Could UK Greens win a seat? (Re-ask)

Back in December, 2007, I asked if the UK Green Party could win a seat in the next House of Commons election; the question produced an interesting multi-national discussion of Green politics that is well worth reading again.

Now that an actual election campaign is underway in the UK, with the likely polling date 6 May, it looks like the answer to the question just might be YES!

A few days ago The Guardian discussed the prospects of the Green candidate in Brighton Pavilion constituency, Caroline Lucas (currently a Green Member of the European Parliament from the regional PR district that includes Brighton in EP elections).

Given that this UK election campaign is also taking place within the context of a debate over a potential future referendum on electoral reform, it is worth asking whether the Greens’ chances of winning even one seat are better with the current first-past-the-post (FPTP) system than with the proposed alternative vote (AV). A recent poll by ICM Research and touted on Lucas’s website, puts the Brighton Pavilion candidates of the various parties with the following vote percentages:

    35 Green
    27 Conservative
    25 Labour
    11 LibDem

While the “instant runoff” procedure of AV normally can be expected to make the final count come down to the top-two candidates, a field such as this is precisely the sort of contest in which AV might favor the third-placed candidate. If LibDem voters are more likely to give their second-preference votes to a third-running Labour candidate than to either of the top two, the sequential-elimination and transfer procedure of AV could result in a final round of counting between Green and Labour. Then Conservative preferences come in to play, and we are down to the question of whether they tend more strongly to lean towards “anybody but Labour” or towards “safe establishment over new politics.”

In any case, it has never been clear to me that parties like the Greens should prefer AV (or IRV as it is often called in the USA) over FPTP. Obviously, any form of proportional representation is superior to either single-seat system.

31 thoughts on “Could UK Greens win a seat? (Re-ask)

  1. I noticed this with some Welsh nationalist blogs I read-they seem weirdly positive about AV. I would have thought that it would be a disaster for them, in that they could get squeezed by transfers between the unionist parties. Perhaps they consider the level of polarisation for other reasons between the unionist parties is enough to dampen this effect, but still.

    I do wonder if it isn’t part of the appeal of AV for the (Scottish) leader of the Labour Party.

  2. I used to be a big fan of AV, but am growing less enthusiastic. Consider the ICM poll:

    35 Green
    27 Conservative
    25 Labour
    11 LibDem

    In this situation, if I was a Lib Dem supporter, I would seriously consider switching to the Greens, the Conservative, or Labour if one of those candidates was as strong second choice for me. In other words, my vote would migrate to one of these other parties just as if there was AV. But if I didn’t care too much about whether one of the top three won, I could choose to “waste” my first preference vote on the LibDem.

    Now a low information voter would not be aware of the poll, and in fact would likely not vote for the Green out of a desire not to “waste” his vote. But being a low information voter really means voting for the two two national parties anyway. I am persuaded by the Spike article referred to in comments earlier that compulsory voting tends to help the two larger parties.

    So I’m generally pro AV and pro runoff, but I’m sceptical of how much it really changes from the FPTP situation, since voters will create virtual AV situations by moving away from the back of the pack parties over the course of a campaign.

    Also, the Greens will concentrate their resources in this constituency since with a FPTP situation, there is a huge value for them in winning a seat, high enough that its rational for them to sacrifice votes in other constituencies. Under AV, the Greens maximize their influence by directing the second preferences of their supporters, so their politicians will concentrate first on maxmizing the number of second preferences they can direct, then in trying to direct them. They will put less resources into winning a seat. So I suspect that AV may in fact decrease the chances of minor parties winning seats in the legislature because it changes the politicians’ behavior, even if it doesn’t much change the behavior of the voters.

  3. I agree with Ed (and the Spike article mentioned). This is one of the main reasons why I oppose AV – it can hurt small parties by raising the effective threshold from Arend Lijphart’s nominal 35% under a plurality system to over 50%. I live in Scotland, where a multiparty system developed in the 1970s under plurality/first-past-the-post voting (while Labour dominated for some time, the rise of the Scottish National Party (SNP) and persistence of the Liberals/Liberal Democrats and the Conservatives meant Scotland had multiparty politics). Labour’s reason for (reluctantly) supporting proportional representation for the Scottish Parliament was to prevent the SNP from winning a ‘manufactured’ majority of seats some time in the future under FPP and claiming a ‘mandate’ for independence. This defensive reason for introducing PR has been admitted by Labour (it is on record – a former first minister said it in a BBC interview), and it is the same basic reason why PR was introduced in so many other countries, if you accept the argumentation of a number of scholars (like Josep Colomer).

    This is why I oppose AV and would (reluctantly) prefer to keep FPP, hoping that conditions for a ‘defensive’ move to PR will develop (further). British voting patterns at Westminster elections have shifted dramatically since about 1970. In the 1950s and 60s, around 90% of the votes went for the two largest parties; in 2005, only 67% did. With PR voting used in other UK elections, and multiparty systems prevalent, I would rather keep FPP than change to AV, which would probably hurt small parties (yes, the Liberal Democrats might gain slightly, but this is a medium-sized party, not a small one).

    Much of this comes down to one’s values. I am not a majoritarian – if you are (and there’s nothing wrong with that!), then it makes sense to support AV, since it appears to promote enhanced majoritarian outcomes. This appears to be what some in the Labour Party want, hoping that they will get a lot of Lib Dem voters’ second preferences (which may not be the case as much as it was in the 1990s). If you prefer what Lijphart calls the consensus model of democracy, which I do, then you should support PR. What frustrates me most in this debate (in the UK) is people and groups (like the UK Electoral Reform Society and the Liberal Democrats) who claim to support a more consensual, pluralistic form of democracy but call for AV in the apparent belief that it is somehow a step in the ‘right direction’. This simply does not make sense – AV appears to be more majoritarian than FPP. There is no reason to believe that AV would be merely some kind of transitional phase to PR. It would probably become entrenched. The supposed extra choice for voters is unlikely to be effective in the sense of allowing these smaller parties to win seats, leading to even more disenchantment with British democracy. Advocates of AV here appear to be taken in by the (superficial) preferential nature of AV and STV. This is dangerously naive. While the Lib Dems may see some partisan seat advantage in AV (which might only be short-term in nature – what happens to Lib Dem autonomy if Labour makes demands on Lib Dem policies in return for directing second preferences their way?), the Electoral Reform Society’s position just seems bizarre. I just can’t explain it. This is why I am not a member, despite my long-standing support for PR. AV and STV are only similar in a superficial way – using preferential ballots. Otherwise, they are in entirely different electoral system ‘families’ which have completely different philosophical underpinnings and value systems.

    While the proposed AV referendum amendment did pass the Commons last night, there is little chance of it passing the Lords before Parliament is dissolved for the election expected in May:

    Dr Thomas Lundberg, University of Glasgow

  4. It would seem that AV is more majoritarian than FPTP only under simple plurality FPTP. In FPTP with runoffs (or in an entrenched two party system and primaries), minor parties are less likely to compete where their odds of making it into the final round are low, and minor party voters are less likely to turn out and influence a runoff outcome.

    Under those conditions, it would seem that Green candidates or other minor party groups should prefer AV over FPTP+runoff, because they are more likely to break through the ranks.

  5. With all due respect to old friends Matt Shugart and Tom Lundberg, you might want to consider what small parties in the UK and elsewhere think about AV before suggesting that first-past-the-post is better for them. And note that the British Electoral Reform Society, deeply grounded in advocacy for PR, backed a yes vote yesterday.

    Elections are about representation, to be sure, but they are much more than that- – they are about having a chance to make your case, to mobilize backers. to influence debate between elections and participate in campaigns based on what you have to say rather than your potential role as a spoiler. They are about your efforts to transform public opinion rather than just operate within it. If you discount all that, there’s a rather weak argument that first-past-the-post is not much worse than AV — but if you accept them as considerations, it’s simply absurd to think that FPP is better for small parties.

  6. Yes, Tom L., PR-STV advocates favoring AV as a step towards their preferred system would be a bit like list-PR advocates favoring FPTP (assuming perhaps some pre-existing multi-seat or multi-round majoritarian system as has been the case at some time in some Continental countries, including France still). In both cases, after all, the formula for casting and counting votes is the same in the single-seat-district version as in the preferred PR version.

    MLatner: “FPTP with runoffs” is an oxymoron, because the existence of a runoff provision permits the second-placed candidate in the initial count to be the eventual winner. I do agree that the calculation for small-party supporters regarding the merits of AV vs. FPTP (i.e. on-round plurality) for an entrenched two-party system are different from those in an existing multiparty system.

    Thus I completely agree with Tom L. that supporters of the representation of small (especially non-regional) parties in the UK should prefer, albeit reluctantly, to keep FPTP rather than move to AV. However, supporters of small parties in the US might consider AV (‘IRV’) their only hope of having any influence in House elections and favor, albeit reluctantly, a move to AV. (I remain unconvinced myself, but were I in the UK or Canada, I would be with Tom on this question.)

  7. It is good to see Rob join the debate. I commented at #6 before seeing Rob’s comment in the moderation queue.

    It really does come down to what you want from your small party. Personally, I would prefer the UK Greens get even one seat, taking advantage of the effective FPTP district-level threshold of about 35%, and have an effective spokesperson for my principles in the legislature. I would much prefer this to the possibility that I have no actual spokesperson, but my party becomes an appendage of the leading center-left party, transferring its preference votes to some of their winning candidates.

    I recognize that reasonable electoral-reform advocates may disagree on this point, and as noted above, I also agree that conditions in the US are different from those in the UK (or Canada): my preferred outcome might be vastly more remote here than in the UK, making the second-best outcome the best we can ever hope for. On the other hand, I have not totally given up hope that Greens might one day elect a member of the House from some district somewhere, and then build on that.

  8. Matthew,

    We’ll have to disagree on this one. The “spoiler” dynamic is much more likely to suppress potential small party support in such situations and not put them in a position like this example where they have a shot to win under any winner-take-all system.

    But more broadly, I simply disagree with you about discounting what it means to be able to run aggressively everywhere and suspect the great majority of serious small party candidates would agree with me — for instance, try making your case to Australian Greens that they would be better off getting rid of IRV and going to first-past-the-post. The Greens ran candidates in every district in 2007, won nearly 10% of the vote and that vote translated into seats in the senate =– and all without being bashed as “spoilers.”

    But hey, we agree on the value of winning proportional representation – -the obvious route to fair representation.

  9. This also comes down to what you want out an electoral system, echoing some earlier comments. Here are some possible values:

    1. Ensuring the government/ successful candidate is actually supported by the majority of the population.

    2. Ensuring that all political factions with some degree of support have spokesmen in the legislature.

    3. Ensuring that all political factions have representation in the legislature in proportion to their support in the population. This is somewhat different than #2.

    4. Ensuring the elected government can actually implement its program.

    5. A small geographical area gets a representative.

    To some extent these conflict, but the political system you favor probably is driven by how you rank these values. AV is good on #5, since it keeps lots of single member districts, good on #4, good on #1 within an individual seat. FTPT may likely be better than AV on #2, worse on #1, and about the same in the other three areas.

    Of course the current system in the US has problems with all five!

  10. Compare France with UK or US – France have a two-round majoritary system, and AV/IRV can be considered a kind of variant of this (wuth more rounds and an authomatical process).

    And small parties have traditionally more stength in France than in UK or US.

    It is possible that, for a given number of votes, a small party had more chances of elect a representative in FPTP than in AV. But, in AV, small parties will have more votes than in FPTP.

  11. In an existing multi-party system, FPTP can yield seats to geographically concentrated small parties, but in a two round system in, say, local elections, two-round (what I meant earlier by FPTP+runoff) doesn’t offer supporters of unpopular(first round) candidates much hope, but they can still be influential voters, and have a better chance at getting their first preference under AV, no?

  12. The US has that other characteristic-a system where gerrymandering is both constitutional and frequent. So even if you build a viable base under FPTP and get one rep elected, there’s a good to excellent chance that base will be cut out from under you before the next election. I believe that happened to the gReen Party in Maine?

  13. I should also point out that in the French version of two-round balloting, any candidate that receives more than 12% of the vote can stay in the second round (except in the presidential election, where its only the top two).

    Originally it was 5%, and up to the beginning of the 20th century, new candidates could even enter the election between the two rounds-in fact this is still possible in municipal elections for communes with leass that 2000 inhabitants. So smaller parties have a somewhat greater amount of leverage between elections that they would not have in a top-two system.

  14. The U.K. and Canada are relatively easy cases. Introducing AV for partisan legislative offices into a society that already has a significant element of multi-party politics is a step backwards. The hard cases are the U.S. and anywhere else with a similarly pure two party system. I’m still on the fence about this, with respect to state and federal offices (I strongly support AV for non-partisan local government).

    This discussion always revolves around the fortunes of small parties to right and left of the two major ones. Suppose, instead, we were talking about independent voters and candidates who appeal to them by positioning themselves in the political center, which is left largely vacant when the two major parties are as polarized as they are now. Would we reach a different conclusion about whether AV helps or hurts alternatives to R’s and D’s?

  15. There’s a serious misstatement in what I said in #14. “with respect to state and and federal offices” should be “with respect to state and federal legislative elections”.

  16. Canadian electoral reformers will applaud Dr. Lundberg’s excellent statement that “the Electoral Reform Society’s position just seems bizarre.” We think so too.

    The final paragraph of Fair Vote Canada’s statement (written by a group, with great deliberation) reads:

    “Is switching from our current voting system to AV for parliamentary elections likely to be a step toward fair voting in the foreseeable future?

    “No. Societies rarely change their voting systems for parliamentary, legislature or council elections. When those scarce opportunities arise by popular demand, proposals for cosmetic change are diversionary and may make the legislatures even less representative. Some established politicians are only too willing to misdirect public opinion in the name of reform. Democrats must be constant in the demand for fair democratic representation for every citizen and nothing less.”

  17. Thanks to Wilf Day for the above comment about Fair Vote Canada’s position on AV. As some other comments above have noted, if there is a movement towards multiparty politics in single-member plurality/first-past-the-post system, you are probably better off letting this pluralism develop to the point that the larger parties begin to fear for their own fortunes and respond with PR. Watch out if they respond with another majoritarian system, which is probably meant to contain the developing pluralism more effectively, preventing these smaller parties from winning seats. AV was introduced in some places to contain the fragmentation of the right in the face of a ‘socialist menace’ – at least I think this was the case in Australia and British Columbia.

    Tom Lundberg

  18. STV was introduced in the Australian senate by a failing Labor government which believed, correctly, that they would lose control of both houses under the existing majoritarian method of electing senators. I am not sure that means STV is an evil system that should be opposed at all costs or that decrying the motives of those who introduced AV in Australia and British Columbia much advances the conversation.

    Majority preferential voting has simply not had the results claimed. In the 1998 Queensland election One Nation won 11 of 89 seats on 22% of the vote.

    During the time since MPV was generally adopted for lower houses Australia has simply not had a stable duopoly. The current party system did not emerge until 1946 after the demise of not one but 2 preceding Urban conservative parties, the Nationalist Party of Australia and the United Australia Party. And the clearest example of a duopoly, Tasmania, where the Nationals do not exist, does not use MPV.

    Queensland tells us that minor parties can win seats under MPV. Tasmania tells us that major parties can lock out minor parties under STV.

    I am not at all sure the one-shot theory, that one reform excludes the possibility of any others, is necessarily true and without it the anti-MPV argument in this thread collapses.

  19. Concerning the debate over whether AV is a counterproductive measure from the point of view of electoral reformers, I will suggest that the most prominent such reformers in Britain have traditionally been the Liberal Democrats, and therefore their party interests may be important to consider also.

    I believe that if the first order elections (to the Commons) were to be conducted under a PR system, the party would either split or shrink. Firstly, the left would peel off to form the main vote for the Greens (as it does in other PR elections already). Secondly, coalition governments would become commonplace, meaning that the Lib Dems would have to choose sides. Most likely the right would either split off or start voting Conservative as a Lib/Lab coalition emerged. The result would be one or two liberal parties that after a transitional period might have fewer seats than the Lib Dems do currently (their influence in government would probably be constant, though).

    With AV, however, the centrally placed Lib Dems should benefit as long as they come in second place, which they do in a lot of places. In the event of an election rout of one of the two main parties, AV increases the likelihood of the Lib Dems displacing the routed party as one of the big two. Their broad base of voters would probably stay with them.

    Another calculation could be that AV would make their favourite preferential system, STV, much more natural for an elected upper house. As a second order election this would probably not lead to a party split, and the Lib Dems would form a crucial bloc that decided which government policies could be enacted.

  20. My clear preference is a single chamber elected by STV (or sortition), possibly with a Bundesrat in federations. In the current thread I am not convinced that MPV would be a bad thing where STV is not on the table.

    I think it is also important to bear the electors on mind. Requiring people to vote by radically different methods in different elections disadvantages the individual elector, particularly if you’re asking them to vote later preferences on one ballot and not to vote later preferences on others. There is also a legitimacy problem, particularly in the US where ‘alternate’ views of events, even election outcomes, get put forward so vigorously.

  21. What Alan said except that, while the Tasmanian Liberals have been very successful in preventing Nationals from winning seats (or even votes) in that State – largely by being uber-rural conservatives themselves – Labor has been less successful in holding up its end of the duopoly (which was an accurate description of the Tasmanian Assembly until 20 years ago) against the emergence of the Greens.

    Mind you, when the Greens are showing 22% in opinion polls (the same as One Nation did in Qld in 1998), it is hard to confidently claim that single-seaters would also shut them out. At some point, one has to deal with the fact that 50% of voters do not support, or even tolerate, one party governing on its own.

    Alan’s point about use of different voting methods confusing voters is supported by Australian experience. NSW had a higher-than-average informal vote in its 1991 state election because there was a concurrent referendum that called for ticks instead of numbers. Optional preferences in NSW and Qld State elections have, I believe, led to slightly higher informal votes under exhaustive AV in the federal House of Reps.

  22. The other thing worth mentioning is the mergence of a ‘non-party’ of country independents, particularly in the New England region of New South Wales. Partly this is a rural rejection of the free market policies forced on the Nationals by their Liberal coalition partner. (A former Labor premier of South Australia famously described the Nationals as committed to socialising their losses and privatising their gains)

    The Nationals used to hold all the federal and state seats in that area. The Division of New England was the jewel in the Nationals’ crown. They held the seat continuously from the foundation of the party in 1920 until 2001.

    The district is now represented by country independents at both federal and state levels. Windsor, the federal MP was elected on preferences in 2001 but secured absolute majorities in 2004 and 2007. The NSW legislative assembly includes 6 independent MPs in a house of 93.

  23. And Bob Katter in North Queensland – former National Party state minister, then federal MP, resigned the party over “economic rationalist” policies. Barnaby Joyce skates very close to dumping the Liberals at times also (even though he is on Tony Abbott’s front bench last time I looked).

  24. And in SA, which goes to the polls in 2 weeks, Dean Jaensch is predicting that the 3 independents in a house of 47 will all be re-elected and will determine who forms the next government. The major parties there can take comfort from this thread in the certain knowledge that minor party and independent candidates cannot be elected under MPV.

  25. So as well as Messrs Greiner, Goss, Kennett, Henderson and Menzies (Mark I) ringing their offices , we can soon add Mike Rann, perhaps.

    Maybe South Australia should get rid of its Complicated System Of Proportional Representation and Adopt Single-Member Electorates to Ensure Stable Government.

    (If anything, given that SMEs in Australia are not that much better at manufacturing one-party [or one-pre-election-coalition] majorities of seats than Hare-Clark has shown itself to be, I would go so far as to argue that it is better to have an actual third party holding the balance of power than one or several Independents. Independents are often nice people – respect to John Hatton and Peter Andren in particular – but they are often very concerned with local issues. When Liz Cunningham brought down Wayne Goss because his Labor Govt hadn’t fixed a school roof in her electorate, the other 88/89 of Qld had absolutely no leverage over her decision. Whereas with STV, a minor party with balance of power will likely have won seats in all or most districts – or at least have branches and party presence in all – so that every voter can lobby “their” local MP over actions the kingmakers take. I mean “their MP” in a real, actual sense, not the nominal, formal sense of SMEs).

  26. I do not think anyone in this thread has said that minor parties can’t be elected under AV/IRV/MPV. However, I do think there are sound theoretical reasons for expecting minor parties generally to have better chances at winning some seats under FPTP.

    For the reasons that Tom noted, independents are a rather different class of electoral competitor–and post-electoral bargainer.

    If an independent has a local base (which is true pretty much by definition for those with a chance at winning), he or she is probably not systematically less likely to be able to win a majority under transferred votes than to win a plurality. However, minor parties–and here I am not thinking of regional parties but of ideological minorities–might be systematically more likely to win a plurality or two, but to get squeezed out in the preference transfer by the initially second or third-ranked candidate.

    I believe the empirical record supports these hunches, although I freely admit that I am far more ignorant of the state-level performance of parties under transferable vote single-seat-district elections than some others in this thread are.

  27. It would be nice to have lots more empirical evidence. While we’re at it, let’s ask for true experiments with random assignment to experimental and control conditions.

    While we’re waiting for that to happen …

    I think there may be some middle ground, and least in terms of theoretical expectations. Where small parties are extremely marginal (candidates get 1% or 2% under plurality), AV might well help them get more, for the reasons articulated earlier by Rob Richie. The U.S. is the paradigmatic case. But where small party candidates are already spoiling elections and winning a few seats (but much fewer than they are entitled to) the adoption of AV might well be a setback.

    There are other factors to consider in addition to the effect on the vote for small parties. If you can accept single-member districts at all, AV is a more democratic way to elect representatives from them, and that should count for something. But so should the prospects (or lack thereof) for future electoral reform. The issue here, whether small parties benefit, is a big part of that. I think it depends.

  28. I will have a look at the last NSW assembly election, since NSW is the most independent and minor party friendly MPV state, to see if the squeezing out effect shows up there. I take MSS’ point that ideological minorities are disadvantaged over regional minorities under MPV. That is one of the main reasons I think MPV inferior to STV.

    Meanwhile, Antony Green on the MPV preference tickets in South Australia (which has an election next Saturday) including extraordinarily well-informed comments by one Tom Round. Independents and minor parties are allocating their preferences in ways somewhat contrary to the main thesis of this thread.

  29. Still trying to figure out why my comment (or Antony’s “comment” – scil. reply) on Fiji was “hilarious”. AG does seem unusually curt of late. Normally he likes to expound at length when asked for eelctoral information. And yes, you could spend 25 minutes googling this stuff, or else you could ask him.

  30. ‘… At best, I wonder if the increased publicity from the debates will only help the Lib Dems hold onto a handful of seats they might otherwise have lost. The reason? People have already put a price on the value of a vote for the Liberal Democrats and in most constituencies have concluded that it’s not worth very much. Consequently, I suspect plenty of voters will ignore Clegg et al and discount their debate performances. It doesn’t much matter how good Clegg is since he’s not going to win the election.

    The Lib Dems suffer from a collective action problem. If you could be sure that lots of other people would vote Liberal Democrat you might be much more inclined to vote Lib Dem yourself. But you can’t be sure of that and, in any case, all these other potential Lib Dems are wondering the same thing and concluding, perhaps with a mournful little sigh, that they’ll not vote for the Lib Dems either.

    Perhaps super performances in the debates will change that but I rather doubt it. It might help the Lib Dems in a handful of seats but as more and more voters take a national, not local, view of the election this won’t do much to boost the Cleggers unless, that is, all of a sudden several million people decide Nick Clegg could be Prime Minister. I must say that I think this improbable….’

    – Alex Massie, “Do Debates Really Help the Liberal Democrats?” The Spectator (Tuesday 30 March 2010).

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 )

Google+ photo

You are commenting using your Google+ 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.