The 2011 LegCo elections are all over but for the remaining entrails of the preference distribution process. The winners (certain now in two seats and virtually so in the third) were all as predicted here before the poll, but Tony Mulder’s triumph in Rumney has occurred in a manner I didn’t consider likely after the primary count. Silly me. Labor’s primary vote from the state election had actually held up remarkably well in the circumstances, but the preference flow was a totally different story, with support for Thorp evaporating, except for that portion of the Green support base that will never support any Liberal. Not only have those who voted 1 for protest conduit Paul Mason rejected Thorp by a margin of over four to one (such strong flows are extremely rare in such elections) but even the supporters of resigned Labor leftie Cate Clark have given the incumbent the thumbs down and preferred Tony Mulder. So much for the Trojan Horse theory.
Thorp is the first incumbent formally endorsed by a major party to lose a LegCo seat since 1982, when Kath Venn had the misfortune to face the voters the weekend after Robin Gray trounced the tattered remnants of the dam-era Holgate mess. To find the last case of an MLC from the governing party losing we need to go back to previous member Joseph Dixon reclaiming Derwent from Labor’s Don Marriott back in 1967. What we have seen this weekend is not that surprising in the circumstances but a very rare event all the same.
Superficially, the Rumney primaries showed all party votes taking a hit compared with state results, as a consequence of the 25% of the vote received by Paul Mason, Cate Clark and John Forster. Thorp was five points below the Labor state figure, Mulder eight points below the Liberals’ and Ann 12 points below the Greens’. That sounded like Thorp had managed the damage and a two-candidate result of about 53-47 in her favour could be expected (consistent with the state results) but the primary results did not reflect what was really going on. In the end, there was an effective “swing” of six points on preferences from the state result and the 53-47 went the other way.
Looking at booth by booth patterns in the Mason, Clark and Forster votes, the combined total of 25 points for those three is very consistent from booth to booth (give or take a few points) except some peninsula booths where it was much lower, and Richmond where it peaked at 32. The vote levels for these three independents in given booths are more or less independent of each other. A linear regression shows that the Mason vote was very strongly influenced by differences between the Liberal state vote and Mulder’s vote, and ditto for differences between the Green state vote and Ann’s vote, but barely at all by differences between the Labor state vote and Thorp’s vote. It may even be that only a few percent of the Mason voters voted Labor at the last state election.
The Mason vote was a lot like the Wilkie vote in character, and if he had been able to get his primary vote up into the low twenties instead of the mid teens he would have had some winning chances. The Clark and Forster votes, however, moved strongly with the changes in Labor and Green votes between the elections and had not that much to do with the Liberal vote. So state-level Labor voters who were no longer willing to support Thorp went mainly to Clark and Forster, and a share of these then preferenced Mulder. But state-level Greens voters who did not want to vote Green this time went mainly to Mason, those doing so preferencing Mulder over Thorp by at least 60:40. State-level Liberals who voted 1 Mason would have preferenced Mulder as well, very faithfully.
Thus while the primary picture looked not too bad for Thorp, the fact was that the eight points of primaries “dropped” by Mulder from the state election result were on loan to Paul Mason and loyally returned to Mulder via preferences. That would be no major problem for Thorp except that the primaries she lost to Forster and Clark included some genuine losses. Even so, she may have just survived on the preferences of lower house-level Green voters, except that these actually turned slightly against her. The ones who voted Green this time around favoured Thorp 53:42 but with the ones who didn’t vote Green opting more strongly for Mulder, the net effect of the Green voter base on Thorp’s chances was actually negative.
In the end it is the preferences of the “soft Green” voters which have ended Thorp’s career. By “soft Green” I mean the kind of voter who is likely to vote Green in preference to the two major parties but would rather vote for a suitable independent. Far from being a handicap, Tony Mulder’s small-l reputation, his zany image and his lack of a party endorsement seem to have helped him get the preferences of such voters ahead of the “embattled” Thorp where a rank-and-file Liberal hack, endorsed or not, may not have actually got there. And it was not just a case of Mulder being the least disliked of the major candidates after voters had run out of other options to preference. At the exclusions of both Forster and, very surprisingly, Ann, he received more preferences than Paul Mason.
Mulder’s win might be seen as a triumph for the Liberal Party, but it is really a triumph only for Tony Mulder. For the Liberal right (and perhaps one particular senator) it could become a serious nightmare. Having won the seat strongly on preferences at an election that saw an Abetz-style endorsed Lib campaign thrashed in the same manner in one of the state’s most Liberal seats, Mulder may have proved some kind of internal-political point. It is doubtful the party can harness his suddenly upsized eccentric political presence in any reliably controllable way. Another aspect of Mulder’s victory that will be interesting to watch is the growing number of MLCs with a career background in police matters.
Although I said Labor would be very relieved if they could manage to just hold the seat, it may turn out they are better off without the liability Thorp came to be. It should be remembered that for most of her career Thorp was successful and well regarded within Rumney, and her fate, though substantively self-inflicted, is also a symptom of the size reduction of the parliament. A Legislative Councillor holding multiple ministries can come under media attack in all her portfolios at once. In a larger parliament, with a lighter load, Lin Thorp would never have got in such a mess in the first place.
Launceston
In Launceston, Labor’s Steve Bishop, running a relatively minor campaign, had a predictably nondescript result, while Lou Clark polled a rather low primary vote, perhaps because she was competing for a similar vote to Rosemary Armitage and had a lot less experience. So it came down to the Liberals’ Sam McQuestin vs Armitage and it was soon obvious on polling night that McQuestin’s lead was way too small. Although the final result was incorrectly reported in some quarters as close, Armitage won easily, 56-44.
McQuestin needed a primary of at least 40 to have any chance against Armitage; he actually polled 34. A primary vote of 40 would have required him to hold all but seven points of the Liberal 2010 state election base. In almost every booth in the electorate he fell short of this goal, but in many he did not fall far short. It would have been a close result (without the winner changing) but for one striking feature in the booth pattern: the blowouts against McQuestin in the three strongest Liberal-voting booths of Newstead (17 points below state result), Norwood (20) and Prospect (20). The vote for Armitage was high in the first two of these. The winner’s margin was greatly bolstered by her ability (and McQuestin’s failure) to gain the support and trust of conservative (in the strict sense of the word) voters, and probably conservative elderly voters in particular. The young and somewhat brash McQuestin appealed well enough to middle-suburban Lib voters but the old money wasn’t interested. As with many other cases, the ability of well established local government independents to be all things politically to all people has proved decisive.
It was speculated (including by me) that Armitage’s views on the pulp mill might give her an advantage in attracting Green support. The Armitage vote was higher in strong Green booths, but it was also higher in strong Liberal booths. It tended to be lower in the nearest thing Launceston has to strong Labor booths. Green voters would have tended to vote for independents irrespective of the pulp mill, so if there was any juice in the mill argument one would expect Armitage (mostly seen as anti-mill) to poll well relative to Clark (pro-mill) in booths that are strong for the Greens. If anything, the exact reverse is true – the booths with the lowest Armitage/Clark ratios were mostly good booths for the Greens (the exception being Hadspen where Armitage’s municipal influence would be diluted), while the booths with the highest Armitage/Clark ratios were more Labor-leaning booths where the Green vote is very low. If there is any pulp mill signal in the Launceston result at all, it must be buried under many layers of noise from other pattens. Frankly I suspect the mill was simply not a factor.
Given the effort the McQuestin camp put into trying to win this seat the result is really not all that crash hot for them. It is likely that the unsubtle scare campaign by the Liberals against the independent candidates turned voters off and made the preference flows even worse for McQuestin, but by that stage they may well have realised Armitage was beating them and it was time to try something desperate. It failed, and yet another city councillor is off to the premier league. Or as some may call it, the local government retirement home.
Derwent
The current state of play in Derwent is that the preference distribution has stalled because after the exclusion of the Greens Phillip Bingley (in an inglorious and rather surprising last place) the next two candidates are too close together to be sure who will go. Jenny Branch, who will most likely come second, is currently 3867 votes behind Craig Farrell with 6751 Flint and Williams preferences still in play. Branch needs about 78% of those preferences to win and while there could well be enough pooling of conservative votes to make the final margin fairly close, that’s much too big an ask. I say this especially because the votes for Flint in particular are not necessarily “conservative” leaning votes. As expected the vote for the three independents was very regional, with Branch being the highest scorer of the three in the Glenorchy and Bridgewater-Brighton suburban booths, Williams leading the charge in the booths around New Norfolk plus Maydena, and Flint predictably beating Branch and Williams in all booths on and some around her own local government area. Flint was so locally well regarded that she topped the poll even ahead of Labor at all of Bronte, Ellendale, Gretna, Hamilton and Ouse.
Rumney can be argued about, because of the Mason factor and the relatively token nature of the Green campaign. But the Green vote in Derwent looks unequivocally awful, and I doubt the candidate was to blame, although he was far from high-profile. I also doubt effort level was the issue. Sure, there are more candidates than in a straight Labor-Liberal-Green contest but recent results such as those in Pembroke and Windermere showed Greens holding the state vote not too badly in multi-candidate contests. This time the Greens lost votes all over the electorate compared to the last LegCo election, increasing their vote marginally at just three booths compared to 2009 and losing more than half of it at six. Even with 13 points dropping off the Labor primary the Greens still managed to shed over a third of their vote either to Labor or to candidates known to have historic links to the Liberal Party. The damage was relatively minor in the booths near the city but in the New Norfolk area booths the Greens shed several points while in the Central Highlands booths the Flint factor pretty much wiped them out (from 20 points to 6 at Bronte and from 19 points to 2 at Hamilton the worst examples.)
Losing votes to Flint, a popular mayor of a small community (and a good speaker in the ABC radio forum, too) is excusable – within and around that community. But losing votes to Jenny Branch because she had an epiphany about the true nature of the Tasmanian Liberal Party immediately after it refused to give her a plum position? Losing votes to Craig Farrell, who outside the New Norfolk area went into the campaign pretty much an unknown? And losing votes to Ray Williams who is an ex-Lib, ex-Shooter and Fisher, ex-CECer who can’t even hold a New Norfolk council seat and spent the campaign lambasting greenies?
What I think is really going on with the Green vote here is that the state 2010 Green vote is being shown up as containing a huge slice of basic anti-major-party voting. Some voters do not like either major party and vote Green on this basis without any deep commitment to Green policies or values, or even without being all that happy with the Greens. A Green vote can be a token stamp of nonconformity. As soon as any other kind of candidate comes along with a reasonably prominent campaign, be it a Wilkie, a Mason or even to a lesser degree a Mulder or Flint, they will switch to that candidate instead of the Green. The Greens can’t get near matching state totals in LegCo at the moment because the state totals they are expected to match include such a high default vote that can be picked off by pretty much anyone who isn’t endorsed by the big two – and not just in Denison where that vote was harnessed by Wilkie, but everywhere. The Green results are still pretty bad, but this explains the bulk of it.
It’s been a long way down for Labor in the Upper House. At the heights of the Jim Bacon era the party was not far short of a majority that it could have used to transform the Tasmanian political system. It held five of the fifteen seats formally, and a sixth informally through Silvia Smith. The first domino fell when Smith was thrashed by Ivan Dean. David Crean was replaced by Terry Martin who then quit the PLP and the party could not recover the seat when he left. Allison Ritchie resigned and Labor gave the by-election a miss only to see it won by the Liberals, and then there were three. Lin Thorp lost to Tony Mulder and then there were two. And finally, Doug Parkinson retired or was defeated in 2012, and then there was only one.
And finally …
Because of the amount of misrepresentation, spin and fibbing in politics, it is increasingly common for quality outlets to guide their readers through the maze with “factchecker” services, rating the truthfulness of claims on a scale from unintended cases of complete accuracy on one side to pants-on-fire on the other. If TT had such a system, some of Will Hodgman’s responses to these results would break the dial.
Mr Hodgman claims “However you measure it, Labor and the Greens had massive swings recorded against them yesterday.” That part, in isolation, is somewhere near true (although if you measure the Green vote in Rumney against the 2005 LegCo poll it is actually up a point, and in a larger, stronger field). What Hodgman neglects to mention, as he goes on to claim a 10% “swing” against Labor compared to last year’s state election, and an 8% swing against the Greens on the same basis, is that by the same unreliable yardstick the Liberals did even worse in the one seat where they scraped up the courage to formally endorse a candidate. The state-to-LegCo primary “swing” against the Libs in Launceston was 12.7%. If Mulder is counted as a Liberal in Rumney the primary vote swing there was less, but it was still comparable to that for Labor and the Greens. Of course, as I explain above, there was really no swing against Mulder at all (just some votes being parked with Paul Mason) but identifying the size of comparable “swings” against other parties is no easy business. The real swing against Labor in Rumney from the state result was about six points, and we don’t know yet where it will finish up in Derwent.
Rather than the result showing a massive swing away from the governing coalition and towards the Liberals, it really shows that independents are far more competitive in LegCo than in the Lower House and that it is hard for any party to match its state primary vote in these circumstances. With Bishop making not that much effort, and Farrell a low-profile replacement, it’s hard to say if the loss of Thorp is a vote against Labor or a vote against Thorp. Reading state messages about any party into these votes is not easy, but if the results show a backlash against one party then they probably show one against all of them.
In these circumstances, Hodgman’s claim that “If yesterday’s Legislative Council election results were replicated in a statewide lower house election, the Liberals would win a clear majority” is quite hilarious. What, when they couldn’t even win on preferences in the most pro-Liberal seat that was contested? When their final two-candidate preferred vote in that seat was lower than their state election primary there? Of course if the LegCo results were replicated literally in a statewide lower house election what you would really get would be a Lower House swarming with independents.
Still there’s worse! Will Hodgman refers to “the bizarre situation which now sees the Green candidate in Rumney effectively kingmaker as to whether or not Lin Thorp survives.” Bizarre? It’s called going to preferences, Will; we do have them in this country, in case you hadn’t noticed. At the last federal election the bizarre situation Hodgman refers to (Green voters deciding the outcome of a seat) happened in about every second electorate, including all five Tasmanian House of Reps seats. Since Rumney was a notional marginal based on state voting, it would have been bizarre if it had not gone to Green preferences.
It seems, not for the first time, that Will Hodgman baulks at any idea of electoral merit more complex than “but more people voted for me, miss” (even though in this case that would mean Lin Thorp keeping the seat). His attitudes (also on display after the last state election) would fit in neatly in the masterfully executed but shamelessly, hugely unfactual Tory-spearheaded campaign that has just (with a major assist from the Lib Dems’ sellouts and naïve ineptitude) scuppered any chance of electoral reform in the UK for a long long time.
This column rarely makes advocative comments but in this case I will make a small exception. Will Hodgman should state for the record whether he personally supports the concept and principles of preferential voting, and whether he can commit himself to never attempting to introduce first-past-the-post.
