In the elections in Kosovo/a, the party of PM Hashim Thaci has retained its plurality.
Thaci’s Democratic Party of Kosovo (PDK) has won 33.5%, while its former coalition partner Democratic League of Kosovo (LDK), whose withdrawal precipitated this election, has won 23.6%. In 2007, these parties won 34.3% and 22.6%, respectively. So not much change.
On DW-TV, seen via Link TV, the following billboard caught my interest. I apologize for the poor quality; it is shot from a paused image on the DVR.
This is clearly (well, maybe “clearly” is not the right word) a billboard for the LDK. I assume it is showing all the candidates on the party’s list.
As far as I know, the electoral system continues to have a 100-seat district, with voters free to cast preference votes for up to 5 (earlier reports had said 10) candidates. It is unclear, from an earlier discussion at F&V, whether this is a fully open or a flexible list. In any case, the billboard shows 110 candidates, counting the party leader.
There are also another 20 seats, elected separately, for minorities. Ten of these are set aside for the Serb minority. However, voting was apparently sparse in Serb regions. (Maybe the billboard above shows 110 because it includes the 10 non-Serb minority candidates. Just speculating. Or it could simply be that parties may nominate more candidates than seats for the principal district.)
While I do not read Serbian, I know enough Cyrillic to know these signs call on Serbs to boycott the election. (Cognates help, too!)
This last photo is for the PDK and reminds us that, for the ethnic Albanian majority, the country is Kosova, not Kosovo.
The DW-TV report mentioned a new party that had come in third, with around 15%. From Balkan Insight, which has a regional (but not national) breakdown of the vote, it would seem that the third party is something called the Alliance for the Future of Kosovo(a?).
As an aside, I am always bemused at how many media outlets will declare that a party has “won” an election when it merely has the most seats–and nowhere near a majority. Of the first five hits in my Google News search, Voice of America, Xhinhua, eTaiwan News, and Radio Free Europe/Radio Liberty all had variations on Thaci or his party “winning.” Only Aljazeera (the one I linked to at the top) got it right: “Party of Hashim Thaci holds on most seats in parliament but fails to take majority amid allegations of ballot stuffing.”
Each of the stories mentions claims of fraud.
> ‘… media outlets will declare that a party has “won” an election when it merely has the most seats–and nowhere near a majority…’
Indeed so, but it ceases to be bemusing (or amusing) when the average person-in-the-street picks up the idea that “[Hashim Thaci/ Tony Abbott/ Tzipi Livni/ David Cameron/ Stephen Harper/ etc] won the election, but didn’t get to be prime minister because the losing parties ganged up against them.”
If journalists could be persuaded to apply one-tenth the number of brain cells needed to report – say – the Olympics (“China won more gold medals, but with more silver and bronze, the USA is ahead”), this wholly unwarranted whiff of illegitimacy would dissipate. In Canada, Australia, etc it probably goes no further than grumbling over a beer, but in other democracies this perception could have serious consequences.
(Cases where the chief minister holds on or gets in by bribing MPs from the opposition party to defect immediately after the election – Queensland in 1983, some States in India and Malaysia – are different. But if you vote for Winston Peters or Nick Clegg or Nick McKim or Tony Windsor with full knowledge that they have left their post-election options wide open, you have constructively consented to whatever coalition may emerge).
Seed planted by Tom Round — 15 December 2010 @ 20:22
In regard to the uncertainty from a previous post about the ballot type in Kosovo, it is indeed an open-list system. The electoral law initially adopted after independence in 2008 stated that voters could cast only one preference vote for a candidate from their chosen party list. If a voter decided to not cast a preference vote, their preference vote was cast for the candidate at the top of the list by default. So in this way, this would have been considered more flexible than completely open.
But a few weeks before the election, the Assembly of Kosovo passed an amendment to the electoral law stating that voters would now be given up to 5 preference votes and candidates would be ranked on their list according to the preference votes deliberately cast by voters, only. This amendment removed the previous part saying that a preference vote would be automatically given to the top candidate on the list if a voter failed to cast a preference vote. So, at the time of the election, voters faced completely open lists it looks like.
For anyone following the situation in Kosovo, the link below is pretty… interesting. I assume they released this report after the elections so the report would seem more legitimate. Interesting timing still though.
http://www.guardian.co.uk/world/2010/dec/14/kosovo-prime-minister-llike-mafia-boss
Seed planted by John G. — 16 December 2010 @ 04:37
John, thanks for this interesting clarification.
Under the old system, was there any mechanism by which party votes without a preference could augment votes of candidates other than one at the number one rank?
If not, then it was a good example of a flexible list that would be very close to fully open. Presumably the party’s first choice would normally have sufficient preference votes to be elected in any case, and the rest of the party’s seats would be filled entirely by order of preference votes (assuming I am understanding correctly).
Are there other “flexible” list cases like that? I know some others, such as Belgium, have a mechanism by which party votes sort of “cascade” down the list in the order preferred by the party, once any reordering based on preference votes for lower-ranked candidates has been taken into account.
Seed planted by MSS — 16 December 2010 @ 17:47
I believe your description of the old law is correct. Party votes without a preference had no effect on candidates not ranked first. Those candidates’ positions were fully determined by preference votes.
I suspect this type of “flexible” list operation will be back in effect in Kosovo sometime within the next couple elections. The reason being because there was talk leading up to independence about delimiting Kosovo’s electoral districts, because Kosovo’s electoral system has historically used just one national district, like they still do. I suspect the designers of the electoral system at the time of independence figured the country would have been broken up into multiple electoral districts by the time this election came about. If this were the case, voters would have a much smaller number of candidates to choose from in which to cast their preference for. Seeing that this was not the case, I suspect lawmakers enacted the new law a few weeks ago (providing voters with 5 preference votes) precisely because voters would be choosing between many more candidates than previously thought.
This is just speculation on my part, but I would bet that we can count on Kosovo to break up into several electoral districts in the near future, and a return to an electoral law offering just one preference vote.
Seed planted by John G. — 20 December 2010 @ 00:18
John: “I suspect the designers of the electoral system at the time of independence figured the country would have been broken up into multiple electoral districts by the time this election came about.”
That’s what the designers of Israel’s single-district PR system thought–over sixty years ago!
The imperative to do this is certainly stronger with open lists than with closed (as in Israel). But even in Kosovo, I suppose they could just go back to a less open (i.e. flexible) list if the open list is to hard to manage with such high magnitude.
This has me wondering: how many cases do we know of that moved from single national district to districted? Iraq, after just one (constituent assembly) election. Not too many others, I suppose. I can think of two cases that have gone the other way, from districted to single district (Colombia senate, Slovakia). Well, the ‘N’ here is pretty small in either direction…
Seed planted by MSS — 20 December 2010 @ 20:40
Germany since 1945? – inasmuch as Weimar had a rule that a list could not win more Reichstag seats at the national level than it had already won at the state and district level. Whereas the Bundestag has pretty much a national PR allocation, apart from any overhang seats. If a party gets, say, 8%, it gets 48 Bundestag seats, even if it won no districts (whereas the Weimar rule would give it fewer than 48 overall if it won fewer than 25 locally and at State level).
Seed planted by Tom Round — 21 December 2010 @ 00:08
Err, fewer than “24″.
Obviously…
Seed planted by Tom Round — 21 December 2010 @ 00:08
Also, while this doesn’t count as “nationwide” in a literal sense, if we are using the term as a synecdoche for “at large”, then South Australia moved from five 4-seat upper house “provinces” to Statewide PR in 1973.
(2 of 4 Legislative Councillors vacated each election, replaced by block-preferential vote akin to the Australian Senate 1918-48. Also extremely malapportioned.)
Seed planted by Tom Round — 23 December 2010 @ 20:30
I should have asked about “jurisdictionwide” rather than “nationwide.” The South Australia case most certainly counts!
Seed planted by MSS — 26 December 2010 @ 15:38
Thanks MSS. Curious that my two examples involved Teutonic jurisdictions…
I’m sure most political scientists would raise an eyebrow if one described the Bundestag’s electoral system as “more at-large” (so to speak) than the Reichstag’s, but so it is, in that particular sense.
NSW also has a State-wide-elected upper house, but before that MLCs were indirectly elected (joint sitting of the lower house plus the non-retiring three-quarters of the Legislative Council) rather than elected from provinces.*
* For some reason, never fully explained to me, Australian practice has been to title upper house electoral districts “provinces” when they elect MLCs using majority-preferential voting (South Aust before 1973, Western Aust before 1986, Victoria before 2005, Tasmania even today), but “regions” when they elect MLCs from sub-State electorates using PR (WA, Victoria today). Only in Tasmania does “majority-preferential” mean “single-member”. Vict and WA formerly elected one of two MLCs at each election – something like the US Senate.
Seed planted by Tom Round — 26 December 2010 @ 19:44
… In fact, one reason Labor in South Australia was happy enough to push for Statewide PR (11 seats per election, at that time the largest district magnitude ever used in Australia until the NSW Legislative Council took the title in 1978) was long-term annoyance with the conservative bias in the electoral weighting for both houses.
Interestingly,
(a) although district lines for the Upper House have gone, the State Constitution Act’s provision for breaking deadlocks between the two Houses (ie, elect two more members, for one term only, from each Upper House district – Victoria formerly had a similar clause, and Canada has a second cousin of it) still remains law. Electing another 2 MLCs by MPV from 5 districts would probably break a deadlock. But electing 2 by Statewide PR would – going by the Territory Senators precedent – change nothing. Odd.
(b) SA is the only (?) jurisdiction I’m aware of, other than Malta, that requires a uniform district magnitude without specifying a particular number.
Seed planted by Tom Round — 31 December 2010 @ 19:48
The Proportional Representation (Hare-Clark) Entrenchment Act in the ACT does not specify a uniform district magnitude, but does require district magnitudes be odd numbers not less than 5. Separate legislation, the federal ACT Self-Government Act, requires uniform quotas, taking the number of MLAs for each district into account.
Seed planted by Alan — 31 December 2010 @ 21:58
The Australian Senate (also, to my knowledge, unique among upper houses) requires the same number of Senators per State but allows Parliament to fix any number not less than six (ie, minimum three seats per half-Senate election). But SA and Malta are rare examples where this applies to a lower house with districts set by (or under) statute, and regularly revised, rather than entrenched forever in the Constitution (indeed, put beyond the normal power of Constitutional amendment, like in the US).
In other cases, eg, Tasmania, Victoria and WA, the relevant Constitution Act provision simply stipulates “N [= five/ six/ seven] seats for every district/ region”, ie specifies a particular number which automatically entails uniformity, whereas Malta’s Constitution says “any number from 5 to 7, as long as it’s uniform”.
The Hawke Govt tried to include a similar rule in the Aust Constitution via its proposed “Fair and Democratic Elections” Constitutional alteration in 1988. The Proportional Representation Society persuaded the Democrats to block that clause in the Senate since it would have meant that if a State had a large prime number of MHRs (say, 31 or 47), the only alternatives would be either single-member districts or all at large (the latter would, ironically, have meant much wider variations in DMs across States).
I do think there is a case for stipulating that the largest district magnitude must be less than (say) twice as many seats as the smallest, or less than the smallest plus two seats, something like that, to put a cap on extreme variations (which often cause anomalies, even when the multi-seaters use PR, as in Finland, and a fortiori when they use winner-take-all, as in the US Electoral College).
Seed planted by Tom Round — 01 January 2011 @ 08:45
Tom (13.), is this why South Australia has a 47-member lower house, to make single-member districts the only feasible option within those rules? Presumably though, that number is set by ordinary legislation.
On your footnote at 10., the Cape Colony upper house also had electoral provinces (1854-74 two large ones, then seven, each electing three by cumulative vote, with gradual modifications, until 1910). Could it be that “provinces” were a mid-19th century fad at the Colonial Office that took hold in Australia?
Russia and Ukraine moved from half-districted to all-nationwide. The Netherlands still has 19 districts for candidate selection, but in practice it went from districted to nationwide when PR was adopted. France recently introduced districts for elections to the European Parliament.
Seed planted by Espen Bjerke — 02 January 2011 @ 05:46
Espen, thanks for reminder re Cape Colony. Ironic to think that a “Colony” in one SA that became a “Province” (Cape) and a “Province” in the other SA that became a “State” (South Australia was not officially a “Colony” and never had convicts), have had subdivisions also titled “Provinces” for their upper house.
As I understand it, certain provisions of the SA Const Act 1934 are entrenched (ie need a referendum to change) but the rest can be amended by ordinary statute. I am pretty certain that the uniform D-Mag rule (as well as the “all districts equal to within 10%” rule and the procedures for regularly adjusting electoral boundaries) are entrenched in the Const Act, but whether “47 MLAs” is (a) entrenched in the Const Act, (b) in the Const Act but not entrenched, or (c) set by an ordinary Act, I know not. The AUSTLII site would be helpful if someone has time. The usual pattern in Aust States is for the size of the Upper House (if there is one) to be entrenched, but not (unlike the US States) the size of the Lower House.
Seed planted by Tom Round — 02 January 2011 @ 08:40
‘Province’ cannot be blamed on the Colonial Office because in each colony except South Australia the legislative council was originally appointed for life by the governor-in-council.
The Queensland council stayed that way until its abolition. The NSW council went through a strange period when a quarter retired every 3 years and the election was by joint sitting of the legislative assembly and the continuing members of the council.
In South Australia the council was elected but on a restricted franchise. I would be surprised if the Province of South Australia was divided into provinces (ultraprovinces?) for electing MLCs but nothing is impossible.
Seed planted by Alan — 02 January 2011 @ 23:45
“Metaprovinces”. Now, back to Kosovo, eh…?
Seed planted by Tom Round — 03 January 2011 @ 04:03
Tom,
To take up your comment on entrenchment:
The size of the Lower House is entrenched by referendum, as are the duration of the Parliament, the number of regions and the numbers of members per region in the Upper House , the inability of the Upper House to block Supply, and a fair bit more. Strangely, the PR method of election for the Legislative Council is not entrenched, though I was told four years ago that it had been.
The Victorian Constitution is here.
Seed planted by Chris Curtis — 03 January 2011 @ 05:20
In negotiating a government after the 2010 parliamentary elections, the biggest party PDK (of PM Hashim Thaçi) ceded the presidency to AKR (5th party) and Behgjet Pacolli was elected in the third round with 62 votes. The opposition (LDK, AAK) walked out (only 67 MPs were present) and challenged the election in court.
The president of Kosovo is elected by parliament (120 MPs). In the first two rounds, a 2/3 majority of all MPs is needed to win, in the third only a 1/2 majority of all. (art. 86 of the Constitution )
A typical rule favoring a consensus winner but enabling a patient bare majority to win.
The Constitutional Court went along with the opposition (decision of 30 march 2011) and ruled that the ballot could not start:
1) as long as there was only one candidacy and
2) as long as less than 80 MPs (2/3) were present because the president can only be elected by 2/3 of all MPs in the first two rounds.
Two of the three international judges (Robert Carolan, USA and Almiro Simões Rodrigues, PT) dissented:
1) The majority can easily start the ballot on their own by proposing a second candidate while only voting for their first candidate (although not that easy: a candidate must be proposed by 30 MPs) Such a ‘sham and mockery’ cannot be the intent of the constitution drafters.
2) One should not confuse the quorum rule – can we start? how many present? – with the voting majority rule. In this case, the voting majority is clear – 2/3 of all MPs in the first two rounds, 1/2 of all MPs in the third – but the quorum isn’t (in general it’s 1/2 – art. 69.3 of the Constitution).
If the quorum in the first two rounds is also 2/3, a 1/3 minority can block indefinitely the first two rounds. This was not the intent of the Constitution: a bare majority of MPs should be able to elect the president in the third round.
On the other hand (what follows is my opinion, not that of the dissenting judges):
if the quorum is lower than the voting majority (1/2 – 2/3) it is possible that the number of MPs present (in this case 67) is higher than the quorum – the vote can start – but lower than the voting majority – the vote cannot be won, the majority cannot be reached even if all vote in favor. In such a case the first two voting rounds are over in no time and the third round with the lower voting majority can too easily be reached. This was also not the intent of the Constitution: the bare majority should feel some incentive to bargain towards a 2/3 majority for a consensus candidate.
Afterwards, a compromise was reached with LDK: Atifete Jahjaga, not a politician but a female police chief, was elected president, this time with 80 MPs in favor and with another (genuine?) candidate.
The deal included LDK entering government, the direct election of the next president and other electoral reforms (on the latter, I have no details).
Seed planted by Bancki — 14 April 2011 @ 08:50
That’s fascinating, Bancki. Thanks! So Kosova will change to premier-presidential, evidently.
(As Bancki’s comment shows, old threads are all re-opened, thanks to a new plug-in that, I hope, will override the gremlin that was regularly changing the settings.)
Seed planted by MSS — 14 April 2011 @ 14:39
A correction to my @13 above:
“The Australian Senate ) requires the same number of Senators per State but (also, to my knowledge, unique among upper houses) allows Parliament to fix any number not less than six”
In fact, come to think of it, not just in the Australian States but worldwide the trend seems to be that the size of the Upper House is fixed by or under the Constitution (either specified numbers per State/Province, whether uniform as in USA/ Switzerland or varying as in Canada, or a set population formula, as in Germany) – either way, ordinary statute cannot change the numbers – while the Lower House (other than in US States) can be enlarged or reduced by statute. Even if there are constitutional parameters limiting the range (eg a set min/ max size, as in Spain, Greece and S Africa, or a min/max population ratio, as in USA/ Ireland), the range is usually generous. Congress could legally set the number of CongressReps at anywhere between 50 and 10,300+.
UK is a standout because the size of the Upper House can be increased easily – it doesn’t even take an Act of Parliament but a simple Royal instrument creating a peer, an executive action.
Seed planted by Tom Round — 15 April 2011 @ 19:32