Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.
Display:
The solution is double voting and an additional-member system.

Give states seats in the parliament proportionally to the square root of population (Penrose rule). Then complete the EP with seats to be assigned to EU-wide party lists for overall proportionality.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.

by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 09:54:55 AM EST
[ Parent ]
Yep to the EU-wide lists. Even electing 22 MEPs is hard on the limit of sensible; but say Ireland's 12 or the even less for all the sub-national constituencies just reduces the expression of democratic will. As for the Penrose rule, that's too far from proportional for me; maybe with a lower power.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Sun Jul 19th, 2009 at 10:11:25 AM EST
[ Parent ]
Proportionality is achieved by the additional members form EU-wide votes.
ceiling(sqrt(EU.population)/200)
    Germany	 France 	 UK	  Italy       Spain	 Poland 
	 46	     41 	 40	     39 	 34	     31 
    Romania Netherlands      Greece	Belgium    Portugal	Czechia 
	 24	     21 	 17	     17 	 17	     17 
    Hungary	 Sweden     Austria    Bulgaria     Denmark    Slovakia 
	 16	     16 	 15	     14 	 12	     12 
    Finland	Ireland   Lithuania	 Latvia    Slovenia	Estonia 
	 12	     11 	 10	      8 	  8	      6 
     Cyprus  Luxembourg       Malta 
	  5	      4 	  4 
for a total of 497 seats. Add to this another 200 or 250 elected on a single EU-wide list and you get a parliament of 697-747 which is comparable to the current 736. Iceland would have 3 seats under this scheme, bringing the total to 500 + 200 or 500 + 250.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 10:24:53 AM EST
[ Parent ]
Still. Don't. Like. It. Rather a lower power and less direct seats for ther minis. BTW, Andorra, Monaco, San Marino: 1, Vatican: 0.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Sun Jul 19th, 2009 at 10:37:05 AM EST
[ Parent ]
Still. Don't. Like. It. is not a very persuasive argument.

This is a parliament - you want to increase the number of possible different voices able to speak in it. Proportional representation along EU-wide ideological lines is preferable to proportional representation along national lines. What is gained by having 64 as opposed to 49 Germans? Not much in terms of representation of views. Whereas having 4 instead of 2 Maltese may be the difference between getting a single -party or a 3-party Maltese delegation. 2/3 of the seats on EU-wide lists will ensure ideological balance.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.

by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 10:42:55 AM EST
[ Parent ]
I mean 1/3 on EU-wide lists.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 11:39:52 AM EST
[ Parent ]
I wonder if a situation similar to overhang mandates in Germany could arise; more MEPs for one EP party from smaller countries than in the proportional vote.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Sun Jul 19th, 2009 at 11:44:48 AM EST
[ Parent ]
The additional-member system tends to give the nationwide seats to smaller parties which can-t make the threshold in constituencies. The parties that are over-represented in the constituencies are the largest ones.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 11:55:09 AM EST
[ Parent ]
However, the question is the possibility of over-representation overall; and in this case, the question is one that would come from the Penrose rule.

Silly example I calculated for myself: right-populist parties that join Europe of the Nations sweep all former communist countries minus Poland at 60% of elected MEPs, while sister parties fail to get on the ballot everywhere else. That's 69 MEPs, or 9.2% tof a total 747. Now, assuming similar ratios in the all-EU list vote, and using the 2009 population projections, Europe of the Nations would deserve 8.3% of the vote, or 62 MEPs.

So, the possibility is there, though I submit it is a minor effect.

*Lunatic*, n.
One whose delusions are out of fashion.

by DoDo on Sun Jul 19th, 2009 at 02:06:03 PM EST
[ Parent ]
Over-representation of small parties is not a problem. The real problem is a steamrolling majority from 40% of the overall vote.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 02:22:31 PM EST
[ Parent ]
The question in this subthread wasn't if it is a problem or not, but if overrepresentation is possible, and how the system would deal with it. Based on the above calculation, I say it is possible, and may be dealt with either with overhang mandates or by reducing the proportional share of other parties.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Tue Jul 21st, 2009 at 05:24:38 AM EST
[ Parent ]
Proportional representation along EU-wide ideological lines is preferable to proportional representation along national lines.

By that argument, why have national lists at all? Also, what do we do with sub-national constituencies: apportition seats according to the Penrose law, or proportionally? And won't this make elections in the smallest countries overly focused on the direct seats, as opposed to the largest countries?

*Lunatic*, n.
One whose delusions are out of fashion.

by DoDo on Sun Jul 19th, 2009 at 11:42:02 AM EST
[ Parent ]
By that argument, why have national lists at all?

Because there is actually an advantage to the combination of smaller local constituencies with a global list for overall proportionality. Electing 46 seats in Germany by proportional representation leads to someon on the #20 slot of one of the big parties being elected. Who actually votes with the #20 candidate in mind? An EU-wide list of 250 candidates has the same problem - someone in position #100 of the EPP or PES list is likely to get elected. Also, I would say that you want to have representation from all combinations of nation+party, even if it is only one from each. That's for giving views a voice. The vote is the overall proportionality.

Also, what do we do with sub-national constituencies: apportition seats according to the Penrose law, or proportionally? I would abolish them. Or institute a similar EU-wide rule where if a country wants to have subnational constituencies they have to use the same Penrose + country-wide scheme as is used in the EU as a whole.

And won't this make elections in the smallest countries overly focused on the direct seats, as opposed to the largest countries?

As opposed to making the votes from the smallest countries all but irrelevant? Iceland will have 1/1500 of the EU's population. Why vote in the EU elections if there are only 750 seats to be filled?

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.

by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 11:52:48 AM EST
[ Parent ]
Also, as was pointed out in my silly question about Iceland and Japan, "culture tends to follow geography and people." Even the very homogeneous USA has a Senate where regional differences are represented without regard to population.
by asdf on Sun Jul 19th, 2009 at 12:45:58 PM EST
[ Parent ]
However, the Senate is a separate body from the House. Just like the European Council or the Bundesrat is separate from the European Parliament resp. the Bundestag.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Sun Jul 19th, 2009 at 02:09:50 PM EST
[ Parent ]
Also, I would say that you want to have representation from all combinations of nation+party, even if it is only one from each.

That still won't apply for Malta. But, you could prescribe something like that for the parties.

As opposed to making the votes from the smallest countries all but irrelevant?

By that argument, why vote in any single voter district in a national election? For disproportional national representation, we could reform the European Council.

*Lunatic*, n.
One whose delusions are out of fashion.

by DoDo on Sun Jul 19th, 2009 at 02:18:06 PM EST
[ Parent ]
That still won't apply for Malta.

No, not in the "every combination" sense. But 4 seats is better than 2 in that respect. For Germany, it is not clear how 99 seats is better than 49.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.

by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 02:23:56 PM EST
[ Parent ]
Not "for Germany". For the single voter. Similar weight of vote and all that.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Sun Jul 19th, 2009 at 02:25:39 PM EST
[ Parent ]
Similar weight of vote comes from the additional member system. If your vote is overrepresented at the constituency level it is underrepresented at the additional-member level.

Let's start from the opposite end. I would vehemently oppose 736-member E_ wide party lists. People would usually vote paying attention to at most the top 2 people in each party's list, or the top one and the top compatriot. It really makes a lot of sense to have constituencies with a small number of seats (using transferable votes to ensure proportionality). But to insist on even-sized constituencies leads to redistricting nonsense including gerrimandering. So you want

  1. small constituencies with STV
  2. a sizeable number of seats elected on overall party lists
The Penrose method evens out the size of the constituencies.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 02:38:28 PM EST
[ Parent ]
If your vote is overrepresented at the constituency level it is underrepresented at the additional-member level.

Why do you think that doesn't make a difference? I haven't made this angle explicit before; but there is also the issue of ideological weights within EP-parties.

It really makes a lot of sense to have constituencies with a small number of seats

What about using the 97 second-level NUTS regions rather than nation states?

*Lunatic*, n.
One whose delusions are out of fashion.

by DoDo on Tue Jul 21st, 2009 at 05:40:41 AM EST
[ Parent ]
...I of course mean first-level NUTS regions.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Tue Jul 21st, 2009 at 05:49:09 AM EST
[ Parent ]
But the point of having additional members assigned at the EU level is that it will force the state-level parties to take a serious look at what their "fellow party members" in other states are like. If it works that way, it should force EP parties to form around political conviction instead of relative position in their domestic politics. That should prevent having EP parties that are total ideological crapshots.

- Jake

Friends come and go. Enemies accumulate.

by JakeS (JangoSierra 'at' gmail 'dot' com) on Tue Jul 21st, 2009 at 06:05:22 AM EST
[ Parent ]
I'm sceptical that that alone would achieve such a change. The 'national delegations' would act like party wings and old boy networks do in any party, and positions on the all-EU list would be haggled for.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Tue Jul 21st, 2009 at 06:26:22 AM EST
[ Parent ]
DoDo:

It really makes a lot of sense to have constituencies with a small number of seats

What about using the 97 second-level NUTS regions rather than nation states?

I would like that, but good luck getting the Member States to agree to it.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Tue Jul 21st, 2009 at 07:05:43 AM EST
[ Parent ]
I think significant resistance would come more in nationalist tones, than for practical reasons. For Belgium, France, Spain, Poland or Italy, the first-level NUTS regions would more or less correspond to the current sub-national EP election regions -- only the number of seats contested would change. For Germany, the NUTS regions are the federal states, thus the new system would bring it closer to what they have in federal elections -- also, much to the liking of the regionally strong CSU, I suppose. (Though, 18 million strong Northrhine-Westphalia still stands out -- the most populous NUTS area, unless I missed one of the Italian ones.) Many of the small countries are a single NUTS area, so again a potential change only in number of seats only. It would make a significant difference in the mid-sized countries, from Sweden to Romania (minus Belgium).

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Tue Jul 21st, 2009 at 08:21:48 AM EST
[ Parent ]
DoDo:
For Belgium, France, Spain, Poland or Italy, the first-level NUTS regions would more or less correspond to the current sub-national EP election regions
Except Spain or Italy don't do constituencies in the EP elections.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Tue Jul 21st, 2009 at 08:48:44 AM EST
[ Parent ]
I was mistaken about Spain, then, so the comment about Germany applies; however, Italy does have EP election sub-national regions.

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Tue Jul 21st, 2009 at 02:57:56 PM EST
[ Parent ]
I would abolish them.

With what argument? (Not that I wouldn't want to abolish them myself, but what is the rationale in your system?)

Or institute a similar EU-wide rule where if a country wants to have subnational constituencies they have to use the same Penrose + country-wide scheme as is used in the EU as a whole.

But a national level Penrose (apportitioning a number of seats as given by Penrose for the whole country), instead of EU level? With that, you are strenghtening the national character of the elections further.

*Lunatic*, n.
One whose delusions are out of fashion.

by DoDo on Sun Jul 19th, 2009 at 02:24:38 PM EST
[ Parent ]
I'm saying, if a country wants to have constituencies they have to have an additional-member system in order not to distort overall proportinality of the national delegation. The penrose-apportinment of the national constituencies is a footnote, but it is necessary when apportionment proportionally to population would lead to extremely large disparities in numbers of seats per constituency (as would be the case in Spain). It is actually an advantage of the Penrose system that it limits the size of the largest constituencies.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 02:28:46 PM EST
[ Parent ]
But, again, if that is your choice, you fix the system to countries, perpetuating "national delegations".

*Lunatic*, n.
One whose delusions are out of fashion.
by DoDo on Tue Jul 21st, 2009 at 05:27:15 AM EST
[ Parent ]
Vatican zero, eh? We could give them one seat, but only if it's filled personally by the pope. ;)

Peak oil is not an energy crisis. It is a liquid fuel crisis.
by Starvid on Sun Jul 19th, 2009 at 12:11:15 PM EST
[ Parent ]
Yeah, why shouldn't we have one Bishop speaking in Parliament? But not the Pope or a Cardinal (members of national parliaments cannot be in the EP so the Vatican curia members would be disqualified).

That way there won't be an argument against an EU-wide Islamic party. Or Bulgaria's Turkist minority party...

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.

by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 12:16:30 PM EST
[ Parent ]
Oh, it has to be the pope himself, in full regalia, or it won't count.

Peak oil is not an energy crisis. It is a liquid fuel crisis.
by Starvid on Sun Jul 19th, 2009 at 12:29:56 PM EST
[ Parent ]
He gets to go to Council meetings...

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 12:45:19 PM EST
[ Parent ]
I like this idea a lot and indeed it would essentially help to solve most of the problems I have with the entry of Iceland. Additionally it would be necessary to appoint commissioner for their capability and not a commissioner per country.

Btu As I have noted further down, as the number of countries in the EU increases it will become effectively impossible to change the institutional setup. That is why all those things have to be done before we extend the EU even further.  

by rz on Sun Jul 19th, 2009 at 12:26:18 PM EST
[ Parent ]
Here's "a lower power"
ceiling(exp(log(c(EU.population, iceland=300000))/1.5)/3000)
    Germany	 France 	 UK	  Italy       Spain	 Poland 
	 64	     54 	 52	     51 	 43	     38 
    Romania Netherlands      Greece	Belgium    Portugal	Czechia 
	 26	     22 	 17	     17 	 17	     16 
    Hungary	 Sweden     Austria    Bulgaria     Denmark    Slovakia 
	 16	     15 	 14	     13 	 11	     11 
    Finland	Ireland   Lithuania	 Latvia    Slovenia	Estonia 
	 11	     10 	  8	      6 	  6	      5 
     Cyprus  Luxembourg       Malta	iceland 
	  3	      3 	  2	      2
for a total of 553 + EU-wide list. The problem I see with allowing flexibility in the power is that you will then get endless bickering as well as not having a clean a priori justification for the choice (which Penrose's square-root does have).

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 10:37:54 AM EST
[ Parent ]
The probability of "endless bickering" is exactly why the two-house system works. You set up a rule: Proportional representation in one house and per-country representation in the other, and then you don't change the rule when some new tiny country is added.
by asdf on Sun Jul 19th, 2009 at 12:48:13 PM EST
[ Parent ]
The European Council operates as the second house - it is not unlike the "Federal Council" (even in name!) of Germany or Switzerland. I am not sure the second house needs to be directly elected, and in fact that was not the case originally in the US but the result of an amendment.

The peak-to-trough part of the business cycle is an outlier. Carnot would have died laughing.
by Carrie (migeru at eurotrib dot com) on Sun Jul 19th, 2009 at 12:50:24 PM EST
[ Parent ]
The point of direct election (or appointment, as in the original rules) is that there is a fixed pair of formulas: 1.) Population, and 2.) Per geographically separate unit.

A given combined formula (logarithmic or whatever) can by correct selection of parameters give the same--or close to it--result for a given set of data, but if the parameters can be fiddled politically, they will. Like rorting or gerrymandering or whatever, you always get it from a political system.

I have wondered whether this issue was at the root of the UK's handing over of some independence to Wales and Scotland, the idea being that with the current arrangement, Britain could potentially claim to have four distinct geographical units.

by asdf on Sun Jul 19th, 2009 at 03:21:21 PM EST
[ Parent ]
The swedish upper house (before 1970 iirc) was appointed by the regional and local assemblies. I think that was a good model as it prevented the creep of powers to the highest level by giving lower levels representation. Since it was the assemblies and not the local executives it also kept the seperation of powers (in contrast to the Councils make-up).

Sweden's finest (and perhaps only) collaborative, leftist e-newspaper Synapze.se
by A swedish kind of death on Sun Jul 19th, 2009 at 04:38:21 PM EST
[ Parent ]

Display:

Occasional Series