In the EP’s view, the equal treatment rules should be strengthened, but the Council wants to weaken them. The EP wants to extend former students’ and researchers’ stay to look for work from 12 to 18 months, and to make Member States wait longer to check on the genuineness of the job search or likelihood of employment. On the other hand, the Council wants several restrictions: to reduce the stay to 6 months; to allow Member States to limit students’ possibility to stay to those who have at least a Master’s degree; to check on the likelihood of employment after 3 months; and to give Member States an option to limit the job search to the areas of the migrant’s expertise. The EP agrees with the extension of students’ employment rights, while the Council wants to revert to the current 10-hour a week limit, and introduce a possible labour-market preference test (overturning Sommer).

As regards family reunion, the EP wants to extend the Commission’s proposal to family members of students as well, while the Council wants to limit the rules to researchers, without requiring Member States to give researchers’ family members immediate access to employment. Similarly, the EP believes that the new mobility rules should be extended to cover unpaid trainees and volunteers as well, while the Council wants to limit them to researchers and students (as at present), and to add very complicated details to the proposal on this issue. The EP supports an even shorter period to decide on applications (30 days), while the Council wants to raise the time limit to 90 days.

Finally, the EP agrees with the Commission proposal to extend the Directive to further categories of migrants and to make all of the rules binding, while the Council rejects both suggestions.

On a few points, the EP and Council are going in the same direction. Both institutions want Member States to have a greater possibility to apply more favourable rules for the persons concerned. The Council wants to go further still, and provide expressly that Member States can have rules on admission of other categories of students or researchers. Also, both the EP and the Council agree  that the current rules on delegating decision-making to research institutions or universities should remain, and that many of the rules on whether to refuse entry or withdraw or refuse to renew the right to stay should be optional, not mandatory (as the Commission proposes).


The EP’s and Council’s positions should be examined in light of the Commission’s impact assessment report for the proposed Directive, which makes detailed arguments for the amendments which the Commission proposed. This report provides evidence that a job-search period after the end of research or studies is a factor in attracting students or researchers, although presumably the Council’s less generous version of the rules on this issue would be less attractive in practice than the EP’s version. There is also evidence that migrants are attracted by further employment rights for students and for researchers’ family members: here the EP and Commission are seeking to amend the rules to this effect, but the Council is not.

Would-be migrants are also deterred by the great variety of national rules and the rules on mobility between Member States.  On this point, the Commission and EP position would certainly address this concern, by significantly reducing Member States’ options to have divergent national laws and providing for a relatively simple extension of a stronger version of the current mobility rules. The Council, for its part, would leave more national divergence in place, and amend the mobility rules in a fiendishly complicated way. Normally, I’m a big fan of the fiddly details of EU immigration law, but this is far too much of a good thing.

What should the priorities be for the forthcoming negotiations? The evidence clearly shows that the EU’s ailing economy would benefit most from more students and researchers, and less so as regards the other categories of migrants. And given the huge resistance of Member States to extending binding rules to the other categories of migrants, this would be difficult to agree, if it’s possible at all. So the EP should trade off one set of amendments against the other.

In particular, it should drop its insistence that binding rules should govern all aspects of the admission and stay of school pupils, trainees, au pairs and volunteers. Instead, it could seek a reasonable compromise, aiming to harmonise only certain aspects where there is a stronger case for common rules. The impact assessment report makes such a stronger case on two issues: where EU programmes aim to facilitate admission, and as regards the exploitation of au pairs and trainees (which also amounts to unfair competition as regards the domestic workforce).

So the future Directive could have binding rules as regards migrants who enter on the basis of an EU programme, but leave Member States discretion as regards these groups of migrants otherwise. And it could apply binding rules as regards equal treatment and transparency to those at risk of exploitation, leaving Member States the option to apply the other rules in the Directive to such groups. The objective of harmonisation could be preserved for the longer term, by means of a standstill clause, obliging Member States which currently apply EU law to these groups to keep doing so, while giving other Member States a renewable derogation. But if it is not possible to agree any binding rules on any aspect of the immigration of these groups of migrants, it would be better to drop them from the text of the Directive altogether, since the EU should avoid pointless legislation.

In return for not insisting on binding rules governing all aspects of the migration of other categories, the EP should demand improvements in the Council’s position on students and researchers, focussing on the changes that are most likely to attract them to the EU: better job-search rules, more employment prospects for students and researchers’ family members, less complex rules on mobility, and speedier decision-making. Requiring all Member States to permit in-country applications for researchers would surely help too. And since the EU also needs to attract researchers who fall outside the scope of the planned rules (because they don’t have a formal agreement with a research institution), the institutions could agree a political commitment to address this issue as part of the forthcoming revision of the EU’s Blue Card Directive, which concerns the admission of highly-skilled workers (on the reform of that Directive, see here). That review could also address what happens at the end of the job-search process for former students and researchers.

Speaking of the Blue Card Directive, it may also be useful to consider whether researchers and students might be more attracted to the EU if there were a similarly emblematic form of residence permit for them: an ‘EU Student Card’ and an ‘EU Researcher Card’, for instance. (I wouldn’t suggest using other colours for this purpose, since the existence of an ‘EU Red Card’ and an ‘EU Yellow Card’ would confuse football fans no end). While this might focus greater attention on the EU as a possible destination for researchers, marketing should not, of course, come at the expense of the substance of the rules concerned.  The EU has an opportunity to develop its immigration regime in the near future to ensure that the people it wants to attract here actually decide to come. It would be foolish to miss this opportunity again.   

Barnard & Peers: chapter 26
Photo credit:

Monday 19 January 2015

From bad to worse? On the Commission and the Council’s rule of law initiatives



Dimitry Kochenov* and Laurent Pech**

*Professor of EU Constitutional Law at the University of Groningen and Visiting Professor of Private Law at the Universita degli studi di Torino

**Jean Monnet Professor of EU Public Law, Head of the Law and Politics Department at Middlesex University London


The rule of law is one of the fundamental values on which the EU is founded according to Article 2 TEU. Faced with a rising number of ‘rule of law crises’ in a number of EU countries, the Commission adopted a new ‘pre-Article 7’ procedure last March in order to address any instance where there is a evidence of a systemic threat to the rule of law. Having criticised the Commission’s initiative primarily on the (unconvincing) ground that it would breach the principle of conferral which governs the allocation of powers between the EU and its Member States, the Council proposed its own solution: a rule of law dialogue between national governments and to be held once a year in Brussels.


Both initiatives, and in particular, the Council’s, appear grossly inadequate to tackle the problem of ‘rule of law backsliding post EU accession’ to quote Frans Timmermans, the First Vice-President of the Commission in charge inter alia of the Rule of Law.


Let us begin with the Commission’s proposal. The rationale underlying its new mechanism is that the current EU legal framework is ill designed when it comes to addressing internal, systemic threats to this principle and more generally, to EU values. The former President of the European Commission himself called in 2013 for a better developed set of instruments that would fill the space that exists at present between the Commission’s infringement powers laid down in Articles 258–260 TFEU, and the so-called ‘nuclear option’ (suspension of a Member State’s EU membership) laid down in Article 7 TEU. Both procedures suffer indeed from a number of procedural and substantive shortcomings, with the consequence that Article 7 TEU has never been triggered whereas the Commission’s infringement powers have proved ineffective to remedy systemic violations of EU values.


Numerous proposals were made like prior to the publication of the Commission’s Communication last March. These proposals would appear however to have been were found too ‘radical’ for the Commission which decided instead to put forward an eminently ‘light touch’ mechanism (previous analysis by Steve Peers is available here). This new rule of law mechanism builds on and complements an already existing – albeit never used – procedure, the ‘nuclear option’ referred above and on the basis of which the Council may theoretically suspend certain EU rights of the ‘guilty’ Member State such as voting rights.


In a nutshell, the Commission’s new mechanism takes the form of an early-warning tool to enable the Commission to enter into a structured dialogue with the Member State concerned to prevent the escalation of systemic threats to the rule of law preceding the eventual triggering of Article 7 TEU. This ‘pre-Article 7’ mechanism does not exclude a parallel recourse to the infringement procedure.


In practice, the Commission’s new rule of law mechanism rests on three main stages:


(1)    The Commission will first have to assess whether there are clear, preliminary indications of a systemic threat to the rule of law in a particular Member State and send a ‘rule of law opinion’ to the government of this Member State should it be of the opinion that there are;

(2)    Commission’s recommendation: In a situation where no appropriate actions are taken, a ‘rule of law recommendation’ may be addressed to the authorities of this country, with the option of including specific indications on ways and measures to resolve the situation within a prescribed deadline;

(3)    Finally, the Commission is supposed to monitor how the relevant Member State is implementing the recommendation mentioned above. Should there be no satisfactory implementation, the Commission would then have the possibility to trigger the application of Article 7 TEU.


The Commission’s new pre-Article 7 procedure is anything but revolutionary. In essence it merely requires any ‘suspected’ Member State to engage in a dialogue with no new automatic or direct legal consequences should the Member State fail to agree with any of the recommendations adopted by the Commission. Undoubtedly, Article 7(1) TEU already and necessarily implicitly empowers the Commission to investigate any potential risk of a serious breach of the EU’s values by giving it the competence to submit a reasoned proposal to the Council should the Commission be of the view that Article 7 TEU ought to be triggered on this basis. The criticism expressed by the Council’s Legal Service, which has criticised the Commission for overstepping its powers, would therefore appear particularly misplaced. The Commission’s framework is procedurally sound, no Treaty change is required and for the first time, a wide range of expert bodies is to be consulted: so far so good one may be tempted to say.


This bright picture however fades a great deal as soon as one focuses on the likely effectiveness of this new procedure, which is based on the presumption that a dialogue between the Commission and the Member State is bound to produce positive results. The validity of this presumption is highly questionable. Indeed, once we move towards really problematic cases, i.e. the countries where the ruling élite has made a conscious choice not to comply with EU values, engaging in a rule of law dialogue is unlikely to be fruitful. Worse still: the confidential nature of the whole discussion to be held between the Commission and the Member State under investigation will prevent a successful ‘name-and-shame’ environment from crystallising. The non-legally binding nature of the ‘rule of law recommendation’ to be addressed to the authorities of the country under scrutiny, and the non-automatic recourse to Article 7 TEU should the recalcitrant Member State fail to comply, further increase the likelihood of ineffective outcomes.


The Council’s negative response to the Commission’s proposal leaves one rather pessimistic about the chance of ever seeing the Commission activating its new rule of law framework. Indeed, rather then supporting the Commission’s rule of law framework, the Council decided instead to establish an annual rule of law dialogue to be based ‘on the principles of objectivity, non discrimination and equal treatment of all Member States’ and to be ‘conducted on a non partisan and evidence-based approach’. The Council’s response is as disappointing as it is unsurprising considering the reported unease of several national governments at the idea of letting any independent EU body looking into rule of law matters beyond the areas governed by EU law. The British government, for instance, has made clear its opposition to the Commission’s framework on three main grounds: It would be superfluous to the extent that the European Council and the Council of Europe would already monitor rule of law compliance within EU Member States; it would undermine the role of the Member States within the Council of the EU and finally, that the Commission and the Council would have already been successful through informal dialogue and lobbying in addressing in recent concerns on the rule of law in Member States.


Suffice it to refer to recent events in Hungary to understand that this last point is rather ludicrous. The point about the possible duplication of existing mechanisms is similarly unconvincing. To put it concisely, if multiple bodies gather data and monitor some specific aspects of EU Member States practice in relation to the rule of law, democracy and human rights, no European body currently subjects EU countries to a specific, country-based and permanent monitoring and assessment of their adherence to the rule of law broadly understood (for an overview of existing monitoring mechanisms within the Council of Europe, the EU and the UN, see this very useful report from the Bingham Centre for the Rule of Law). For instance, the Council of Europe’s Venice Commission, whose work is unanimously praised, is primarily a consultative body. In the end, the criticism directed at the Commission’s proposal essentially stems from the reluctance of some national governments, especially those whose rule of law records are highly questionable, to accept any potential effective form of supranational monitoring which could result in the adoption of legally binding recommendations and/or sanctions. 


Viewed in this light, it is hardly surprising that while the Commission’s proposal suffers from many a flaw, the Council’s response goes nowhere near enough what is required to address current challenges. The latest buzzwords are used to hide an unwillingness to meaningfully act. For instance, the Council calls for an evidence-based approach but what will this mean in practice and who will in charge of collecting this evidence and analysing it? Similarly, the dialogue is supposed to take place in the Council ‘following an inclusive approach’, the substance of which is nowhere explained. More fundamentally, the Council is seeking to use a soft instrument, which has regularly been criticised precisely for its ineffectiveness when used by the EU to promote its values abroad. To put it concisely, the EU has set up close to forty ‘human rights dialogues’ with third countries but evidence of substantial and concrete achievements is thin on the ground. One would have hoped a different, stricter approach for any Member State whose authorities have made a conscious political choice of undermining EU values.


To conclude, the Commission and the Council’s initiatives may leave one deeply disappointed considering the serious nature of the internal challenges faced on the rule of law front. When comparing the two initiatives, one may however argue that the Commission’s is much less half-hearted and, thus, at least less counter-productive, than the Council’s, which does not simply represent the triumph of empty rhetoric over genuine action but also unfortunately undermines the future legitimacy of any Commission attempt to trigger its new pre-Article 7 procedure. For a more detailed analysis, we would refer interested readers to our forthcoming Schuman Foundation policy paper, which is due to be published this spring in both English and French.


Barnard & Peers: chapter 9
Photo credit: The Economist

Saturday 17 January 2015

Is the ECB’s OMT programme legal? The Advocate-General’s Opinion in Gauweiler

Alicia Hinarejos, Downing College, University of Cambridge; author of The Euro Area Crisis in Constitutional Perspective (OUP)

On the 14th of January, AG Cruz Villalon delivered his Opinion in Gauweiler (C-62/14) on the legality of the Outright Monetary Transactions (OMT) scheme of the European Central Bank (ECB). In his view, the OMT programme is, in principle, in compliance with the Treaties, as long as certain conditions are observed if the programme is activated in the future. The case has important implications for the constitutional framework of EMU and the role of the ECB, but also for the relationship between the German Constitutional Court (the Bundesverfassungsgericht) and the Court of Justice of the EU. Indeed, this is the first time that the Bundesverfassungsgericht has ever asked the Court of Justice for a preliminary ruling.


The ECB is in charge of conducting monetary policy for the euro area and its role is very narrowly defined in the Treaties. This role, however, has evolved and expanded substantially in recent years, as the ECB has announced or adopted various ‘non-standard’ measures in response to the euro area sovereign debt crisis. The OMT programme is one of these measures: it was announced in September 2012 in a press release and, so far, it has never been used.

The idea is that the ECB will buy government bonds from euro countries in trouble, i.e., when nobody else buys these bonds, or their yield is becoming so high that the Member State will not be able to cover interest payments on newly issued bonds, thus having no more access to credit and risking default. Crucially, the Treaty prohibits the ECB from acquiring government bonds directly (Art 123 TFEU) as this would amount to monetary financing, or becoming a direct lender of last resort to a Member State. Instead, the ECB would buy government bonds in the secondary market—that is, from an institution that has bought these bonds first from a Member State—rather than from a Member State directly. While the ECB had already done this before, with the OMT programme there would be a formal element of conditionality as well, as the Member State in question would need to obtain financial assistance from the European Stability Mechanism or the EFSF and comply with its conditions (i.e. macroeconomic reforms negotiated between the Member State and the troika: the Commission, the ECB and the IMF).

The applicants before the German Court argued that the ECB had overstepped its Treaty role by creating a programme that should be viewed as a tool of economic, not fiscal, policy; it was also alleged that the programme violated the prohibition of monetary financing. In an exercise of ultra vires review, the German Constitutional Court’s preliminary response was to consider the OMT programme illegal under EU law. For the first time ever, the national court then referred the case to the CJEU. In the referring court’s view, the Court of Justice may either declare the OMT scheme contrary to the EU Treaties, or provide a more limited interpretation of the programme that is in accordance with the Treaties. The German Court provided certain indications as to what those limits should be.

The case is sensitive for various reasons: although not yet used, the mere announcement of the OMT scheme played an important role in getting the euro area out of the acute phase of the crisis, and offers a credible defence against similar future scenarios. A declaration of illegality, or the placing of substantive limits on the programme, may jeopardise post-crisis recovery. Additionally, the reference is the first ever submitted by the German Constitutional Court, and its tone is quite bold; there is clear potential for conflict between the two courts, with consequences unknown for EMU (on this aspect of the case, see this earlier blog post). Moreover, the case touches on the nature and legitimacy of the role of the ECB as an independent expert, and on the dichotomy between the original, rule-based conception of EMU and the evolving, more policy-oriented EMU that rose out of the crisis.

The AG Opinion

AG Cruz Villalon delivered a carefully argued Opinion that, first, acknowledged and unpacked the significance of the exchange for the dialogue between the German Constitutional Court and the Court of Justice, and, second, considered all concerns put forward by the national court. In doing so, the AG came to the conclusion that the ECB is free to create and implement a scheme like OMT, as long as it abides by certain limits in doing so. Crucially, these limits are far more permissive than those suggested by the German Court.

(1)    The relationship between the two courts

The German Constitutional Court has been very vocal on the question of limits to European integration, vowing to exercise its ‘emergency jurisdiction’ in different scenarios in the past: in order to protect human rights enshrined in the German Basic Law (Solange saga), to ensure that EU action is not ultra vires, i.e. does not go beyond what is allowed in the Treaties (Maastricht, Honeywell), and to protect Germany’s constitutional identity, which has so far included a particular conception of democratic legitimacy and the protection of national parliamentary powers (Lisbon and various post-crisis decisions).

In Gauweiler, the case at stake, the German Court exercised its ultra vires jurisdiction, coming to the interim conclusion that the ECB’s actions went beyond the powers given to it in the Treaties. Following its undertaking in Honeywell, the German Court referred the matter to the Court of Justice before reaching a final decision. Space precludes more careful consideration of this point, but it should be noted that ultra vires and constitutional identity intertwine in this case: first, because the German court used its conception of democratic legitimacy to ‘sharpen’ its ultra vires jurisdiction, in the sense that, for the first time, it was citizens’ right to vote that gave them standing to challenge EU action for going beyond EU primary law. And second, because the German Court went on to suggest that further review on the basis of constitutional identity would or may follow a Court of Justice’s decision that the OMT scheme is not in fact ultra vires: whether the OMT scheme could violate the constitutional identity of the Basic Law would depend on the Court of Justice’s specific interpretation of the scheme in conformity with EU primary law.

AG Cruz Villalon engaged with the case-law of the referring court on limits to European integration and acknowledged the background and significance of a reference that was worded in very bold (some would say almost aggressive) terms by the German court. Indeed, this discussion may be seen as the most diplomatic part of the Opinion.

The AG emphasized the ‘functional difficulty’ of the reference: in short, that the Court of Justice should not issue a preliminary ruling requested by a national court if that request ‘already includes, intrinsically or conceptually, the possibility that it will in fact depart from the answer received’ [36]. This, the AG continues, is not the intended or proper use of the preliminary ruling procedure. But was this such a situation? In this respect, it is problematic that the German Court may still conduct its own and independent ‘identity review’ after the Court of Justice has conducted its ultra vires review. Nevertheless, the AG relied on the principle of sincere cooperation to argue that trust is required in this situation: the Court of Justice should provide a constructive ruling, ‘on the basis of a particular assumption regarding the ultimate fate of its answer’ [66]. So there we have it: since both courts are under a duty to cooperate sincerely and to trust each other, the Court of Justice should give the requested ruling to the German court, trusting that the latter will, in turn, ‘do the right thing’. The AG was very clear as to what he considered that to be: ‘it seems to me an all but impossible task to preserve this Union, as we know it today, if it is to be made subject to an absolute reservation, ill-defined and virtually at the discretion of each of the Member States, which takes the form of a category described as ‘constitutional identity’. That is particularly the case if that ‘constitutional identity’ is stated to be different from the ‘national identity’ referred to in Article 4(2) TEU.’

(2)    The legality of the OMT scheme

The German court’s concerns regarding the legality of the OMT programme can be summarized as follows: first, the programme is a measure of economic, not monetary policy, and as such beyond the remit of the ECB. Second, a programme of this kind amounts to monetary financing of a Member State, which Art 123 TFEU prohibits. It would allow the ECB to become lender of last resort to a country in financial difficulties, and it would transform EMU into a transfer union—something not foreseen in the current Treaties.

Is it monetary policy?

The AG started by considering the nature of the OMT scheme as a measure of monetary or economic policy. The applicants had argued that the scheme should be classified as an economic policy measure with the aim of saving the euro by changing certain flaws in the design of monetary union, i.e. by pooling the debt of euro countries. They also emphasized the effects of the attached conditionality on Member States’ economic policies. All this, they argued, placed the OMT scheme beyond the merely supporting role that the ECB may have in economic policy, according to the Treaties. The German Constitutional Court agreed, based on various features of the OMT scheme: its conditionality and parallelism with ESM and EFSF financial assistance programmes (as well as its ability to circumvent them) and its selectivity (in that OMT bond-buying would only apply to select countries, whereas measures of monetary policy typically apply to the whole currency area).

The ECB, on the other hand, argued that the aim of the scheme ‘is not to facilitate the financing conditions of certain Member States, or to determine their economic policies, but rather to ‘unblock’ the ECB’s monetary policy transmission channels’ [104]. In other words, the crisis was making it impossible for the ECB to pursue monetary policy through the usual channels. The proposed bond-buying would ensure that credit conditions return to normality, and that the ECB is able to conduct its monetary policy again. Additionally, the ECB argued that the element of conditionality was necessary to ensure that the OMT scheme would not interfere with the programme of macroeconomic reform agreed between the ESM and the Member State in receipt of financial assistance.

The AG started by considering that it is within the ECB’s considerable discretion to adopt ‘non-conventional’ measures of monetary policy in exceptional circumstances. He accepted that it was the ECB’s intention to pursue monetary policy when announcing the OMT scheme and then proceeded to analyse whether the features of the OMT programme bore out this initial aim. After addressing each of the German court’s arguments, it came to the conclusion that the OMT scheme was indeed a measure of monetary policy—with one caveat: the AG saw a problem in the fact that the ECB made bond-buying through the OMT scheme conditional on the Member State’s compliance with a programme of macroeconomic reform adopted within the framework of the ESM or EFSF, and the fact that the ECB plays a very active role in the negotiating and monitoring of this programme with the Member State. This double role of the ECB (first within a framework for financial assistance which constitutes economic policy, according to Pringle, and then in its bond-buying role within the OMT) would tip the OMT scheme beyond the boundaries of the ECB’s powers: monetary policy with, at most, a supporting role in economic policy. The AG thus considered that, if the OMT were to be activated, the ECB would have to distance itself from the Troika and the monitoring of the conditionality for financial assistance immediately.

Is it proportionate?

Once the AG was generally satisfied as to the monetary nature of the OMT scheme, he reviewed its proportionality; the fact that this was a non-conventional use of competence made the proportionality assessment the more essential.

The OMT programme is an incomplete measure (as not all its features were specified in the ECB press release, and the programme has never been implemented). The AG considered that the programme’s basic features were known and could be put through an initial proportionality assessment, but that a full review of proportionality will only be possible once or if the OMT programme is ever fully regulated. The result of that initial proportionality assessment was positive: the basic configuration of the OMT programme passed the tests of suitability, necessity (the AG considered that the limitations suggested by the referring court would likely render the programme ineffective) and proportionality stricto sensu. The broad discretion granted to the ECB had a bearing on the application of the proportionality test. In sum, the programme was considered proportionate in principle, subject to the ECB complying with the requirements of proportionality (among them the duty to give reasons) if the programme is ever implemented.

Is it against the prohibition on monetary financing?

Once the nature of the OMT programme had been discussed, the Opinion turned to the possible circumvention of the prohibition on monetary financing of Member States, which is a further manifestation of one of the principles underlying EMU, namely fiscal discipline. While the Treaty makes it illegal for the ECB to buy government bonds directly from a Member State, the referring court argued that, although OMT bond-buying would take place in the secondary market, this amounted to a circumvention of the same rule. This circumvention would undermine fiscal discipline and would make certain Member States responsible, ultimately, for the debts of others, which is banned by Article 125 TFEU.

The AG considered that the prohibition of monetary financing (as a manifestation of fiscal discipline) was one of the features of the constitutional framework of EMU that contributes to the attainment of a higher objective, the financial stability of the monetary union (Pringle). Exceptions to this prohibition must thus be interpreted restrictively, and a formalistic approach must be avoided: the focus must be on the substance of the measure, and not on whether the bond-buying occurs directly or in the secondary market.

The referring court had identified various technical features of the OMT scheme as running counter to this prohibition: the ECB’s lack of preferential creditor status and waiver of rights, its exposure to excessive risk, the disruptive effects of holding the bonds until maturity, the fact that bond-buying in the secondary market would take place on a large scale and only a short time after their issue (making it too similar in its effects to buying bonds directly from the state) and that the ECB’s action would encourage new investors to buy newly issued bonds. In very broad terms, the German court’s view was that these features amounted to a circumvention of the prohibition of monetary financing because, even though the bond-buying would take place in the secondary market, it would disrupt the market and undermine fiscal discipline to an intolerable degree.

The AG disagreed on all counts but one; after discussing the effects of each technical feature, he considered that they were not disruptive enough of the normal functioning of the market and of fiscal discipline to fall foul of the Treaty. Again, with one caveat: if the ECB ever implements the programme, the timing needs to allow for actual formation of a market price in respect of government bonds before the ECB buys them. If the ECB does that, according to the AG, the technical features of the OMT programme do not endanger fiscal discipline to a disproportionate degree, and as such they do not have the potential to make Member States responsible for each other’s debts or turn EMU into a transfer union.

Final Remarks

The AG Opinion in Gauweiler is thoughtful and carefully argued. His discussion of the German court’s case-law and the problematic of the reference is measured, while still seeking to protect certain elements of the Court’s jurisdiction that he considers essential to the integrity of the EU legal system. It will be interesting to see how the Court of Justice handles the matter in its decision but, just as importantly, how the German Constitutional Court reacts to the latter.

The Opinion is less diplomatic when it comes to the legality of the OMT scheme: it rejects almost all concerns put forward by the referring court, and it does so from a particular conception of the independence of the ECB and the role of courts in controlling its activities. In this regard, the Opinion can be said to continue in the Pringle vein of ratifying the move from a rules-based EMU to a policy-based one in the wake of the crisis. Yet despite the wide margin of discretion enjoyed by the ECB, the Court has a crucial role to play in protecting the constitutional framework of EMU and of the Union. In his Opinion, the AG discharges this task by grounding an important part of the analysis on the technical features of the OMT and their effects: this is particularly clear when it comes to the question of whether the programme is compatible with the prohibition of monetary financing, where the discussion turns on technical matters rather than on more abstract ones such as the nature of EMU, its evolution, and the role of solidarity within its constitutional framework. While this may seem like a shame, it is also understandable: this broader debate is of paramount importance, but the Court (or any court) may not be the most suitable forum for it.

[Update: the Court gave its ruling in June 2015; see analysis here.] 

Barnard & Peers: chapter 19