“Shadow” MPs

In the earlier thread on my disdain for the z-word in reference to mixed-member systems, one comment (by Chris) suggested that a better term might be “Shadow MP”. The logic for this term is that this is what the district loser who wins via the list often does: “shadows” the winner as a second representative of the district. This makes sense, although the use of a term “shadow government” for the opposition in Westminster-type systems might render “shadow MP” confusing (as Alan suggested). ((That is, because not all “shadow” MPs are from an opposition party; they can be from the governing party in districts won by the opposition.))

Here is a good example of actual shadowing in practice!


Jacinda Ardern (“List MP based in Auckland Central”) has her office just three houses down from that of the electorate MP, Nikki Kaye.

Kaye’s office is at the left of the photo–see the blue sign behind the parked car at the left; Ardern’s is at the red sign just beyond and to the left of the motorcyclist’s head.

Here are close-ups of the offices and signs of the two MPs.


It is interesting that in New Zealand, many legislators elected via the party list, such as Ardern, refer to themselves as “List MP for” (or “based in”) whatever the district name is. However, others simply refer to themselves on the signs at their offices as a party MP. ((The example shown at that link, Tim Groser, did contest the district as well, both in 2008 and 2011.)) Surprisingly, I did not see one sign that said “Zombie MP for…”

34 thoughts on ““Shadow” MPs

  1. Ironic given that a big reason for supporting MMP over STV was that the former was seen (by the 1987 Royal Commission and others) as preserving the direct link between MPs and their constituents that only SMDs can provide (although the Commission did note the German practice of defeated districts candidates and/or List MPs “shadowing” the district winner). Voters would supposedly be confused if there were 3, 5 or 7 MPs for their area. Now NZ has it anyway, but without any intra-party choice (to be fair, the Commission considered the latter a bad thing so it succeeded at least in one respect).

    In Queensland, where there are only SMDs at State level, there is still ample confusion about which MP to approach. A voter who resides in one electorate, works in a second, drops her children at school in a third, and drives past a prominent billboard outside the electorate office of the MP for a fourth district, can get plenty confused about which is “her” “local” MP.


  2. Notably, there is no long-term connection to the electorate (the NZ term for district/riding/constituency) for the ‘shadow MP,’ at least not in lower-profile electorates. Ms Ardern was the Labour candidate in the very safe National seat of Waikato in 2008 (which no one expected her to win), but was elected as she was placed 20th on the Labour list.

    In 2011, she moved up to 13th on the list, as well as to the more high profile and marginal riding of Auckland Central (National took it from Labour in 2008), where she lost by only 717 votes (2.16%).

    Notably, Ms Ardern is not the only elected List MP who contested Auckland Central–the Green Party candidate was also elected from the list (and indeed, it appears that almost half of Ms Ardern’s electorate vote consisted of Green tactical votes, based on the “party vote” numbers). Given that the Greens have only 14 MPs, I wonder whether their “shadow MPs” represent themselves as the “list MP” for a single electorate, or whether they handle constituency duties for more than one electorate.

    Epsom, where the ACT Party won its only seat in 2011 (on a large number of National votes intended to get ACT into Parliament despite their lack of 5% of the party vote), is another riding with three MPs–an ACT electorate MP, and National and Labour list MPs. The absurdity of the “one electorate win and you get list seats too” rule is demonstrated by the votes in Epsom, where ACT received 44.1% of the electorate votes and a measly 2.55% of the party vote (just 939 of nearly 37,000 votes), compared to the Nationals’ 37.8% of the electorate vote and 64.52% of the list vote.


  3. There is sort of an Australian analogy in governments where the upper house is proportional and the lower house is majoritarian. Major party senators and MLCs typically take responsibility for lower house districts not held by their own party. Unfortunately the expression is ‘patron senator’ or ‘patron MLC’ which is not a very useful expression, but there it is.

    It does however reinforce Ed’s frequent contention that the Australian practice of proportional upper houses and majoritarian lower houses has some quite strong resemblances to MMP in the lower house. In Tasmania, where the pattern is reversed, assembly members do not patron upper house districts as far as I know.


  4. > ” ‘patron senator’ or ‘patron MLC’ ”

    The ALP uses “duty Senator/ MLC” which makes it sound like unenthusiastic drudgery – marginally better than sounding like characters from “Zorro”, I guess.

    One could make a case that if upper house members are used to make a de facto MMP system, party preselections should take account of that. Eg, not only should the upper house ticket be selected by a ballot of all financial members, but members enrolled in electoral districts that have been won by The Party for three or all of the last four elections get 1 vote; those in districts won by The Party for one or two of those elections get 2 votes; those in districts The Party has lost for all of the past four elections, get 3 votes.


  5. I’d argue quite strongly that both pre-selections and leadership contests should be by vote of all party members. I’m not sure the accident of where they live should increase their vote. I also agree with Senator Faulkner that the courts should be able to restrain inner party skullduggery.

    First, Party rules must be subject to the courts. Currently, Rule C1 “ Legal Status of Rules” states that the rules of the NSW Branch of the ALP are not judiciable. There are serious doubts if this rule is even enforceable, nevertheless, that rule should be abolished, and replaced with a rule that makes it expressly clear that any action or decision within the NSW ALP can be challenged in a court of law.

    I further believe that the rules and decisions of all political parties should be judiciable, and that State and Federal Governments should consider making a party’s eligibility for public funding contingent upon it.


  6. If anything, though, you’d want it to be reversed, because you want Senate candidates who will draw a strong above the line vote. Basing your candidate list on the votes of safe Coalition electorates draws a strong risk of losing votes to the Greens on the left.

    In the US, party conventions give more delegates to states that have voted more strongly in favor of the party (for the Democrats, you receive bonus delegates based on the state’s votes in the two most recent presidential elections as well as the most recent gubernatorial election, iirc). It’s enough to bump New York ahead of Texas in number of delegates overall, though elected official ‘superdelegates’ may add to that.

    It’s understandable that a state that isn’t going to vote Democratic not have as much weight, though it does seem rather unfair on one person one vote grounds. Personally, I think if delegate allocations are going to be awarded based on voting performance, rather than population, the swing states should be those most overrepresented (even if they’ve voted GOP in the last election), or else the states where the party’s vote has seen the most improvement over the past two cycles.


  7. Alan, Chris, hmmm, good point, I’d forgotten about US conventions rewarding their stronger States with more delegates. (I assume superdelegates would exacerbate that even further, that eg, a State that increases its Democratic vote from – say – 50% to 60% is going to increase its number of Democratic officeholders by a margin greater than one-sixth, ie may even have a clean sweep).

    My point is that if Australian parties see their federal and state upper house tickets as a way of representing their supporters who don’t have a district MP they voted for, then it’s similar logic to MMP.

    It seems to be very hard to find single-member district champions who actually stick consistently to the logic of single-member districts. (I pointed this out to some of my anti-STV friends in my former political party who wanted me to abandon “my” “local” MP on election day – “he’s gone, mate, the polls are hopeless” – and instead drive 30-40 minutes across town to campaign for a candidate in a more winnable seat.)

    In fairness, I have heard opponents of STV argue that Hare-Clark in Tasmania is only workable because there’s an upper house to provide single-member representation on local issues. (Curiously, the two chambers’ boundaries don’t coincide. I understand this is common for US State legislatures but cross-cutting is unknown in WA and Victoria – upper house regions are formed by grouping whole lower house districts).

    However, the fact that Tasmanian MLCs are usually elected as Independents, without party labels, and with strict campaign spending limits, no doubt makes them better at assiduously pursuing local issues than any partisan MP can be, whether elected by Hare-Clark, party ticket PR, or SMDs.


  8. It suprises me that no one here has criticised ‘shadowing’ on grounds that it may give certain constituencies more influence, and that pork-barrelling etc. might therefore be encouraged by MMP.


  9. JD (#8), interesting! The only studies I can think of offhand on “pork” activities in MMP systems suggest that only those elected in the nominal tier have pork incentives. However, neither study explicitly addresses the “shadowing” question. (I am referring to Lancaster and Patterson, CPS 1990, and Stratman and Baur, AJPS 2002. I promise more on this question by Shugart and others in the future.)

    Tom (#7), I do not know the general practice in US bicameral state legislatures, but in California, it is mandated that every state Senate district fully contain within its boundaries two Assembly districts.


  10. It depends on the state–in Texas, it’s definitely not the case. Counties with the population for more than 2 representatives have all representatives contained within the county, and while Senate seats are also supposed to follow county lines as closely as possible, they don’t tend to in large urban areas.

    Other states compound the issue with the concept of ‘floterial’ districts which float over smaller primary districts. New Hampshire is the most profligate user of this system–just within Hillsborough County, there are some voters who only get to elect two representatives in a single district, others who elect 11 representatives in a single district, and others who elect 5 representatives in 2 districts (their 2-member primary district and a 3-member floterial district shared with three other primary districts), all using plurality at-large voting. Their House of Reps has 400 members for a state population of 1.3 million (an average of 3,300 per Rep), almost 4 times as large as the cube root. The 24 Senators each represent a more normal 55,000.

    How this system passes muster under one man one vote I have no earthly idea.


  11. “In fairness, I have heard opponents of STV argue that Hare-Clark in Tasmania is only workable because there’s an upper house to provide single-member representation on local issues. (Curiously, the two chambers’ boundaries don’t coincide. I understand this is common for US State legislatures but cross-cutting is unknown in WA and Victoria – upper house regions are formed by grouping whole lower house districts).”

    Each Tasmanian MHA electorate has the same boundaries as the corresponding federal electorate (which are underpopulated compared to the rest of the Commonwealth as it is). The average total population per electorate is 100,000, smaller than our single-member House of Rep districts in Texas by nearly 70,000 (though certainly nowhere near New Hampshire’s 3300 per rep). NSW’s SMDs have nearly 80,000 people, while there are only 20,000 per MHA in Tasmania; the electorate of Murray-Darling is over 7.5 times larger than the largest electorate in Tassie, with just 1/5 the number of MLAs. Considering all this, the ‘no local representative’ argument sounds rather feeble.


  12. I think the idea, which I do not agree with, is that having “one local representative” for every person and only one representative is that it means that “any” candidate can be voted out and that everyone has the same level of direct representation. If the local representative doesn’t respect the people, he is voted out. If there are five representatives, none of them “belongs” to everyone and few of them can be voted out.

    I think that theory falls apart in safe districts where the incumbent will still pick up 60% of the vote or more even if he eats babies on national television or in any district where the 40% or more of the people who didn’t vote for the winner may feel that they have no effective representation at all.


  13. What l meant @7 is that when offered a choice of legislative representatives to approach, most citizens seem happy to take any who can assist them. Australian political parties certainly don’t think that their supporters living in hostile electorates are adequately serviced by “their” “local” MP (how could they be, when s/he belongs to Team Satan?!) and so designate duty Senators/MLCs. And clearly few – or no – voters in Auckland Central have been telling Ms Ardern “Pass off you unterloper! Wiv ulriddy got our virry own local Impee un the person uv Muss Nukki Kaye! You’re awskin us tuh commut thuh ilictoral uquavilint uv udultery uf we upproach you ibout constutuincy mitters!”
    What’s especially interesting is that duty Senators/MLCs and list MPs are not directly removable by the voters in that district, yet are still marketed by their parties as helping to provide adequate local representation. Shouldn’t, then, a second or third (or fifth, or seventh) MP for the area who’s actually elected by the residents thereof provide at least equivalent quality representation? Or is it essential that (a) “the” local MP be clearly first in the hierarchy, (b) voters not have a choice that might see party colleagues campaigning against each other, and/or (c) there be only 1, or at most 2, extra parliamentarians shadowing the electorate, whereas STV-PR adds too much of this good thing?


  14. One of the great criticisms levelled at PR-STV in Ireland in the last few years is that TDs are too responsive to their constituents, leaving them little time think great thoughts, and that seeking election under PR-STV is generally an undignified prospect for persons of stature and quality-a closed national list, or single-member districts, would allow for “parachuting” of such fine individuals (who mercifully but inconveniently lack the necessary common touch) into the Dáil.

    What I find curious is that persons who would accept and even celebrate this sort of competition in almost every other aspect of society reject it when it comes to political representation.


  15. Ireland uses some unusually small magnitudes,3, 4,and 5, which may contribute to the perception that TDs spend too much time on constituency work. The Tasmanian assembly had magnitude 7 until the duopoly tried to exclude the Greens by reducing the magnitude to 5. The ACT has 2 five-seaters and one seven-seater. Malta uses magnitude 5. Northern Ireland uses magnitude 6. The Australian upper house magnitudes range from 5 in Victoria to 21 in New South Wales.

    The solution to the irish TD problem may be larger electorates with a uniform magnitude of 5 or 7. The eminent persons who are just dying to serve their country could easily be elected in larger districts and could then have eminent careers meditating on the good, the true and the beautiful.


  16. I would agree that larger districts might help Ireland. But they also have a senate that doesn’t seem to do much of anything. That seems like a good place to put those “eminent persons” who cannot get elected due to lack of appeal to their constituents.


  17. They’re trying to phase out 3 seaters in Eire where possible, and have had up to 9 seaters in the past. Now that Fianna Fail is a smaller party, now would be the time to increase magnitude and therefore proportionality.


  18. The large magnitude impact on time spent on constituency affairs can go both ways. In a 3-seater, you have to get a quarter of voters to get a quota, so you have to please more of your constituents. In a 9 seater, you just have to keep 10% of them happy. The first would encourage broad high quality constituent service; the second encourages pork and special interest votes to keep your 10% of voters happy.

    One pro-STV argument not heard often enough is that it encourages internal competition. Members of the same party must compete to earn the higher preference from their party’s supporters (be that in predilection for a group ticket, or else at the general election). Members of both parties have to compete to offer the best constituency services, the best legislative ability and the best service in forming the government (be that as a minister, as a minor party with coalition bargaining power, or based on opposition performance). This internal competition between sitting MPs for local votes is something which few systems have.


  19. @ Chris, 18: New York has internal competition between legislators about once a decade. Since the state seems to lose two seats at a clip, the State Legislature usually jams to sitting Democrats into one seat and two sitting Republicans into an another and then draws nice safe districts for everyone else.


  20. Pingback: Should New Zealand do away with by-elections? | Fruits and Votes

  21. Pingback: MMP weekend: Germany and New Zealand 2017 | Fruits and Votes

  22. Just wanted to point out that in 4 years, Jacinda Ardern has gone from being a “shadow MP” for Auckland Central to the electorate MP for Mount Albert to the Prime Minister. Quite a meteoric rise.


    • …and now she’s in the Time 100 and arguably the moral leader of the free world. Not bad for a shadow MP. Not bad at all.


    • An alternative history might be: in 2011 Jacinda Ardern was #13 on the Labour list; in 2014 was #5 on the Labour list and had 42.1% of the nominal vote in Auckland Central where Labour got 21.6% of the party vote; in 2017 became Labour party leader.

      That narrative makes the rise less meteoric and the presence of an office in Auckland Central electorate unsurprising. But “shadow MP” does make for a catchier headline and a livelier F & V discussion.


  23. Firstly, MMP in Germany does not produce “shadow” MPs. It produces MPs who ran in a constituency “elected directly” or “elected from the list.” In either case they serve the constituency; the voter does not perceive the difference, as found in many studies.

    Secondly, Jacinda is not the first “shadow” to become PM in a Commonwealth country. Nicola Sturgeon was elected as a list MSP (losing the local contest) in 1999 and again in 2003; she was a list MSP when she became Depute Leader of the Scottish National Party in 2004. She was elected locally in 2007, 2011, and 2015.

    And it is unremarkable in Germany. Louis Massicotte wrote in 2004 “Of the seven federal chancellors in office since 1949, three (Adenauer, Erhardt, and Schmidt) were constituency members and two (Brandt and Schröder) were list members. Helmut Kohl was successively a list member (1982-1990) and a constituency member (1990-1998). On becoming chancellor during the life of a Parliament, Kurt Georg Kiesinger (1966-1969) had no parliamentary mandate and could not get one because of the absence of by-elections.

    “Helmut Kohl’s case is interesting. He was the Minister-President of Rhineland-Palatinate (1969-1976), and in 1973 was elected federal president of the CDU. He was a candidate for the chancellor’s office in the 1976 election, and he ran both in the constituency of Ludwigshafen, his hometown, and as list frontrunner in the Land. Ludwigshafen was not a safe party seat and he was defeated there, but he entered the Bundestag through the list. The same scenario took place in 1980, 1983, and 1987. In 1990, in the euphoria of reunification, he succeeded in getting elected in Ludwigshafen (while still taking care to put his name on the list) and repeated this feat in 1994. In 1998, when his party lost power, he was personally defeated in his constituency but was saved by the list. Nothing indicates that his constituency setbacks in any way undermined the personal authority of the longest-lasting party leader and government leader in the history of the Federal Republic of Germany. Two Kohl biographies, including one full of details on his rise to power, make no mention of his successes or failures in this area.”


    • Germans may not use the term, but they have the phenomenon: list MPs who were candidates in a constituency (and typically maintain an office there).

      Is there no difference? There is indeed evidence of differences, for instance in allocation of speaking time in the Bundestag (Proksch and Slapin), and in committee assignments (my and coauthors’ in-progress Party Personnel, as well as Lancaster and Patterson, among others).


      • Sure. I don’t know what that changes in my point, other than generalizing beyond “second” MP. There is always one constituency MP for each constituency, by definition under MMP. There may be zero, one, two, or conceivably more list MPs who had contested the same constituency. The latter can be called “shadow MPs”, although I am by no means wedded to that term, and in the first paragraph of the post above, I already noted a potential confusion from the term. In fact, in my academic works I have never used it. As the original post here stated, I proposed it in light of the tendency I observed of NZ constituency and list MPs’ offices being close to one another. And because we need something better than ‘zombie’, which actually has appeared in the academic literature on MMP and MMM, much to my disdain.


      • So I suppose my response then, as a PR supporter who ranks MMP third (behind STV and flexible regional lists), is… What is the point of continuing with elections in single-seat districts? Why not just have closed lists with variation of the order based on localised voting patterns? It’s like abolishing the monarchy but then stipulating that candidates in presidential elections must demonstrate direct blood descent from Ulffgar The Conqueror in 936 CE.
        I’m wary of PR supporters buying into (what I see as) the myth that restricting voters to one single MP per district, denying them choice, promotes better constituency service. (Exhibit A: Ireland, The Republic of). Therefore am pleased to learn from Wilf that MMP doesn’t in fact do this. But the incentive for good constituency service should be a fairly meaningful stick or carrot. Under STV or open/ flexible regional lists, it’s the difference between “You get elected” vs “you’re out of the legislature” (or “you get a full term” vs “you might get to fill a casual vacancy as the highest-polling runner-up”. With MMP, the difference seems to be “You, um, get to brag about being a constituency MP as opposed to being a ‘mere’ list MP and you, um, get more speaking time in the Chamber and a better choice of plum committee assignments. But it doesn’t stop you ascending to the top job”. Loath as I am to agree with the NZ Business Council, they seem correct that List MPs are effectively irremovable by the voters and to have them claim to be “respresenting” a district that has – if one follows the same logic as Chapter I of the MMP Elections Statute – “rejected” them at the ballot box.


  24. EDIT: … effectively irremovable by the voters – but they still get to claim to be “representing” a district…


    • “The NZ Business Council seem correct that List MPs are effectively irremovable by the voters.” That’s why virtually no one in Canada proposes closed-list MMP, the favourite is open-regional-list MMP, the second choice being “semi-open” or “flexible” or “open-list with party option.” The other option discussed is Baden-Wurttemberg no-list MMP (or “best runners-up” MMP).


  25. I suppose ‘shadow MP’ is an obvious construct from ‘shadow minister’, in the sense that a shadow MP shadows the MP in the way that a shadow minister shadows a sitting minister.

    I believe the Canadian term for a shadow minister is ‘critic’, although Round’s Law of Canadian Political Terminology tells us it may well be ‘parliamentary eviscerator’ or something.


  26. Pingback: MMP in NZ: An example of “best of both worlds” in action | Fruits and Votes

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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.