Sunday 28 June 2015

The law of Grexit: What does EU law say about leaving economic and monetary union?

Steve Peers 

A Greek referendum on whether to accept its creditors’ offer is currently scheduled for next week. It’s not clear at this point whether the Greek voters’ refusal to accept the offer would necessarily lead to Greece leaving the EU or EMU, or at least defaulting on its debts. In fact, it is not clear what would happen if Greek voters decided to accept the offer, since it was still under the process of negotiation when the referendum was announced, and may no longer be on the table at the time of the referendum.

However, since a wide range of outcomes are possible, it’s useful at this stage to look at the legal framework for departure from economic and monetary union (EMU) – and in particular whether Greece would have to leave the EU if it left the single currency. (See also my previous blog posts, before and after the last Greek election, and Ioannis Glinavos’ recent analysis of whether Greece could be forced out of the euro).

The starting point is that the EU Treaties contain detailed rules on signing up to the euro, which apply to every Member State except Denmark and the UK. Those countries have special protocols giving them an opt-out from the obligation to join EMU that applies to all other Member States. (I’ll say that again, more clearly, for the benefit of those who claim otherwise: there is absolutely no way that the UK can be required to sign up to the single currency. That would not change in any way if British voters decided that the UK should stay in the EU).

But there are no explicit rules whatsoever on a Member State leaving the euro, either of its own volition or unwillingly, at the behest of other Member States and/or the European Central Bank (ECB).  There’s an obvious reason for this: the drafters of the Maastricht Treaty wanted to ensure that monetary union went ahead, and express rules on leaving EMU would have destabilised it from the outset. Put simply, legally speaking, Greece can’t directly jump or be pushed from the single currency.

In practice, though, its continued existence in the single currency could be made very difficult, as Ioannis Glinavos pointed out, either by the ECB restricting or ending emergency assistance (ELA) to Greece, or by the ECB limiting or removing Greek access to payment systems. It’s possible that any such moves would be legally challenged by the Greek government, and perhaps by other litigants too. It could be argued that they are in breach of EU monetary law as such, and/or that they breach an implied rule that Member States cannot be forced out of monetary union.

But let’s imagine that some sequence of events leads to Greek departure from the official legal framework for EMU nonetheless. This could lead to the fully-fledged introduction of a national currency (the ‘New Drachma’, or somesuch). It could instead lead to some informal link with the single currency – for instance a Greek ‘version’ of the euro, or the use of the euro as Greek’s official currency in practice without participating in the legal framework of EMU. Several countries outside the EU (such as Montenegro) take the latter approach. None of these actions are legal (for a Member State) as a matter of EU law.

For that matter, the less extreme possibility of Greece defaulting on Greek debts without leaving EMU (if that were feasible in practice) is not provided for in the Treaties either. Moreover, other Member States and the EU institutions are arguably legally obliged to refuse debt relief for Greece, in accordance with the Treaties’ no bail-out rule: as the CJEU said in Pringle, this rule allows Member States to loan money to Greece in return for conditions and an appropriate rate of interest. But they cannot simply assume responsibility for Greek government debts. Forgiving those debts would have the de facto result of assuming them – although it might be possibly argued that the letter (but surely not the spirit) of EMU law would allow this as long as the Greek debts were not formally transferred to the EU institutions or Member States. It might also be argued that a Greek default on such debt would be a situation of force majeure, which could be accepted by creditors without this amounting to a breach of the no bail-out rule.

However, the no bail-out rule does not apply to the private sector, which explains the ‘haircuts’ already imposed on private banks, or to international bodies or third States. So Greece could default on its loans to the IMF without infringing the no bail-out rule (although that would surely breach some other legal rule). Thanks to the gods of irony, the IMF is the biggest supporter of Greek debt relief. And equally, without infringing that rule, Greece could refuse to pay back any loans that Putin might be foolish enough to give it.

Of course, the reason we got to this position in the first place was a series of legal breaches: Greece joined the single currency on the basis of allegedly inaccurate economic data (for the debate on that issue, see here), and was not punished (as EU law provides for) when it started to run debts and deficits well above the legal limits of EU law.

So if Greece does leave the EMU framework, and/or default on its debts in violation of EU law, is it obliged to leave the EU, as some have suggested? On the face of it, it’s certainly illegal for a Member State to leave EMU unless it also leaves the EU. But having said that, there would still be no legal obligation for Greece to leave the EU if it defaulted or left EMU.

Why is that? The main legal reason is that the Treaties have a specific legal regime on withdrawing from the EU: Article 50 TEU, as discussed in detail here. Article 50 says that a Member State ‘may decide to withdraw from the Union, in accordance with its own constitutional requirements’. This is manifestly a voluntary choice. There are no rules in the Treaty stating that a Member State ‘shall’ withdraw from the Union in any particular circumstances.

Nor is it possible to throw a Member State out. Article 7 TEU allows a Member State to be suspended for breaching key principles such as human rights, democracy and the rule of law. But there is no provision allowing a Member State to be fully expelled from the Union against its will.

So implicitly but necessarily, the Treaties rule out any expulsion from the EU and any requirement to leave it, in any circumstances.  But the Treaty drafters didn’t provide for States to leave EMU and/or default on debts either. If those States can’t be forced to leave the EU in such circumstances, what is the legal way forward?

Solutions to the tragedy

Classical Greek tragedies often ended with a ‘deus ex machina’ (‘god out of the box’). The playwright had manoeuvred the characters into an impossible situation, and the only way to resolve the plot was by the introduction of a radically new plot element – a god or goddess who could use his or her divine powers to resolve all of the problems which the characters faced. The normal rules of narrative are suspended.

In my view, this is where we stand with Greek participation in the EU’s single currency. Whether or not Greece stays in EMU, a new approach to the legal framework is necessary to try and address the Greek position.

I see four main possibilities. First of all, the Treaties could be amended to try to regulate the situation, if necessary with some degree of retroactivity. There could, for instance, be a new general power for the Eurozone States in the Council and/or the European Council to adopt measures to address the legal consequences of Greece departing EMU. Legally, this is the tidiest solution; but politically, it’s the most difficult one, since the Greek issues would get bound up with the British ones. It’s possible that the Treaty amendment process would fail due to issues related to Greece, rather than the UK – or the other way around.

Secondly, it could be argued that the implied powers of the EU (most obviously, Article 352 of the TFEU) could be used to address the situation. This is a difficult argument since the Treaty drafters considered EMU to be ‘irrevocable’. However, the CJEU has taken a generous approach to measures aimed at saving EMU that many people believed were clearly ruled out: financial assistance in Pringle, and the ECB’s bond purchasing programme in Gauweiler (discussed by Alicia Hinarejos here). It might equally take a generous approach to the legality of any measures aiming to clean up the enormous mess that a ‘Grexit’ would make.

Thirdly, some Greek law-makers have suggested that the Greek debts might be illegal, on the basis of a theory of ‘odious debts’ that violate human rights. As noted above, though, the CJEU has insisted on the conditionality of financial assistance, and it has also repeatedly refused to answer questions from national courts about the legality of those conditions. So at first sight, it looks difficult for this argument to succeed as a matter of EU law, although the Court has not ruled on this issue as such yet.

Fourthly, there’s a novel argument that I haven’t seen suggested before: Greek participation in the euro was invalid in the first place, because of the allegedly inaccurate economic statistics used at the time. The CJEU could declare in the same ruling that all of the legal commitments relating to Greek participation in EMU in the past remain legal, so as not to disturb legal certainty (there’s plenty of precedent for CJEU rulings like that). There are two possible variations here: a) if Greece is still participating in EMU, its participation must be retained for the same reasons of legal certainty; or b) if Greece has left EMU, its departure is legal because the original participation was invalid.

But in either case, a crucial exception to the ‘legal certainty’ rule can justify debt relief for Greece. It’s arguable that due to the essential illegality of the legal framework in which Greek debts were incurred, the no bail-out rule did not fully apply, leaving the creditors and Greece free to negotiate a realistic amount of debt relief. (True, the no bail-out rule does apply to non-Eurozone States too; but Greece borrowed far more than it would have done due to its illegal participation in the euro). If Greece has left the euro already, it could in future benefit from the slightly different regime for financial assistance to non-Eurozone States.

Although Greece would still be formally required to try to join the single currency in future, the EU tends not to pressure countries (like Sweden) which have no real intention of joining. Realistically, no one would pressure it to join for a very long time.

All of these solutions provide, in one way or another, that ‘it was all a dream’: either the debt or the euro participation never existed in the first place, or the Treaty or EU legislation retroactively apply to address the issues, or the Treaty means something quite different from what it was generally thought to mean. It is always preferable to avoid such an approach to the law, but it’s hard to see how any other type of solution could work in this case. Legally, simply put: Greece allegedly should not have joined the euro; it should not have been allowed to run up huge debts; it cannot leave EMU; and it cannot be forced to leave the EU. Economically and politically: Greeks have suffered more than enough; Greece can never pay its accumulated debts while taking austerity measures which depress its economy; but taxpayers of other Eurozone States understandably would like to see their money back.  

These illegalities and economic and political conflicts cannot be resolved within the current framework, so we need to revise it radically. Of the suggestions considered here, the fourth solution has the most appeal: it is consistent not only with the classical tradition of Greek tragedy, but disturbs the current legal framework as little as possible while offering solutions (a fully legal Grexit, effective debt relief) that aim to resolve the situation as best it can be managed. It is impossible to find any solution that would satisfy every legitimate demand, but in my view this approach is the least bad alternative.

Barnard & Peers: chapter 19

Cartoon: Peter Schrank, Independent on Sunday 


  1. Hmm....interesting analysis

    A bit disappointing though in that it doesn't address the legal tender aspect of the EU treaties and regulations.

    For instance Article 128 TFEU states:

    "Article 128

    (ex Article 106 TEC)

    1. The European Central Bank shall have the exclusive right to authorise the issue of euro banknotes within the Union. The European Central Bank and the national central banks may issue such notes. The banknotes issued by the European Central Bank and the national central banks shall be the only such notes to have the status of legal tender within the Union.

    2. Member States may issue euro coins subject to approval by the European Central Bank of the volume of the issue. The Council, on a proposal from the Commission and after consulting the European Parliament and the European Central Bank, may adopt measures to harmonise the denominations and technical specifications of all coins intended for circulation to the extent necessary to permit their smooth circulation within the Union."

    And this article is supported by "Council Regulation (EC) No 974/98 of 3 May 1998 on the introduction of the euro" and "Council Regulation (EC) No 2169/2005 of 21 December 2005 amending Regulation (EC) No 974/98 on the introduction of the euro"

    The regulation in question (as amended in 2005) states in Articles 10 and 15:

    "Article 10

    With effect from the respective cash changeover dates, the ECB and the central banks of the participating Member States [note that the "participating Member States" are listed in the Annex to the regulation and includes Greece] shall put into circulation banknotes denominated in euro in the participating Member States. Without prejudice to Article 15, these banknotes denominated in euro shall be the only banknotes which have the status of legal tender in participating Member States."

    "Article 15

    1. Banknotes and coins denominated in a national currency unit as referred to in Article 6(1) shall remain legal tender within their territorial limits until six months after the end of the transitional period at the latest; this period may be shortened by national law.

    2. Each participating Member State may, for a period of up to six months after the end of the transitional period, lay down rules for the use of the banknotes and coins denominated in its national currency unit as referred to in Article 6(1) and take any measures necessary to facilitate their withdrawal."

    So based on the treaties and regulations, in Greece ONLY the euro can be legal tender and therefore any attempt by the Greek government to issue a Greek Euro or New Drachma as legal tender either alongside the euro or in replacement of the euro would be illegal without changes to the necessary treaty articles and secondary legislation. Just as you noted that the Greek government and othe litigants might well legally challenge ECB moves to restrict or end the ELA to Greece or limiting or removing Greek access to payment systems, so too would other litigants (likely private citizens whose bank accounts, savings and salaries would be affected) could and would likely legally challenge the Greek government over any attempt to make any currency other than the euro as some form of legal tender.

    I've imagined that the most likely "Grexit" scenario would not be a true Grexit, but rather a limbo situation whereby the Greek government is forced to issue IOUs/promissory notes/long-term bonds (something which even Ireland did in 2009-2010 in relation to Anglo-Irish Bank and had to convert them into long term bonds in 2013) to pay for stuff including salaries.

    1. Good point, but it's another argument in favour of the basic rule that Greece can't leave or indirectly be forced out of the single currency. In fact I would base my argument on the Treaty obligation to join the euro and fix rates irrevocably, since that rule is in primary law, not secondary law. And your approach is still trumped if it was illegal for Greece to join the euro, as I suggest might be the case.

    2. Indeed as I note in the final (third) part of my reply, the prospect that Greece was never legally in the euro in the first case would make for a very interesting and indeed landmark CJEU case.

      That said, the arguments that Greece can't leave or be indirectly forced out is based on both primary and secondary law and for either to occur it would require changes to both (either the treaties and regulations are amended OR Greece leaves the EU entirely to ensure that it can have a different currency as legal tender)

    3. Johnny Chapman9 July 2015 at 00:21

      The fact that only the Euro can be legal tender in Greece does not prohibit Greece from introducing a new Drachma but not legislating for it to be legal tender. There is nothing uncommon about this. The pound sterling is not legal tender in Scotland and neither are Scottish banknotes. Scotland has no currency as it's legal tender.

    4. That's an interesting suggestion, Johnny, I was wondering if something like that could be a fudge in the event of a semi-Grexit. In normal times I doubt that the CJEU would be willing to accept that sort of thing but it is very flexible when it comes to monetary union issues.

    5. That suggestion by Johnny is not correct though and what he is suggesting is actually quite uncommon. The pound sterling IS legal tender in Scotland. It has always been so.

      What is NOT legal tender in Scotland are the banknotes themselves (whether Scottish banknotes or Bank of England banknotes - see: At one point Bank of England notes up to the value of one pound were legal tender in Scotland but then the BoE discontinued those notes:

      However ALL coins produced by the Royal Mint ARE legal tender in Scotland and indeed throughout the entire United Kingdom and coins of a value of £1 or greater are legal tender for payments for ANY amount with only coins of 50p or less have limits on what amount of debt they can be used collectively to pay (see: and indeed are the ONLY legal tender in Scotland.

  2. continuing.....

    I think the current debate in the media has overlooked this possibility save for a few articles like this one:

    In it the author notes that nearly everyone believes the sequence of events will go like this:

    1) Removal of ELA from the banking system
    2) Immediate bank run and collapse
    3) The only way to get the banking system up and running again is to print a local currency
    4) And they’re gone.

    But he no longer thinks that reasoning is valid based on the events in Cyprus and proposes this sequence instead:

    1) ELA is not completely removed, but restricted to an amount somewhat less than the liquidity needs of the banking system
    2) Acceleration of deposit flight, beginnings of a bank run
    3) Controls placed limiting permitted deposit withdrawals
    4) Compulsory writedown of bonds and/or large deposits of the banking system to rebuild capital
    5) As a necessary consequence of above, capital controls.
    6) Country is still in the Euro, but with an extremely inconveniently arranged banking system and a lot of angry depositors.

    He describes less as an ejection/exit and more like putting a country in a penalty box/sin bin.

    He also notes that "the consequences for Greece would be nearly as bad as those of actual Euro exit, but sufficiently less awful to prevent them deciding to unilaterally leave. The consequences for the rest of the Euro area would be greatly mitigated. The consequences for the ECB and Commission would be … they’d be really really great, because this way, they get to square the circle of not having any member states leave the Euro, but also not getting into a situation where they find themselves having to fund the entire Greek budget indefinitely while always losing the battle on conditionality..

    ..My guess is that relative to Cyprus, the losses to bondholders and depositors would be smaller (because there isn’t a real problem of solvency at the Greek banks at present, and a government default, although it would cause losses, would not render the system totally insolvent), but the capital controls could be deeper and longer lasting..."

    And it would get around the tricky business of getting the two-thirds of the ECB Governing Council (comprised of 6 board members and the heads of the 19 eurozone national banks) required to agree to a complete cutting off of the ELA to Greece.

    We've already seen where Cyprus had to put in place capital controls for 2013-2015 but there was no "Cyprexit", despite this being a key assumption for the "Grexit" scenario (that capital controls will somehow mean that Greece must exit the single currency).

  3. Final continuation…

    With the ECB recently announcing that it is only going to maintain the ELA to Greek banks at current levels this has the effect of restricting the ELA need to being somewhat less than the needs of the Greek banking system without removing it altogether. The upshot is that capital controls for Greece are more likely down the road. And as with Cyprus in 2013-2015 (and Ireland in 2009/2010) there would likely be no need to amend the treaties to reflect the situation since Greece would still legally be in the euro and the euro would remain the only legal tender in Greece, even with capital controls and if persons got some kind of promissory notes from the government that everyone in the media would be screaming are the "new drachma" and a sign that Grexit has finally arrived.

    A Greek default on debt also seems likely, but this should not be surprising as in the 1990s (as this article notes:, the consensus then was that euro area members that got into trouble would not be bailed out and that this WOULD lead to sovereign defaults and that markets would price the sovereign debt of euro countries differently based on their public finances and the possibility of default.

    That article incidentally also notes that the popular conception of the no bail-out rule is incorrect when one reads the relevant articles (Articles 122 and 125 TFEU) which, respectively, explicitly allows the EU to grant financial assistance in periods of 'severe difficulties caused by natural disasters or exceptional occurrences beyond its control' and merely stated that the Union and its member states “shall not be liable for or assume the commitments” of other countries.

    So what is often called the "no bail-out rule" or clause apparently is more like a "no assumption of other's debts rule" or "no debt laibility transfer/assumption rule". It would indeed allow for bailouts.

    The idea that Greece was never legally in the eurozone in the first place and thus the euro isn't legal tender there anyway is an interesting one (though that scenario shouldn't be termed "Grexit" but more something like "Grellegality"). That would certainly make for an interesting CJEU case and the outcomes and ramifications could be quite wide reaching.

    1. On Article 125, the CJEU said in the Pringle case that the 'no bail-out' rule doesn't prevent loans being given as long there are conditions and interest attached. Already I think many say this breached the spirit of the Treaty rule but in my view it is consistent with the letter of it. The CJEU ruling in Gauweiler on the ECB purchases of bonds on the secondary market (getting around the Treaty ban on purchases on the primary market) is seen similarly. So as I suggest we could follow this chain of reasoning to say that Article 125 does already allow for debt relief, just not the formal transfer of one Member State's debt to another one. But the political resistance to debt relief in the creditor countries (although it's hard to see how all the debt can ever be paid back, and a managed debt relief would surely be tidier and collect more of the money than a default) is so entrenched that a CJEU ruling on 'Grellegality' (no-one will thank you for coining that word!) like the one I describe might be necessary to ensure that it takes place.

    2. Very good points. Ruling that Greece was never legally in the eurozone in the first place and therefore many of the debts incurred after Greece joined the euro are now questionable would quite likely spur Greece's various creditors into trying to secure a realistic debt relief deal with the country.

      And nobody need thank me for "Grellegality" ;-)

      I actually don't like the terms "Grexit" or "Graccident" anyway, so if "Grellegality" can help discourage the use of those terms I'm all for it.

  4. This is really off the wall, and I can tell I am in the company of better financial and legal minds than myself, but I want to put to you my fairy godmother proposal which makes sense to me, mainly because, at times, i reside on a different planet:

    Given that the ECB has the sole right to issue Euros and authorise the issue of Euro coins, what if there were a phantom benefactor, European or otherwise, Nation or organisation, that chose to do one of the following:

    1. Loan sufficient funds, in Euros, on generous conditions, to Greece specifically earmarked as a stimulus package. Repayments could be specifically tied, in the first instance, to the tax receipts from Greece's best performing industries.

    2. Purchase, in Euros, a bundle of Greece's State assets at an inflated price, but which, nevertheless, would be a reflection of their projected future value. Once again, the lion's share to comprise a stimulus package. Repayment could be as above, but perhaps the lender could expect an on-going return from Greece's taxable income, after a prescribed length of time, for a prescribed period.

    3. By all counts, Greece has already undertaken most of the structural reforms necessary, and seems to have also undertaken a seismic shift in the national mindset; however, a measure of 'austerity' may still be necessary. By this, I certainly don't mean a blind adherence to a policy of cuts in spending which reduce Greece to 'third-world' status, but an approach which seeks to eliminate waste and State funded largesse, while preserving the basics of a dignified life for the most vulnerable in Greece.

    Along with these, the lender/purchaser and/or borrower/vender could negotiate an appropriate (independent?) monitoring agency to oversee arrangements, report on progress and offer remedial solutions as necessary. Now, in my world, while this may be seen to cut across the grain of what many would regard as economic orthodoxy, it has the merit of being in the "win-win" mould of economic theory rather than that of the current predominant paradigm of "winners and losers". This current approach does not even deserve to be called "utilitarian".