Talk:Enlargement of the eurozone/Archive 2

Archive 1

Archive 2

Archive 3

ERM-II applied retroactively ?

To be frank, I think the linked story is just a Polish wishful bogus. The ERM-2 criteria has never been applied retroactively, during the past 20 years. Plus its not at all reliable to believe it will, just because a Polish political adviser think such a new criteria interpretation is possible without changing the treaty. This is much more doubtful speculation, rather than assuming and reporting at Wikipedia how the ERM-2 criteria has been evaluated previously in all past 20 years, while assuming a status quo of course will apply throughout the next 2 years. If you want to report that a ERM-2 criteria change is being considered, we need additional references backing-up such a story. It is not sufficient just to back this up with your linked article, which only reflect a launched proposal by a loud-thinking creative Polish political adviser. Danish Expert (talk) 06:46, 15 August 2014 (UTC)

In anyway, your previous speculation that Romania has to join ERM-II on 16 May 2016 at the latest to join the Eurozone on 1 January 2019 is wrong. First, Slovenia is a clear-cut example of how a country does not need to have been part of the ERM-II for two years when the convergence report is adopted, to fulfill the exchange rate criterium. Slovenia joined ERM-II on 28 June 2004 and the Eurozone on 1 January 2007. Its economic convergence was assessed on 16 May 2006, i.e., before two years in the ERM-II had elapsed. Second, we do not know when the convergence report will be published in 2018. Hence, 16 May 2016 is a completely incorrect date, and it demonstrates how wrong it can be when one does original research. --Glentamara (talk) 07:28, 15 August 2014 (UTC)

Thanks for highlighting the 2006-example, which I previously somehow had overlooked. In regards of the Romania topic, I admit my edit to insert the "target date imply ERM-II entry before 16 May 2016" was a bit too hasty and thoughtless of me, and accept it was removed again. As you correctly point out, we can not be certain if the two-year reference period will begin+end on 23 April or 30 April or 16 May (as they indeed flex this even for the ordinary scheduled convergence reports). And your 2006-example is a proof, that the 2-year ERM-II membership requirement does not need to be fulfilled in the convergence report - but instead is something that only needs to be complied with when the ECOFIN council subsequently make their binding euro adoption decision. As the ECOFIN counsil meets once every month, it is impossible to define a stone-written date, although I think we can conclude for all countries that it can be no later than November (due to the concerned state needing minimum 1 month for dual price display, and the minting process plus pre-distribution to the stores). In order for my disputed Romania line to be correct, it should instead just say: "Adoption target imply ERM-II entry at the latest in 2016". Because your provided 2006-example is still no proof for the Polish claim, that a state no longer needs to be an ERM-II member for at least 2-years, in order to adopt the euro. In the 2006 convergence report they wrote this exact evaluation phrase (both for Slovenia and Lithuania):

• Slovenia entered ERM II on 28 June 2004 and has so far spent 22 months in ERM II. By the time of a possible Council decision in July 2006, the tolar will have participated in ERM II for more than 24 months. Slovenia fulfills the exchange rate criterion.

The 2004 convergence report indeed proof, that the ERM-II criteria was not met in 2004, at a time when both countries only had been ERM-II members for 2 months. Therefore we can only claim an existing precedence, that the EC+ECB somehow require 24-months ERM-II membership, and the 2006-example is only a proof that they (somehow surprisingly, although we can note this is also the case for the EDP-criteria) stretched their handling of the ERM-II criteria into basically two subcriteria, with different reference periods. The first subcriteria is "currency stability in the 2-year reference period (16 may 2004 to 15 May 2006)". The second subcriteria is "24 months ERM-II membership (apparently measured with the cut-off date being equal to the date of the ECOFIN council giving their final approval at the meeting where they fix the conversion rate). I was not aware of this special 2006-decision, before we started this discussion. So thanks again. However, still this appliance of a second reference period for the second ERM-II subcriteria, does not mean the EC+ECB has decided completely to convert the 24-month ERM-II membership requirement into just a 1-day ERM-II membership requirement applying retroactively for two years (as suggested by the Polish political adviser). We have no previous precedence for such a thing.

Finally, instead of displaying that Romania "possibly join ERM-II in 2014-15" based on a speculative forecast reference, I think it would be less speculative to replace it with my (now rephrased) more observational and neutral line: "Adoption target imply ERM-II entry at the latest in 2016". However, if the 2014-15 ERM-II target period indeed can be backed up by additional references, then I will of course accept this info as preferable to keep. If you still think "at the latest in 2016" can not be displayed without a reference, then I would actually prefer we replace "possibly 2014-15" with the phrase "not set". Danish Expert (talk) 10:11, 15 August 2014 (UTC)

I was not the person who referenced the Polish adviser. Actually I more or less agree with you in that case. I think it is ok to write "at the latest in 2016". What I was opposed to was the previous formulation where "16 May 2016" was mentioned without any reference. Best regards, --Glentamara (talk) 10:25, 15 August 2014 (UTC)

It may well be wishful thinking; I'm not convinced it would be possible to be approved with a 1-day ERM II membership either. (Note however that the article doesn't say say that the Polish think that the criteria should be reinterpreted as you suggest. The sentence "the rules do allow ERM-2 to be applied retroactively" was written by a Reuters journalist.) However, I'm also not convinced that a state couldn't be approved with <2 years of membership. As far as I am aware there is no evidence to support your claim that 2-years as a ERM-II member is an absolute necessary condition and has been enforced as such in the past. If you read the treaty, nowhere does it say states must be ERM II members for 2 years. What it actually requires is "the observance of the normal fluctuation margins provided for by the exchange-rate mechanism of the European Monetary System, for at least two years, without devaluing against the euro" As written, this could be satisfied by unilaterally observing the exchange bounds set by ERM II without actually being an ERM II member. Conversely, just being an ERM II member isn't enough, you need to actually keep your exchange rate stable for 24 months.

If you read the 2006 convergence report on Lithuania, prior to discussing the 24 months in the ERM II, it says

"The litas has been at the ERM II central rate for the period of two years covered by this assessment" and then "Lithuania fulfils the exchange rate criterion."

You used the last sentence to argue that the point was that Lithuania would fulfil the criteria by the time the Council decision was made because by then they would have been members for 2 years. However, the sentence is written in the present tense not the future tense. It says they currently fulfil the criteria, which would not be true under your interpretation. Lithuania could have withdrawn from the ERM II the next day. The way I read that passage, they are arguing that they satisfy the criteria because their exchange rate has been stable for more than 2 years.

Are there any examples of a state which otherwise satisfied the criteria, and which has maintained a narrow exchange rate band for 2 years, but which was deemed to be non-compliant simply because they weren't an ERM II member for long enough? (In the 2004 report Slovenia and Lithuania had other issues holding them back.) If not, then there is no evidence to support your interpretation. We can't claim it is necessary if there is no evidence or precedence that it is actually necessary. Yes I don't have any proof that it isn't necessairy, but you don't have any proof to the contrary either. The reality is we don't know how such a situation would be handled since it has never arose as far as I'm aware. Hence we are in the domain of OR and speculation. Given that the speculation adds little of value to the article, it is better to refrain for adding it.

Personally, I have an awfully hard time believing that Bulgaria would be rejected if they otherwise complied with the criteria just because they hadn't formally been a ERM II member for a full two years, even though they have pegged their currency to the euro for more than a decade. TDL (talk) 19:08, 15 August 2014 (UTC)

Just two comments. First, I think it is quite clear from both the treaty text (don't miss the additional Protocol No 13 regarding the convergence criteria) and other sources that the convergence criterium on exchange rate stability means that you - in principle - need to participate in ERM-II for at least two years without any severe deviations. With that said, I also think - and this is pure speculation - that the Commission is a bit flexible and focus more on the purpose of the criteria than merely on the actual criteria themselves, especially when it comes to the exchange rate stability. Second, what you write about Bulgaria is interesting because if you only need to be within ±15 percent fluctuation band and you don't need to participate in ERM-II itself, then Bulgaria already fulfills all convergence criteria. Even more interesting, this would mean that several other member states fulfill the exchange rate criteria. This would imply that also Sweden fulfills all convergence criteria and could eventually end up in the Eurozone unintentionally. --Glentamara (talk) 20:32, 15 August 2014 (UTC)

As I said above, I don't share your interpretation of the treaty text that 2 years of ERM II membership is required. A140 and P13 only require the "observance of the normal fluctuation margins provided for by the exchange-rate mechanism" and "respected the normal fluctuation margins provided for by the exchange-rate mechanism" not actual ERM II membership. A literal reading of this says that only 2 years of adhering exchange rate stability is required. Perhaps this interpretation is wrong, but I think it is definitely vague enough to be open to different interpretations.

However, the treaty text does require ERM II membership ("and of its participation in the exchange-rate mechanism"), so Sweden and Romania couldn't qualify without some time in ERM II. Also, there are additional requirements beyond the convergence criteria for euro adoption under A131. As per [1], Sweden can't join because "it has not made the necessary changes to its central bank legislation". The 2014 convergence report says: "Bulgarian law does not comply with all the requirements for central bank independence, the monetary financing prohibition, and legal integration into the Eurosystem. Bulgaria is an EU Member State with a derogation and must therefore comply with all adaptation requirements ...".

Anyways, we agree on the main point, that the Commission can (and has) chosen to implement the 2 year ERM II membership requirement with flexibility. If we can't say with 100% certainty that they would be denied entry if they didn't have 2 years of ERM II membership, we shouldn't say that they need to join by such and such a date. TDL (talk) 00:13, 16 August 2014 (UTC)

OK, after digging through some old convergence reports I have found an example of retroactive application of ERM membership criteria in the 1998 report. Italy joined ERM in 25 November 1996 while Finland did on 14 October 1996. In March 1998, with both currencies in the ERM <24 months, the EC concluded that both had "not experienced severe tensions during the review period and has thus, in the view of the Commission, displayed sufficient stability in the last two years." And later "By the time of the decision by the Council in May 1998, the Italian lira and Finnish markka will have participated in the ERM for about 17 and 18 1/2 months, respectively ... For the purpose of this examination, the stability of the Italian lira and Finnish markka is assessed as if the two currencies had participated in the ERM with their current central rates for the full two-year period." The ECB said "in an ex post assessment, the [lira/Finnish markka] has been broadly stable over the reference period as a whole." Even when Council approved, they were still well short of the 24 months. TDL (talk) 00:13, 16 August 2014 (UTC)

I don't want to prolong this discussion too much, but as you can see on the website of the Commission [2] they write that one of the convergence criteria is:

"Exchange-rate stability, through participation in the Exchange Rate Mechanism (ERM II) for at least two years without strong deviations from the ERM II central rate;"

Even though it is interesting to discuss the treaty texts, we shouldn't interpret the treaties ourselves since that is original research. It is clear from the website of the Commission that the criterium requires that you have participated in ERM-II for at least two years. That said, I don't think that the criteria are set in stone. We both agree that there is some flexibility in the application of the criteria. Nonetheless, the main rule is that the currency should participate in ERM-II for at least two years. --Glentamara (talk) 07:48, 16 August 2014 (UTC)

A possible explanation for the two 1998 incidents might be, that the Council did not irrevocable fix the euro conversion rates on their May 1998 meeting - where the country selection for participation was decided. According to the joint communication, the irrevocable fix only took place on 31 December 1998. As I replied yesterday, it appear (based on the 2004+2006 convergence reports) as EC+ECB measure the 2-year ERM requirement shall only be met once the "euro exhange rates gets irrevocably fixed by the ECOFIN Council". According to the European Council's May 1998 meeting conclusions, they indeed concluded that Finland and Italy had fulfilled all four criteria, highlighting they complied with the +/-15% currency stability criteria in the two-year reference period, and did not explicitly mention the second 2-year ERM membership criteria. However, this doesn't mean it doesn't exist, as the conclusions did not go into details - but only presented the main findings. I agree with Glentamara, however, that we might still have some interpretation uncertainty attached to whether or not it perhaps also could be good enough just to comply with the 2-year ERM membership criteria "ahead of the euro adoption day", and not just "ahead of the ECOFIN councils final approval and irrevocable fix of the euro conversion rate". Reason why I insist we also have the 2-year ERM criteria, is not only because of the protocol line cited above by Glentamara, but mainly because of this direct cite from the 2004 convergence report page 19:

• "None of the countries examined in this year’s Convergence Report participated in ERM II for the full reference period from October 2002 to September 2004. Consequently, they did not participate for at least two years before the convergence examination as laid down in Article 121 of the Treaty."

Contrary to your reply above, I think the 2004 convergence reports actually proof that both Slovenia+Lithuania complied with the +/-15% currency stability criteria in the two-year reference period from Oct 2002 - Sep 2004. The way I read the reports, the reason why the EC+ECB overall deemed their currency not met the overall "currency criteria" was because they only had been member of ERM-II for 3 months on the evaluation time and for 6 months ahead of the aspired euro adoption day 1 Jan 2005. If we look back on all 19 euro adoptions, it apply for all 19 cases they had been ERM members minimum 2 years ahead of their euro adoption. Danish Expert (talk) 09:54, 16 August 2014 (UTC)

Sorry for counter-arguing myself here, but after taking a look at the Cypriot convergence evaluation in December 2006, it appear they complied with all economic criteria (despite only being ERM members for 20 months ahead of 1 Jan 2007), and only failed to comply with the legal criteria. On the other hand, it can also be argued the imagined euro adoption date at this late evaluation time had to be 1 Jan 2008 (and not 1 Jan 2007) to allow for the practical preparations to take place, and in that case it was a certainty the country would have ERM partcipation for 32 months. This criteria is really messed-up! From a logical approach towards this, I would say it make most sense if they do not have a second 2-year ERM membership subcriteria, but only one criteria (as suggested by TDL) measuring if all ERM members have maximum +/-15% currency exchange rate fluctuations towards the euro in the past two years. To have the second subcriteria simply does not make sense. The more I think of it, I start to feel TDL is correct, that it is just a coincidence when we have all 19 out of 19 euro members with minimum 2-years of ERM-participation prior of their euro adoption. In that case however, the Romanian reference indicating possible ERM-membership in 2014-15 does not make sense, because they would then risk being forced to join the euro ahead of their target date (provided they comply with all economic criteria upon the automatic convergence checks performed twice a year). Or would it be legally possible for them to comply with all criteria in May 2016, except for the "legal compliance criteria", which they then deliberately postpone for years, so that it is only met in May 2018? Or would the show of such a deliberately delaying intend, be considered a breach to their commitments in the EU accession treaty?

To conclude, I no longer think we should change anything in the article as it stands now, in regards of Romania. For the sake of reflecting some needed "compliance evaluation clarity" in the article, we however need to update the euro convergence table not only with a "legislative compliance" column (with input values yes/no), but also change the ERM requirement from "2 years" to just being "yes". Danish Expert (talk) 10:54, 16 August 2014 (UTC)

I wouldn't read too much into the 2014-15 date. It could just be the source speculating, as they don't quote anyone. My understanding (based on reading past convergence reports) is that a state could block their adoption of the euro by not bringing their legislation into compliance. A140 on the convergence criteria says "These reports shall include an examination of the compatibility between the national legislation of each of these Member States, including the statutes of its national central bank, and Articles 130 and 131 and the Statute of the ESCB and of the ECB.", so this seems to be a fundamental part of the eligibility. How the EU would react if they didn't bring their legislation into compliance we can only guess, but I don't see that really being any different than refusing to join ERM II.

As for updating the table, I agree with adding a "legislative compliance" column, but I'm not sure what to do with ERM II. Though I personally don't think that 2 years of ERM is really required, I'm just not sure if the sourcing is sufficient to actually say that in article space. It is all very vague, which is probably intentional. The more vague the treaty text, the more leeway there is for decisions based on politics rather than on facts. But it also means that we shouldn't make clear statements when in reality the situation isn't clear. Perhaps it is simplest to leave it at 2 years, with a footnote explaining that states have been admitted in the past if they maintained stable exchange rates but had <24 months in ERM II. TDL (talk) 18:20, 16 August 2014 (UTC)

Once again, it is crystal clear from the website of the Commission that one of the criteria is to participate in the ERM-II for at least two years. The website [3] reads:

Exchange-rate stability, through participation in the Exchange Rate Mechanism (ERM II) for at least two years without strong deviations from the ERM II central rate;

If you claim that the Commission is wrong, then you have to provide a source proving this and that is more authoritative than the Commission, which I doubt you can find. Interpretations of the treaties that cannot be confirmed through secondary sources are nothing else than original research, that is, something we should not be doing on Wikipedia. That said, we could of course state in the article that there have been exceptions in the history, which show that there is some flexibility in the application of the criteria. This does not, however, change the fact that the exchange rate criteria implies a main rule that means that you need to participate in ERM-II for at least two years. --Glentamara (talk) 19:12, 16 August 2014 (UTC)

@TDL: Yep, I was aware of A140, but still a little uncertain in regards if a state - for situations where they comply 100% with all 5 enconomic convergence criteria, legally can be forced by the EC also to comply with the legal convergence criteria - per the legal obligations signed in their EU accession treaty. On the line that: "Hey guys you actually comply - then we hand you a deadline on X.W.ZZZZ to fix your legal issues - which eventually if not met could end up as a no-treaty compliance issue for the CJEU to rule and fine you upon." Currently, I tip to believe a country indeed can escape their legal euro adoption obligation (without risking CJEU fines), by simply refusing to work actively to ensure legal convergence on the needed areas. Looking back on the past history, we however (as far as I remember) never had such a situation playing out for years. We had Cyprus in December 2006 complying with all economic criteria but without legal compliance, and then they fixed their remaining legal issues just 3 months later (and most other states also fixed their legal issues in the same year as they experienced 100% economic convergence). The more I think on this issue, I start to think you are correct in your assessment that the EC can do nothing to force a state, to deliver a removal of the remaining legal issues preventing the needed "legal convergence compliance" to happen. If I was Romania, I however would not wish to test this out for multiple years.

For the table changes, the addition of a legal convergence column is also supported by the fact, that the ECB convergence reports actually both use the term "economic convergence" and "legal convergence", which approve we also can speak of a legal convergence criteria being yes/no to the question: Do its national laws comply with the ECB statute, ESCB statute and articles 130+131 of the Treaty on the Functioning of the European Union ?. For the sake of avoiding the addition of too many columns, I prefer we only add 1 overall legal compliance column, instead of 3 subcriteria columns for the legal convergence.

On the Romania topic, giving it a second thought, I think the "possibly 2014-15" data indeed is speculative + incorrect, and propose we remove this by simply adding the phrase "not set". It does not make any sense to believe that Romania would have any desire to join ERM2 more than maximum 2 years ahead of their official euro adoption target. When they just chose to have a 1 Jan 2019 euro adoption target, then we should only expect them to join ERM at the earliest in 2016, and if not being afraid for "legal court cases" (and criteria interpretation), they will probably delay it to 2018. The point is, that Romania has received advise by multiple economic experts, telling them they should not seek a fixed ERM rate (and thus euro adoption), before the state has reached "real economic convergence" on other relevant parameters (GDP per capita, labor wages, labor productivity) which are not part of the EC's convergence criteria and which they only expect to be met (with gap to the EU-average reaching acceptable levels) minimum 5 years down the road, then it does not make any sense why Romania anyway impatiently would desire several years in advance to their euro adoption, now to remove their "floating exchange rate" advantage from the equation. Currently, the reason why economic experts advice Romania to delay their euro adoption, is exactly that their currency need some more years to benefit from the depreciation opportunity (ensuring they do not loose competitiveness through their "economic catching up process", which historically has been proven to be associated with higher levels of inflation and loss of competitiveness if you fix your exchange rate too soon. So the EIU forecast about 2014-15 ERM membership, does not make any sense. Moreover, after a Romanian google search, I found no other sources mentioning the existence of such a "2014-15 ERM membership" target. On the contrary all sources say, that Romania does not have a target for ERM membership, and in fact (funny enough) they all claim Romania only will become an ERM member 2 years ahead of its euro adoption, for the only reason to meet the requirement to comply with the "minimum 2 years ERM2-membership" criteria: [4][5][6][7][8]. Danish Expert (talk) 19:54, 16 August 2014 (UTC)

@Glentamara: In regards of the previous mis-interpretation about the existence of a "2-year ERM2-membership criteria", which not only can be found repeated by our Wikipedia article but also by this 2008 OFCE article and all the Romanian criteria assessment articles, we most likely can blame this solely on the confusing EC-website line about this. I think we should correct the Wikipedia article for this issue, in the attempt to stop the rolling wave of misinterpretation of what this criteria actually is about. To be on the save side, we can always attach an observational note explaining the previous cases where the EC+ECB approved euro adoptions for states with less than 2-years ERM2-membership, and noting its just a coincidence that 19 out of 19 states had minimum 2-years of ERM2-membership ahead of their euro adoption date. Your proposal that our article shall continue reflecting "minimum 2-year ERM2-membership is required" and not clarified to say that "ERM2-membership is required along with a past record of minimum 2-years currency stability within +/-15% from the parity exchange rate" is not something I prefer. I prefer it the other way around, but will leave it for TDL (and perhaps the opinion of others to decide which approach is the best). In all circumstances, no matter which approach we pick, I agree with you that we need to attach the observational note. Danish Expert (talk) 19:54, 16 August 2014 (UTC)

Just to clarify, I support any clarifications of the criteria, but it should still be clear that the main rule is that you need to participate in ERM-II for at least two years, even though there have been some flexibility in the past. I am not opposed to any proposal which means that we mention the previous examples of countries which have been allowed to join the Eurozone even though they haven't participated for more than two years at the time the convergence report was adopted. I also support that we add the legal criterium in the table. However, even though there is a legal criterium, it does not mean that it forms part of the convergence criteria. Protocol No 13, whose title is "Convergence criteria" does not mention the legal criterium. It is also clear from A140 that the convergence criteria and the criteria that need to be fulfilled to adopt the euro are not the same. The convergence criteria are rather a subset of the criteria that need to be fulfilled to adopt the euro. --Glentamara (talk) 20:15, 16 August 2014 (UTC)

@Glentamara: Technically speaking, you are correct that the treaty and its protocol (both originally written in 1991) do not call it "legal convergence". As I pointed out, the ECB convergence reports, however, clearly have coined the term "legal convergence". Defining this as a term and criteria in the assessment report, in addition to the "economic convergence criteria" (which the protocol plus past EMI+ECB evaluation practice, together has defined how to be understood). What I attempt to say, is just that the ECB convergence reports have paved the way for us also to speak about existence of a "legal convergence criteria", because they have coined the term "legal convergence" and explained on which points a "legal convergence" is required - for this additional criteria to be fully met. Danish Expert (talk) 21:42, 16 August 2014 (UTC)

Agreed with DE. The terminology "legal convergence" is regularly used in the convergence reports so I don't see why would shouldn't utilize their terminology, given that they are the authority on the matter. TDL (talk) 00:01, 17 August 2014 (UTC)

Again, your interpretation is not unanimously supported by sources. Cherrypicking one source to claim something is "crystal clear", and ignoring all those that contradict your POV, does not reflect WP:NPOV. My argument has been backed by several secondary sources which suggest that different interprets are possible. We have convergence reports by both the Commission and ECB, and decisions by Council, which concluded that the exchange-rate criteria has been met without 2 years as an ERM member. The treaty is a legally binding document, so if the rule in the treaty was "you must be an ERM member for two years", as you suggest, the Commission can't just ignore the treaty, make up their own rules, and say "oh well, close enough". That would be illegal. If you claim that the Commission/ECB/Council were all wrong in their legal documents and conclusions, you need more compelling evidence then an informal website. We also have discussions in the convergence reports which make it clear that there is disagreement on how the criteria should be interpreted. In fact, there is a minority opinion that not only is 2-years of ERM not required, but that NO ERM membership is required:

Here the ECB says they have not taken a formal position on how the criteria should be interpreted, that there is a majority that believes some ERM membership is required (but no mention of 2 years) and that there is a minority opinion that NO ERM membership period is required at all:

• "... the EMI does not at this state consider it appropriate to give a precise ex ante operational content to the measurement of exchange rate stability according to Article 109j of the Treaty, which could also be mechanically be applied to forthcoming periods. Regarding the Treaty provisions of membership of the ERM, there is a strong majority position within the EMI Council that the requirement of ERM membership applies. A minority takes the view that exchange rate stability based on sustainable underlying economic fundamentals is more important than the institutional setting within which stability is achieved."

Here the ECB states that there is a minority opinion that ERM membership is irrelevant to convergence:

• "In the opinion of two members of the Council, the EMI should not base its Report on a particular interpretation of ERM participation, but should submit a thorough Report for the EU Council to assess de facto sustained exchange rate stability in the context of the economic fundamentals in all Member States. These members underline that the exchange rate criterion can no longer be applied straightforwardly because the form of the ERM which existed when the Treaty was prepared came to an end in August 1993 with the decision to widen the permitted fluctuation margins from ±2.25% to ±15%"

When sources are vague and experts disagree on how the treaty should be interpreted, we need to present both perspectives. To be accurate and reflect WP:NPOV we need to carefully say something like:

"The exchange rate criteria requires "the observance of the normal fluctuation margins provided for by the exchange-rate mechanism of the European Monetary System, for at least two years, without devaluing against the euro". Normally this is satisfied by membership within the ERM II for two years, however in the past states have been deemed to have satisfied this criteria with shorter membership periods, provided that their exchange rate has been stable for the full 24 months."

This way we don't take a position one way or the other on what the "real" criteria is, since sources demonstrate that this is open to interpretation. TDL (talk) 23:42, 16 August 2014 (UTC)

Arbitrary break

I've found an excellent source which describes the Commission's interpretation and application of the exchange rate criteria in more detail. See Appendix D3 of the 2002 Report. They say that "the conditions to be respected in fulfilling the exchange rate criterion" are:

1. "Participation in the ERM II at the time of the assessment is mandatory."
2. "Participation in the ERM II for at least two years is expected, although exchange rate stability during a period of non-participation before entering ERM II can be taken into account."

This is in agreement with what I was arguing above: 2 years in ERM is "expected" though not required. Perhaps if we use this "expected" wording everyone will be happy? Something along the lines:

"The exchange rate criteria requires "the observance of the normal fluctuation margins provided for by the exchange-rate mechanism of the European Monetary System, for at least two years, without devaluing against the euro". The Commission has interpreted this to mean that a state must be an ERM II member, and two years of participation is "expected" though not necessary as the "exchange rate stability during a period of non-participation before entering ERM II can be taken into account."

Any objections to this? TDL (talk) 06:16, 17 August 2014 (UTC)

It's ok for me to use the formulation "expected". Regarding the legal criterium, I'm not opposed to the idea to mention it in the article as an additional criterium you have to fulfill to adopt the euro. We could also include it in the table. However, we should not say that it is one of the four convergence criteria. Even though you can speak of legal convergence, it doesn't mean that it is one of the four convergence criteria. The treaty article 140 and the protocol No 13 are clear on this point. They say that the Commission/ECB should evaluate the legal aspects and then it says that it should also evaluate four economic criteria, after which the four convergence criteria are listed. These are then specified in the protocol on convergence criteria. Note that the legal criterium is not mentioned there. There are only four convergence criteria. This is also explicitly confirmed by the website of the Commission [9], the ECB [10] and other official websites [11]. They don't list the legal criterium as one of the convergence criteria. --Glentamara (talk) 07:04, 17 August 2014 (UTC)

@TDL: Thanks for digging the new ref up. Think we are getting close to consensus. The ref is great, and should definitely be added. However, I tend to see it as a confirmation of the fact, that they evaluate "exhange rate convergence" as two subcriteria, as we also currently reflect in template:Euro convergence criteria. So for the template column, I propose for the first subcriteria (as formulated by your point 1 above), that we replace the required ERM reference value "min. 2 years" with "yes" - and thus stop to note the dates for ERM-2 membership and only input a Yes/No. For the second "exchange rate convergence" subcriteria, the required reference value is still "maximum +/-15%" for the reference period always being 2-years backwards from the time of examination. I think it is most correct if we implement it this way, and then add the explaining note why the ERM2 reference criteria is just "yes" and not "2 years". My argument for this position, is that its actually only the second subcriteria dealing with "2-year evaluation" (and that we since the EC-2002 comment has assessment precedence for ECB+EC basically having changed the phrase "can be taken into account" to "will be taken into account" - and then it makes sense to assume this will be continued). The first subcriteria just check if the member state was an ERM2-member as of the time of the assessment. I would prefer our template to reflect the story in this way (being identical to the past ECB evaluation practice). Danish Expert (talk) 07:16, 17 August 2014 (UTC)

Just found this very interesting ECB convergence policy statement issued in December 2003. I have copied 3 paragraphs being relevant to our current ERM2-criteria debate:

1. With regard to currency boards, the ECB does not consider them to be a substitute for participation in ERM II, implying that countries operating a currency board will be required to participate in ERM II for at least two years before the convergence assessment that is made before a country can finally adopt the euro. However, countries that operate a euro-based currency board deemed to be sustainable might not be required to go through a double regime shift, i.e. floating the currency within ERM II only to re-peg it to the euro at a later stage. Such countries may therefore participate in ERM II with a currency board as a unilateral commitment, enhancing the discipline within ERM II. However, the ECB has stressed that such an arrangement will be assessed on a case-by-case basis and that a common accord on the central parity against the euro will have to be reached.
2. The Treaty sets out the minimum period of participation in ERM II prior to the adoption of the euro.[note: See the third indent of Article 121(1). The fourth indent of Article 121(1), which states the interest rate criterion, also refers to participation in ERM II as follows: “the durability of convergence achieved by the Member State and of its participation in the exchange-rate mechanism of the European Monetary System being reflected in the long-term interest-rate levels”.] A protocol attached to the Treaty elaborates on the criterion for participating in the exchange rate mechanism, stipulating that a Member State must have respected the normal fluctuation margins provided for by the exchange rate mechanism without severe tensions for at least the last two years before the examination. In particular, the Member State must not have devalued its currency’s central rate against the euro on its own initiative within this period.[note: Protocol on the convergence criteria referred to in Article 121 of the Treaty establishing the European Community.] A minimum stay of two years in the mechanism prior to the convergence assessment leading to final adoption of the euro is therefore expected.
3. The assessment should be based on the principle of equal treatment with the current Member States. Therefore, no additional criteria for the adoption of the euro by the new Member States will be introduced, while at the same time there will be no relaxation of the criteria laid out in the Treaty, including the criteria concerning the sustainability of nominal convergence. With regard to exchange rate stability, the criterion refers to participation in ERM II for a period of at least two years prior to the convergence assessment without severe tensions, in particular without devaluing against the euro.

To be honest, I start to get a little dizzy. Seems like we have opened Pandora's box. The 2003-policy declaration by ECB now clearly support the argumentation posted by Glentamara, while the EC-2002 appendix support the argumentation by myself and TDL. Should we now use the EC-opinion or the ECB-opinion? Or perhaps a note reflecting both? The above first half of paragraph 3 (the equal treatment principle), is something which both ECB+EC have agreed and confirmed consistently throughout all convergence reports in 1996-2014, which is why I pay so much attention to their past "evaluation precedence". What they did in the past (in regards of interpretations on how the criteria mentioned in treaty+protocols should be understood), obligates them to do the exactly same in the future. Paragraph 1 was new to me, and might explain a few cases with less than 24 months ERM2-membership? As of 9 April 2003 this special paragraph 1 applied for the 4 countries: Bosnia and Herzegovina, Bulgaria, Estonia and Lithuania. To clarify this matter all to the bottom, I therefore think we also need to attach a rather long note including the observations in previous cases (i.e. 19 out of 19 states adopting the euro had participated minimum 2-years ahead of euro adoption, but among those X,Y,Z had not participated for a full 2-years ahead of the publication of their final approving EC+ECB convergence assessment reports, although all 19 had participated for a full 2-years upon the Council's date of irrevocably fixing of their euro exchange rate as the last step in their euro adoption approval process.) Can you support my proposal for a long note, and what should we do? Danish Expert (talk) 10:51, 17 August 2014 (UTC)

ERM membership data for eurozone states

For the purpose of getting a complete overview for our debate, I just worked-out this data table. It is not intended for the article, but only for the talkpage.

Country	ERM membership	Euro-Based Currency Board	Euro adoption	Time of ERM-membership (ahead of euro adoption)
Denmark	13 March 1979	-	-	-
Bulgaria	-	1 Jan 1999	-	-
Lithuania	28 Jun 2004	2 Feb 2002	1 Jan 2015	10.5 years
Greece	14 March 1998	-	1 Jan 2001	33 months (25m ahead of CR)
Slovenia	28 Jun 2004	-	1 Jan 2007	30 months (22m ahead of CR, 24m ahead of IF)
Cyprus	2 May 2005	-	1 Jan 2008	32 months (24m ahead of CR)
Malta	2 May 2005	-	1 Jan 2008	32 months (24m ahead of CR)
Slovakia	28 Nov 2005	-	1 Jan 2009	37 months
Estonia	28 Jun 2004	1 Jan 1999	1 Jan 2011	6.5 years
Latvia	2 May 2005	1 Jan 2005	1 Jan 2014	8.5 years
Belgium	13 March 1979	-	1 Jan 1999	20 years
Germany	13 March 1979	-	1 Jan 1999	20 years
Spain	19 June 1989	-	1 Jan 1999	9.5 years
France	13 March 1979	-	1 Jan 1999	20 years
Ireland	13 March 1979	-	1 Jan 1999	20 years
Italy	25 Nov 1996	-	1 Jan 1999	25 months (17m ahead of CR, 25m ahead of IF)
Luxembourg	13 March 1979	-	1 Jan 1999	20 years
Netherlands	13 March 1979	-	1 Jan 1999	20 years
Austria	9 January 1995	-	1 Jan 1999	4 years
Portugal	6 April 1992	-	1 Jan 1999	7 years
Finland	14 October 1996	-	1 Jan 1999	26 months (18m ahead of CR, 26m ahead of IF)

Among the 19 eurozone members, the following 3 were only barely meeting the 2yr EBCB/ERM-membership criteria: Finland, Italy, Slovenia. Danish Expert (talk) 13:15, 17 August 2014 (UTC) - updated table above to correct the mistake pointed out below: 14:05, 17 August 2014 (UTC)

Thank you for this useful table! However, Greece joined ERM on 14 March 1998 (ref: the French central bank), not on 31 December 1998. --Glentamara (talk) 13:35, 17 August 2014 (UTC)

Bytheway, the same source (from the French central bank) says that:

As regards the Italian lira and the Finish markka, which had participated in the ERM for just under two years at 2 May 1998, the Heads of State or Government stressed that, over the previous two years, these currencies had appreciated in the months preceding their joining the ERM, and had then participated without severe tension in the mechanism. Moreover, these currencies had participated in the ERM for over two years at January 1999

I think this summarizes the situation. The official convergence criterion means that you have to participate in ERM-II for at least two years, but both the Commission and the Council of the European Union (as well as the European Council) can in the end decide different, i.e., there is some flexibility for them to take into account the real situation. --Glentamara (talk) 13:45, 17 August 2014 (UTC)

Agree with you on this. I speculate, that either they have adopted an unwritten rounding up option, so that all countries complying with the EBCB/ERM-membership for more than 1.5 years ahead of the Convergence Report (CR) will be rounded up to 2-years compliance - if everything else looks fine. Or as I speculated earlier, they simply adopted the approach to measure the length of EBCB/ERM-membership - not upon the publication of the CR - but instead upon the Irrevocable Fix (IF) of its euro currency exchange rate by the Council. When the 3 states had their IF, they had been ERM members for respectively: Slovania=24m, Italy=25m and Finland=26m. I tend to believe it is the latter (IF cut-off date), that applies as the "extended option", simply because the 2006 report explicitly mentioned this date - concluding the 2yr membership would be reached by this IF date. To sum up, I now agree that we keep the official 2yr ERM-membership criteria. However, we still need to add a clarifying note, where we mention the "IF observation" and that if a state is ERM-member for less than 2yr it can still comply if it prior of its ERM-membership (and for the remaing part of the 24m period going backwards) had established an EBCB. This last observation is important, in particular for Bulgaria. Danish Expert (talk) 15:00, 17 August 2014 (UTC)

Basically, I agree with you, but I can't really find where ECB states that "EBCB states" (like Bulgaria) don't need to participate in ERM II for at least two years? Could you post the cited relevant text here again? Thank you, --Glentamara (talk) 15:34, 17 August 2014 (UTC)

I don't necessary see any contradiction between the ECB and EC opinions. Both state that 2 years is "expected", which does not mean required. Regardless, if the ECB and EC have interpreted the criteria differently that just emphasizes the fact that it is open to interpretation. I continue to believe that it is best to stick to the "expected" formulation as I proposed above, as this is what is used by both sources and it avoids taking a hard position one way or the other. I'm not opposed to keeping the "2 year ERM criteria" in the table though.

As for your suggestion about a note on the "IF observation", I'd be ok with that provided that we don't say they were allowed to join BECAUSE the two years passed before IF, since we don't know that to be the case. All we know is that this happened to be the case. It could be a coincidence, or it could be meaningful, we really don't know. TDL (talk) 23:29, 17 August 2014 (UTC)

@Glentamara: Paragraph 1 in my reply above from 10:51 17 August 2014 explains the matter, copied from the 2003 ECB convergence policy statement. However, upon my second read I can see, that I accidentally misread the second line. I first thought it was saying that ECB wouldn't require the regime shift from EBCB to ERM-II necessarily to happen 2 years ahead of CR. Now I can see, it only say that EBCB are not required to be dissolved - as part of a double regime shift - once the state enters into ERM-II, but on a case-by-case basis the EBCB can be approved by ECB to continue so that the state has both "ERM-II + EBCB" ahead of euro adoption. Sorry for misreading it. I agree with you now, that there are no special exemptions (or advantages) applying to states with an EBCB. Criterion wise they are treated equally with those who did not have any EBCB. Danish Expert (talk) 11:34, 18 August 2014 (UTC)

@TDL: We have consensus now for your "expected" line above, plus how to deal with it in the convergence table. The EC versus ECB concern I had, was because it seemed like the second bold line in paragraph 3 of my previous reply (With regard to exchange rate stability, the criterion refers to participation in ERM II for a period of at least two years prior to the convergence assessment without severe tensions, in particular without devaluing against the euro.) did not explicit mention the "although exchange rate stability during a period of non-participation before entering ERM II can be taken into account". You are right though, that the two sources not necessarily contradict each other, as both uses the term "2 year expected". I support we use the EC-formulation (and if you think its not contradicting, we could perhaps source it by both the EC-source and ECB-source, because they supplement each other with the ECB-source clarifying some other relevant aspects). Also agree with you, that it is best to describe the 3 incidents of "only 2 year ERM2-membership ahead of IF" as observations, and not a stone-written always applying measuring approach, meaning that other states might not benefit from the same exemption rule (because it might depend on additional subsub-criteria also to be in a positive territory, to grant such an exemption). I think its best, if you now write the note. If you do not add the IF observation straight away, then I will add it straight away, and then you are welcome to reformulate it without any discussion here. I think we now agree on all aspects. Danish Expert (talk) 12:04, 18 August 2014 (UTC)

OK, I've had a go at implementing this at Euro_convergence_criteria#Criteria. I'm not sure it's worth getting into all the caveats on this page though, since we really only offer a rough outline of the criteria. TDL (talk) 18:40, 19 August 2014 (UTC)

Think its OK for now. I noticed though (from your latest Czech revert), that you do not agree that we based on the previous EC+ECB assesment practice, should conclude that: "ERMII-membership is required for minimum 2 years ahead of euro adoption day, along with a measured currency stability existing in the 2yr reference period ahead of the publication of the Convergence Report". Multiple other sources have jumped to this conclusion. In example:

• ECB spokesperson Tina Zumer stating in 2013: If Croatia join ERMII in 2016 they can at the earliest have euro adoption on 1 Jan 2019.
• Polish Ministry of Finance stating in 2009: "A failure to meet the above-outlined conditions would delay the moment of entry to the ERM II beyond 2009Q2. Nevertheless, it would not exclude the possibility of adopting the euro in 2012, provided that the Polish zloty has stabilized before July 2009 and that assessment of the exchange rate criterion is based on the 24-month period beginning just before ERM II membership, as was already the case in other countries. With other prerequisites fulfilled, entering the ERM II in the second half of 2009 would then allow Poland to adopt the common currency in line with the "Road Map timetable" [targeting euro adoption on 1 January 2012]. Should that condition not be met either, euro area membership would have to be postponed beyond 2012."

I could easily find and add minimum 8 more examples of respectable persons/institutions having jumped to the same conclusion, that "euro adoption at the earliest can happen on the 1 January following the date when you have been an ERM2-member for 2 years." Looking back this was (as we already discussed) indeed true for the previous 19 out of 19 eurozone members. Yet you continue to insist, that we are not allowed to assume this is the case, in the same way the more than 10 other respectable sources assumed this is the case, simply because EC+ECB did not publish something "stone written" in regards of how they measure this criterion. Personally I would have preferred to copy the same criteria interpretation being used by all respectable persons/institutions (except from your funny little Polish journalist from Reuters, who posses a minority view on this issue), but of course attached with a note that this ("euro adoption at the earliest can happen on the 1 January following the date when you have been an ERM2-member for 2 years.") currently is the general prevailing criteria interpretation - and that EC+ECB had not officially clarified if it was true or not (despite some of their spokespersons using this interpretation while speaking to other journalists during various international euro adoption conferences). I sense the only thing that will change your opinion about this, would be if we can dig-up evidence of ECB/EC concluding in some of their previous assessment reports that "as X upon publication of the Convergence Report had only been a member for less than 1 year, X did not comply with the exchange rate criterion". If I can dig up such an example, would you then be ready to accept we add it as a part of the observational note, and on that basis accept we in the article adopt the same "criterion interpretation" as currently utilized by the vast majority of secondary sources? Danish Expert (talk) 13:46, 20 August 2014 (UTC)

If you want to say a specific date is the "earliest possible", then we need authoritative source which actually says "ERMII-membership is required for minimum 2 years ahead of euro adoption day" and it is impossible to join with <2 years. Neither the EC nor the ECB source says anything like that, or relate the 2-year period to euro adoption date. Yes some sources make that assumption, but others make other assumptions. Here is another example from the Bank of Romania which said they planned to join ERMII in 2010 or later, and adopt the euro as early as 2012. [12] That would imply a 2017 date for the Czech Republic rather than your claimed 2018. Are you 100% sure that if the Czech Republic joined the ERMII effective 1.1.2015, they would be denied entry if they otherwise qualified for 1.1.2017? What if they joined on 2.1.2015? We really don't know how such a situation would be handled, but I have a hard time believed that they would be told "nope, you were 24 hours too late, wait until next year." All the convergence criteria are applied with some flexibility, so I don't see why the Council couldn't (in principle) allow a state to join if they are short of the 24 months.

You have made sever hypothesis now on how the 2-year provision is enforced. First you assumed that two years pre-convergence report release date was necessary. Next you guesed that two years pre-council approval was required. Then you argued that two years pre-IF was required. Now you are suggesting two years pre-adoption is required. In the first two cases, your argument has been proven to be wrong. This demonstrates the danger of drawing conclusions from an argument based on a logical fallacy as you are: correlation does not imply causation. Yes in all cases to date the 2 years has passed prior to euro adoption, but that does not mean that Council is forbidden from allowing states that would not reach the two years prior to adoption to join. It could just be a coincidence. Saying something is "impossible" is a very strong statement which allows for no wiggle room. However, perhaps if we say something like "a decision by the Czech Republic to join ERM-II in mid 2015 would imply that 1 January 2018 would likely be the earliest possible euro adoption date" then this would be acceptable to both of us. TDL (talk) 19:36, 20 August 2014 (UTC)

There are authorative sources (Commission and ECB) that confirm that you must participate in ERM II for at least two years to fulfill the exchange rate stability convergence criteria. However, we need to distinguish between the convergence criteria, as outlined in the treaties and the attached protocols, on the first hand, and the decision the Council takes in accordance with article 140 TFEU, on the other hand. As TFEU-140 states, the Council shall make its decision on the basis of the convergence criteria, but there is certaintly room for political considerations. Hence, once again, as I have mentioned several times before in this discussion, there is some flexibility for the Council/Commission to make minor "exceptions". To stress the main point: we need to distinguish between the formal convergence criteria, on the first hand, and the practical rules used to let member states adopt the euro, on the other hand. They are not necessarily the same in practice. However, it is difficult for us to predict what political decisions the Council may take in the future. We only know for sure how the convergence criteria look like according to the primary law and we have to assume that the Council uses these criteria as its benchmark when it decides which member states should join the eurozone. --Glentamara (talk) 20:00, 20 August 2014 (UTC)

Partly agree and disagree. Treaty+Protocol criteria were written in 1991. EMI=ECB was subsequently tasked to further develop and define the criteria within the Treaty+Protocol framework. This mean, that the criteria has only been sharply defined by the developed ECB assessment practice crafted out in their Convergence Reports during the years (i.e. they developed the HICP statistic only in 1995 to measure the inflation criteria). As I mentioned earlier, we can be however always be absolutely certain about one thing (repeatedly confirmed by all convergence reports and ECBs policy statement in 2003): The thing we can be certain about is the principle about "equal treatment", which mean their interpretation of the criteria will never be relaxed or stricter for those states who later apply for euro adoption. This is why I keep on speaking so loud about "ECB assessment precedence" (because this bind and define the criteria). All in all, we should describe the criteria (which we already do), not according to the exact words by the treaty+protocol, but according to the developed ECB assessment practice. ECB has been granted power to interpretate, and so far they have done this in a good clear way, except for the ERM criteria. I agree its probably fully intended they did not clearly define how they measure the ERM 2yr membership requirement, as this gives them decision room in the future to push it the way they desire. My legal understanding of this however is (and here I fully agree with Glentamara), that the treaty+protocols direct note of ERM-membership for 2yr, mean ECB never completely can disregard this 2yr limit. Legally they are only allowed to define how this shall be measured, which mean they can never convert it to just 1yr. Plain and simple, somehow there needs to be minimum 2yr ERM membership. As I reported in this section of our debate, there is only 3 interpretation options complying with the treaty+protocol: either the 2yr is measured ahead of euro adoption day OR the 2yr is measured ahead of the irrevocable fix of the exchange rate OR they simply just round up the measurement (so "minimum 2yr membership ahead of the publication of the convergence report" in the real world mean "minimum 1.5yr membership ahead of the publication of the convergence report"). For the 3 boundary cases (Finland+Italy+Slovenia), it can be observed that all 3 measurement interpretations could be true, so we can not out from this observation alone, conclude which exact interpretation it is they have set their mind on. It however seems that the majority of secondary sources (as I replied above) believe the ECB interpretation mean this 2yr ERM membership criteria is measured just to mean "prior of euro adoption", which is why I now tend to believe this is the case (per the presumption that the majority usually never is wrong). Danish Expert (talk) 21:27, 20 August 2014 (UTC)

@Glentamara: You have misunderstood my point. I wasn't speaking of the treaty provisions, I was speaking of the implementation of the provisions by the Council. The question we are discussing is, "is it impossible for a state to adopt the euro with <2 years in ERM II"? We don't have any authoritative sources that say it is forbidden for Council to "flex" the criteria to allow for states with <2 years in ERM II to adopt the euro. I agree with you that Council is allowed to make "exceptions", and there is lots of precedents for this, thus it is not impossible. Perhaps it is improbable, but there is no evidence that it is impossible. Hence we should not claim that it is impossible in the article. That's why I suggested the "likely" wording above.

@DE: The problem is we don't have a precedent here. As far as I'm aware, there is no case where a state was denied entry solely due to being short of the 2 years by the date of euro adoption. Conversely, there is no example of a state adopting the euro with <2 years. (If I'm wrong and there is a case this would help clarify the situation.) Even if we did have an example, trying to draw conclusions from a few data points is a faulty generalization. There could be underlying hidden variables that you aren't taking into consideration. Also, you are creating a false dilemma to try to deduce the rules. All we can do is speculate how they would handle such a situation. Perhaps your speculation is correct, but it is still speculation. TDL (talk) 01:09, 21 August 2014 (UTC)

Ok, if that is the case, then I agree with you on this point. --Glentamara (talk) 05:34, 21 August 2014 (UTC)

@TDL: The Romanian example in your previous reply, is not an example of a contradicting view. Its a power point presentation, where you deduce too much when interpreting what the (2010-12) and (2012-14) parenthesis mean on the last slide. Personally I suspect the last line "Euro zone entry (2012-2014)" was likely to have meant "Approval for Euro zone entry (2012-2014) with adoption on the subsequent 1 January". The nature of power points is to cook down the words, so you can not conclude from this it is a contradicting example. On the contrary the source clearly states in its four last slides:

• Participation for at least two years in ERM II, when the national currency should be highly stable against the euro, is compulsory for adoption of the euro.

So far, you was only able to find the Polish journalist source to support your opinion that: "1 days ERM2-membership ahead of the publication of the convergence assessment report might be enough - provided the currency could be proofed to have been stable within a narrow band with narrow short-term interest rate differentials in the preceding 2yr period ahead of the publication of the convergence report". My point is, that this opinion represents a minority view on how the ERM2 criteria should be interpreted. If we check the entire pile of secondary reliable sources (research articles + ministry of finance reports + national central bank reports), I am rather sure we will have more than 90% (if not 100%) supporting the view that "ERM2-criterion compliance can not be met if you did not enter ERM2 at the latest 2yr ahead of the applied for euro adoption day". I agree with you, however, that ECB (as the primary source) never have written this in stone, so what we deal with here is a pile of secondary interpretations. My only speculation here is, that all the reliable sources (central banks and ministry of finance) before making their convergence publications and publish their "national euro changeover plans" actually communicated and asked ECB for expert advice in regards of how they should understand the ERM2-criteria, and then based their opinion on what the ECB had told them in this regard. Meaning, that the fact all secondary reliable sources agree on a certain criterion interpretation is not a result of "group thinking", but a result of this interpretation actually being the unwritten but correct one - an interpretation the ECB employees crafting the periodic assessment reports orally confirm when asked for advice but something they just never clarified in writing. I still think we need minimum 1 reliable secondary source agreeing with the stated opinion by your Polish journalist (that minimum 1 days ERM2-membership is enough), before we can mention in the article that such alternative interpretation views exist. Your Romanian power point does not qualify as proof, for the reasons I stated above. During the next days, I will read and check all the historic convergence reports for ERM2-precedence, to see if some of them provide a precedence example saying "ERM2-criterion not met because it had only been a member for [less than 1.5yr] ahead of publication of the Convergence Report". If such a precedence example can be found, it would solve our dispute. If not, then I can accept we leave the info open-ended for the readers to judge, but only for as long as we refrain from explicitly reporting in the Wikipedia article (or leaving the impression) that "some sources interpreted the ERM2-criterion could be met after only 1 day of ERM2-membership ahead of publication of the Convergence Report provided a currency exchange rate stability existed in the 2yr period preceding the publication of the Convergence Report". This last italic line is wrong, as we do not have any reliable sources to support such claim. Danish Expert (talk) 05:43, 21 August 2014 (UTC)

If you don't like powerpoint, here is the same deadlines discussed by the IMF in a formal report: [13].

You seem to not be understanding my point. I've never suggested that 1-day ERM membership is sufficient. In fact I said exactly the opposite: that I doubt that it is (though I've seen no evidence one way or the other). So this is a straw man. What I said is that I doubt that a state would be denied entry if they had 23 months and 29 days in ERM II. Is there some line in some article that you object to that you think is giving the impression that 1-day of ERM memberships would be sufficient? I thought this is why we used the "expected" wording: to emphasize that the guideline called for 2 years, but that it isn't a hard and fast rule and Council has flexibility in how it is applied. I don't see that phrasing giving the impression that 1 day would be sufficient. TDL (talk) 06:46, 21 August 2014 (UTC)

OK, based on your supplementing much better source, I now accept NBR in 2006 wrote the following sentence:

• "NBR's policy guidelines call for entry into ERM2 by 2010-12, and euro adoption by 2012-14."

However I still highly suspect, this is likely just to be a published statement affected by a translation mistake going from Romanian to English. I suspect the NBR statement should have been translated to the following:

• "NBR's policy guidelines call for entry into ERM2 by 2010-12, and euro adoption approval by 2012-14."

As you can see the minor add of the word "approval" would mean, that the sentence suddenly comply with the prevailing criterion interpretation (stated by the Polish MoF and ECB spokesperson), as "approval by 2012-14" would lead to adoption either 1 Jan 2013 or 1 Jan 2014 or 1 Jan 2015. However, I acknowledge that it might also be the Rumanian NBR having misinterpreted the criterion into the incorrect minority view reflected most recently by your Polish journalist. If we shall evaluate the quality of the statements, the one that I cited above made by the Polish Ministry of Finance has the highest quality, as it go into clarifying details what the final definitive deadline for ERM2-entrance would be in order to have a shot for meeting their specific euro adoption target. The Romanian statement is more vague, and most likely either the product of translation mistake or mis-communication.

My objection about the discussed: "Is 1-day of ERM-membership enough, provided the currency in the 2yr period ahead of the assessment time did not fluctuate by a big margin while at the same time not experiencing any severe tensions (measured by shortterm interest rate differentials, potential interventions and/or set up of financial external assistance instruments to counter tensions)", is mainly related to my objection towards the statement made by your Polish journalist: "The rules do allow ERM-2 to be applied retroactively. Poland could be deemed to have met the criterion if its currency stayed within the prescribed range for two years, even if it had not declared it was entering ERM-2." I object to the bold part of the line, as it imply he think ERM2-membership for 0 or 1 day in theory could be enough for Poland to comply with the "ERM criteria".

When Poland sets their euro adoption target after next general election in 2015 I am positively certain Poland will set its euro adoption targets in accordance with the ECB assessment practice, so that the requirement of minimum 2yr ERM2-membership ahead of the euro adoption day would be met. In example, the most likely thing to expect, would be that Poland will set a target for ERM2 entry in 2016 with the target of euro adoption being 1 Jan 2019. I say (along with the Polish Ministry of Finance and ECB spokesperson), that adoption of the euro on 1 Jan 2019 would not be possible, if they only join ERM2 on 2 January 2017 or later. For the sake of proving we are all correct about this (except the Polish journalist), I hope this scenario will actually play out (which of course also could happen if Poland join ERM2 2 Jan 2017 - which then would mean the earliest possible euro adoption day would be 1 Jan 2020).

When we have no preceding examples of ECB stretching the criteria even more wide, in example to mean: "2yr ERM-membership ahead of euro adoption - while applying a rounding principle for the measurement so that any membership above 1.5yr ahead of the euro adoption day in our eyes would be equal to 2yr." When we have no precedence for such a thing, we should not speculate such a thing would be possible. As stated above, I however respect your call for a precedence case of a country not complying with the ERM2-criterion because of its ERM2-membership on the envisaged euro adoption day would have been less than 2yr. If we can find such a case, this observation should also be added to the article, to tell readers that ECB in a previous case judged "ERM2 membership for less than 2yr ahead of the aspired euro adoption day to be insufficient to comply with the criteria". I already approved your formulated criteria line, with the observation "what was previously enough", and now just suggest to extend the observation with a line saying "what was previously not enough" (in regards of the length of the ERM-membership). Danish Expert (talk) 10:23, 21 August 2014 (UTC)

First interesting case to report, is the Greek ERM-compliance assessment on 24 March 1998. The ECB convergence report (March 1998) notes the state had no ERM-membership in the 2yr reference period stretching from 1 Mar 1996 - 27 Feb 1998. The state joined ERM on 14 March, and as the report was published with a data cut-off limit for legal compliance on 24 March 1998, ECB had the option also to use the new Greek ERM central rate as part of the assessment whether or not currency developments for the Greek Drachma complied with the ERM criteria. However, they did not. Instead they only analysed the state's currency developments according to the recorded average rate in March 1996, which is a standard procedure to be applied if you was not a ERM-member in the 2yr reference period.

The conclusions from the final approving Council meeting states: "Greece does not fulfil any of the convergence criteria mentioned in the four indents of Article 109j(1). For the third indent: The currency of Greece did not participate in the ERM in the two years ending in February 1998; during this period, the Greek drachma (GRD) has been relatively stable against the ERM currencies but it has experienced, at times, tensions which have been counteracted by temporary increases in domestic interest rates and by foreign exchange intervention. The GRD joined the ERM in March 1998.".

From this observation we can conclude, that states as minimum needs to have been an ERM-member before the 2yr reference period ends in the assessment report. It is not enough to become an ERM-II member ahead of euro adoption day, and even not enough to become member ahead of the CR publication (if it only happen a few weeks ahead of the publication). This is a precedent case, proving that: "Every time it happen later than the end of the 2yr currency exchange reference period noted by the CR, it will per definition (according to the applied measurement approach) be too late."

Of course the above observation is only to a small extend limiting the theoretic opportunities, as states are allowed to request renewed criteria compliance checks, basically by the end of each month during the year. However, as the political approval process and minting preparation (incl. frontloading banks with coins), all together is required to start at the latest 3 month ahead of euro adoption (with the standard process in fact normally starting 7 months ahead), we can conclude no state will ever join the eurozone with less than 3m of ERM-membership ahead of the euro adoption day. As this was only a small step forward (and not really a surprising fact), I will of course continue to search and report if there is additional interesting observations throughout the history. Stay tuned. Danish Expert (talk) 13:44, 21 August 2014 (UTC)

If you're going to take the position that any source which refutes your POV has a "typo" or is wrong, then you shouldn't be surprised that you reach the conclusion that there are no sources which refute your POV. This is called a confirmation bias. I could just as reasonably claim that there is a typo in the Croatian or Polish source and hence there are no sources to support you POV and a "vast majority of secondary sources" take the opposite position. Every time I prove your argument wrong, you just move the goalpost. If I found a source saying the same thing in Romanian, no doubt you'd say that the person was misquoted. If I found an audio tape of him saying it, you'd say the tape was altered.

If you object to what the Reuters journalist said (which you refer to as the "Polish journalist" though the article has no name on it so I'm rather perplexed why you think the author is Polish or why their ethnicity is relevant) then you should contact them. I'm not here to defend them. But I've never suggested that we should claim in the article that 1 day of ERM membership is sufficient, and you have not been able to point to anywhere in article space where this is actually claimed, so it completely baffles me why you keep suggesting that I "speculate such a thing would be possible". Conversely, you have invented quite an elaborate theory including "rounding principles" that are completely made up with no basis in fact. Perhaps they use ceiling rounding? In that case 1.1 years would be enough. We could invent all kinds of "principles" but just because the fit the facts doesn't make them true. You are trying to say in article space that something is impossible based on your flawed logic and deductions. This is very much speculative, and yet you are pushing this speculation while claiming you are opposed to speculation.

I also disagree with your point about Greece. Once again you are drawing conclusions unstated by the source and ignoring alternative plausible explanations. We cannot conclude that they failed due to insufficient ERM time because, as Council clearly points out, the drachma experienced "tensions". How would Council have reacted if there was no tension like the lira or markka? We will never know. If you can find an example where the EC/ECB deemed a state had not converged due solely to "insufficient ERM time" then I would support the addition of a note about that. But even in that case we can't conclude that it is a hard and fast rule. Council could still approve them. TDL (talk) 17:35, 21 August 2014 (UTC)

Sorry if you feel attacked, this was certainly not my intention. Together we are all helping to push the fact finding forward and share thoughts about the matter, which all help contribute to improvement of the article. To sum up and avoid repetitive arguments, I accept a minority criterion-understanding might have been present in the mind of the Romanian central bank, but as I stated above, if we dedicate 1 day of our life to compile a list of all previous criterion-understandings, I am rather sure at least 90% of them will agree with the 2 sources I started to list (from the Polish ministry of finance and ECB spokesperson). My assumption however is, that if we compiled such a list this would still not change your opinion, as you would continue arguing that although being the prevailing majority assessment, it still was only collected from secondary sources, and by the end of the day only EC+ECB primary sources would count when decisions were made, and hence our article should still refrain from reflecting the majority opinion about how far this 2yr-membership-criteria in best case can be stretched, if its only based upon a collection of secondary sources like central banks + ecb spokesperson + research articles + national euro changeover committees. I sense that you insist, the Wikipedia article should only be specific about this if the primary EC+ECB source is specific (which it is not). Am I right or wrong, that this is how you feel? For the purpose of moving things forward (and not to use time on things you would anyway reject), instead of compiling this "majority opinion list", I started just to check how much precedence info we can extract from the convergence reports. Please note, all my proposed explanations what the unwritten ECB assessment criteria could be was just posted to emphasize that they are not making arbitrary decisions. I agreed they most likely by full intend decided not to reveal in writing exactly where the borderline for the length of ERM2-membership has been drawed, so that it provide them flexibility for future interpretations. My interpretation proposals was just to emphasize, that they can not do implement arbitrary interpretations that will variate over time. Once they made one decision based on a certain argument, this binds them to continue this same assessment practice in the future, so there needs to by the end of the day there needs to be some kind of clear definition. For as long as we do not know, I just politely suggested we should let the article reflect the majority opinion (by secondary sources) where this ultimate borderline has been drawn in the sand, and (yes) speculated the majority opinion by secondary sources, most likely has been developed after oral consultation with the ECB convergence assessment office crafting the convergence reports. So reflecting this "majority opinion by secondary sources" I think we would actually reflect the unwritten truth (at least as of todays political climate).

As I sense you insist we should refrain from mentioning "the majority opinion by the secondary sources about the minimum length of ERM2-membership ahead of euro adoption", even if we can compile a list proving 90% have agreed on the same opinion, I deliberately gave up this idea. So for the sake of reaching a compromise, my spotlight has now been directed only to attempt finding historic assessment precedence examples to map how long the ERM2-membership needs to be. I think you should not reject this example to fast. By reading through all subsequent convergence reports, we can see that states passed the ERM-criterion both with "tensions" + "high interest rate differentials" + "market interventions". In fact, when Greece in 2000 passed the ERM-criterion, they also had such things. The major only difference for Greece between 2000 and 1998, was that in 2000 they had now been an ERM2-member for 2yr ahead of euro adoption. I think the example is water proof, and also is save to conclude that for a state to pass the ERM2-criterion it is an undisputed fact, that it in all circumstances would need to have an entry into ERM2 at the very latest on the same day they apply for euro adoption. This however does not mean, that I now suggest this would be soon enough, as I still think the most logical think would be to expect it needs to be 2yr ahead of euro adoption at the latest. The highlighted Greek example, however sets the outer borderline for how much ECB in the future can stretch their current interpretation limit without actually breaking the criteria as defined legally by the treaty (or break the principal about equal treatment). ECB needs to respect both. Hence, there is a limit for how far they can go, in regards of interpreting how much will be needed for states to reach an ERM2-criterion compliance. Danish Expert (talk) 19:58, 21 August 2014 (UTC)

"when Greece in 2000 passed the ERM-criterion, they also had such things" - Council said "the Greek drachma (GRD) has not been subject to severe tensions" when they admitted Greece.

My argument has nothing to do with the quantity or quality of the sources and everything to with drawing unstated conclusions from those sources. I think Glentamara made an excellent point above. Even if it was 100% certain that the convergence criteria required 2 years in ERM II (which I still don't accept), that still wouldn't 100% rule out Council approving a state joining the euro. The treaty only requires Council to make the decision "on the basis" of the criteria. Thus we can't say with 100% certainty that it would be impossible for a state to be approved with <2 years. Your point about past states being admitted with "tensions" and "high interest rate differentials" just emphasizes this point. Council can and has approved states joining even when they don't strictly meet the criteria (which specifically requires that a state must have been in ERM "without severe tensions").

If you want to look for past precedents and mention them in an explanation of the criteria, I think that's a good idea as it helps readers understand how the criteria has been applied in the past and thus how it will likely be applied in the future. But again, we can't draw far reaching conclusions for these observations. If country X joins with 1.5 years and country Y is rejected with 1.4 years we can't say, "1.5 years is the minimum to join", since that is WP:SYN. All we can do is describe the history. As I said above, as long as we are careful and frame things with words like "likely" or "based on past precedence" rather than flat out saying that it is "impossible" I'm fine with it. But we shouldn't make concrete conclusions stating something is impossible when that isn't certain. In politics, anything is possible if you have friends in the right places.

Rather than arguing about the underlying truth, which we're not making any progress on, why don't we focus on any changes to the article that you feel are necessary. Perhaps we can make progress that way. Write the sentence/paragraph that you would like to see in this article or some other article and I'll respond with a counterproposal. TDL (talk) 00:34, 22 August 2014 (UTC)

In the Greek example, the keyword here in the third subcriteria of the exchange rate criterion, is that there needs to be "absence of severe tensions". When they described Greece in 1998 they noted "temporary tensions", which in principal was good enough (as other states later on having "temporary tensions" managed to pass the goalpost). So I insist the precedence example is valid, and that Greece already complied with the remaining part of the 3 subcriteria when the council wrote "Greek drachma (GRD) has been relatively stable against the ERM currencies but it has experienced, at times, tensions which have been counteracted by temporary increases in domestic interest rates and by foreign exchange intervention." I can proof other countries to whom this situation was true, actually managed to pass the "exchange rate criterion".

Our problem though, and here you have a valid point, is that ECB never quantitatively sharply defined their 1st (minimum length of ERM-membership) + 2nd (exchange rate needs to be close to the central rate) + 3rd subcriteria (absence of severe tensions). The only sharply defined subcriteria, is the 4th (absence of devaluation by realigning the ERM-central rate against the euro). My agenda here is two-headed. First I now work to see if I can find relevant precedent examples for the purpose of slightly extending the criteria-note. Second I also work to see how we shall reflect the 4 Exchange rate subcriteria in the convergence table.

For the first agenda, my ongoing search (which will be completed tonight) is the attempt to find any relevant precedence examples extracted from the past convergence reports, helping us to define the minimum length value for the first of the four subcriteria (ERM2-memberhship). In that regard, I agree with you we shall be careful to filter away any false-negatives (meaning that we need to pay attention to the existence of all 4 Exchange rate subcriteria and not just 1). My quest is therefore to find border drawing examples, so that we can note we have a precedent example showing its (1) possible or (2) impossible overall to comply with the overall exchange rate criterion if the length of your ERM2-membership was (1) X or (2) Y. Such precedent examples would show the borderlines of possibility and impossibility in regards of the minimum length of ERM2-membership (being required, along with compliance with the 3 other exchange rate subcriteria). For the distance between X (minimum 2 years membership ahead of euro adoption day) and Y (membership at the latest on the last day of the 2 year backwards going reference period in the published convergence report), I agree with you this is so far still a matter of undefined interpretation room (although 90% of secondary sources seem to agree the Y impossibility line should be drawed on the following day after we have drawed the X guaranteed possibility line - and not "on the last day of the 2 year backwards going reference period in the published convergence report"). For further debate on how to reflect the 4 exchange rate subcriteria in our convergence table, I have set up a separate section below, to go more into depths on this issue. I will return later tonight, to post my findings and search results for both debates. We will soon reach a final conclusion! :-) Danish Expert (talk) 08:58, 22 August 2014 (UTC)

Yes, the criteria says "severe" tensions. You will notice that for every state Council concluded that "their currencies have not been subject to severe tensions". For Greece they concluded that "it has experienced, at times, tensions". You are guessing that this means that Council found these tensions to be not severe. However, Council never said this. This is merely your guess at what Council was thinking. That future states were admitting with "tensions" is no different than that states were admitted with <2 years in ERM. Council does what it likes after considering the criteria. Council isn't a logic machine that follows black and white rules that you are attempting to ascribe to it. And again: precedence does not prove that something is impossible. It only proves that there is a precedence which Council may or may not choose to follow. TDL (talk) 22:15, 22 August 2014 (UTC)

OK, you have a valid point that the Council is a political machine, so I agree we should refrain from using their conclusions as data reference. The conclusion of the 1998 EC convergence report however qualify as reference. Here is the two exchange rate conclusions they wrote about Greece:

• The Greek drachma entered the ERM in March 1998 but did not participate in the mechanism during the two years under review; the drachma was relatively stable against the ERM currencies in the review period but at times experienced tensions which were counteracted by increases in domestic interest rates and by foreign exchange intervention. Greece does not fulfill the exchange rate criterion.
• The firm commitment, as expressed by the Greek authorities, to participate in EMU and progress in lowering inflation and government deficits has promoted exchange rate stability. However, the Greek drachma experienced considerable selling pressure in October 1997, as financial disturbances spread from emerging markets in Asia. The Greek Drachma was defended successfully by temporary interest rate increases and foreign exchange intervention.

If the Greek ERM2 central rate had been applied for retroactive analysis in March 1998, it would have showed exchange rate fluctuations in a band from -12.6% to -15.6% from the central rate. If the Greek D-mark central rate (157.5) had been used, the fluctuations had been within a much narrower +-1.6% band. Or in other words, the low rates against ECU/euro was solely the result of ERM2 picking a way to high central rate from the very beginning, and not the result of exchange rate instability caused by severe tensions. The note in the EC convergence report about tensions in October 1997, was not related to local currency stability issues but due to general market pressures arising as part of a global financial crisis, and can be found repeatedly noted for many of the same currencies that the same report approved for euro adoption. Hence, this proof my point, that Greece was in full compliance with the two other exchange rate subcriteria (no devaluation, and no severe tension). As it did not comply with the first exchange rate subcriteria "ERM2-membership at the latest on the last day (27 Feb 1998) in the convergence report", this was the sole reason why EC concluded Greece did not comply with the exchange rate criterion in March 1998. Danish Expert (talk) 05:43, 24 August 2014 (UTC)

Instead of the above Greek case, I have now found a much stronger precedent case that narrow down the interpretation space even further. The EC convergence report 2004, published 20 October 2004, concluded collectively (including for the 3 states recently having joined ERM2 on 28 June 2004 - with membership only during the last 95 days of the reports review period from 1 Oct 2002 to 30 Sep 2004):

• "None of the 11 Member States with a derogation has participated for two years in ERM II during the review period. As a result, none of them meets the exchange rate criterion."

This is the first clear cut example, that you need to be ERM2-member for a certain minimum amount of time in order to comply with the exchange rate criterion. Being ERM2-member during just the last 95 days (or 3 months) of the review period, has been assessed to be insufficient! At the other side of the boundary we have Italy (as previously reported), where EC evaluated a membership throughout the last 461 days (more than 15 months) of the review period, to be sufficient. So we have now narrowed the window of uncertainty! The two precedent rulings from 1998+2004, conveniently fit with the assumption, that EC only exempt shorter than 24 months membership in the backwards-looking review period, if the state will have been a member for minimum 24 months upon its euro adoption day. The 3 states being disregard in 2004, would only have been members for either 6 or 18 months ahead of their applied for euro adoption day. However, I agree the "24m membership ahead of euro adoption requirement" so far only can be labeled as qualified speculation backed by secondary sources and no contradicting rulings so far by the European Commission. I admit, that the precedent cases can not predict with 100% certainty what the minimum membership length shall be. In example, it could also be imagined that EC perhaps suddenly decide that having been an ERM2-member just for the last 12 months (more than 50%) of the review period would be sufficient. To conclude: I propose we extend the observational note in the article to say: "A membership only during the last 95 days of the convergence assessment's review period, has previously been ruled to be insufficient by the European Commission". Danish Expert (talk) 06:39, 24 August 2014 (UTC)

Yeah that seems reasonable. I'll go extend the note on Euro convergence criteria with this example. TDL (talk) 23:24, 24 August 2014 (UTC)

Request for extended criteria note (still about ERM2-membership length)

Finally, I now also managed to find the last missing-link we have been searching for. In EC convergence report 2006, with data-cut off on 17 November 2006 and the exchange rate fluctuations observed for the reference period 1 Nov 2004 - 31 Oct 2006, it is concluded that Cyprus+Latvia+Malta did not comply with the exchange rate criterion due to an insufficient record of ERM2-membership. The 3 states all joined ERM2 on 2 May 2005, meaning they had been members for the last 548 days (or 18 months) of the reference period. The report also take note they would have been members for 19 months once the Commission adopt and approve the report. This case is very interesting, because in 1998 Italy was approved despite only having ERM2-membership throughout the last 461 days (or 15 months) of the reference period. Hereby we can conclude, they do not operate a certain time limit according to the reference period. Instead it can only be the original hypothesis they apply, where their needs to be minimum 24 months ERM2-membership at the latest upon the "Councils irrevocable fix (and approval) of the euro adoption" or alternatively on the subsequent "1 January euro adoption day". To reflect this last finding, I have just updated the criteria explanation at Euro convergence criteria to this new version below. Hope you can all approve it. Danish Expert (talk) 01:43, 25 August 2014 (UTC)

• Exchange rate: Applicant countries should not have devalued the central rate of their euro pegged currency during the previous two years, and for the same period the currency stability shall be deemed to have been stable without "severe tensions". As a third requirement, participation in the exchange-rate mechanism (ERM / ERM II) under the European Monetary System (EMS) for two consecutive years is expected, though according to the Commission "exchange rate stability during a period of non-participation before entering ERM II can be taken into account." In example, Italy was deemed to have converged with only 15 months of measured ERM II membership, as of the last day in the review period of the convergence report. Although it has not been confirmed by the European Commission or ECB in writing, it is generally assumed they require a minimum of 2 years ERM membership ahead of either "euro adoption day" or the day of the final political approval with "irrevocable fix" of the currency exchange rate. This theory was further supported, when the European Commission concluded collectively for Cyprus, Malta and Latvia, that their membership within the last 18 months of the review period ending on 31 October 2006 was insufficient, noting it would have been equal to only 19 months of membership ahead of final approval or 20 months ahead of the "euro adoption day". As of 2014, all nineteen successful applicants had surpassed two years as an ERM member, both by the day their currency exchange rate was "irrevocably fixed" and by the day they adopted the euro.

You are once again jumping to conclusions which are unstated by sources. I agree with your statement that "they do not operate a certain time limit according to the reference period" (which is what I have been arguing all along) but strongly disagree with your conclusion that 24 months before euro adoption is necessary. None of these sources say anything to that effect. This is just your hypothesis. Until some source is produced that says that the 24 months is measured before euro adoption day we should not say it. Patterns do not prove rules. TDL (talk) 03:26, 25 August 2014 (UTC)

Yes, a source would for sure be preferable. In the absence of a source, I however still believe we are allowed (within a certain room) to apply "observed logic". We have several primary sources noting that both EC+ECB repeatedly have claimed to adhere to the "equal treatment principal", meaning that no states are allowed to be evaluated stricter or more lax, compared to precedent cases. Then how is it possible, while knowing 2 years ERM membership is expected as an isolated criteria, that Italy's ERM-membership for 15 months was sufficient and subsequently Cyprus+Malta+Latvia's ERM2-membership for 18 months insufficient???

Such a "Cyprus+Malta+Latvia" vs Italy observation can only comply with the "equal treatment principal", if the measurement approach did not focus on "length of membership within the review period", but instead focused on either "length of membership upon the day euro adoption is finally approved and the exchange rate irrevocably fixed" or "length of membership shall be equal to minimum 2 years when measured backwards from the applied for euro adoption day". Otherwise, the EC+ECB would be guilty of breaching the "equal treatment principle" in the case "Cyprus+Malta+Latvia" vs Italy. This observation is moreover underlined by the fact, that the "3 states" posted far less "currency fluctuations" and "negative pressures" throughout their 2yr review period, compared to what Italy did; meaning they had a far better performance on the other subcriteria (fluctuation close to central rate with absence of severe tensions). My underlined part of the formulated italic note above, took care to describe, that the neutral observations fit with the noted theory for the de facto applied measurement approach, which has been generally assumed by secondary sources although never confirmed by the primary EC/ECB sources in writing. After this elaboration on the argument behind the reasoning, I want to know if you still insist we should omit the underlined observation in the note above? And if yes, why? Best regards, Danish Expert (talk) 21:58, 25 August 2014 (UTC)

First of all, the text you wrote claims that the European Commission "noting it would have been equal to only ... 20 months ahead of the "euro adoption day". I can't find anywhere that the EC points this out, though I didn't read the entire report. It does point out the time as of Council approval, but that would suggest the opposite of what you argue: that the two years is measured until Council approval not euro adoption date.

Yes the EC and ECB have stated that they have an "equal treatment principal", but does Council have to follow this principle? I doubt it. They can pretty much do as they like. If you look at the 1998 ECB report, I can't find anywhere that they conclude that "Italy meets the exchange rate criteria". It is really written from a purely observational perspective with no conclusions: "it has participated in the ERM for around 15 months of the two-year reference period". It was Council who decided that they have converged, even though they had <2 years. So I don't think it is fair to say that the two cases violate the "equal treatment principal" since they were treated equally by the ECB/EC (and Council's decision is a political one.) And even if it was, it would still be speculation to claim they did it because of the two years prior to euro adoption day. TDL (talk) 05:58, 27 August 2014 (UTC)

As I mentioned earlier (although it probably drowned in my many words), I accepted your point that we can not use what the Council discides/does as a source, due to being a "political vehicle". However, I then argued, and will now repeat again, that we have the 1998 EC convergence report clearly assessing that Italy met the "exchange rate criterion" despite only having been a member for the last 15 months of the review period, as per the two cites below:

• As regards the criterion mentioned in the third indent of Article 109j(1), the currency of Italy, although having rejoined the ERM only in November 1996, has displayed sufficient stability in the last two years. For these reasons Italy has achieved a high degree of sustainable convergence. Consequently, Italy fulfills the necessary conditions for the adoption of a single currency.
• Although the lira has participated in the ERM only since November 1996, it has not experienced severe tensions during the review period and has thus, in the view of the Commission, displayed sufficient stability in the last two years.
• The Italian lira has participated in the ERM from 25 November 1996 onward, i.e. for longer than 15 months by end of the review period in February 1998. The Italian lira was more than 2.25 % below its central parity against the median currency for a total of 96 days in the review period. However, since re-entering the ERM, the Italian lira has been within ± 2.25 % of its central rate against the median currency and was only 0.4 % above it by end-February 1998.

In my proposed underlined note in italic at my previous reply (the one you deleted from the article), I never claimed it was "24m ahead of euro adoption day", but reported and argued it had to be either "24m ahead of euro adoption day" or "24m ahead of the the final approval and irrevocable fix of the currency exchange rate". My argument goes, that it is the European Commission who needs to comply with the "equal treatment principle" when reporting their assessments. My argument further more goes, that our Convergence table shall never be filled with Council data but instead filled by "convergence assessment data" provided by EC. You are correct to note, that ECB often/always refrain to conclude, but please note that this is only for precautionary reasons, because they know that per the treaty text it is the Commission who shall file a recommendation to the council if the state comply or not - so just to be save ECB sometimes stay silent - leaving the conclusion part fo the Commission. My entire collection of "precedent assessment cases" reported here in this ongoing debate, was all extracted from EC convergence reports. Hence, its fair to conclude all theses cases needs to fit into a pattern that comply with the "equal treatment principle". Because of the Italy vs. "Cyprus+Malta+Latvia" incident (both extracted from EC convergence reports), we actually have basis to reach the conclusion that I reported above. Now that I clarified this, did it then change your opinion? Or do you still insist we should omit the underlined observation in my earlier written note? And if you still insist this should be omitted, why? Danish Expert (talk) 18:54, 27 August 2014 (UTC)

Yes, but the first quote is not actually from the convergence report itself, but rather from the Decision adopted by Council approving their adoption of the euro (which just happens to have been published jointly with the report.) The second quote speaks only of Italy's "tensions" IMHO and the third says nothing about whether Italy meets the criteria. Though looking closer I see that this Decision was made at the recommendation of the EC, so perhaps there is a case to be made about precedence set by their recommendations rather than the conclusions of their report.

However, the main point is that you are creating a false dilemma by artificially limiting the possible reasons why the EC could have reached a different conclusion to only two of many many alternatives, with no evidence to justify this assumption. Perhaps the 2 years is measured to the date that euros start to be minted? (That would seem to me to be the drop-dead date for deciding whether to go ahead.) Perhaps it is 2 years by the end of any dual circulation period? Perhaps it is 22 months by euro adoption day? I could come up with a million different hypothesis to justify the different conclusions that are specifically engineered to fit the facts and that thus you can't disprove. And if you are able to disprove them, I could just keep adding more and more epicycles as you have to make the theory more and more unfalsifiable but also more and more speculative. Just because I can contrive a theory that fits the evidence doesn't make it true. Since we can't list every possibility (nor even think of every possibility) we shouldn't give an incomplete list of the possibilities without evidence to back it up.

Even if we suppose you are correct, what would happen if a state joins ERM II on 31.12.2014, gets Council approval to join the euro and has their rate irrevocably fixed in mid 2016, and then pulled out of ERM II on 29.12.2016? Legally, how could the switchover process be halted at that point? So from a practical perspective I don't see how it could be a requirement since it is unenforceable. TDL (talk) 23:21, 27 August 2014 (UTC)

The drop-dead date for deciding whether to go ahead is not when minting starts. The drop-dead date, is when the "euro adoption is finally decided and the exchange rate irrevocably fixed". At the very same time this happens, the state however sign an agreement with ECB how to behave in the transition period ahead of their euro adoption, so its impossible for a state to jump out of ERM2 in the transition period! This fact is also reflected by the ECB press release about Lithuania's euro adoption:

• The European Central Bank (ECB) and Lietuvos bankas agreed to monitor developments in the foreign exchange market of the Lithuanian litas against the euro until 1 January 2015.
• The agreement to monitor the litas is in the context of ERM II. Participation in ERM II and observance of the normal fluctuation margins for at least the last two years is one of the convergence criteria to be fulfilled ahead of euro area accession.
• The conversion rate of the litas is set by way of an amendment to Regulation (EC) No 2866/98, which will become effective on 1 January 2015.

Although I get your point from your reply above, that we can not with 100% certainty extract the "measurement approach" for the 2yr requirement from the precedent cases made in the "EC convergence reports", I think the 3 cited ECB press release points above, in combination with a low level of applied logic, can only lead to the conclusion, that ECB has now helped us to further describe and source the de facto applying "measurement approach". The moment a state is "finally approved and get their exchange rate irrevocably fixed", which happened on 23 July 2014 for Lithuania, their euro future and destination is irrevocably locked, and this situation will automatically imply a continued ERM2-membership until euro adoption day - with ECB and the national central bank sharing a joined responsibility to monitor everything runs smoothly in the transition phase. So after 23 July, nothing can now happen for Lithuania that will reverse their ERM2-membership or euro adoption! Hence, this is presumably also the winning argument used by EC, why they chose to implement this more lenient 2yr "measurement approach", as I described in the below disputed part of my criteria note:

• Disputed part of criteria note: Although it has not been confirmed by the European Commission or ECB in writing, it is generally assumed they require a minimum of 2 years ERM membership ahead of either "euro adoption day" or the day of the final political approval with "irrevocable fix" of the currency exchange rate. This theory was further supported, when the European Commission concluded collectively for Cyprus, Malta and Latvia, that their membership within the last 18 months of the review period ending on 31 October 2006 was insufficient, noting it would have been equal to only 19 months of membership ahead of final approval or 20 months ahead of the "euro adoption day".

To turn the argumentation upside down, we have no sources or precedent cases that contradicts the above conclusion. Moreover, the noted "measurement approach" makes fully sense from a logic point of view. Any other theory simply does not make sense, and can not be supported by sources. In addition, the noted "measurement approach theory" can be supported both by the official ECB source i posted above, plus as mentioned earlier: The secondary published interpretations by the Polish ministry of Finance + by the ECB spokes person clarifying the Croatian case + The French central bank (per the link posted by Glentamara noting that both Finland+Italy with less than 2yr membership ahead of "assessment day" were approved for euro adoption because they had been ERM2-members for 2yr ahead "euro adoption day"). I think we now have collected enough circumstantial evidence to proof the case without reasonable doubt. As I respect your argument, that we however still only deal with "circumstantial evidence", I took care in my note to say it was not a "written confirmed rule" but a "theory supported by precedent EC assessment cases". I still accept to keep it this way. For the sake of increasing the clarity and understanding of the lines, I am however willing to tweak the formulation (or add a linked note), so that we also integrate the ECB source I posted in this reply - and noting what they noted in this. If we add this proposed additional ECB source (using Lithuania as example) to the note, can you then approve we keep the disputed part of my written note? Danish Expert (talk) 06:24, 28 August 2014 (UTC)

No. Once again this reply is filled with numerous unsubstantiated assumptions that you present as facts which you then attempt to tie together with a string of logical fallacies in an effort to draw conclusions that are never stated by the sources.

Latvia did not "sign an agreement with ECB". The source says simply that they "agreed to monitor developments". Nowhere does the source say that they are forbidden from withdrawing from ERM2, so this is just more hypothesizing by you to try to reinterpret the facts to try to make them fit your theory. And none of your quotes say anything about the 2 years being measured to euro adoption day. While I understand that you believe your argument "makes fully sense from a logic point of view", it most certainly does not. I've pointed out the logical fallacies that your argument rests on time and time again. Since you haven't seemed to understand, let's try a counterargument the is based in the exact same "logic" you use:

• "In 8 out of 8 cases the ECB and EC have both found that states whose names start with the letter H did not meet the convergence criteria. There are no examples of states whose name starts with the letter H ever joining the eurozone. Hence, the EC does not allow states whose name starts with the letter H to join the eurozone. Therefore, Hungary must change its name before it can join the eurozone."

The above is the exact illogical argument you are making. They are both "theory supported by precedent EC assessment cases". You have no sources or precedents to disprove it. It makes just as much sense as the theory you are pushing, and is supported by exactly the same number of sources: 0.

This discussion has gone on for a week and 100,000kb and it is just going in circles. I've given you ample opportunity to support your theory with sources, but you have been unable to do so. I've made a very good faith effort to try to explain to you why your argument is illogical, and why even if it was logical it would still be unacceptable since it is WP:OR, but you either are not listening or don't understand. Nothing is accomplished by us arguing about the validity of your hypothesis because at the end of the day it is still just a hypothesis, unsupported by any sources. I am becoming quite tired with this discussion, and I have much more productive things I could be doing with my time. As such, I see little point in continuing with it. I'm not going to reply to this thread further unless you are able to find a source that actually says what you argue: that the two years is measured to euro adoption/irrevocable fixing day. In the absence of any such sources, I remain strongly opposed to your attempts to insert this unsourced theory into article space so please do not. If you wish to pursue the issue further I suggest starting a discussion at WP:NORN or WP:DRN where perhaps an uninvolved third party can give some input. TDL (talk) 08:45, 28 August 2014 (UTC)

As this entire debate has proofed, I am certainly not wasting your or others time. On the contrary my inputs greatly helped to develop both the criteria note, technical insights about the convergence assessment approach, and we had a fruitful exchange of views based solely on the value of arguments. I concealed to some of yours, and you concealed to some of mine. For this latest debate entitled "Request for extended criteria note", we are not so far away from each others standpoints, as you seem to suggest. I agree with you, that we have not got a "written rule" for how they exactly "measure and judge" the ERM2-membership length (backed by primary EC/ECB sources). Yet, I still proposed still to keep the disputed part of the extended note, while calling the disputed extension line an "interpretation theory", which was backed by majority of secondary sources (France central bank + Polish ministry of finance + ECB spokesperson), while also being supported by the logical circumstance that ECB's euro adoption approval always entail the settlement of a joint agreement for the state to ensure continued ERM2-membership and pursuit of exchange rate stability throughout the entire transition phase from "euro adoption approval" to "euro adoption day", and finally it is an "interpretation theory" which not (as of yet) has been contradicted by the 29 previous Commission verdicts extracted from all of their past EC convergence reports. As I see it, the winning argument why it should be allowed, is the below two comments by ECB which fully supports existence of the noted "interpration theory":

• The European Central Bank (ECB) and Lietuvos bankas agreed to monitor developments in the foreign exchange market of the Lithuanian litas against the euro until 1 January 2015. The agreement to monitor the litas is in the context of ERM-II.
• Participation in ERM II and observance of the normal fluctuation margins for at least the last two years is one of the convergence criteria to be fulfilled ahead of euro area accession.

Please note, I highlighted the word "accession". If ECB had written "ahead of assessment for approval of euro area accession" or "ahead of approval for euro area accession", it would not help us in any regard. But in fact it said "ahead of eurozone accession" (equal to "ahead of euro adoption day"). So what we basically discuss, and what we now must await other editors opinion about, is whether or not it can be allowed for a wikipedia article in this unique situation to accept noting this "disputed content" - that EC interprets that minimum 2yr ERM2 membership shall be fulfilled ahead of euro area accession - as an existing "interpretation theory" (as proposed by me, and backed by the 4 sources in this reply plus existence of no contradicting assessments in the EC convergence report), or if we completely should refrain to note existence of this "interpretation theory" - while only noting previous assessment observations (as argued by you). I am fully ready to leave this final question for Glentamara and other editors to decide. I thank you all for a good fruitful debate. Best regards, Danish Expert (talk) 18:18, 28 August 2014 (UTC)

Timeline

Debate about the proposed timeline chart to display exchange-rate regimes in EU member states, was moved on 23 Sep 2014 into the dedicated template talkpage.