THE CORE

Fruits & Votes is the Web-log of Matthew S. Shugart ("MSS"), Professor of Political Science, University of California, Davis.

Perspectives on electoral systems, constitutional design, and policy around the world, based primarily on my research interests.

Also experiences with growing many varieties of fruit (always organic) and other personal interests. Please see the Mission Statement for more. (There is also an explanation of the banner.)

Other "planters" have been invited to contribute. Please check the "Planted by" line to see the author of the post you are reading.

Join the conversation. Comments are always open. Except, that is, when Word Press mysteriously shuts them down, which happens with distressing frequency.

Core principles:

Henry Droop on the "moderate non-partisan section"

Madison on "dangers from abroad" and "the fetters... on liberty"

The Head Orchardist's other sites:

PRESERVED FRUIT
orchard blocks
  • All
  • FRUITS
  • VOTES
  • wide open spaces
  • 01 March 2007

    Planted by MSS
    Planted in: ELECTORAL SYSTEMS & REFORM; STV

    In the previous thread on Thailand’s possible retrogression back to MNTV or even SNTV, Suaprazzodi asked why more countries do not use the single transferable vote, so we can find out how it might work outside of the narrow context of a few former British colonies.

    I suspect there are various reasons why STV is not widely used, mostly having to do with party leaders not liking the way it undermines their own authority within their parties. After all, Irish politicians have tried more than once to get rid of STV, only to have voters reject their efforts at referendum. And in most Australian jurisdictions, it is used with an above-the-line vote that makes it relatively close to a closed-list system by giving party leaders the ability to determine how the votes of above-the-line voters (typically 90% or more of the total) are transferred.

    Of course, these reasons would not be relevant to a context like Thailand, where the apparent intention of the current authorities is to undermine parties. However, in such a context, STV has another disadvantage: if those choosing the new electoral system want a highly personalized system as a means to allow the election of local (and usually conservative) notables, STV is less desirable than SNTV, because while the latter (and MNTV, too) allows election of candidates with extremely narrow bases of support (see some good examples in the Afghanistan subdomain), STV forces them to compromise and gain second (and third, etc.) preferences.

    Thus STV may be undesirable to most elites for precisely the reason that its supporters say it is so good for voters: It empowers the wrong kinds of people. Voters!

    (I should note that I am rather agnostic about STV. I like its voter-empowering aspects, but the small magnitudes needed to make it practical are not sufficiently favorable to the kinds of parties I tend to favor, and it is a bit too favorable to “friends and family” networks and localism for my taste. But it sure beats any NTV system–including cumulative vote–or plurality!)

    Anyway, regarding STV’s use outside of that small subset of jurisdictions:

    It is worth mentioning the use of STV in Estonia as the country was, for a short time, the exception to the rule that STV was only used in countries with a British imperial background.

    STV was used for local elections in December 1989 and for a national election in 1990. The use of STV clearly occurred at a time of significant social, political and regime change within Estonia and Eastern Europe more generally. The choice of STV for elections was a
    compromise decision made by the major political parties in Estonia at that time. These were: the Communist Party of Estonia (CPE), Popular Front of Estonia (PFE), and the Joint Council of Work Collectives (JCWC). Taagepera (1998) comments on the reason for this compromise as follows:

    “It [STV] was adopted because it satisfied the Communist need to avoid party
    lists and labels while still leading to vague proportional representation. The
    district magnitude (seats per constituency) was left to the discretion of local
    powerholders, which satisfied the JCWC, because they could and did chose
    one-seat districts in the north-east, reducing STV to Australian-type alternative
    vote” (p.30).

    The rules surrounding the use of STV also varied from those used in other countries. As noted
    in the above quote, the number of candidates elected in each constituency varied from one to
    five candidates with the number of candidates from each seat being determined by local county
    and local authorities. A legacy of the Soviet era was that a turnout of at least 50% was required
    before an election result could be considered valid. In addition, the electoral rules also required
    that:

    “a candidate must attain 50% of the first place votes in order to win election. If
    not, then a runoff election was to be held in which the top two contenders were
    to be pitted against one another (if there were ties between three or more
    candidates for first place and two or more candidates for second place, then the
    number of second place votes received acted as a tie breaker, and so on”
    (Ishiyama 1996, p.493).

    In the context of the socio-political upheaval in Estonia during the time-period when STV was
    used it is not entirely surprising that the electoral system was changed. A number of reasons
    have been posited for this decision. Taagepera (1998) notes that some parties criticised STV
    for weakening emerging party structures whilst also commenting:

    “many Estonians were uneasy about their inability to figure out how the votes
    were converted into seats, and they suspected opportunity for fraud, although
    no formal complaints were lodged” (p.31).

    Ishiyama (1996) also notes that STV was perceived as inhibiting the emergence of political
    parties and noted that this made STV “dangerous for a country beset by regionalism and the
    potential for ethnic conflict” (p.499).

    The quoted text–and sorry for the poor formatting–is a quotation from a study prepared for the Scottish Parliament, “The Single Transferable Vote in Practice,” by Stephen Herbert. It is available in PDF.

    Propagation: Seeds & scions (11)


    Greens for Greens â„¢ grafted Why not STV?

    11 ideas sprouting »

    1. I like STV because of the small district magnitudes and that voters living in the districts don’t have just one politician to represent them when problems arise but at least a minimum of 3 and maximum of 5 in Ireland. I don’t think STV districts should be no smaller than 3 and no larger than 9 in any given country.

      I think this electoral system is no doubt the best for local governments like city and county, but maybe not for at the national level for some countries.

      The disadvantage of STV is when a candidate is declared elected and has a surplus. It’s is very hard to understand the very complicated mathmatical allocation of the surplus. Some systems have all the votes are transfered at a fractional percentage of the surplus or a weight sample. The counting of the votes and translating them into seats is very complicated.

      What other disadvantages can anyone here reading this blog can think of STV?

      I can think of some more disadvantages of STV. It could not be used in most third world countries where voters have a low literacy rate as well as low numeracy or could it?

      It would be fascinating if a culturally diverse society tried it out to see if different candidates would reach other to other ethnic groups.

      Would this system be more applicable to Fiji than say the Alternative Vote is now?

      Seed planted by Suaprazzodi — 02 March 2007 @ 04:04

    2. Ireland is an excellent example of a country divided, if not by ethnicity, by religion. The treaty which created the Republic and the Province of Northern Ireland (I’m working quite hard to escape the intellectual confines of the Irishcatholictribalghetto here) required STV on both sides of the border. The Republic, as we know, still uses STV.

      The Province abolished it in favour of STV as soon as they were allowed under the treaty, Catholic/Nationalist representation in the provincial parliament fell to zero and stayed that way until the 1970s. Not only did minority representation disappear overnight, but moderate Protestant/Unionist politicians were almost immediately replaced by more extreme tendencies within the Protestant/Unionist community. Obviously the course of social and political development north and south of the border is not determined by the electoral rules on their own, but STV certainly opened up space for competing views in the Republic that were excluded by FPTP in Northern Ireland.

      On a separate issue, the only example I can quote is a student election where I was actually the returning officer in the early 90s, but it is possible to retain ticket voting without turning STV into closed list. I designed a ballot paper where all tickets and all candidates were placed randomly and you did the usual 1, 2, 3 thing. You could vote for a couple of individual candidates and then a ticket, or you could express preferences between different tickets. I was mildly proud of the thing. It answers Matthew’s concerns.

      Seed planted by Alan — 02 March 2007 @ 13:54

    3. Alan, I find your suggestion very intriguing. I’m not sure I entirely understand it though. A “ticket vote” generally means that you make a single selection that specifies all your preferences. I can easily imagine how it makes sense to choose first some individuals, then a ticket–it would simply choose those candidates as the first thru Nth preferences, then follow the ticket for all the following preferences. But what happens when you select a ticket and then an individual, or a ticket and then another ticket? Once you specify a ticket, doesn’t that use up all your remaining preferences?

      The idea of choosing between tickets does suggest a strange way to allow pseudo-STV with large district magnitudes. If one only allowed ticket votes, and somehow limited the number of tickets available, you wouldn’t run into the problem of ballot-choice-overload that you would get with regular large-magnitude STV. I’m not sure if there’s any practical way of doing this. Just as a theoretical example, uou could use something like the system in international figure skating to create a limit of tickets-per party based on the vote counts in the last election. And then I suppose mandate a particular ticket-selection primary process to make manipulation difficult. Clearly all parties would group their candidates at the top of their ticket(s), but lower preferences could be distributed differently, and of course different tickets would have different orders. (Incidentally, I know there are a number of serious faults with this system–feel free to bring them up if you’d like. Or suggest a different system along the same general contraints–STV with only tickets, and a limited number thereof.)

      Is there perhaps an existing preference system that’s appropriate in large-magnitude districts? I’ve never heard of one, but I’m sure someone here is just itching to enlighten me.

      Seed planted by vasi — 03 March 2007 @ 06:08

    4. In elections for the NSW legislative council, parties are limited to an order of preferences among their own candidates. The citizen gets to choose which tickets receive preferences if is their vote does not exhaust with the ticket. So if I cast a ticket vote for the Labor party, and my vote does not exhaust with labor candidates, the remainder will transfer to my second ticket and so on.

      The NSW electoral commission’s how to vote advice explains it better than I have.

      The NSW system still requires you to choose between ticket voting and individual voting, an entirely unnecessary rule. Allow a mix of ticket voting and individual voting and you overcome STV’s magnitude problems without giving political parties sole choice of who gets elected.

      I once filled out every square from 1 to 243 on a Senate ballot paper because there was no ticket that matched my preferences.

      Seed planted by Alan — 04 March 2007 @ 05:40

    5. On STV, you say “the small magnitudes needed to make it practical are not sufficiently favorable to the kinds of parties I tend to favor, and it is a bit too favorable to “friends and family” networks and localism for my taste.” It also weakens party structures (and principles) by giving an advantage to those candidates who most compromise and gain second (and third, etc.) preferences.

      But on the subject of magnitudes, those who look only at Ireland’s current Dail may be misled. Throughout Ireland a large number of towns have nine-seater districts, as the city of Cambridge Mass. still does. The Irish Dail originally had a 9-seater, three 8-seaters and five 7-seaters, that together made up almost half of the Dail’s 147 elected members. Tasmania ran quite nicely with all 7-seaters. Northern Ireland finds 6-seaters quite practical. With a 9-seaters, a party winning 6% in the district has a chance of taking the seat on second preferences, which could well be a party winning only 3% or so across the jurisdiction.

      While I’m not an STV fan, it can be more proportional than the current picture in Ireland would suggest. Ireland gradually cut its magnitudes in half, not to make STV more practical, but to give Fianna Fail a bigger bonus. That, of course, is another flaw of STV: it’s too easy to “tullymander.”

      Seed planted by Wilf Day — 04 March 2007 @ 13:27

    6. I’d contest that STV is too easy to tullymander. All that’s required is a rule against small magnitude districts, and possibly against even magnitude districts. One of the tullymander practices was even magnitudes in opposition areas and odd magnitudes in government areas.

      It’s as valid as saying that MMP can be manipulated (as it can) by fiddling the proportion of list MPs and district MPs, the threshold, and the way overhang seats are dealt with. A fair system of MMP obviously does not manipulate those rules. A fair system of STV does not use tiny magnitudes manipulated by the government.

      PR as a whole is bedevilled by multiple myths about instability, party control, capital-centred MPs and a number of others. Let us not now make up some new ones about STV.

      Seed planted by Alan — 04 March 2007 @ 14:58

    7. “PR as a whole is bedevilled by multiple myths… Let us not now make up some new ones about STV.”

      Right on to that. I suppose I am so guilty, and for a blog dedicated to myth-busting about electoral systems, I should not allow that!

      Thanks for the reminder, Alan.

      I am intrigued by your comment above, in which you said:

      “requir[ing] you to choose between ticket voting and individual voting [is] an entirely unnecessary rule. Allow a mix of ticket voting and individual voting and you overcome STV’s magnitude problems without giving political parties sole choice of who gets elected.”

      I am not sure I understand what this is getting at, in terms of practical system design.

      Seed planted by MSS — 04 March 2007 @ 22:35

    8. “I’d contest that STV is too easy to tullymander.”

      Tell that to the Irish. Or the Tasmanian Greens, who were almost wiped out when the two main parties shrank the house from 7-seaters to 5-seaters, side-swiping women’s representation in the process.

      But even if political norms somehow prevent such partisan behaviour through leaving all this to an independent Boundaries Commission, you are giving them a most unusual task: rather than just identifying communities of interest, they are deciding District Magnitudes and hence proportionality.

      At this very moment the British Columbia Boundaries Commission is drawing up a map to be voted on at the coming second referendum, and no one can say precisely what the average DM will be.

      The same phenomenon occurred in the run-up to the coming local government elections in Scotland: there will be books written on how DM was juggled for those elections.

      Seed planted by Wilf Day — 05 March 2007 @ 02:47

    9. Maybe they should just settle for a standard district magnitude of 5 members in all districts like Malta does.

      Seed planted by Suaprazzodi — 05 March 2007 @ 05:03

    10. In terms of ballot design. All STV ticket voting systems now in use require you to choose between voting only for a ticket or only for individual candidates.

      There’s actually no reason for that rule, except the self-interest of the parties. They overplayed their hand in NSW with the result that voters now get to preference between (but not within) tickets. If voters could mix individual and ticket votes the entire problem would go away. Magnitude is an issue with STV, but so is letting citizens, not parties, decide who gets elected.

      A system that uses standard magnitude is obviously better. As far as ‘telling it to the Tasmanian Greens’, I told it quite a lot to the Tasmanian government and opposition when they they conspired to cut district magnitude from 7 to 5 on the theory they could exclude the greens from the assembly. It didn’t work. The Greens hold 4 seats out of 25 in the current assembly. That is a higher proportion than they held before the magnitude reduction. The equivalent would be raising the MMP threshold vote to exclude inconvenient parties.

      A group excluded from representation by manipulating STV magnitudes still gets to influence the election through second preferences. A group excluded from representation by manipulating the MMP threshold has its votes literally thrown away.

      High democratic dudgeon is a fine thing, but I’d prefer my votes counted somewhere in the election rather than tossed in the trash.

      Seed planted by Alan — 06 March 2007 @ 03:59

    11. Why not STV?

      Scion grafted by Greens for Greens â„¢ — 17 March 2007 @ 22:39

    RSS feed for comments on this post.

    TrackBacks

    To graft a scion to this planting, please use the following URL:
    http://fruitsandvotes.com/blog/wp-trackback.php?p=1158
    (Non-MT bloggers click here to send pings.)

    Grafted scions that are not compatible with this planting's stock will die or be pruned out by the Orchardist.

    About the comment form

    Please note that the name you enter below and the first several words of your comment will appear on the right sidebar of the blog's front page, under "Propagation." New propagators might want to look at the comment policy.

    Please do not enter long URLs into the seedbed. Either mark them up using html hyperlinks or convert them to a "tiny URL." Thank you!

    Seedbed

    The soil is ready for planting:

    `

    FRUIT FEEDS
    PROPAGATION
    Recent comments.

  • Does STV have anything to do with absence of “free votes” in Ireland? (12)
    • Alan: What Tom said, except that I’d add that the major parties in Australia have a habit of subverting their own rules by imposing...
    • Tom Round: JD, because a government body has an electoral roll stating that “These people are registered supporters of the Democratic Party,...
    • JD: Tom, I’m not sure I understand why primaries the secret ballot. Alan, how is that different from a (closed) primary?
    • Alan: I’m not a fan of primaries, for the reasons Tom states. I am a fan of requiring parties to nominate candidates by a ballot of all party...
    • Tom Round: It would indeed be ironic if one reason discouraging parties from allowing free votes was an electoral system that could enable voters...
  • BC election 2013 (8)
    • Chris: The federal Liberal party hate the Conservatives more than they hate the NDP. They think Trudeau fil will get them a majority government,...
    • Ed: Its been explained to me that BC politics seems complicated, but is actually pretty simple: everyone gangs up against the NDP, but the...
    • MSS: I am struck by the degree of malapportionmen t in BC. For instance, the Peace River South winner’s 46.4% was only 3,904 votes, whereas...
    • MSS: The Green Party won the Oak Bay-Gordon Head seat, with 40.1%. It was not close, with incumbent Liberal Ida Chong having only 29.7% and the NDP...
    • MSS: I guess this is why they still have actual elections with actual voters casting actual ballots! How could the pollsters be so wrong?
    • Vasi: Well that was surprising! Once again, the polls in a Canadian election were off, and the incumbents do much better than expected.
    • Tom Round: BC: where a 39% sub-plurality is enough to elect a government for five years (absolute majority of seats, no upper house) but a 57%...
    • JD: Oh, how different (and more interesting) things would have been had STV been approved…
  • Final MMP Review report is out (11)
    • Suaprazzodi: Perhaps they should put the amended version of MMP to a referendum. They should ask questions like do you want the list percentage...
  • CROSS-POLLINATION

    FRUITS

    morn_blms_corralito.jpg

    The Fruit Blog (Fruit & fruit breeding)
    Daley's Fruit Tree Blog
    Orchards Forever
    The Orchard Keeper
    The Ethicurean
    The Jew and the Carrot
    Small farms ("real people & real food")
    Life begins at 30 (Farmers markets, etc.)
    Banana
    Festival of Trees
    Rare Fruit News Online
    Cloudforest Cafe


    VOTES

    bulgaria_protest copy

    Comparative democracy

    Psephos (Adam Carr's data archive)
    Electoral Panorama
    World Elections
    African Elections Database
    M. Herrera's Electoral Calendar
    Electoral Geography (Data archive)
    Michael Gallagher's data archive
    Election Finance (Blog, data archive)
    IFES
    Election Law (Rick Hasen)
    VoteLaw (Edward Still)
    Ballot Access News

    Electoral and Political Reform

    The FairVote Blog (US)
    Make my vote count (UK)
    Wilf Day (Canada)
    democraticSPACE (Canada)
    Citizens Assembly Blog (dormant)


    POLITOLOGY

    Blogs of political analysis

    PoliBlog
    Arms and Influence (dormant)
    Outside the Beltway
    Political Science Weblog (abstracts)
    Ideological Cartography (Adam Bonica)
    Frontloading HQ (Josh Putnam)
    FiveThirtyEight
    Vote View (Keith Poole)
    The Monkey Cage
    A Plain Blog About Politics (Jonathan Bernstein)
    Political Arithmetik (dormant)
    Polls & Votes
    Pollster.com
    Polysigh
    Reflective Pundit
    Rustbelt Intellectual
    Simon Jackman
    The semi-presidential one
    Josep Colomer
    Chapel Hill Treehouse (dormant)
    Political Behavior (dormant)
    Dart-Throwing Chimp
    Countries at the Crossroads (Freedom House blog)
    Jacob T. Levy

    REGIONAL ANALYSIS

    Canada

    The Mace
    ThreeHundredEight
    Crawl Across the Ocean
    Idealistic Pragmatist

    Europe

    Centre for European Politics
    Dr Sean's Diary
    A Fistful of Euros
    Political Reform (Ireland)
    UK Polling Report
    British Politics & Policy (LSE)

    Latin America

    Bloggings by boz
    Two Weeks Notice

    S.W. Asia & E. Mediterranean & N. Africa

    Informed Comment Global Affairs
    Lisa Goldman
    Michael J. Totten
    Yaacov Lozowick
    Marc Lynch (@FP)
    Ahwa Talk

    Africa

    La Constitution en Afrique

    E. Asia

    Frozen Garlic (Taiwan elections)

    New Zealand

    Kiwiblog
    No Right Turn

    OTHER SOCIAL SCIENCE BLOGS

    Crooked Timber
    Statistical Modeling
    Social Science Statistics
    Cold Spring Shops
    Marginal Revolution
    Brad DeLong
    Greg Mankiw

    SUN & MOON

    CURRENT MOON

    NEWS

    ABC

    BBC

    CBC

    Democracy Now!

    Deutsche Welle

    El Tiempo

    Guardian

    Haaretz

    Hindustan Times

    The Independent

    Irish Times

    NZ Stuff

    RFE/RL

    ORGANIZATIONS

    About/disclaimer

    California Rare Fruit Growers

    Center for Voting and Democracy

    Californians for Electoral Reform

    Society for American Baseball Research

    Link TV

    SCION EXCHANGE

    HARVESTS
    ORCHARD SERVICES

    Powered by WordPress