Template talk:Euro convergence criteria/Archive 1

Archive 1

Archive 2

Archive 3

When?

Which date are these figures from? 91.208.174.15 (talk) 10:16, 11 January 2011‎

I have just updated the entire criteria template with a new set of economic stats. Titles and footnotes now clearly also explain the year for the data, and how the reference values and evaluation process works. The previous debt+deficit data in the template showed the criteria and values for Fiscal Year 2011, and the final evaluation of criteria compliance as of 30 March 2012. As these data and the evaluation results have been displayed and well-known for around 4 months, it was however more interesting now to let the table show a forecasted outlook for the next evaluation as of 31 March 2013. In order to ensure a continued availability for the old previous data, I have saved them into this new file: Template:Euro convergence criteria (April 2012).

For the current template version with new data, I have uploaded the most recent forecast values for HICP inflation, deficit and debt (as per the outlook for next evaluation time: 31 March 2013). These new forecast data were all easy to extract from the EC Spring Economic Forecast, and will be fast to extract and update again, when the EC Autumn Economic Forecast report gets published in November 2012. In regards of the average values for "Annual long term interest rates", the ECB source also publish numbers on a monthly basis (being relatively easy to extract and add). So it is now possible to update the table with new figures during the upcomming year, and thus follow if the outlook turns in a red or green direction halfways through (by November 2012). Just be careful to read and understand all the hidden notes. Danish Expert (talk) 21:08, 4 September 2012 (UTC)

Next version of the template

In order to keep the yearly assement data in some nicely saved templates, one of us should around October 2013 copy-edit the final data from this template, and save it in a new template named: {{Euro convergence criteria (April 2013)}}; to be displayed at the Euro convergence criteria article. Based on the experience with how the forecasted data developed in 2012, I have to admit there will be no point to create a new "forecasted convergence table" for next year, before we at least have the November 2013 forecast report published. Because the HICP values and identification of the 3 lowest scoring countries for a reference year, are basicly only possible to predict with a reasonable amount of certainty, at the point of time where we are halfways through the reference year. So we have to wait until the European Commission in November 2013 will publish both the recorded stats for Q2-2013 and Q3-2013, and the forecasted data for the remaing two quarters Q4-2013 and Q1-2014. :-) Danish Expert (talk) 14:00, 16 October 2012 (UTC)

Another argument supporting that we should stick with my proposed template creation schedule above, is the fact that the criteria table for compliance evaluation through all months in 2013, would also be the one containing the 2012 budget deficit + debt figures (and with recalculated HICP+interest rate reference values according to the month in 2013 where the compliance check is performed). So later modifications of this table will first have to be created, if any country ask for a renewed compliance check in May-December 2013, and in that case it would be confusing for the readers if we had already created the new Convergence Criteria template for future checks in 2014. To say it short and simple: The next 2014-version of the template with forecasted figures will only be appropriate to create in November 2013. Danish Expert (talk) 14:41, 15 November 2012 (UTC)

After additional consideration, and experience with the ongoing updating task here in 2012 (as reported in detail by this other talkpage chapter), I will now go one step further and completely abandon my original idea of running this table with forecast HICPs. When I first introduced the change back on 1 September 2012 it appeared to be a splendid idea. Because it was interesting to track how values were going to be for Latvia and Lithuania just around six months later. Looking back, I however have to admit it did not help to clarify their situation, and perhaps created more confusion instead of helping to sort things out. The problem is, that it is simply impossible to forecast the HICP and "interest rate" reference values six months in advance, because the HICPs are so sensitive towards tax hikes (and in some countries energy prices), that we have very high uncertainty about the forecast figures, even for a forecast of oneyear-averages being based upon recorded data for the first 6 months and only forecasted for the remaining 6 months. Bottom line is, that the forecast reference value for "interest rates" can currently be either 3.5% or 4.0% or even 5.5%, all depending on how the last digit of HICPs will end for the 7 countries currently standing a chance to being picked for the group of 3 benchmark countries with the lowest HICP at the next ordinary evaluation time on 31 March 2013. With such a huge level of uncertainty, I think it is better now completely to avoid displaying these forecast HICP figures in our template. The previous approach with forecast HICPs, is moreover also on the borderline to violate WP:FUTURE. So I will now drop the idea, and instead introduce the new data approach detailed below.Danish Expert (talk) 18:50, 22 December 2012 (UTC)

New data approach and schedule for new template versions

As a good final solution for my many considerations outlined above, I am now ready to introduce a new version and data approach for the template. In the upcomming months, the HICP column and "Interest rate" column will only display actually recorded average data for the past 12 months. Each time we start a new calendar year, it will be time to create a new "Euro convergence criteria (20XX)" template to be displayed in the main article Enlargement of the eurozone. This template can then be used to upload recorded 12-month data for HICP and interest rates during the entire calendar year in concern. It will of course for the first three months of the year (until recorded data are released mid April), show a forecast for the debt+deficit in the past full calendar year, but these forecasts have been prooved to be highly reliable, and when the previous November forecast for debt+deficits will be replaced by a new February forescast -it will be very close (if not identical) with the final values to be published two months later. So this is acceptable. The described approach will also make much more sence, when comparing with the fact that all non-euro countries can ask for a renewed compliance check after each month in the year. So having a template that actually maps the ongoing criteria situation month by month in each year, will be much better and informative to the readers of Wikipedia. I will introduce the new approach, as soon as the next HICP data for December 2012 will be published on 16 January 2013. Stay tuned. :-) Danish Expert (talk) 18:50, 22 December 2012 (UTC)

Today I have adjusted the template to the new data approach described above. The "hidden notes" in the template have also been updated accordingly, and finaly the template was also renamed to reflect the change from "Euro convergence criteria" to the new title "Euro convergence criteria (2013)". As per the new data approach, the template is only allowed to feature data being recorded in 2013 (with debt+deficit data constantly referring to Fiscal Year 2012). The moment the 2014 calendar year starts, we shall then create a new similar template to display the data for all convergence checks in this specific year. All old year templates are supposed to be listed in the euro convergence criteria article, as a place to map the historical record of all the official convergence checks performed in previous years. Danish Expert (talk) 16:10, 13 January 2013 (UTC)

No, this is a very bad idea. Who cares what Sweden's debt to GDP ratio was in 1997? These templates are huge, and filling the Euro convergence criteria article with year-after-year of data doesn't help a reader understand the topic. I don't doubt you put a lot of effort into making this table, however you need to understand that wikipedia is WP:NOTSTATSBOOK. We need to WP:DISCRIMINATE between notable, encyclopedic content and stats that only hurt the readability. I'm honestly not trying to give you a hard time, but these same issues keep coming up over and over again. Not only that, but much of the content is WP:OR. The fact that you provide details on how you calculated the figures should be enough evidence that you haven't provided a WP:RS for the figures. TDL (talk) 04:01, 14 January 2013 (UTC)

For the sake of clarity, I want to point out and remind you, that my recent update of the data approach for the template was in its entirety working to reduce the amount of WP:OR and WP:FUTURE. If you check the edit history, you will find that I have managed to find new direct WP:RS for all monthly HICP and monthly interest rate figures for all countries in concern, meaning that these figures are no longer calculated for each country. We now have direct RS for these figures, thanks to me. The only thing currently still being calculated is the ECB reference limits, which are calculated according to the published ECB method. Uncertainties for this approach have been clearly highlighted by bottom-note text in the legend for "Criterion". You are correct to point out that it would be preferrable if we had a direct RS from ECB for the monthly re-calculated reference limits, but as this is unfortunately not the case, we have to accept this small deviation from the official WP policy and do this calculation by ourself. All input data for the calculation are by-the-way also provided by the Eurostat RS, so the only WP:OR is that we need to concider if any of the data in the equation should be excluded according to the so-called "outlier criteria". I acknowledge we are operating in the borderline area of what the wikipedia policy can accept, but think it is fully justifiable when considering the template currently notes this slight uncertainty being something the reader should be aware of. Along the way, it is also notable content to follow the monthly update of figures to map if Latvia or Lithuania will comply with the criteria for euro adoption in one of the months during 2013. The new data approach for the table will serve that purpose. Danish Expert (talk) 10:47, 14 January 2013 (UTC)

It's great that you've found sources for some of the data. But the data which isn't sourced is still WP:OR, and shouldn't be included in the encyclopedia. Even if you're following the ECB's method, this is still OR. We can't just "deviate" from policy because we can't find a source. TDL (talk) 18:23, 14 January 2013 (UTC)

Actually we now have a RS for all 5 specific data columns for all countries. For the moment we only miss a RS for ECB's monthly "reference limit" for HICP + interest rate. Obviously we need to know these values in order to determine if the background colour in the table for the given month should be red or green. This is why we have no choise other than calculate them ourself on basis of the raw data published by Eurostat. Otherwise we can not perform the compliance check every month, and will be kicked back only to update the table once every second year when ECB publish their official evaluation report. Alternatively we can also make the background colours on the basis of the last reference limits officially calculated by ECB, but as I have already proofed with my publication of the table of reference limits for 2012 and 2013 you can see such an approach would be highly inaccurate. So either we calculate the reference limits each month as I suggest we do for the "temporary template", or else we do not update the table at all (but only one time every second year when ECB publish their report). Again I think that when we have a RS for the raw data in the equation, then it can be supported we calculate these values ourself, for as long as we just continue to note that an uncertainty about the final reference limits exist (as currently indeed is carefully noted by the Criterion legend). I genuinly think many readers are interested to read our updated table, and the information value of the article would greatly decrease if we only show 1 or 2 year old data (and not the most recent data). So I think we should continue to update the template each month, as we have previously done in the past half year. Best regards, Danish Expert (talk) 20:48, 14 January 2013 (UTC)

Yeah, I understand. The problem is that we aren't supposed to publish unsourced facts. It could be argued that this is a routine WP:CALC, but it seems to be a pretty borderline case so I'm not sure what the best way to proceed is. TDL (talk) 22:25, 15 January 2013 (UTC)

Country specific parameter

After thinking about it, these historic details might be notable on individual countries articles (ie Sweden and the euro#Status) to show their progression towards fulfillment of the criteria. We could have one row for every year or something. My thought is to use these existing tables, but add an optional parameter which allows us to output only the country of interest's row. This would also allow us to sync the individual country's articles with the main table as well. What do you think? TDL (talk) 07:46, 14 January 2013 (UTC)

Yes this would indeed be a good idea (if possible), as I noted most of the "country page" stats are never/seldom updated. My only uncertainty about your proposal, is how to write the code. I know its possible to implement a country parameter, so that it would work in the way of showing titles cells plus the specific "country row" being selected by the parameter. But then we still would have the title row listed below for "candidates" and "potential candidates" which we do not really want. Perhaps a workaround can be implemented where these seperation rows are not coded with a ! but as an ordinary | row (with background colour matching the ! color)? And then we should of course also remember to code for the special situation if the country specific parameter input is being set to ALL (or not being set), where obviously the entire table in such situations should be displayed (which we need for our two main articles about the subject). I will not have time to work with implementing your proposal for this new parameter code, during the next month. You are welcome to do it, if you have more time than me. Danish Expert (talk) 10:47, 14 January 2013 (UTC)

Agreed, you put a lot of work into keeping this main template up to date, so it would be great if we could repurpose the data for the country specific articles. I can do the coding, but it might take me a few days to get around to it. TDL (talk) 18:23, 14 January 2013 (UTC)

Thanks, I will appreciate your help on this area and think your proposal is great. Danish Expert (talk) 20:48, 14 January 2013 (UTC)

I finally got this done. You can check out Sweden and the euro#Status for an example. It would be nice at some point to create one of these tables for each of the previous ECB reports so that we can add the data to each state's article to show how they've progressed towards the targets. However, that will take a lot of time... TDL (talk) 02:58, 28 January 2013 (UTC)

Beautiful. Good work. I have just also moved all parenthesises with an exact "date stamp" for the data, so that it is now also displayed by the country specific table rows. If possible (for layout reason), I would prefer if we only in the "country specific table" can merge the "Year" column and "Country" column into basically just a "Year" column. When the country specific article already say the article is about the convergence and euro adoption for the specific country being covered, I dont think it is needed to repeat the country name in each row of the table. If possible, I would instead prefer only to have the following cells+data listed:

Year
Reference limits
Apr 2012
Reference limits
Jan 2013

Is it possible to show the first column of the country specific table like that (without changing the layout of the total table)? If this indeed is possible, I think it would be awsome if you also condence the first column of the country specific table to this improved layout. When we have the layout entirely settled and fixed, I can promise you that I will gradually start to create the historic tables. We first need a final master version, before I start the copy machine and work out the historic ones. :-) Danish Expert (talk) 15:19, 28 January 2013 (UTC)

Ah yes, good point about moving the data date down. I hadn't thought about that. As for merging the columns, yeah that makes a lot of sense and should be not that hard to implement. I'll make the change when I get a moment. I'm trying to get the table functioning at first, but once we're happy with how it's looking I plan on offloading most of the nitty-gritty formatting stuff to a new template that can be called by each of the yearly templates so that all this stuff doesn't have to be hardcoded into each version of the template. TDL (talk) 20:25, 28 January 2013 (UTC)

OK, everything is offloaded to other templates, so generating historic tables should be much simpler. Any time we want to change the layout of the tables, we can just modify one of the other templates and it will apply to every version of the table automatically. TDL (talk) 03:14, 18 February 2013 (UTC)

New subarticle to show all prehistoric official "convergence check" templates ?

My previous proposal was to store an example for each year, showing the Convergence check as of April/December. Per your previous WP:NOTSTATSBOOK argument, I am ready to reconsider how often we should permanently store the data in dedicated templates. First of all, I however want to point out that the wikipedia policy allow for STATS to be included in articles for as long as some explaining article text is also provided to put the STATS into context. This is exacly what I had also done in the previous version of the euro convergence criteria article, so I do not agree with you that it was appropriate to remove it all for good (per WP:NOTSTATSBOOK). For the sake of other comparative examples, I will also mention to you that a plethora of sub-articles mainly comprising stat-tables already exist as a wikipedia standard for country specific football articles (Seasonal league stat tables and/or all historic match statistics for national football teams). This is just to illustrate that sometimes it is indeed okay to create subarticles with historic stats at Wikipedia. However, I do agree with you that we should be carefull not to have a too long main article with all pre-historic data. So one solution could be, that we only show the latest Euro convergence criteria template in the euro convergence criteria article, and then create a sub-article featuring all the pre-historic tables (based on ECB's official evaluation reports, being published every twice year or on the request of a member state). If we do it like that, we will always have a RS from ECB for the permanently stored data, and we will at the same time ensure that readability of the euro convergence criteria article is not destroyed. In my point of view, it is notable to keep a version of all old official "convergence check" tables, in the form of available sub-articles for readers both to see at what size reference limits were in the past and which criteria were the most difficult for previous applicants to comply with. Perhaps if we implement your proposal of including country specific code and list all pre-historic country specific data at the country specific pages, then you might have point we do not need also to have pre-historic sub-articles with the entire prehistoric template being visible. But on the other hand, as per your launched idea above the template would either way then already exist (for the purpose of listing country specific info at country specific pages), and so it would only be a small step also to let them be entirely visible with all countries at a pre-historic sub-article (for those readers seeking this kind of overview). I look forward to your response in this case. My great contributions and efforts to improve both the Euro convergence criteria article, Enlargement of the eurozone article, and the relevant templates in concern, deserve to be praised and should not be hauled out as something being completely out of the field. Danish Expert (talk) 10:47, 14 January 2013 (UTC)

See WP:OTHERSTUFFEXISTS. Just because something is done elsewhere on wikipedia, doesn't make it correct. I'm not fundamentally opposed to an article discussing the historic values, however I think it will be difficult to find independent sources to demonstrate the WP:NOTABILITY of the subject. My suggestion would be to start a discussion at the Wikipedia:Notability/Noticeboard to see whether other editors think the topic is notable enough to warrant an article. That would save you from going through all the work to create the article, only to have it deleted.

Again, please don't take offense that your work has been reverted. That happens to everyone here. I'm not trying to criticize you personally , we just disagree on content. TDL (talk) 18:23, 14 January 2013 (UTC)

If you think we should submit it for a notability check, then please go ahead as you suggested. As a first case example of a pre-historic subarticle, we already have the section you deleted: Fulfillment of criteria in April 2012. Instead of having one subarticle per template, I would be ready to stack the templates together in the one and same subarticle (if you prefer it like that). This would mean we would only have one historic subarticle featuring all 11 official ECB templates from: 1996+1998+2000+2002+2004+2006(May)+2006(Dec)+2007+2008+2010+2012. If you are concerned about space and/or the degree of details, then I am ready to accept that we cut down all historic templates only to show data-lines for the non-eurozone EU members (while only allowing for the current data template in the main article to feature data for the candidates and potential candidates).

In regards of notability sources, I can tell you that each of the countries that currently aspire to comply with the 5 criteria for euro adoption in fact had articles published each time around the time an official ECB status report was published, where the journalist referred to the report and provided the data for how far the country was from applying with the 5 criteria for euro adoption (In example, I can give you this notability proof for Bulgaria in 2008, to document that I am correct about my notability claim). So we indeed have notability for this subject. As I mentioned above, the Template for April 2012 is basicly also functioning as a seasonal scoreboard for the convergence criteria, which in principle can be compared with a football leagues seasonal scoreboard. I insist this comparision can be made, and that many people will be interested to check the old data score boards. I am 100% sure we can find a Wikipedia policy which provide full support for these seasonal stats to be posted, as long as it is part of a broader context and comply with the notability criteria. Qualification for euro adoption is indeed a part of a broader context (providing info about the expansion prospects for the eurozone), and of notable interest both for the specific countries in concern and other readers who is interested to follow the expansion prospect for the eurozone. The historic stats help to show how far the countries are from qualifying with the 5 criteria, and give a fast explanation for the areas being most problematic for the countries, which they need to focus on in order to comply in the future. The historic stats will also reveal that most of the applicants were pretty close to qualify for euro adoption in May 2008, and then subsequently got their convergence exploded far away from compliance due to the Global Financial Crisis kicking in around September 2008. This is why the historic stats indeed are interesting and notable, as they help to put the eurozone expansion (and its forward going speed and progress) into a broader perspective. Danish Expert (talk) 20:48, 14 January 2013 (UTC)

I doesn't bother me whether you seek input on the notability of the topic before creating the article or not. I was simply advising you of that course of action because I suspect that such an article could be deleted due to a lack of notability. I was only trying to save you the hassle of going through all the work of creating the article, only to have it disappear. Proceed as you see fit.

My concern is really only with the main article, and keeping it from becoming cluttered with stats, so go ahead and structure the historic article as you like. I'd agree that a single historic article would be better than one for each year though. For the templates, I think the best approach is to have the most current version at this title, and name the historic ones with a date in brackets afterwords like you did for 2012. Then, every year we can copy this table to the "(XXXX)" template and update this version. That way, all the links to this table are always up to date without having to change any links. TDL (talk) 22:25, 15 January 2013 (UTC)

UK HICP Inflation Rate

Forgive me for being a bit dim but if the Maximum allowed HICP Inflation Rate is 2.8% surely the UK currently satisfies this (being 2.8%). Spudgfsh (talk) 19:17, 18 February 2013 (UTC)

[This is a reply written 1s before the respond by TDL below - my phone rang while writing - I wont cheat you from my reply though - so it his hereby moved a little bit up in the rank]. Thanks for your note. It is a temporary coding bug in the related Template:Euro convergence criteria row. TDL is the great master behind all the coding stuff on the template. We should applaud him and be very greatfull for his untired work to improve and automize the table. I am sure he will also soon fix the coding bug. It is because the code say "if <" and it should instead say "if <". I have forgot the ASCI code for the sign, so we have to wait for TDL to help fix it. I am sure he will within a few days, as he is very active and a great coder. Best regards, Danish Expert (talk) 20:33, 18 February 2013 (UTC)

Good, point, That's due to the difference between a "<" and a "<=" in my code. Although technically the UK could be 2.81%, while the limit is 2.79%, in the absence of more precise data we should probably mark it as green. It should be fixed now. TDL (talk) 20:32, 18 February 2013 (UTC)

Thanks a lot for all your good fast work here. I will just briefly chime in to confirm that the ECB method explicitly states for the HICP convergence evaluation that: "ECB will always perform rounding offs to only one digit after the punctuation for the HICP figures, and that any rounding-off advantage will always be granted to the benefit of the country". ECB perform their evaluation by calculating the 12m-average of monthly recorded annual HICP figures based on raw data for the HICP indexes. The raw data (as showcased by the HICP index) carry 2 digits after the punctuation, but the final result published both by Eurostat and ECB in regards of "monthly recorded annual HICP figures" and their 12m-average will always be rounded off to only 1 digit after the punctuation, in order to reflect the uncertainties involved when the statistical agencies are determining these inflation figures. So in January 2013 everything from 2.800-2.849% will be green in the eyes of ECB. Danish Expert (talk) 21:04, 18 February 2013 (UTC)

In regards of the "interest rate convergence evaluation method", we might also have the same "rounding-off benefit rule". But as far as I remember, the ECB method unfortunately has not explicitly stated anything about such rule being in force for the interest rate evaluation, so this will for the moment remain as an unclarified "method uncertainty". The reaseon why we can not be 100% certain about ECB's applied procedure for interest rates, is that we here actually have no statistical uncertainty justifying why ECB should round-off the figures from 2 digits to only 1 digit after the punctuation; but despite of that ECB have chosen to display the figures with only 1 digit in all their published convergence reports. As ECB only publish the figures with 1 digit after punctuation, I think we can reasonably assume the same "rounding-off benefit rule" also will apply for the "interest rate" evaluation. In order to be true towards the official ECB method, I think we should also start to round-off the interest rate figures in the table. I will do that, the moment we get the February data released. Danish Expert (talk) 10:12, 19 February 2013 (UTC)

Valid comparison?

Is this table supposed to show which states currently fulfill the criteria? If so, it has to compare the current values and the current reference value. If it uses an old reference value instead, the current values would still be correct, but the color showing whether the country fulfills the criterium may be wrong in some cases. Ambi Valent (talk) 22:08, 16 July 2013 (UTC)

Agreed, I've reverted to the April numbers to sync with the reference values. Sometimes people get over eager to do half an update... TDL (talk) 03:14, 17 July 2013 (UTC)

Debt to GDP

in the title row it states "declining" while with concerned countries it says "decreasing". Someone who knows how to may change this. --134.176.205.44 (talk) 15:23, 4 December 2013 (UTC)

 Done - Thanks for the heads up! TDL (talk) 18:28, 4 December 2013 (UTC)

Sources for deficit and debt-to-GDB

Which sources are you using for the Budget deficit to GDP and Debt-to-GDP ratio? TDL (talk) 22:05, 13 February 2013 (UTC)

During the first quarter of the year, the source will always need to be the latest Economic Forecast report published by the European Commission (which obviously "only" is a forecast - but as all previous forecasts have revealed to be fairly accurate - this "forecast" uncertainty for our "fiscal data" is something I consider to be acceptable). This mean, that we are currently having the European Commission's Nov.7 report as source, and that it soon will be replaced and updated by the superceeding Feb.22 Winter Forecast report. The moment Eurostat releases their officially recorded 2012 data on 22 April 2013, we shall of course at that point of time get rid of the forecast reports, and for the remainder of the year only use a link to the Eurostat database. For the historic archiving of data, it is however necessary that we at some point of time ultimately again remove the link to the the Eurostat database, and replace it with the "Economic report" being published with the "last year fiscal data" as they were reported at the time of the data also being displayed or evaluated. This is per the official wikipedia data policy, and also per the official ECB data policy, that the values applying for evaluation purposes shall always be reported as how the data looked at the point of time of evaluating, with subsequently revised figures not being allowed/able to rewrite the previous assement history, but only able to found the basis for new assements in present time. Danish Expert (talk) 04:03, 14 February 2013 (UTC)

Right, so we don't actually know whether these criteria are presently satisfied or not. You can speculate that based on past forecasts that it's probably a good guess, but we certainly need to indicate that this is speculation based on a months old forecast. TDL (talk) 06:55, 14 February 2013 (UTC)

I agree (although by principle I have to highlight here at the talkpage, that the figures are not "just" a forecast - but more correctly should be referred to as an qualified estimate, as the figure currently is based upon recorded data for the first three quarters of the year, and only for the remaining fourth quarter the figures have been added according to the government's fiscal budget according to the agreed budget law - only being subject to some minor potential forecast adjustments). But okay, I can accept we call it a forecast rather than an estimate (to reflect that it is not final figures). You did a good move their to add a note about it. Thanks. Danish Expert (talk) 12:57, 15 February 2013 (UTC)

Why the values in the template do not correspond with the given Eurostat source [1]? E.g., for Hungary, the Budget deficit to GDP is 1.9%, while 2.0% in the source. The Debt-to-GDP ratio is 79.2%, while 79.8% in the source. 78.139.0.254 (talk) 20:30, 9 December 2013 (UTC)

New values for 2014?

Will anyone put 2014 data into the tables soon? I would have done so long ago if it wasn't for OR (I would have to calculate the reference values myself for HICP inflation and interest rate, where I have the values from Eurostat, and I would have to take prediction values from the newest EU forecast from February 2014 for debt and deficit). It's a bit frustrating that I'm not allowed to and the ones who have the information I miss won't do it either - so that it's now being outdated by nearly a year... Ambi Valent (talk) 14:30, 24 March 2014 (UTC)

It comes down to sourcing. If you can source the new reference values, then by all means do an update. But if we can't source the reference values, then we shouldn't be guessing at what they are. If that means the data must be out of date by a year, then so be it. There will be a new report put out by the ECB/Commission in May, so we will definitely have an update then. Prior to that, Danish Expert found reports published by the Polish Ministry of Finance here which contain their estimates of the reference values. So this would be a viable source for updated reference values. We discussed how to do this at User_talk:Danish_Expert#Follow-up_on_the_ECC, but quite honestly I haven't found the time to get around to it yet. But if you'd like to tackle the update, feel free. TDL (talk) 18:55, 24 March 2014 (UTC)

There are numbers from the Bank of Lithuania here: http://www.lb.lt/convergence_criteria_and_their_implementation Do they count? Ambi Valent (talk) 00:34, 26 March 2014 (UTC)

Unless you protest, I would use the following numbers: for HICP inflation and long-term interest, the monthly Eurostat numbers (as before). For reference values on these, the numbers from the Bank of Lithuania. For deficit and debt per GDP 2013, the EC winter forecast numbers (those were used for the candidate states, for the EU states they'd be replaced when the final new numbers are published at the end of April). Would that be okay? Ambi Valent (talk) 13:01, 26 March 2014 (UTC)

The Bank of Lithuania source is another option. I'd say the Polish source is a bit better because:

• a) it's published every month so we could keep it up to date. THe Lithuanian source seems to be only updated avery 6 months or so. And I suspect that once they get approval to adopt the euro in a couple months they will stop updating it.
• b) the sources stay alive. For the Lithuanian source, they keep overwriting the data, so it makes WP:V difficult.

The Polish MFA sources are a bit more complicated to decipher, but if you look at a sample Polish MFA report [2] you just need to look at "Wykres 1" ( Figure 1) on the lower left corner of the first page. The black line and numbers are the reference values. Then it's just a matter of determining what month they are valid for. That's the very first line of the report that says: "W styczniu 2014 r" (In January 2014).

Otherwise sounds good! TDL (talk) 18:25, 26 March 2014 (UTC)

I'm sorry, but I won't use the Polish numbers. I was already sceptical since I couldn't follow what exactly was written in the report, but I found it also gives me wrong numbers. So unless you find me correct reference values that could be computed in the prescribed ways from the Eurostat numbers and that I'm allowed to use, or I'm out of this. (The Lithuanian numbers agree, but if that's not good enough for you...) Ambi Valent (talk) 21:01, 28 March 2014 (UTC)

Well, unless I'm missing something the calculation works for me. In February 2014, the Polish MFA gets HICP=1.3% which is identical to the Latvian Bank. Doing the math myself, I get (-0.3% [BG] -0.2% [CY] +0.0% [LV])/3=-0.166%. That gives a reference value of -0.166%+1.5%=1.33% which is within rounding of both sources. (In January, the PMFA got 1.6% which again agrees with (0.0%+0.1%+0.1%)/3 + 1.5% = 1.566%.)

For the LTIR, the PMFA gets 5.5% and the LB gets 5.4%. Unfortunately, Eurostat doesn't report a LTIR for Latvia for February, so it's not clear why there is a discrepancy. Presumably they are using preliminary data for Latvia that differs slightly.

But if you'd rather wait until the ECB report comes out next month for an update, that's fine with me. TDL (talk) 20:11, 19 April 2014 (UTC)

Excessive deficit procedure column

I think the choice of deadline for EDP is not the one the EU uses: http://europa.eu/rapid/press-release_IP-14-627_en.htm

While the chosen deadline is the same as for the HICP inflation and interest rate, it appears that the EDP only needs to be derogated before the final Council decision about satisfying the Convergence Criteria. I think with the decisions happening in July, and EDP derogations happing in June, the column value would be consistent with EU decisions if e.g. 30 June was chosen as deadline instead. Ambi Valent (talk) 15:29, 6 July 2014 (UTC)

Yeah I wondered about using a different date for the EDP. The problem with that is that it gets a bit confusing if we use different dates for different columns. I think it's better to have data that is all valid at a given snapshot in time (ie the date when the ECB report comes out). Also, say we update this table in 6 months, then what date do we use for the EDP? Still HICP/LTIR + 2 months even though there will be no derogations in the window? It seems a bit arbitrary.

What if we put a note in the cells of any state who had an open EDP on April 30 but received a derogation before 30 June? TDL (talk) 18:51, 6 July 2014 (UTC)

I've indicated states which subsequently had their EDP abrogated with yellow cells. Let me know what you think. TDL (talk) 03:48, 7 July 2014 (UTC)

Shall we completely kill the "current data" concept for the convergence criteria table" ?

One year ago we had a long unsolved dispute about this question. My personal argumentation was, that despite of having acknowledged this was in the very grey area of what the Wikipedia policy could allow, we could from a pragmatic standpoint approve a future use of "estimated reference values" published on a monthly basis by the Polish Ministry of Finance, as the illustrative data input for the "current reference values" in a "current data" table. Although (as I put it back then), this should of course be followed by a big fat disclaimer note emphasizing the Polish estimated reference values might be wrong, simply because it build on unverified assumptions/predictions about which states would be categorized as outliers and thus disregarded as part of the three states to be used for calculation of the reference limits. Plus, I argued back then, and still argue, that for the same reason, our pragmatically approved "current data approach" for the current table, should never push us to allow for the "Polish estimated reference values" to be used for the "historic convergence tables" (having a need strictly and solely to be sourced by the EC convergence report).

As per the comments in the New values for 2014? debate between TDL and Ambi Valent during April 2014, it now appear, that last years resistance for my proposed "current data" idea now vanished. In our most recent "layout debate", which entirely was about the "historic convergence tables", I on purpose delayed the debate about "current data", to be dealt with only by this subsequent dedicated debate.

If we decide completely to drop my previously suggested "current data" concept (as I summarized by the paragraph above), it would mean we only allow for future updates of the convergence table once a new EC convergence report is published (which is only likely to happen next time, in June 2016).

For your consideration to this question, I can inform the "Polish estimated reference values" are calculated by the Polish Ministry of Finance, under the assumption, that starting from April 2014 all states with a HICP criteria value recorded to be equal to (or more) than 1.4% below the corresponding HICP criteria value for the eurozone, automatically will be estimated to be an "HICP outlier". In the past, the Polish data approach was the same, except their enforced "outlier cut-off limit" had been set to be 1.6% (based on the 2004 EC convergence report, where Finland with a HICP criteria value of 1.7% below the eurozone average had been deemed not to be an outlier). The tricky part here, is that strictly speaking its incorrect to assume that all states on the wrong side of the "significant" limit automatically will be classified as outliers. ECB+EC have clarified in their earlier convergence reports, that in order for a state to qualify as an "HICP Outlier", it need in addition to posting a "HICP criteria figure" being "significantly below" the eurozone average, also to receive a judgement by the European Commission that this "value significantly below the eurozone average" has been caused by "exceptional circumstances". Hence, this is likely why Finland being 1.7% below the average was found not to be an outlier in 2004, while Cyprus being 1.4% below the average was deemed to be an outlier. So its undisputed, that the "Polish estimated reference values" can never be regarded to be fully accurate - as they are only observational estimates, and of course they do not know how the additional EC assessment for "findings of exceptional circumstances negatively affecting the HICP price index" will turn out in the next yet-to-be-published EC convergence report (which is the only source of stone-written reference values). As for the possible identification of "interest rate" outliers (to be retracted from calculation of the interest rate reference limit), the Polish approach is just to assume, that the same states found to be outliers in the latest convergence report will continue to be outliers at least until the time of the next convergence report being published (so again this is also just an estimate - which might from time to time be incorrect).

With all these reservations about the accuracy of the "Polish estimated reference values", the question of course is whether or not it is appropriate to allow for the "current data" concept (based on the Polish estimated reference values, but accompanied by a big fat disclaimer note informing readers about its potential inaccuracy), to be applied here at Wikipedia? The argument for, is that it is better to have "current data" in a table with estimated reference values, rather than have "absence of current data for a 2yr long period". The argument against, is that the "estimated reference values" can be wrong - and that despite adding a big fat disclaimer note informing on this - the "estimated reference values" would perhaps by casual readers be misunderstood to mean they were "factual reference values". My own drive for this "current data" concept has cooled down a lot lately, to a level where I am 50% for and 50% against. So to say it short, I will leave this question open for all other editors to decide. Best regards, Danish Expert (talk) 11:51, 10 September 2014 (UTC)

I have no objections to filling this table with current data, as long as the reference values are sourced to the Polish MFA (or some other equally reliable source). As long as we have some disclaimer as you suggest that says that these are estimates and not official values from the EC, I don't see a problem with using them even though their methodology might not be exactly the same. Ideally I think it would be nice to keep it updated, but it's a big job to update every month so I'm fine with leaving it as is as well. TDL (talk) 05:38, 12 September 2014 (UTC)

OK, after sleeping long time on the question for a week, I now tip to the side that we should keep it as it is for a year. The official June 2014 assessment still holds for the 2 fiscal criteria - more or less without changes - up until next summer, and when we switch to "current data" I think we should also place NA inputs for all those of the values not sourced as of the applied "current data assessment month" (i.e. NA for all "HICP sustainability", "Severe tension" and "Legal compliance"), as we would not know if these would still have the same values as the Commission found in their previous official assessment. For these reasons, I think its best to keep our "2014 historic template" displayed for a year, in the format sourced solely by the 2014 EC convergence report. However, after 1 year it will be relevant to switch to the "current data" table format - as both of the "fiscal criteria" will have significantly changed (plus we no longer need to flash on the main page how Lithuania qualified in the past). If nobody else steps in, I will stand ready to help implement the considered "current data" format in June 2015. Danish Expert (talk) 11:59, 17 September 2014 (UTC)

That seems like a reasonable compromise to me. TDL (talk) 21:32, 20 September 2014 (UTC)