REAL COSA 2020: A new model

[Miestră Schivă, UrN-GC]

The Free Democrats will be going into the next election on a platform of having a Senäts Committee agree on a compromise proposal for reform of the Ziu and the electoral systems of its two houses. Notwithstanding this, we would like to suggest for consideration a new format for a "Real Cosa", as existed in Talossa 1997-2002. I have @MPF to thank for the inspiration here.

The essential points would be:

• A "circular" party list. Seats would now be allocated automatically by the SoS, one each to every candidate on the list, in order, and when the list comes to an end, it begins again from the beginning. Example: a party which has 5 members on the list but wins 8 seats would end up giving 2 seats to the first 3 people on the list, and 1 to everyone else.
• The current "1/3 allocations off list" would be changed to "a party may add 1/3 extra members, rounded up, to its list after the election". If your original list had 5 people on it you could add another 2 later. If you only had 1 person on your list, you could only add 1 later.
• Existing provisions barring MCs from individually holding more than x% of the Cosa would still exist, meaning there would be a limit to how many seats could be held by a 1-person party.
• Seats would be allocated by a rounding system known as Webster/Sainte-Laguë, which benefits smaller parties.
• The basic idea is for a 20 seat Cosa, which would mean that - based on the votes from the last election - a party would need at least 3 votes to win any Cosa seats. If this threshold is considered to stringent, the model would work well with more MCs. If one or more parties were "tied" for the last seat, they would all get a seat and thus the Cosa would be slightly expanded.

Based on this, the results of the last election would have been: FreeDems 8 seats, AMP 5, ZPT 3, PNP 3, NPW 1, MTGA 0. I can recalculate it for a larger Cosa if anyone's interested.

What do you think, sirs?

[Açafat del Val]

I don't see a practical benefit to, nor a practical difference of, a smaller-number Cosa if we still allow members to hold more than one seat at a time. In reality, by shrinking the seats but allowing plural seat-holding, we are really just excluding smaller parties as well as overall political participation. I think that one-person parties are a GOOD thing in the context of Talossa, and that they make Talossa a greater democracy.

If we want a "real" Cosa - which I support, to be clear - then we have to get rid of plural seat-holding. One seat, one member.

If that is unpalatable, then we should stick to the current EM200 method.

The half-measure of reducing seats but allowing plural seat-holding does more harm than good.

[esbornatfiglheu]

I'm with Açafat on this.  Let's bring the number of seats down further, on a 1 person 1 seat to fill.  My suggestion would be 15. 

A desirable side effect of this could be that it deflects activity downward into the provinces, which I know has been something under discussion.

[Miestră Schivă, UrN-GC]

I have no issue in principle with a 15-seat Real Cosa, if you have no problem with at least a 4-vote threshold for gaining Cosa seats. (The numbers from the last election would have been FreeDems 6 seats, AMP 4, ZPT and PNP 2 each, NPW 1).

As to the no-double-seats issue, again I'm happy, if you're happy with parties not being able to recruit enough warm bodies to fill seats and thus vacancies. The circular-list idea I still think is a good solution to the not-enough-warm-bodies problem.

[esbornatfiglheu]

I think the "warm bodies" test is an important one.  If a party can't get enough bodies for their seats, they've got a snowball's chance in Phoenix of being able to govern.

It was the problem that assailed the first MRP government in the Republic, as an example.  Not enough folks to keep the lights on.

[Miestră Schivă, UrN-GC]

The problem is that if you need warm bodies to fill all seats, and you just don't have them, there's a lack of representation which distorts the election. Look what has happened to this Cosa when the ZPT didn't fill their seats. I don't disapprove of having no Right-wing opposition in practice, but I do in principle because it impedes the democratic legitimacy of the Cosa when the Right aren't represented. For that matter, I think even the AMP would have struggled to find 4 warm bodies since their leader sits in the Senäts.

(Parenthetically, the warm-body problem is one reason I have been sympathetic to the idea of a unicameral Ziu, although I acknowledge that the Senäts is more or less doing its job right now)

[Açafat del Val]

We could amend the OrgLaw (or maybe we're due for a new OrgLaw altogether??) to allow for this 'Living Cosa' and then just leave the number of seats to be fixed by statute. Then we can increase or diminish the seats as needed and, if really needed, dissolve the Cosa prematurely so that we can clear out the cruft (so to speak).

Not enough warm bodies? Decrease the number of seats, then dissolve the Cosa for early elections.
Not enough seats? Increase the number of seats, then dissolve the Cosa for early elections.

Having said that, what will we do to ensure that one-man parties remain viable? Not to beat a dead horse, but I quite like that Mximo (and others) can serve in the Cosa all on their own merit.

[Miestră Schivă, UrN-GC]

I am personally in favour of a Cosa size that can be set by statute (with a proviso like the Australian constitution, that it be no smaller than twice the size of the Senäts)

[esbornatfiglheu]

Setting the size by statute makes perfect sense to me as well.

[Béneditsch Ardpresteir]

Another practical difficulty, but maybe for the good, might be that a winning party would probably have to have hired hands to work as ministers, for not all 10+ members might not be active enough to do the duties of the presently 8 ministerial roles. In a way, practically all elected members would be ministers.

[Miestră Schivă, UrN-GC]

Well and good; but a Cosa of variable size needs to have an electoral system which can handle various sizes, either Real or the current system. I submit that the "circular list" proposal can handle any size of Cosa.

[Miestră Schivă, UrN-GC]

Okay, here's some model legislation to bring in this model.

Amendments to OrgLaw IV

Section 1 The Cosa is the national legislative assembly, and is composed of a number of seats apportioned among political parties based on their performance in the General Election. This number shall be 15 unless otherwise set by law, and shall be no fewer than twice the number of Senators, minus one. The Coså may administer itself as it sees fit.

Section 4 Vacant seats occurring between elections shall be filled in accordance with law.

Amendment to El Lexhatx B.2.3

2.3 The ballot must also include, for each party contesting the election, a ranked list of citizens to whom the party intends to award Cosa seats. If a party does not submit a candidate list to the Secretary of State before the election, the party leader shall be considered the only person on that party's list.

2.3.1. Pursuant to Organic Law IV.2, the Secretary of State shall apportion seats to parties on the basis of their vote totals by use of the Largest Remainder Method, using a Hare quota (https://en.wikipedia.org/wiki/Largest_remainder_method).

2.3.1.1. If one or more parties are "tied" for the last seat or several seats, then the Secretary of State shall be entitled to assign one seat to all of the said parties and therefore to temporarily expand the size of the Cosa to more than the number otherwise set by law.

        2.3.2. Up until 10 days after the opening of the First Clark, a party leader may amend their list by adding names to it, in ranked order after the original names. The maximum number of names that may be added is 50% of the number of names originally on the list, rounded up.

2.3.3. After all lists are finalised and before the conclusion of the First Clark, the Secretary of State shall announce the distribution of each parties' seats as follows:

2.3.3.1. One seat will be given to each candidate on the list, in ranked order.

2.3.3.2. If all members on the list have received a seat and there are still additional seats, the process in 2.3.2.1 shall be repeated until no more seats are available.

2.3.3.3. If any person assigned a seat as above either
- already has the maximum number of Cosă seats allowable;
- declines their seat(s);
these seats will be reallocated according to the criteria above. 

2.3.3.4 If a party cannot assign all of their seats under the criteria above, the additional seats are forfeited.

2.3.4. Any vacant seats occurring between elections shall be reassigned according to the procedure in 2.3.3 above.

[Miestră Schivă, UrN-GC]

I should not that I'm not married to the Saint-Laguë votes->seats method, but that's what we use in New Zealand. Perhaps a largest-remainders method with a Hare quota would be closer to what we're used to in Talossa: https://en.wikipedia.org/wiki/Largest_remainder_method. But I do think we need a recognized mathematical votes->seats formula with a smaller Coså that, if possible, gives a small advantage to smaller parties.

Actually, you know? I've run the numbers and it turns out that "the largest-remainder method with the Hare quota" gives exactly the same results not only for a 20 and 15 seat Cosa, but of the actual calcuations that were made for the last Cosa. The only difference is that a 15 seat Cosa using this method would mean you'd need 6 votes to get a seat, as opposed to 5 votes with the Sainte-Laguë system. So, with that in mind, I think I'll officially counsel use of LRM+Hare, for the sake of simplicity and continuity, given the warning about the vote threshold.

(One way around the vote threshold would be to simply declare that "the total number of Cosa seats shall the lowest number so that the Hare quota is less than X". So, if we just declared a 2 vote threshold, the last election would have distributed 40 seats; a 3 vote threshold would have meant 27 seats, etc.)

I should also point out that, given Lex H.14 as it stands, the maximum seats/person would be 2 with a 15 seat Coså and the same turnout as last time. I have no real problem with that.

[Miestră Schivă, UrN-GC]

Those calculations in full