Fiji: “open” list PR, M=50

Fiji has a new constitution. Among the provisions is a new electoral system that will consist of a single nationwide district of fifty seats, using “open-list” proportional representation (OLPR).

By my reading of Art. 53 of the constitution, ((I thank Jon Fraenkel of the Victoria University, Wellington, for sending me a copy.)) it is not clearly open list, in the proper sense of the rank-order of candidates being determined solely by individual preference votes. Yes, it says “open list”:

53.—(1) The election of members of Parliament is by a multi-member open list system of proportional representation…

However, it does not go on to define what this means. A variant of the “flexible” (or semi-open) list system would meet the plain wording of the text. It seems to require that voters give a preference vote:

the total number of votes cast for each political party contesting the election, which shall be determined by totalling the number of votes cast for each candidate of that political party;

However, that does not preclude its being what I call (in my chapter in Gallagher and Mitchell, The Politics of Electoral Systems) a latent list, as in the Netherlands and Estonia. These are both variants of flexible list that require individual preference votes yet still have a fixed order that is overcome only by candidates who obtain some quota of preference votes.

If it really is an open list, Fiji will join Colombia as perhaps the only country to use such a system in a single nationwide district. There it is used in this manner for the Senate only (M=100); the House of Representatives is open list, but districted. ((Actually, in Colombia, parties may present either closed or open lists. Most have chosen open. See Pachon and Shugart in Electoral Studies, 2010.)) There are other cases of OLPR in districts at least as large as Fiji’s, including the Sao Paulo district for Brazil’s Chamber of Deputies (M=70) and the statewide districts several Brazilian state legislative assemblies (one of which has a magnitude of 120, I believe). But Fiji’s single district would be larger than the largest district in Finland (M=34, currently). OLPR in such large districts is problematic because of the overwhelming choice of candidates and challenges of clear ballot design. ((Brazil uses all electronic voting in recent elections.))

If it is flexible (latent) list, it would join Netherlands and Slovakia (both M=150) as cases of nationwide application. However, in the Netherlands, choice and complexity are somewhat reduced by the use of nominating sub-districts.

In any case, Fiji will be an interesting addition to the set of very high-magnitude preference-vote systems.

There will also be a 5% threshold for a list to obtain representation.

7 thoughts on “Fiji: “open” list PR, M=50

  1. Ahem.

    The constitution of Fiji is the constitution of 1997 which is temporarily in abeyance while a band of thugs purport to occupy various offices of state by force.

    The document described as a constitution is a decree by the junta. No referendum. No parliament. No constituent assembly. The opposition has rejected this document as a sham designed to continue the rule of the incumbent thugs. Ditto the international community and Fiji’s partners in the Pacific Islands Forum.

    All that being said there is a full copy here.

  2. If I trusted the Fijian “government” to play by normal rules, I wouldn’t mind such a system. While I would prefer STV over party list systems, this system is still far better than either single member districts or closed lists. Though I admit that I am still waiting for a catch.

    If I had to design a party list PR system it would work something like this:
    –Candidates, and not parties, form lists. Obviously parties would put out their own lists, but in theory all lists would just be a bunch of candidates pooling votes. Independents can run as well.
    –Electors vote for a candidate, any candidate. I could live with some form of “Short Ballot” that just lists the Lists and independents on a single page that can be requested in lieu of the full ballot.
    –Any candidate who receives a quota on his own is elected; whether he is a list candidate or an independent.
    –Votes for List candidates are then pooled as list votes. The highest N candidates from the list are selected for every N quotas the list receives. Highest being determined by number of votes and then list order if there are ties.

  3. I like Mark’s system, and I came up with a similar system, with the idea of using it in US Senate elections if the US ever went to a nationwide electorate for the Senate (which mean too many slots to be filled by STV).

    It was basically the same as Mark’s system, except that votes for the list were automatically attributed to the highest ranked unelected candidate on the list. Candidates were elected individually once their votes exceeded a electoral threshold (# of slots/ # candidates + 1). Votes they had accrued in excess of the threshold were then transferred to the next unelected ranked candidate on the list. Periodically in the count -when no candidate passed the threshold but there remained seats to fill?- the lowest ranked candidate on each list would be eliminated and their votes transferred to the top ranked unelected candidate on the list.

    Voters could also if they wanted blackball candidates on the list on which they voted for a candidate, which in practice would mean that when votes were transferred to a candidate, a number of votes equal to the number of outstanding blackballs would be moved on to the next ranking candidate, which also would remove the blackballs for future counts.

    Essentially this was a disguised version of STV with above the line voting, designed to handle large magnitude districts, and with the default of voting for individual candidates instead of the party list. But it might be easier for voters to swallow than straight STV.

    By the way I agree with Alan about Fiji.

  4. I must be missing something because the PDF draft constitution that Alan linked to talks about a parliament of 45 seats (indexed to rise with Fiji’s population) and four large districts?

    Ed, the usual cycle is that (a) pundits consider STV “too complex” and say “Let’s just use lists instead, they’re much simpler”, then (b) attempts to allow voters a choice of candidates within lists become so complex that other pundits say “Let’s just use STV instead, it’s much simpler”…

    • What I understand from Jon is that there have been several variants of the electoral-system proposal. The Yash Ghai commission proposed closed lists (though I am not sure of the districting). Earlier this year it was changed to open lists, and it was indeed a four-district model. The version of the constitution just released explicitly says one district and “open list”.

      I should note that if any of that is incorrect, it is likely my faulty interpretation rather than Jon’s, although I am going off a PDF document of the constitution that he referred to as being “just out” so maybe it has not been put up yet on the website that Alan linked to.

      And, yes, I recognize that the legitimacy of this constitution is contested. I am just referring to what is in the document, not to its legal standing or likely success in facilitating an actual transition back to democracy.

  5. The draft I linked is on the website of the regime. There seems to be some confusion, even within the regime, over the content of the new ‘constitution’.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s