Twilight of the independents

Don’t feel too bad if you missed it, but there was an election in Australia on Saturday. Two seats in Tasmania’s Legislative Council were decided; Rosevears, to the north-west of Launceston, and Huon, south-west of Hobart. Voting was originally scheduled for May, but was postponed due to Covid-19.

The Tasmanian upper house is based on 15 single-member districts, of which two or three go to the polls every year, giving each member a six-year term. Usually they attract very little attention outside of Tasmania (and not a lot even there), but this one marked a historic milestone.

Results are not final, but it appears the Liberal Party has won Rosevears, whose independent member was retiring, while Labor has beaten the sitting independent in Huon. (Official figures are here; Kevin Bonham has detailed analysis and commentary.) Assuming those results hold it means that for the first time ever the two major parties – Labor with five seats and the Liberals with three – will hold a majority in the Council between them.

There’s been a lot of talk in recent years about voter dissatisfaction with the major parties, of which one indicator has been the rise of independents. In the federal House of Representatives, for example, only one independent won a seat in more than forty years prior to the 1990 election; they were similarly thin on the ground in state lower houses. But since then they’ve become much more common, with up to five at a time federally and six for two terms (2003-11) in the Legislative Assembly of New South Wales.

It’s possible that the independent surge has now passed its peak. Their numbers are generally down from a decade ago, although at the most recent elections the trend in their vote is upward. Even if it has, though, that doesn’t mean the major parties have recovered, since votes and seats have also been lost to minor parties. (We might have a more detailed look at this in another post.)

In Tasmania, however, there’s no room for doubt. As recently as 2012 the independents held 13 seats in the Council, with the two majors having just one each. Their decline since then has been precipitous.

The shift has been political as well. Most of the independents used to lean conservative; in Bonham’s analysis of their voting record in 2011-12, he found that only three were left-leaning, while six of them were to the right of the sole Liberal member. Those that remain, however, are more on the left, producing a reasonably reliable centre-left majority of nine to six.

The Tasmanian upper house is the only one based on single-member electorates, but in other respects its history matches those in the other states. Like them, it was designed as a brake on democracy, with a bias in favor of rural and propertied interests; universal suffrage was not introduced until 1968, and it remained badly malapportioned until 1998.

That produced conservative majorities, but they were composed of independents because the Liberal Party generally refrained from endorsing candidates. In the forty years prior to 2009 only one Liberal sat in the Council (for two terms), although a number of independents had Liberal backgrounds or connections. (Ben Raue has a graph of representation over time.)

More recently, however, the independents have been under siege from both sides – the Liberals have become more inclined to run and fair electoral boundaries have made Labor more competitive. This time around the Liberals seem to have picked up a seat formerly held by a left-leaning independent; in Huon, against a right-leaning independent, they chose not to run, but Labor won the seat instead.

So one more piece of Tasmanian exceptionalism comes to an end. Its geographical polarisation, however, remains, and was strengthened by Saturday’s result: seven of the nine left-leaning seats will now be in the south, and five of the six right-leaning seats in the north.


10 thoughts on “Twilight of the independents

  1. So. It took 169 years for this chamber to have a majority for the two major parties combined. How long, I wonder, until there’s a majority for one single party?
    And yet the trope that “single-member electorates produce stable majority governments, while Hare-Clark produces unstable hung parliaments” will still keep running, because it’s a simple meme.
    I actually heard a professor of government,who had taught at a university in Tasmania, claim at a seminar that Hare-Clark made it “impossible” for one party to win a majority. Actually, if we take the last fifty years as the benchmark, Tasmania’s lower house system has produced no more hung parliaments (on election night [*]) than Queensland’s has, with single-member seats and (for half of that time) optional preferences, ie the closest to a pure first-past-the-post system [**] of any Australian parliament.
    The reason why the Tasmanian upper house took 169 years is because voters (assisted, true, by much rural vote-weighting) wanted it that way. Perhaps the same reasoning explains the Tasmanian lower house, too, and its disappointing refusal to conform to Professor LF Crisp’s idea of a proper parliament? (Less so Queensland, since the voters have less control over its party composition – there are many more kilometres of electoral boundaries to get in the way).
    [* An important qualifier because 1983 in Qld did not see a National Party majority on election night but did have one the following morning because Bjelke-Petersen induced two right-wing Liberals – one of whom was later exposed as corrupt by the Fitzgerald Inquiry – to jump parties. The Qld Libs at the time were out of coalition with the Nats, indeed deeply hostile to them, and some small-L elements were toying with a temporary alliance with Labor, if only for long enough to repeal the Bjelkemander and bring in a public accounts committee. Put another way, the voters declined to elect National-endorsed candidates to a majority of seats, despite single-member electorates.]
    [* In the sense of aggregate effects on party behaviours. At the individual level, it is untrue that “OPV is just FPTP”,. Indeed, it’s _less_ like FPTP than mandatory-full-preferential is, because it gives more respect to the voter’s decisions about allocating preferences. However, at overall level, it produces results closer to FPTP inasmuch as the candidate with most first preferences doesn’t always win, but has a much better chance of being last man standing, especially with more opposing candidates on the ballot.]

    Liked by 1 person

  2. Malta’s Parliament is elected by a system very similar to Tasmania’s House of Assembly, and representation there has been monopolised by two parties, effectively guaranteeing a majority to one or the other. Yet another very similar system in Ireland produces a radically different party system.

    Liked by 1 person

    1. True, J-D. Ireland had a very Australian-style political contest (Fianna Fail vs Fine Gael-Labour coalitions), complete with preference-swap deals among the latter, for its first six decades. It started to turn into a multi-party system in 1981, and has generally had hung parliaments since.
      Interestingly, though, in many cases the balance of power was held not by minor parties but by independents with strong localised power bases: eg Jackie Healy-Rae, a sort of Hibernian Bob Katter who not only drove a bargain with Fianna Fail in 20022 that would make Karlene Maywald’s eyes water with envy, but even included a non-disclosure clause so it couldn’t be made public! And those sorts of candidates, who can draw 17% to 25% across a county, will usually win a seat even if that county is divided into three, four or five single-member electorates.
      Most of the journalistic and even much of the academic literature on proportional voting systems tends to focus overwhelmingly on party-list versions, which to be fair are numerically dominant, but if one is selecting an election system as a matter of policy, that shouldn’t prejudice the decision. Non-preferential, tick-one-single-box voting systems with double-figured number of seats per electorate (sometimes even triple figures with nationwide electorates) are of course more likely to produce unforeseen post-election coalitions than preferential voting systems with 3, 5, or 7 seats per district.
      The latter will still give broadly proportional results but still gives an incentive for allied parties to work together before polling day to pool their electoral strength, which means in turn that even if no one party wins a majority of seats on paper, the resulting government will not cause anyone to faint. (Even if the WA Nationals do their usual playing coy “Oh, no, we don’t have any permanent alliance with the Liberals; after deciding afresh, with a completely open mind, after each election, we’re going with… [Tony Abbott/ Colin Barnett].” It’s never a cliffhanger.)
      Ironically, the one time we do have cliffhangers in Australia – apart from the rare example of the _Liberals_ being the balance-of-power minor party, as they were for a few brief shining hours in Queensland in October 1983 – it’s usually independents (again, not minor parties *) winning single-member seats holding the balance of power, and in that case there is usually no advance preference deal (nor post-election distribution of the Independent’s preferences **) that will let us predict which way a Rob Oakeshott will head. Whereas in Tasmania (or the ACT), even when the Greens win seats and hold the kingmaker position, enough of their candidates’ surplus preferences will have been distributed to make it publicly clear that their supporters favour Labor over Liberal.
      (* I think it is fair to class any One Nation MPs who get elected as de facto “Independents”. Almost none get re-elected a second, let alone third, time under the Hansonista party label. Indeed most can barely be in the same room as their former party colleagues after six to twelve months in Parliament).
      (** It is common to say that “we can’t know the two-party preferred vote if any Independent or minor-party candidates are elected, or if Labor runs third in some electorates behind two conservatives” but I can’t see why a 2PP couldn’t still be calculated by allocating every ballot to its higher choice between (a) the Labor candidate and (b) whichever other candidate won, or ran second, in every division).


      1. You say party list systems are “more likely to produce unforeseen post-election coalitions than preferential voting systems with 3, 5, or 7 seats per district.” I wonder if that’s really true? It’s possible, but I’m not convinced it makes a difference.


      2. Hi Charles, I’d say (1) list PR systems are more likely overall to produce coalitions (since they tend to be used in electoral districts with a lot more seats than STV systems are, typically in the teens per district as opposed to 3-4 Ireland, 5-5 in Malta, Tasmania and the ACT – 7-seaters in the latter having been phased out over the years). For that reason STV is more likely than other PR systems to give the party with most (after-preferences) votes a majority of seats even if it’s slightly below 50%.
        STV minimises vote wastage by transferring preferences for minor-party candidates who have no chance of winning. Party-list systems almost never do that: instead, they minimise vote wastage by having more seats available and thus a lower effective quota. So in Tasmania or the ACT, you need 16.7% to win a seat, but candidates/ parties who don’t win a seat still end up among the 85% or so whose vote elects somebody. In, say, Finland, by contrast, you might need only 5-6% of the vote to win a seat, but if you don’t win a seat then your votes go nowhere.
        Furthermore, (2) when list-PR systems do produce coalitions, these tend to be unforeseen. (But this doesn’t seem to annoy Europeans; different cultural expectations, I guess, the way Brits think it’s a travesty of democracy that Australians don’t know the final legal result for every seat by 11 PM on election night). Impressionistically, overall, European countries seem to shuffle the Cabinet deck with each election: last parliamentary term it was A, B and C in government, D and E in opposition; this time it’s A, C and E forming the Cabinet, B and D in opposition.
        Some list PR systems could in theory allow coalition partners to pool their support before an election. Eg Israel’s provision for joint lists so surpluses over the quota can be combined (not that these seems to have any Duvergian effect on Israeli government-formation). Likewise, MMP could in theory allow a large party to throw its smaller coalition party a constituency here and there via electoral pact so it will remain in parliament even if it falls below the 5% threshold. That Merkel’s CDU-CSU bloc allowed its FDP allies to disappear from the 2013-2017 Bundestag – when standing down for the FDP in one single constituency out of 299 could have netted their alliance an additional 30 MPs – seems to me incomprehensible as a rational political strategy, but then NZ Labour seems sanguine about a similar fate overtaking its NZF allies if Shane Jones misses out in Northland, so what do I know.


      3. Yes, there’s a trade-off there; the list system allows you elect far more people per district, lowering the effective quota & therefore giving more people their first choice, but at the cost of throwing away any votes that still don’t elect someone. MMP in Germany requires you to win 3 constituency seats, not just one, in order to get list seats if you fall below 5%, so it’s harder to game – I’m not aware of anyone having tried it there. Maybe the Germans have acquired a distrust of dirty tricks. Labour in NZ could certainly try to do it with NZ First, but I can’t see why it would – I think it’d be quite happy to see them disappear.


  3. ‘True, J-D. Ireland had a very Australian-style political contest (Fianna Fail vs Fine Gael-Labour coalitions)’

    I suppose there must be some points of resemblance, but the relationship between Fianna Fail and Fine Gael and the historical basis for the division between them is something that has no parallel in Australia that I can think of. In Australia the Liberals and the Nationals work together against Labor because (despite their undeniable differences and disagreements) they are closer together on most points. As far as I can tell (although I’m no expert), the reason Fine Gael and Labour worked together against Fianna Fail for so long was the pragmatic one that neither had much chance against Fianna Fail alone (whereas Fianna Fail, for a long time, did have realistic prospects of winning elections by itself). I’ve seen where more than one person has written that in modern times it’s Fianna Fail and Fine Gael which are closest to each other and have little contemporary justification to remain separate parties (which is obviously not how the two parties themselves see it, as they have obstinately remained separate no matter what those commentators say).

    Meanwhile, Labour versus the Nationalists in Malta looks much more like Labor versus the Liberals in Tasmania.

    ‘Interestingly, though, in many cases the balance of power was held not by minor parties but by independents with strong localised power bases’

    Another example of something that seems to happen a lot in Ireland but seldom or never in Malta or Tasmania, despite the similar electoral systems: my point being that the impact of electoral systems on party systems is (obviously) important but is also often exaggerated, as if electoral systems alone automatically determine party systems, when clearly they don’t.


    1. Thanks J-D. Yes, I think absolutely right on the looseness of the relationship between electoral systems and party systems. As to Ireland, the reason Labour historically worked with Fine Gael was that they were both relatively non-nationalist – for different reasons (Labour from a socialist perspective, Fine Gael from a middle class/liberal one), but both oriented away from Fianna Fail’s emphasis on the national question. Once the Good Friday agreement reduced the salience of that issue, it was always likely that Fine Gael & Fiana Fail would draw closer together.


      1. Do you think the Nationals and the Liberals are (at their cores) that much closer ideologically? They both agree that governments should not be handing out taxpayers’ money to lazy bludgers who don’t work, but the Nationals think that that’s because governments should be handing out taxpayers’ money to fine salt-of-the-earth farmers who’ve worked damn hard, while the Liberals think (in principle) that governments shouldn’t be handing out taxpayers’ money to anyone at all, unless by way of (a) military contracts or (b) tax rebates/ refunds.
        The Libs believe that success is at least 95% determined by hard work and at most only 5% by bad luck or societal discrimination, so if you’re not a millionaire, that’s your own fault. The Nationals don’t dispute that may be true for the city slicker but want it minuted that the man on the land may well suffer outrageous misfortunes that are utterly beyond his control (fires, droughts, flooding rains) and so should be bailed out by society when the fates infict calamities that he doesn’t deserve.
        A caveat here that the right wing of the Libs seem to be becoming more like their counterparts in the US Republican Party (big government nationalists who think tax dollars damn well should go to the individuals, and just as importantly to the religions, that deserve them) even as some Nats are becoming a bit pinker on issues like fracking, spending cuts and (whispers softly) climate change. Indeed Nats in Victoria and, more recently, South Australia have sometimes allied with Labor. (Although not, as noted above, the WA Nats despite their regular post-electoral Hamlet-like soul-searching over which side of the chamber to work with).


Leave a Reply to Elizabeth Warren-Truss Cancel reply

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

You are commenting using your 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.