[EM] Preferential voting system where a candidate may win multiple seats

[Juho Laatu]

On 4.7.2013, at 13.55, Kristofer Munsterhjelm wrote:

> On 07/04/2013 08:55 AM, Juho Laatu wrote:
> 
>> In principle ability to vote for persons helps populist candidates.
>> My best understanding is that in Finland, that uses open lists, well
>> known candidates (from sports, TV etc.) probably have slightly better
>> chances to win a seat when compared to countries using closed lists,
>> but that difference is not big. Also closed lists can be populated
>> with well known figures to get "populist votes" (in addition to
>> nominating experienced politicians).
>> 
>> Also campaining could in principle be more populist in open lists,
>> but I don't see big difference here either. In FInland the level of
>> populism differs more between parties than between the candidates of
>> a single party.
>> 
>> All in all, I believe the risk of excessive populism is not big in
>> ranked methods either.
> 
> Both closed list and open (and ranked) methods have their failure modes. Closed list fails when the party leadership becomes unaccountable and insulates itself, and then the voter is forced to either vote "my way or the highway" - i.e. to accept the leadership's ranking or to not vote for the party. Person-based methods fail when it produces an incentive to be excessively populist.
> 
> In a way, that's a mirror of the general balancing act of democracy. If it is too "representative" as opposed to direct, then the powerholders might just run away with the power and mockingly say to the voters that they have no choice but to vote for one of the powerholders. If it's too direct, then it can amplify too much and oscillate around various policies if not degenerating entirely to populism.
> 
> I suspect that the solution to this particular problem lies not in getting the balance right, but somehow setting up the right feedback system so that public discussion and opinion convergence can move beyond populism. That said, I think I favor ranked multiwinner methods if I have to choose: the populist objection seems to be employed to exaggerate the negative results of giving the people more choice.

Your expression "representative vs. direct" captures the small difference between multi-winner methods that allow parties vs. voters to decide which candidates of each party will be elected. Most democracies are representative democracies, so there will be some level of "representative isolation" between voters and decision making. One just has to decide how much direct power to give to the voters, and how much one expects them to be capable of making decisions themselves.

In closed lists it is also important that the party structure can change. If the system is parameterized so that the old parties will stay in power forever, we may be in trouble. But if the system allows parties to grow, emerge and diminish depending on what kind of candidates each party puts on the list, then we are ok.

> 
>>> The leveling seat algorithm is... peculiar.
>> 
>> You said that you don't like methods that lead towards a two/three
>> party system. In other words the method should allow also small
>> parties to survive. I note that typically small districts are one key
>> reason why small parties do not get any seats. If you e.g. have a
>> district with 3 seats, it is obvious that only two/three largest
>> parties can win there. The leveling seats (that are allocated based
>> on support at national level) could fix that problem, but I
>> understood that n Norway they don't apply to the smallest parties.
>> Therefore the 5% threshold probably effectively reduces the chances
>> of the smallest parties to get their proportional share (at national
>> level) of the seats. It does not make sense to the voters to vote for
>> parties that most likely will not get any seats in their district
>> anyway.
>> 
>> I don't know what the situation in Norway actually is today. My
>> comments here are thus just general comments on how multi-winner
>> election methods usually work.
> 
> The Norwegian threshold is at 4%. If parties get sufficient local support, they still get seats; the 4% only regards leveling seats.
> 
> As a concrete example, in the 2009 parliamentary election, the Liberal Party ("Venstre") achieved a support result of 3.9%, just below the threshold of 4%. In the previous election, their support reached 5.9%. As a consequence of going below the threshold, the party lost 8 of its 10 MPs.

The Norwegian system has thus two levels that determine how many seats the smallest parties will get. Here is a list of constituencies of the parliamentary election (taken from Wikipedia).

County Seats
Østfold 9
Akershus 16
Oslo 17
Hedmark 8
Oppland 7
Buskerud 9
Vestfold 7
Telemark 6
Aust-Agder 4
Vest-Agder 6
Rogaland 13
Hordaland 15
Sogn og Fjordane 5
Møre og Romsdal 9
Sør-Trøndelag 10
Nord-Trøndelag 6
Nordland 10
Troms 7
Finnmark 5
Total 169

If a voter expects some party not to reach the 4% threshold, then voting for such party makes sense only in the largest districts, e.g. in Oslo that has 17 seats, and where some 6% of the local votes (or maybe less with Sainte-Laguë) might be enough to get one seat. In the smallest constituencies it does not make sense to vote for that party. This further reduces the probability of reaching the 4% threshold.

If the voter expects that party to get more than 4% of the votes, then it makes sense to vote for that party also in the smallest constituencies (in the hope that the party will get a leveling seat).

I'd say that the system tends to eliminate parties with less than 4% national support (because of the threshold, although the threshold covers only the allocation of leveling seats).

New parties may emerge if they can first get sufficient support in some of the largest constituencies, and then reach the 4% threshold, and then gain support also also in other parts of the country. That favours parties that are popular in the large constituencies (typically large cities).

> 
>>> I'd like to get rid of both leveling seats and election threshold.
>> 
>> If you want to achieve exact proportionality (also for small parties)
>> I think it is important that proportionality will be counted at
>> national level. Also in ranked methods it is not enough if each
>> district does its best alone since the small number of seats per
>> district will distort proportionality at national level. From this
>> point of view the leveing seats (or any construction that aims at
>> providing proportionality at national level) is good, and thresholds
>> are bad.
>> 
>> - - -
>> 
>> I note that you can achieve national level proportionality in list
>> based methods also without leveling seats. In Finland there was a
>> proposal that was alrady once accepted by the parliament but then
>> cancelled by the current government. This proposal counted the
>> proportionality first at national level, and then allocated a
>> predetermined number of seats to each district so that at the same
>> time also the calculated national proportionality numbers were met.
>> This means that the last seats in some districts were slightly
>> "forced" to correct parties, to meeth the national proportionality
>> target. All methods that try to reach multiple targets, like
>> political proportionality and geographic proportionality at the same
>> time will have some "rounding errors". In the Finnish proposal those
>> rounding errors were thus solved by slight distortion in who and
>> which party wins the last seat in each district, instead of using
>> e.g. leveling seats to capture the rounding errors.
> 
> That doesn't sound so different from leveling seats. In the Norwegian system, you give each county an extra seat, but this seat is assigned based on the difference betweeen the seats so far allocated (on county by county basis) and the national apportionment. I'm not sure how the algorithm decides which county gets which party's seats, but it's not a simple biproportional thing.

Yes, the end result is probably very similar. The fact that each leveling seat is tied to one county further reduces the difference (since there are no "countyless" seats).

> 
> I know that it has led to very counterintuitive results (e.g. a party getting a seat with only 300 votes). Thus, I would favor actual biproportional representation, because it tends to take seats away from a party where that party has weak support, and give seats where the party has strong support, thus significantly reducing the chance of getting counterintuitive results. The biproportional representation algorithm could even be limited so it can't change more than n seats, where n is the number of counties, thus retaining the balance between national and local representation. It might be a bit unintuitive, though, that the seats are floating: e.g. a county might get more than one "leveling" seat because that reduces the majority that has to be overruled.

I tend to give highest priority to reaching exact proportionality between parties at national level. I also don't like any threshold like concepts since I believe most democracies are stable enough to allow also one or two representatives of some minor groupings in the parliament. That think that would do no harm in stable countries like Norway. It is more important to allow all interest groups to get their own reprsnetatives.

Also regional proportionality and avoidance of weird results where candidates with very few votes (to the party in the county) can be elected are important, but to me that is less important than making the poilitical balance exact and allowing all groups to be represented. I'd thus push the unavoidable rounding errors to these parts of the system (maybe to the persons first and to regional proportionality only after that).

> 
>> You had interest in guaranteeing that the "lost" votes of small
>> parties will go to parties that are similar-minded. If one counts
>> exact proportionality at national level the number of lost votes will
>> be quite small. That alone might be enough for some needs. The
>> traditional way of voting for one party or one candidate only could
>> thus be enough, and there would not be need to have ranked votes for
>> this reason. (Ranked votes could be there for other reasons, like to
>> support party internal proportionality.)
>> 
>> Ranked methods may also be quite heavy for the voter if there anre
>> tens of candidates to rank.
> 
> I don't think there would be a need to rank tens of candidates or parties. Given enough seats, the list method is quite fair to the voters that vote for parties that get at least one seat. So unless there's a balancing system to pull in a more centrist direction for each county (electing C instead of L or R in a one-seat situation), then voters who vote for large parties wouldn't have to rank at all. The voters who vote for smaller parties would only have to rank up to a large party.

In Finland ability to handle numeros candidates (maybe 200 per district) is a quite strong requirements since people are used to that. In closed lists there is no need for each party to list more candidates than one can maximally get, but in open lists one must have a wide range of candidates to choose from. Elections serve also as primaries in Finland in the sense that if some candidate gets a reasonable amount of votes this time, then he may already become well known, and possibly will be elected in the next election.

Another reason why political parties want to nominate as many candidates as they can (in Finland) is that they want to have candidates from each town, from each profession etc. This is because many voters are anyway interested in voting a candidate that is in some sense close to them. Roughly, every additional candidate brings at least the votes of his family, relatives and friends to the party.

If a closed list system will be changed to an open list or ranked system I'd expect also the number of candidates to grow (unless artificially limited somehow).

> 
> DAC/DSC variants may change this situation a bit, but I don't think it would be necessary to rank many parties in general.
> 
> > For the needs of Finland I have been
>> interested in methods that would combine lists and ranked votes in
>> another way. The idea is that ranking would be used within one party
>> only. This approach does not allow the ranked votes to be used to
>> improve the proportionalty between parties (as in your original
>> proposal). One reason why ranking would be party internal only is
>> that in this way also bullet votes to one single candidate and votes
>> that do not list all the (tens of) candidates of one single party
>> would support the intended party with their full strength (the whole
>> strength of each short vote will go to the intended party). It might
>> be too tedious for the voters to first rank all the candidates of
>> one's own party, then all the candidates of the next best party, then
>> the third and so on. One could thus achive (almost exact) national
>> proportionality by counting it at national level, and still keep
>> voting simple, also when there are tens of hundreds of candidates (in
>> each district). The actual paper ballots could be large lists of all
>> candidates, or just simple paper ballots with maybe three slots/boxes
>> for the numbers of three candidates (of one party).
> 
> That's like open list, but with ranking instead of just giving an extra vote to a single MP.

Yes, plain open list allows voters to choose which candidates will win, but the outcome is quite random (only roughly proportional). A ranking based aproach would fix this problem.

> Ideally, open list would let the voters who have a definite reason to prefer another ordering express that (and have that influence the final list), while letting everybody else default to the ordering set up by the party itself.

Note that in Finland the system is purely open list in the sense that there is no party given preference ordering. The party has thus no say on which ones of its candidates will be elected (except that they could do some strategic tricks like nominate only one candidate from the western part of some district in the hope af making many voters vote him that way).

In my proposal above a bullet vote would be a vote to one candidate. If that candidate will not be lected, that vote will anyway be counted in the total number of votes to the party nationally and in the district in question (but it would leave the preference order of the other candidates unspecified).

> However, the question of how to do the balancing there is difficult. On the one hand, voters that don't change the ranking may be expressing an explicit wish to have the ordering the party leadership provided. On the other, the voters may simply be saying "it doesn't matter to me". That's like the tension between average Range and sum-total Range.
> 
> If it's too tedious to rank the candidates of a party, I'd permit equal rank and truncation. Thus, anybody who wants to vote Approval-style can do so.

Yes, equal ranking could be supported too. It is not essential since ranking does not do much harm to the voter. In some ballot types equal ranking might be easier than full ranking.

Since I want to make voting so easy that all citizens will vote, the system should allow also basic bullet votes and short votes. Actually I think already ballots that allow voters to rank max three (or maybe five) candidates would offer most of the benefits of ranked votes. The results would be quite accurate already with votes like "my favourite with no real chances of getting elected" > "my second favourite taht might get elected" > "another good candidate that also has good chances of getting elected".

(In Finland the number of candidates will be high (since people are already used to that). If one wants to support also short votes, this combinaltion easily leads to the idea of supporting party internal (and district internal) ranking only. Or are there other better approaches to cover these needs?)

> 
>> If one wants to be more exact with national proportionality between
>> parties (without forcing voters to rank also the candidates of other
>> parties), one solution could be to allow the parties to determine
>> their relationship to other parties. For example left wing parties
>> could join together so that their extra votes would support other
>> left wing parties and not the right wing ones. This gets however
>> quite complex, and parties may not be interested in agreeing which
>> parties are cosest to each others. Another approach would be to allow
>> voters to rank also parties in addition to ranking candidates, but
>> also that gets quite complex.
> 
> That solution doesn't seem to do very well where it's been tried. For instance, in Fiji, this led the parties to make deals among themselves to the detriment of the voters. Wikipedia mentions this:
> 
> 'Vice President Ratu Joni Madraiwiwi expressed his own misgivings about the voting system on 3 November 2005. He said it made the work of political parties much easier and denied freedom of choice to voters, as a vote for a political party was ultimately a vote for any other party to which that party had decided to transfer its preferences. "In hindsight, it would perhaps have been preferable to leave the voter to make up his own mind," Madraiwiwi said.'
> 
> But unless there are very many parties, the party-sharing solution doesn't really do anything that ranked voting couldn't. The system is equivalent to having a ranked ballot where every voter votes according to the preference order of the party they voted first. Thus you'd still need some kind of ranked voting method to translate those rankings into an allocation, and that's the hard part.

I favour systems that are so simple that regular voters can easily understand how they work.

Juho