Centrist strategic voting

This image was picked from a bunch of images on Google Image. This post is not about Canada. If you’re interested in Canada, it’s unlikely you’ll get ANYTHING out of this post. Canada is just incidental to this post. It is very cold there a lot of the time though, so it’s odd more people live there than live here. But that’s just the kind of provincial small mindedness that Troppo readers hate, but in which Troppo’s leadership collective CONSTANTLY INDULGE. SHAME.

This is my response to Peter Dempster’s proposals.

I can see one important merit of them. Electoral politics is inherently polarising because electoral politics involves politicians beating other politicians to qualify to be politicians in the first place – by getting into parliament – and then joining the team that helped them get to parliament in beating the other team. So it’s good to have stabilising influences such as what’s been proposed.

But while I like this intention, it seems to me that the idea has some difficulties intellectually and has no chance practically. The results of the survey that Peter quotes are interesting and informative, but I don’t think they should be read naïvely.1 For instance 45 percent of people say they’d consider voting for a new centrist party, but we know how many people do vote that way when people try to establish such parties.

This leaves aside the question of where the ‘centre’ is. For me the ALP are a thoroughly centrist party. The concrete policy decisions the Government is making so far are relatively centrist also, but the right now has the problem the left had from the 1960s to the around the mid 1980s, which is that they have an unreasonable faction. 2 In the same vein, I’m not sure how easy it is to pick a single dimension of ‘centrism’. For instance on military matters there’s nothing centrist about me. I’m in favour of doing almost anything to avoid getting into a war. Not anything – I’m not a pacifist – but almost anything. There’s nothing centrist about that.

Moreover I’m quite left wing when it comes to income distribution, but allergic to the self-righteousness of the left. I don’t hold this against them personally, because the more I’ve thought about politics in the way I now think about it, the more I blame the system not the victim. Self-righteousness works politically. It works to recruit and hold adherents, and rev them up against their opponents. Being more low key and resigned about things in one’s manner doesn’t.

On the practical side, I also don’t think political journalists know much at all. They hang around parliament and engage in lots of pub-talk. Then they continue the pub-talk on Insiders, and Outsiders and Leftsiders and Rightsiders and Purplesiders and Lemon Merangsiders. I think we have a pretty good idea of who’s a moderate and who’s not from political journalists’ and others’ reporting right now.

More generally, I think the idea has precisely no chance of going anywhere. Even if quite a few people could give it a lot of public profile. The engine of political engagement is emotion, not reason. The proposal embodies an entirely abstract idea, and people routinely fail to defend abstract ideas at the ballot box.3

In speaking with people who are seasoned campaigners on both sides about the idea of a party based on giving sortition or random selection a bigger presence in our democracy – for instance by selecting the upper house or some part of it by lot – their advice to me is that it would definitely bomb unless one had squared off some policies to campaign on – which, to be consistent, one imagines one would do with some randomly selected or otherwise representative body.

If one campaigned for this, most people wouldn’t get it. And most of those that did have lots of other things on their mind. Do they want to vote Green? Which of the parties do they feel happiest with? Do they feel vengeful towards the existing lot for the way their electricity prices have gone up? In that context abstract ideas typically cut across those considerations. If I help ‘centrists’, will that get my electricity prices down or should I vote for the candidate who says that working families are stretched to breaking point by rising electricity prices and we have to Eaze the Skwese?

There’s also the chicken and egg problem. Lots of small businesses start thinking that their idea is so great that people will beat a pathway to their door. I know because I thought people would queue up to get the same home loans from Peach Home Loans that they could get from Westpac or any other lender. It turns out that, not only is it very difficult to get known, but even then, there are various psychological reasons why people are wary and will head down to their local branch. Still it plugs away in its low key way. Does anyone out there want a home loan? Our Going Troppo Liberator Loan comes with free access to Troppo for the duration of the loan – with a home loan from the bank of your choice and $1,000 in your bank account.

Finally, where is the evidence that strategic voting of the kind Peter envisages can take off in a substantial way? That’s a genuine question, I don’t know, but I would expect it’s a pretty boutique kind of operation (though the graph the Troppo elves chose to illustrate the piece suggests otherwise, so perhaps I’m wrong). Perhaps more to the point,  I can’t really envisage any way that’s practical and consistent with our traditions of preventing large scale cheating. If someone agrees to vote in a deal with me, how do I know they’ve done so? And how will that effect my confidence in them and theirs in me?

  1. I hope you like that little gizmo over the “i” in ‘naïve’. I do! But I digress.
  2.  I was going to say it’s an ideological faction and to some extent it is, but that’s dignifying it somewhat. Opposition to greenhouse gas abatement isn’t really ideological, it’s part tactical and part a reaction to another ideology which one might call ‘political correctness’ – anyway, that’s just a quibble – not very interesting. .
  3. My favourite example is the election of 1977 when there was a small swing away from the Fraser Government after its landslide win in 1975. One Liberal MP had a swing to him. Phillip Lynch who’d been engaged in various shady land deals (I may be being unfair here, he may have been innocent, but there was a fair smell about him at the time as I recall.)
This entry was posted in Democracy, Political theory, Politics - international, Politics - national. Bookmark the permalink.

27 Responses to Centrist strategic voting

  1. Alan says:

    Canadians vote strategically because the electoral system forces them to. They have a three and a half party politics in a two party electoral system. At the last general election, for example, very large numbers of NDP voters were forced to vote Liberal to prevent the election of another Conservative government. There is no Canadian option for centrism. The Liberal Party of Canada, the New Democratic Party of Canada, the Bloc Québécois and the Green Party of Canada (which between them typically hold around 3/4 of all seats in the federal house of commons) would all be regarded as dangerously radical by the ALP. The Liberals, for example, have just legalised cannabis.

    The system of preferential voting that we use in Australia largely overcomes the need for strategic voting. A system of proportional representation in the house as well as the senate would do immeasurably more to strengthen the centre and deliberation than voting for a slate of centrist candidates. As would either a random assembly or incorporating some random representatives into one or both houses.

    Machiavelli proposed random procurators who would sit in the main legislative council with no right to vote, but the right to exercise a collective veto. Procurators would be about a tenth of the house. They would be elected randomly. They would have very short staggered terms with a ban on public appointments or contracts after they left office. . Parliamentary procurators could perhaps be introduced as a test run for fuller forms of dicastery.

    • Nicholas Gruen says:

      Thanks for the point on Machiavelli. They had quite a lot of randomness when he was about in Florence.

      Can you point me to a source?

      I’ve had a similar idea of putting randos in the House of Representatives with only a procedural vote. The idea is to clean up Question Time. But selection by lot provides a huge repertoire of potentially useful hacks.

      • Alan says:

        When Machiavelli proposes a constitution for a revived Florentine Republic, he very subtly – indeed, almost surreptitiously – incorporates tribune-like offices, the provosts (proposti), into his plan. Machiavelli wrote the “Discursus on Florentine Affairs” (1519–20)30 in response to Giovanni de’Medici’s (Pope Leo’s) solicitations for advice on converting Florence from a de facto principality to a genuine republic, since the leading Medici, now Church prelates, will leave behind them no legitimate heirs to serve as princes of Florence.

        McCormick. Machiavellian Democracy (p. 103). Cambridge University Press. Kindle Edition.

        • Nicholas Gruen says:

          Fantastic, thanks so much!

          • Alan says:

            The big advantage of the provost model is that it could be introduced on a trial basis, for 1 or 2 parliaments to see how it worked.

            15 provosts. They get a procedural but not substantive vote. 10 of them can delay a bill for a year. They serve 3 months. 5 retire and get replaced every month.

            • Nicholas Gruen says:

              Any of these things could be trialled. In fact I’d like to see most restrictive security legislation ‘trialled’ by default which is to say with clauses which have the legislation lapse if it’s not renewed, though that’s an aside.

              On this, I wouldn’t call a 1 year block on a bill ‘procedural’. I’d also want more provosts than that to tame randomness.

              • Alan says:

                I picked 15 as 1/10 of the house of representatives.

                That would mean 60 provosts a year, which is quite a lot. We would have to provide them with salaries, right to return to previous jobs, and probably, given the very short terms, with accommodation and family travel as well.

                • Nicholas Gruen says:

                  Chickenfeed IMO.

                  I’m intrigued that you want them to have such short terms. I rather like the idea that they grow in experience over a period. In my model of a citizen’s chamber they’ve got 6 year terms and a third rotate every two years. Not that there’s any science behind that.

            • Moz of Yarramulla says:

              3 months is a long time off work, I think many people would struggle with that. Casual employees and “independent contractors” would almost certainly lose their … can’t call it a job, “source of income” during that time, and there’s no way to prevent that. Good luck telling Uber they have to compensate a random person with the Uber app installed for whatever that person has lost by spending 3 months in Canberra, but more often it’ll be some hapless shelf-stacker or cafe employee who comes back to discover that whatever goodwill kept them in regular shifts has vanished.

              I suggest pay would have to be “whatever an MP gets”, and ditto allowances etc. If that provokes outrage… well, they would say that, wouldn’t they.

              • Alan says:

                It is almost standard in Europe for constitutions to provide that you get leave to run for and sit in parliament and you get your old job back when the campaign ends or you leave office. Ex-President Mariano Rajoy recently resumed work as a land register in Alicante.

                I join Machiavelli in proposing very short terms and high rotation because I do not want the provosts to become professional politicians. machiavelli, and rather supporters of sortition, go into some length arguing why random terms should be short. I want the provosts to check legislation, not legislate. I am not aware of any random office in a historic republic that lasted for longer than a year. Almost all lasted for much shorter periods. The council presidency at Athens, which functioned rather like our governor-general, rotated every 24 hours.

              • Moz of Yarramulla says:

                I wasn’t aware provosts would have to run for office. That would be unworkable, I think. A wide-area election campaign for a 3 month term with an election every month? madness!

                • Alan says:

                  Provosts would not have to run for office and nothing in my comment suggested they would. I mentioned the way things work in most European countries merely to show that there are ways around the problems you mentioned.

            • Moz of Yarramulla says:

              I do like the idea of the provost model, partly because those people would have no motivation to keep secrets from the public that they represent. And the internet means that the usual gatekeepers would struggle to maintain that role.

              The problem with a small number having significant power is that they can effectively kill some bills and wreck others “nice budget you got there, I hope it still looks this good in 12 months time”. But with only 15 of them, for a paltry million dollars or so you could have a good go at whatever they call bribing MPs these days. It makes acting against a bill much easier than acting for one. You’d probably find a whole new class of government appointments, since the usual internal rewards wouldn’t work, but the revolting door would be hard to justify. 3 months as a provost then straight into a paid sinecure on a board somewhere…. hmm, it’d want to be coastal because it certainly smells fishy.

              Perhaps better to do it as “can defer for 3 months, renewing the deferral up to 4 times” so that you need a pretty solid consensus that a bill is junk before it can really be held off.

              • Alan says:

                I would ban a provost from any public office or contract for quite a long time after they left office.

                • Alan says:

                  And 10 million in undetectable payments or rewards is actually not enough. To retain control of 2/3 of the provosts for an entire year at the rate of 1 million per provost will cost 40 million.

                  More if you want to keep the block on for longer than a year.

                  More if Nick proposes a more reasonable formula for setting their numbers than I have come up with.

              • Moz of Yarramulla says:

                The thing is that most provosts would be more affordable than a million, and you only have to bribe 2/3 of them, once, to kill a bill (as I read it: a single veto lasts 12 months). So 10 provosts, $100k each… it’s money for jam in the mind of the provosts.

                Sure, for something important like a budget they’d probably want more, but if it’s something controversial you’d probably have half of them on side already. And for trivia… again, some would probably vote to block it without payment.

                Imagine gay marriage… about 1/3 of the population was against it, leaving you 5 provosts to buy from the remaining 10. A campaign could probably get some of them, and threats some more (lawful threats like “your local community will be upset when they find out that you want gay men teaching their kids about sex” – the content doesn’t have to be true for the threat to be effective, as we saw)

                • Alan says:

                  This is frankly getting to be a fairly silly subthread. It would just as easy to purchase MPs and indeed entire political parties with the kind of money you re suggesting.

                  • Nicholas Gruen says:


                    I think the issues I raised about the Australian Ballet raised more serious questions – but not insurmountable ones of course.

                    As someone who often suggests that we do something differently, I often encounter these kinds of (silly) objections. They seem contrary.

                    I think that, partly, they arise from people trying to reconcile in their own mind why it might be that something that makes sense doesn’t exist – thinking that there must be a good reason for it. But only a infinitesimal fraction of the infinitude of possible things in the world exist. Of those, there are quite a lot of potentially good things – including near self-evidently good things.

                    As here.

                    • Moz of Yarramulla says:

                      You don’t agree that it would be easier to influence 10 of 15 newbies rather than 60 or so professional politicians who have a long-term interest in what they’re doing?

                      I agree that professional politicians are corruptable in known ways, what I’m thinking about is the downsides of the veto.

                      But I understand it’s easier to say “silly” than to come up with reasonable arguments. Because after all, if you ut this as a serious proposal in public no-one is going try to poke holes in it, or bring up objections that you find silly? Talk about falling at the first hurdle…

                    • Alan says:

                      I don’t agree that it would be easier to bribe 10 provosts than 10 MPs and the relevant figure, as noted previously is not 10 provosts but 40. You may notice that the higher figure suggests to a reasonable reader that the suspensory veto is not a one-off event.

                      I am assuming provosts are subject to the same rules on bribery and financial disclosure as other public officials. I did not see it as necessary to mention that. Only a single provost has to come forward and say Moz of Yarralumla tried to bribe me and the game is up.

                      I am assuming that, as in all cases of direct democracy, the budget is exempt from the suspensory veto.

                      For the record I am assuming all other assumptions that tend to make this a rational proposal.

                      So far you have read a perfectly reasonable comment about how to protect provosts from employer retaliation as requiring them to run for office. Now you’re reading the veto to mean that bribing 10 provosts gives you control of the body of provosts forever. These are not reasonable objections.

  2. Nicholas Gruen says:


    Another point which I should have already made. When you say Canadians vote strategically, that’s similar to Australian’s voting differently in the House of Representatives and the Senate. That’s unilateral strategic voting, which I’d accept was at the levels indicated there. People understand the logic of ‘wasting a vote’ and some respond accordingly.

    Here we’re contemplating bilateral strategic voting – which is much harder to organise and, it seems to me, both unverifiable and enforceable (I would imagine you would have no cause of action against someone who broke their word – that electoral law – both common law and statute would prohibit that kind of thing in the process of outlawing the selling of votes. (It was to interdict it that the ‘Australian ballot’ secret ballot was invented. This is to be clearly distinguished from the Australian Ballet which was not outlawed, but was established much later.)

    • Alan says:

      As fas I know the Australian Ballet was not banned in Canada even during the worst excesses of the Harper regime.

    • Alan says:

      More seriously, strategic voting, also known as tactical voting, is a fairly well understood phenomenon. A Canadian whose sincere preference is NDP is more or less forced to vote Liberal in some ridings if they want to ensure election of a progressive candidate. An Australian would simply vote 1 NDP 2 Liberal.

      Split voting between the two houses is a slightly different phenomenon.

  3. Moz of Yarramulla says:


    how to protect provosts from employer retaliation as requiring them to run for office

    You said when I mentioned that the notion of “employer” is irrelevant to a disturbingly large number of people: you get leave to run for and sit in parliament and you get your old job back when the campaign ends or you leave office

    So … you were talking about running for elected office in the context of provosts losing casual employment.

    I don’t agree that it would be easier to bribe 10 provosts than 10 MPs and the relevant figure, as noted previously is not 10 provosts but 40.

    You originally said 15 provosts. They get a procedural but not substantive vote. 10 of them can delay a bill for a year.

    So where does the ’40 provosts’ number com from? To me, if 10 of them can delay a bill for a year, that means bribing 10 of them once is enough to delay a bill for a year. Given the regular corruption scandals in Australia, and the ongoing issues with elected members acting in the interests of their funders, I’m not at all sure that appointed figures would in this one case be unusually unlikely to act corruptly. In NSW we have seen Eddy Obeid et al found to have engaged in widespread corruption, and are going through a series of state government dismissing councils and appointing administrators to carry out tasks that the councils refuse to (approving Westconnex, merging councils etc). That provosts wouldn’t do those things…. is obvious to you, but not to me.

    It makes sense to exclude budgetary bills from that, but again, you apparently implied that without mentioning it so it was up to me to imagine what you meant. I’m sorry that I once again failed. Look, it’s the internet, all I have to go on is what you write.

    • Moz of Yarramulla says:

      where does the ’40 provosts’ number com from?

      Amusingly the only place I can see 40 provosts being necessary is in my suggested modification. from above:

      Perhaps better to do it as “can defer for 3 months, renewing the deferral up to 4 times” so that you need a pretty solid consensus that a bill is junk before it can really be held off.

      That would require 10 different provosts on each of four occasions = 40 provosts… but that’s me “being silly”, not Alan’s entirely reasonable suggestion that any 10 provosts, acting only once, can suspend any bill for a year. Over to you, Alan and Nick.

    • Alan says:

      Indeed, it is you being silly by extracting the most unreasonable reading you can.

      The conditions of return to employment are an analogy. They mean only that analogous conditions would be appropriate for provosts. They do not alter the proposed way in which provosts would be selected. Analogous reasoning is entirely normal.

      I mentioned 40 provosts., If you apply the most abstract and siphicitacted methods of mathematical analysis, you may notice I prose I propose 5 new provosts a month, which means 60 provosts a year. A reasonable reader would discern where the figure of 40 comes from and understand that I am not proposing a one-off veto.

  4. Peter Dempster says:

    My unstated assumption is that the major parties organise themselves around the centre, although not necessarily close to the centre. Pulling too far one way or the other is dangerous, since centrist voters may be lost to the other side. Thus, with candidates self-organising relative to the centre, the journalist’s task is simply to identify the 50:50 slate that is best able to meet in the middle.

    ‘Meeting in the middle’ would be an iterative process. If a political settlement plays out in such a way that centrists are unhappy with the result, or change their minds when they see the result, expect political parties to adjust their positions in light of the new information about where the centre is located. Related to that, my guess is that voter preferences are about values and outcomes, not specific policy settings and strategies. The centrist agenda is to elect people who will conscientiously target centrist outcomes, by whatever means that works and making changes as necessary.

    On this view, the centrist settlement emerges from an iterative political process. The centre is not something that needs to be closely pre-defined by a journalist, or by anybody for that matter.

    I don’t know enough about voting behaviour to comment on the role of self-righteousness, emotions and abstract ideas. But the idea of voting for a 50:50 slate of moderate candidates does not seem particularly abstract, no different from the various forms of protest voting. Having a gutful of politics-as-usual is a common emotional state.

    My guess is that centrists are simply realists. They recognise that many issues are complex and difficult, that they are often not willing or not able to engage effectively with the issues, that their best option is to delegate the decisions to people would will work through the issues with respect and open-mindedness. My hope is that the proposed mechanism, involving political journalists and public debate around polarisation and moderation, provides the basis for an awareness campaign. Watch this space.

    I can’t reconcile that journalists are not up the task but that we already know the result.

    On cheating, I puzzled over a trading mechanism for a couple of years. The available means of simultaneous or conditional voting are clumsy and restrictive, most obviously by postal voting, possibly also at a pre-poll station. Plausibly, however, the transactional approach is restrictive and unhelpful; centrist voters may happily accept a considerable imbalance of cross-voting as the price of delivering a strong political message to both sides of politics.

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.