Not ‘zombie’ but what?

Colleagues who study mixed-member systems know my disdain for the term, zombie, for candidates who lose a district contest but win a list-PR seat.

In a nutshell, what I do not like about this term is how it conveys a notion that candidates who do not enter parliament via the nominal tier (winning a district individual contest) are somehow “dead” and can enter parliament only by being resurrected through the sorcery of the party-list mechanism. Oh, the horror of it!

There is nothing illegitimate about winning via the list. Nor is it desirable to have the tiers segregated. (Some mixed-member systems ban dual nomination, I realize.) To the extent that there are benefits to mixed-member (MM) systems, they come largely from the incentive of list nominees to pay attention to a locality even if they don’t emerge on top in a local contest. ((See the thread from December, 2005, for a discussion of issues with dual candidacy.)) Candidates who win via the list rather than the nominal tier were not “killed” by district voters; they have won by one of the two methods that make a mixed-member system what it is–like it or not. We need a better term than the pejorative one that’s almost becoming standard. The alleged illegitimacy of dual-nominated candidates who win via the list is an empirical question, and as analysts we should avoid terminology that presupposes illegitimacy. ((It seems quite clear that the issue of perceived “zombie” illegitimacy comes up a great deal in Japan (where I think the term, zombie, was first used in this context) and in New Zealand (where dual-nominees who win via the list are sometimes called by the also pejorative “back door MP”. On the other hand, in Germany, the divergent paths to parliament seem to be a non-issue. Probably the alleged illegitimacy in Japan and New Zealand is a legacy issue–that these countries had pure nominal electoral systems before adopting MM, unlike Germany.))

A simple replacement term for ‘zombie’ is not obvious. I’ve suggested acronyms: DNLW for dual-nominated/list winner, DNDW for dual-nominated/dstrict winner, etc. But ‘DNLW’ is not as snappy as ‘zombie’ and might struggle to win acceptance.

I have thought about taking a term from the sports pages: ‘wild card’. You have your division/district winners, and then you have those who enter the playoffs/legislature despite coming second or lower in the division/district. The latter are wild card teams/candidates. However, I am not entirely serious about this “recommendation”.

I could see using something like a series of codes for nomination (N=nominal, L=list, D=dual) and then indicate with (1, 0) whether the candidate won or lost, or in the case of dual winners, whether the win was on the Nominal or List tier:

L1: List only, won
L0: List only, lost
DL: Dual nominated, won list seat
DN: Dual nominated, won nominal-tier seat
N0: Nominal tier only, lost
N1: Nominal tier only, won

I suppose convincing analysts to adopt a scheme like this does not have a high probability of prospering. But that does not mean it would not be valuable.

My goal is to develop an easy to read scheme that is analytic, not pejorative about whether one way of running and winning is more legitimate than any other.

25 thoughts on “Not ‘zombie’ but what?

  1. I think the problem could easily be overcome by having a free list, as opposed to closed lists, where the order of the list is entirely determined by the voters. Anyone who doesn’t vote for a candidate on the nominal tier is very unlikely to vote for that same candidate on the list vote. The candidate would need a lot of support from outside their local district in order to win. A system of approval voting or score voting to determine list order would be best to order the candidates. The initial list would be ordered by the party, and used only in the final seat allocation to break a tie. Closed lists seem to be at the heart of the issue.


    • Dominic, I think you mean open list. What you propose seems similar to what I sketch in my December, 2005, post (linked in footnote above). I did not know it at that time, but Bavaria actually has such a system.

      Chris, the US municipal systems you refer to are not mixed-member, as both tiers are nominal and majoritarian.

      (Dominic, again: “free list” is a term sometimes applied to “panachage” systems such as in Luxembourg and Switzerland, where voters may cast votes for candidates of multiple parties; they are not constrained by the lists parties present, hence “free” to make up a list.)


  2. In the Japanese context, I think that many voters actually did see their SMD vote as cast against a preferred loser as much as one cast for a preferred winner. That’s part of the “lesser of two evils” logic of FPTP, especially for voters whose favorite has no chance. In the first MM election in 1996, many voters seemed upset that they were unable to keep someone out of parliament.

    That’s of course the beauty of closed-list PR – voters get who they get. I think the annoyance with MMM and dual candidacy was that voters thought they had the opportunity to exclude people, only to find they could not.

    I completely agree that there’s nothing illegitimate about zombies. But I like the term anyway. Zombies are people too!


    • Mike, would that problem have been alleviated by MMM without the “best loser” provision, or is it inherent to any MMM (and MMP)?


  3. I like the idea of the phrase “wild card.” However, “wild card teams” are generally only used in North American sport (meaning the US and Canada), where there are no mixed-member systems (aside from the atrocious city council elections in some localities where some members are elected in single-member wards and others are elected at-large). Academics outside NA would tend to associate “wild card” with playing cards, and perhaps not understand the analogy.

    Perhaps an analogy could be drawn to Joe Lieberman/Lisa Murkowski type candidates who lose their primaries but go on to run in a general election, and the “sore loser” laws that many states have to prevent this, though in calling them “sore losers,” it hardly makes them more legitimate than “zombies.”

    As I understand it, list MPs in Germany tend to present themselves as a sort of shadow representative for the district in which they lost, so perhaps “shadow MP” might be a better term.


  4. I am not the world’s biggest fan of MMP and/or closed lists but be damned if I’ll let FPTP supporters – of all people – criticise either as undemocratic.


  5. Anyone watching any of the tennis grand slams knows the expression ‘wild card’. Ditto various flavours of football.

    Zombi, comeback and backup are about equally awful because they imply that list MPs were not there and were then let back in. Shadow MP has those problems and the additional problem of leading to confusion in places that use the expressions ‘shadow minister’ or ‘shadow cabinet’.


  6. Mike @@2, query how much practice did Japanese voters get a kicking the rascals out before the 1993 reforms? The consensus was that SNTV favoured “quota squatting” incumbents and was blamed for promoting stagnation, with voters reluctant to switch candidates in case their party of choice lost all its seats.


  7. Re “zombies” – MMP backers could point out that not all district-defeated candidates get in via the list. Each party has a finite number of seats. No more than every Minister voted out under FPTP has a colleague or party organisation in another constituency obliging enough to create a quick vacancy.


  8. Tom @ 7: Japanese incumbency rates were considerably lower than, say, U.S. rates, but that’s not saying much, I guess. Pre 1993, incumbents frequently lost and sat out a term before returning. Steve Reed has referred to this equilibrium as a game of “musical chairs.” Frequently, the odd-man-out would place first in the next election, having had a full term to cultivate the electorate.


  9. Using an open list instead of a closed list might help things, everyone was elected by either winning a constituency or by winning a seat through the open list count.

    My preferred terms would be Local Member for people elected from districts and National (or State/Provincial/Regional) Member for those seated without winning a district.


  10. In cases like the Italian senatorial pre-1993-system or the actual Romanian system where there is no ‘list tier’ and winners-by-PR are assigned to a not-so-single-member-district, ‘zombie’ might do.


  11. Does ‘mixed-member’ imply at least one list tier, rather than different nominating districts in one chamber?

    Would the Puerto Rican system, with some elected fptp (in smds for the house and two seaters for the senate) with 11 elected island wide by SNTV be considered mixed member (esp with the urging for voters to vote their party’s top listed candidate in the SNTV, though retaining the option to vote for one of the others)?


  12. How does the Swiss panachage system work? Do voters have as many votes as there are seats to be field? Isn’t it a bit akin toward the STV system?

    MMP and MMM system where there are two front doors can create problems especially with a country like NZ which was use to FPTP, which is so simple, and the country wasn’t use to the idea of closed party lists which make things more complicated.

    Perhaps the MMP and MMM zombies should be called the biggest losers or worst winners because even though they lost in a single member district, and won a list seats.

    As for Japan, the new electoral system is better than SNTV, but it seems as if the MMM system could be tweaked to make it better. It seems to me that if Japan had used SNTV, why wasn’t anybody suggesting STV system or an open party list PR system, but I would think Japan would be best served with a close party list system as to reduce patronage.

    Which electoral system reduces patronage spending? Which is the best electoral system to insure the nation’s resources are evenly distributed nationwide and not on some small little district?.


  13. Chris, by the definition used in Shugart and Wattenberg (2001), yes, it is only mixed-member if one tier is elected by nominal rules and the other via lists.

    I am not aware of definitions in the literature that disagree fundamentally, although some might focus on the inter-party dimension of the rules rather than the intra-party: one tier majoritarian, one PR.

    In most cases, these get us to the same place. The differences would come in with something like the old Korean rules where the list tier was also majoritarian. I suppose if you use the second definition, AND if you classify SNTV as “PR” you could wind up with the Puerto Rican system as “mixed-member.” I think that would be inaccurate, and I don’t think I have ever seen that system called mixed-member (then again, then most ignore it entirely).


  14. Pingback: Fruits and Votes

  15. @ MSS, 19

    I read both of those descriptions, but I am not sure I understand how it works. If a voter changes around one party’s list, then the voter wants the party to earn seats and influence who will fill those seats. But if a voter creates his own list composed of candidates from different parties, how does that list determine which parties earn seats? Or does making one’s own list effectively state that the voter is more interested in picking candidates for seats awarded by other voters than in choosing which parties are awarded seats?


  16. Mark R @19:

    As I understand the Swiss/ Luxembourg system… say your Canton elects 10 deputies. You submit a list with up to 10 names (no name more than twice). If all 10 are, say, Socialist Party [*] candidates then the Socialist Party list is credited with 10 votes for the purposes of inter-party allocation. But if, say, only seven are Socialist candidates and the other three names are Liberals, then you’ve given three of your party votes to the Liberals. I think if you write-in names of individuals who are not on a party list, then it works like a cross between MNTV and cumulative voting. Some Cantons allow full cumulation, I believe.

    In other words, 100 voters submitting lists with the same seven Socialists and the same three Liberals will affect inter-party numbers exactly as if 70 had voted a straight Socialist ticket and 30 had done the same for the Liberals.


    [*] Not only is the main German centre-left party called “the Social Democratic Party” while the main French centre-left party is called “the Socialist Party”, but within Switzerland, the same political party is called “the Social Democratic Party” in German but “the Socialist Party” in French. Interesting.


  17. @ Tom, 22, Thanks. That makes sense. It seems I skipped the part about voters having as many votes as there are candidates and just assumed it was one man one vote with open lists.


  18. Adding to the [*] of Tom Round: the Flemish centre-left party is called ‘Socialist Party’ and the far-left party ‘Labour Party’ (PvdA) ; while in the Netherlands it is the other way round – and both in Dutch!


  19. Pingback: “Shadow” MPs | Fruits and Votes

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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.